1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Handle short writes when writing local log-files.
8 In practice, only affects FreeBSD (8 onwards).
9 Bugzilla 1053, with thanks to Dmitry Isaikin.
11 NM/01 Bugzilla 949 - Documentation tweak
13 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
16 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
18 PP/02 Implemented gsasl authenticator.
20 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
22 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
23 `pkg-config foo` for cflags/libs.
25 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
26 with rest of GSASL and with heimdal_gssapi.
28 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
29 `pkg-config foo` for cflags/libs for the TLS implementation.
31 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
32 properties get this fed in as external SSF. A number of robustness
33 and debugging improvements to the cyrus_sasl authenticator.
35 PP/08 cyrus_sasl server now expands the server_realm option.
37 PP/09 Bugzilla 1214 - Log authentication information in reject log.
38 Patch by Jeremy Harris.
40 PP/10 Added dbmjz lookup type.
42 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
44 PP/12 MAIL args handles TAB as well as SP, for better interop with
45 non-compliant senders.
46 Analysis and variant patch by Todd Lyons.
48 NM/04 Bugzilla 1237 - fix cases where printf format usage no indicated
49 Bug report from Lars Müller <lars@samba.org> (via SUSE),
50 Patch from Dirk Mueller <dmueller@suse.com>
52 PP/13 tls_peerdn now print-escaped for spool files.
53 Observed some $tls_peerdn in wild which contained \n, which resulted
54 in spool file corruption.
60 PP/01 Solaris build fix for Oracle's LDAP libraries.
61 Bugzilla 1109, patch from Stephen Usher.
63 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
65 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
68 TF/02 Fix a couple more cases where we did not log the error message
69 when unlink() failed. See also change 4.74-TF/03.
71 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
72 lock up or crash if it happened to be inside a call to libc when it
73 got a SIGUSR1 from exiwhat.
75 The SIGUSR1 handler appends the current process status to the process
76 log which is later printed by exiwhat. It used to use the general
77 purpose logging code to do this, but several functions it calls are
80 The new output code in the SIGUSR1 handler is specific to the process
81 log, and simple enough that it's easy to inspect for signal safety.
82 Removing some special cases also simplifies the general logging code.
83 Removing the spurious timestamps from the process log simplifies
86 TF/04 Improved ratelimit ACL condition.
88 The /noupdate option has been deprecated in favour of /readonly which
89 has clearer semantics. The /leaky, /strict, and /readonly update modes
90 are mutually exclusive. The update mode is no longer included in the
91 database key; it just determines when the database is updated. (This
92 means that when you upgrde Exim will forget old rate measurements.)
94 Exim now checks that the per_* options are used with an update mode that
95 makes sense for the current ACL. For example, when Exim is processing a
96 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
97 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
98 must specify per_mail/readonly. If you omit the update mode it defaults to
99 /leaky where that makes sense (as before) or /readonly where required.
101 The /noupdate option is now undocumented but still supported for
102 backwards compatibility. It is equivalent to /readonly except that in
103 ACLs where /readonly is required you may specify /leaky/noupdate or
104 /strict/noupdate which are treated the same as /readonly.
106 A useful new feature is the /count= option. This is a generalization
107 of the per_byte option, so that you can measure the throughput of other
108 aggregate values. For example, the per_byte option is now equivalent
109 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
111 The per_rcpt option has been generalized using the /count= mechanism
112 (though it's more complicated than the per_byte equivalence). When it is
113 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
114 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
115 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
116 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
117 Note that using per_rcpt with a non-readonly update mode in more than
118 one ACL will cause the recipients to be double-counted. (The per_mail
119 and per_byte options don't have this problem.)
121 The handling of very low rates has changed slightly. If the computed rate
122 is less than the event's count (usually one) then this event is the first
123 after a long gap. In this case the rate is set to the same as this event's
124 count, so that the first message of a spam run is counted properly.
126 The major new feature is a mechanism for counting the rate of unique
127 events. The new per_addr option counts the number of different
128 recipients that someone has sent messages to in the last time period. It
129 behaves like per_rcpt if all the recipient addresses are different, but
130 duplicate recipient addresses do not increase the measured rate. Like
131 the /count= option this is a general mechanism, so the per_addr option
132 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
133 example, measure the rate that a client uses different sender addresses
134 with the options per_mail/unique=$sender_address. There are further
135 details in the main documentation.
137 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
139 TF/06 Removed a few PCRE remnants.
141 TF/07 Automatically extract Exim's version number from tags in the git
142 repository when doing development or release builds.
144 PP/02 Raise smtp_cmd_buffer_size to 16kB.
145 Bugzilla 879. Patch from Paul Fisher.
147 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
148 Heavily based on revision 40f9a89a from Simon Arlott's tree.
151 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
153 PP/05 Variable $av_failed, true if the AV scanner deferred.
154 Bugzilla 1078. Patch from John Horne.
156 PP/06 Stop make process more reliably on build failure.
157 Bugzilla 1087. Patch from Heiko Schlittermann.
159 PP/07 Make maildir_use_size_file an _expandable_ boolean.
160 Bugzilla 1089. Patch from Heiko Schlittermann.
162 PP/08 Handle ${run} returning more data than OS pipe buffer size.
163 Bugzilla 1131. Patch from Holger Weiß.
165 PP/09 Handle IPv6 addresses with SPF.
166 Bugzilla 860. Patch from Wolfgang Breyha.
168 PP/10 GnuTLS: support TLS 1.2 & 1.1.
170 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
173 PP/11 match_* no longer expand right-hand-side by default.
174 New compile-time build option, EXPAND_LISTMATCH_RHS.
175 New expansion conditions, "inlist", "inlisti".
177 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
179 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
181 PP/14 fix log_write() format string regression from TF/03.
182 Bugzilla 1152. Patch from Dmitry Isaikin.
188 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
190 PP/02 Harmonised TLS library version reporting; only show if debugging.
191 Layout now matches that introduced for other libraries in 4.74 PP/03.
193 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
195 PP/04 New "dns_use_edns0" global option.
197 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
200 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
201 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
203 TK/01 Updated PolarSSL code to 0.14.2.
204 Bugzilla 1097. Patch from Andreas Metzler.
206 PP/07 Catch divide-by-zero in ${eval:...}.
209 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
212 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
213 format-string attack -- SECURITY: remote arbitrary code execution.
215 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
216 time unintentionally subject to list matching rules, letting the header
217 cause arbitrary Exim lookups (of items which can occur in lists, *not*
218 arbitrary string expansion). This allowed for information disclosure.
220 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
221 INT_MIN/-1 -- value coerced to INT_MAX.
227 NM/01 Workround for PCRE version dependency in version reporting
230 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
231 This fixes portability to compilers other than gcc, notably
232 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
234 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
235 makefiles for portability to HP-UX and POSIX correctness.
237 PP/01 Permit LOOKUP_foo enabling on the make command-line.
238 Also via indented variable definition in the Makefile.
239 (Debugging by Oliver Heesakkers).
241 PP/02 Restore caching of spamd results with expanded spamd_address.
242 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
244 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
245 Improves build reliability. Fix from: Frank Elsner
247 NM/02 Fix wide character breakage in the rfc2047 coding
248 Fixes bug 1064. Patch from Andrey N. Oktyabrski
250 NM/03 Allow underscore in dnslist lookups
251 Fixes bug 1026. Patch from Graeme Fowler
253 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
254 Code patches from Adam Ciarcinski of NetBSD.
256 NM/04 Fixed exiqgrep to cope with mailq missing size issue
259 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
260 is logged, to avoid truncation. Patch from John Horne.
262 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
263 Patch from Jakob Hirsch.
265 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
266 SQL string expansion failure details.
267 Patch from Andrey Oktyabrski.
269 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
270 Patch from Simon Arlott.
272 PP/09 New lookups functionality failed to compile on old gcc which rejects
273 extern declarations in function scope.
274 Patch from Oliver Fleischmann
276 PP/10 Use sig_atomic_t for flags set from signal handlers.
277 Check getgroups() return and improve debugging.
278 Fixed developed for diagnosis in bug 927 (which turned out to be
281 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
282 Patch from Mark Zealey.
284 PP/12 Bugzilla 1056: Improved spamd server selection.
285 Patch from Mark Zealey.
287 PP/13 Bugzilla 1086: Deal with maildir quota file races.
288 Based on patch from Heiko Schlittermann.
290 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
291 Patch from Uwe Doering, sign-off by Michael Haardt.
293 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
294 variable declaration deep within a block. Bug and patch from
297 PP/15 lookups-Makefile IRIX compatibilty coercion.
299 PP/16 Make DISABLE_DKIM build knob functional.
301 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
302 Patch by Simon Arlott
304 TF/03 Fix valgrind.h portability to C89 compilers that do not support
305 variable argument macros. Our copy now differs from upstream.
311 TF/01 Failure to get a lock on a hints database can have serious
312 consequences so log it to the panic log.
314 TF/02 Log LMTP confirmation messages in the same way as SMTP,
315 controlled using the smtp_confirmation log selector.
317 TF/03 Include the error message when we fail to unlink a spool file.
319 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
320 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
321 for maintaining out-of-tree patches for some time.
323 PP/01 Bugzilla 139: Documentation and portability issues.
324 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
325 Handle per-OS dynamic-module compilation flags.
327 PP/02 Let /dev/null have normal permissions.
328 The 4.73 fixes were a little too stringent and complained about the
329 permissions on /dev/null. Exempt it from some checks.
330 Reported by Andreas M. Kirchwitz.
332 PP/03 Report version information for many libraries, including
333 Exim version information for dynamically loaded libraries. Created
334 version.h, now support a version extension string for distributors
335 who patch heavily. Dynamic module ABI change.
337 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
338 privilege escalation vulnerability whereby the Exim run-time user
339 can cause root to append content of the attacker's choosing to
342 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
345 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
346 If dropping privileges for untrusted macros, we disabled normal logging
347 on the basis that it would fail; for the Exim run-time user, this is not
348 the case, and it resulted in successful deliveries going unlogged.
349 Fixed. Reported by Andreas Metzler.
355 PP/01 Date: & Message-Id: revert to normally being appended to a message,
356 only prepend for the Resent-* case. Fixes regression introduced in
357 Exim 4.70 by NM/22 for Bugzilla 607.
359 PP/02 Include check_rfc2047_length in configure.default because we're seeing
360 increasing numbers of administrators be bitten by this.
362 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
364 PP/03 Bugzilla 994: added openssl_options main configuration option.
366 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
368 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
370 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
372 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
373 without a peer certificate, leading to a segfault because of an
374 assumption that peers always have certificates. Be a little more
375 paranoid. Problem reported by Martin Tscholak.
377 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
378 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
379 NB: ClamAV planning to remove STREAM in "middle of 2010".
380 CL also introduces -bmalware, various -d+acl logging additions and
381 more caution in buffer sizes.
383 PP/09 Implemented reverse_ip expansion operator.
385 PP/10 Bugzilla 937: provide a "debug" ACL control.
387 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
389 PP/12 Bugzilla 973: Implement --version.
391 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
393 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
395 PP/15 Bugzilla 816: support multiple condition rules on Routers.
397 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
398 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
399 ignore trailing whitespace.
401 JJ/02 prevent non-panic DKIM error from being sent to paniclog
403 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
406 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
407 Notification from Dr Andrew Aitchison.
409 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
410 ExtendedDetectionInfo response format.
411 Notification from John Horne.
413 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
416 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
417 XSL and documented dependency on system catalogs, with examples of how
420 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
423 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
424 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
425 configuration file which is writeable by the Exim user or group.
427 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
428 of configuration files to cover files specified with the -C option if
429 they are going to be used with root privileges, not just the default
432 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
433 option (effectively making it always true).
435 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
436 files to be used while preserving root privileges.
438 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
439 that rogue child processes cannot use them.
441 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
442 run-time user, instead of root.
444 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
445 Exim run-time user without dropping privileges.
447 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
448 result string, instead of calling string_vformat() twice with the same
451 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
452 for other users. Others should always drop root privileges if they use
453 -C on the command line, even for a whitelisted configure file.
455 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
457 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
463 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
464 $data_path, and $header_path variables; fixed documentation bugs and
467 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
468 exipick to access non-standard spools, including the "frozen" queue
471 NM/01 Bugzilla 965: Support mysql stored procedures.
472 Patch from Alain Williams
474 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
476 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
477 Patch from Andreas Metzler
479 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
480 Patch from Kirill Miazine
482 NM/05 Bugzilla 671: Added umask to procmail example.
484 JJ/03 installed exipick 20100323.0, fixing doc bug
486 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
487 directory. Notification and patch from Dan Rosenberg.
489 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
491 TK/02 Improve log output when DKIM signing operation fails.
493 MH/01 Treat the transport option dkim_domain as a colon separated
494 list, not as a single string, and sign the message with each element,
495 omitting multiple occurences of the same signer.
497 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
498 Bugzilla 985, 986. Patch by Simon Arlott
500 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
501 Patch by Simon Arlott
503 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
504 MBX locking. Notification from Dan Rosenberg.
510 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
512 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
514 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
516 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
518 NM/04 Bugzilla 915: Flag broken perl installation during build.
524 TK/01 Added patch by Johannes Berg that expands the main option
525 "spamd_address" if it starts with a dollar sign.
527 TK/02 Write list of recipients to X-Envelope-Sender header when building
528 the mbox-format spool file for content scanning (suggested by Jakob
531 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
532 (http://www.dcc-servers.net/) support via dccifd. Activated by
533 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
535 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
536 by Mark Daniel Reidel <mr@df.eu>.
538 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
539 When building exim an external PCRE library is now needed -
540 PCRE is a system library on the majority of modern systems.
541 See entry on PCRE_LIBS in EDITME file.
543 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
544 conversation. Added nologin parameter to request.
545 Patch contributed by Kirill Miazine.
547 TF/01 Do not log submission mode rewrites if they do not change the address.
549 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
551 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
552 log files in place. Contributed by Roberto Lima.
554 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
556 TF/03 Bugzilla 615: When checking the local_parts router precondition
557 after a local_part_suffix or local_part_prefix option, Exim now
558 does not use the address's named list lookup cache, since this
559 contains cached lookups for the whole local part.
561 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
562 Robert Millan. Documentation is in experimental-spec.txt.
564 TF/04 Bugzilla 668: Fix parallel build (make -j).
566 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
568 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
569 Patch provided by Jan Srzednicki.
571 TF/05 Leading white space used to be stripped from $spam_report which
572 wrecked the formatting. Now it is preserved.
574 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
575 that they are available at delivery time.
577 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
579 TF/08 TLS error reporting now respects the incoming_interface and
580 incoming_port log selectors.
582 TF/09 Produce a more useful error message if an SMTP transport's hosts
583 setting expands to an empty string.
585 NM/06 Bugzilla 744: EXPN did not work under TLS.
586 Patch provided by Phil Pennock.
588 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
589 Patch provided by Richard Godbee.
591 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
592 acl_smtp_notquit, added index entry.
594 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
595 Patch provided by Eugene Bujak.
597 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
598 accept(). Patch provided by Maxim Dounin.
600 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
601 Patch provided by Phil Pennock.
603 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
605 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
606 Patch provided by Brad "anomie" Jorsch.
608 NM/14 Bugzilla 622: Added timeout setting to transport filter.
609 Patch provided by Dean Brooks.
611 TK/05 Add native DKIM support (does not depend on external libraries).
613 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
614 Patch provided by Graeme Fowler.
616 NM/16 Bugzilla 851: Documentation example syntax fix.
618 NM/17 Changed NOTICE file to remove references to embedded PCRE.
620 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
623 NM/19 Bugzilla 745: TLS version reporting.
624 Patch provided by Phil Pennock.
626 NM/20 Bugzilla 167: bool: condition support.
627 Patch provided by Phil Pennock.
629 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
630 clients. Patch provided by Phil Pennock.
632 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
633 Patch provided by Brad "anomie" Jorsch.
635 NM/23 Bugzilla 687: Fix misparses in eximstats.
636 Patch provided by Heiko Schlittermann.
638 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
639 Patch provided by Heiko Schlittermann.
641 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
642 plus update to original patch.
644 NM/26 Bugzilla 799: Documentation correction for ratelimit.
646 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
647 Patch provided by David Brownlee.
649 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
651 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
653 NM/30 Bugzilla 888: TLS documentation bugfixes.
655 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
657 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
658 Unlike the original bugzilla I have changed all shell scripts in src tree.
660 NM/33 Bugzilla 898: Transport filter timeout fix.
661 Patch by Todd Rinaldo.
663 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
664 Patch by Serge Demonchaux.
666 NM/35 Bugzilla 39: Base64 decode bug fixes.
667 Patch by Jakob Hirsch.
669 NM/36 Bugzilla 909: Correct connect() call in dcc code.
671 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
673 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
675 NM/39 Bugzilla 911: Fixed MakeLinks build script.
681 TK/01 Add preliminary DKIM support. Currently requires a forked version of
682 ALT-N's libdkim that I have put here:
683 http://duncanthrax.net/exim-experimental/
685 Note to Michael Haardt: I had to rename some vars in sieve.c. They
686 were called 'true' and it seems that C99 defines that as a reserved
687 keyword to be used with 'bool' variable types. That means you could
688 not include C99-style headers which use bools without triggering
689 build errors in sieve.c.
691 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
692 as mailq or other aliases. Changed the --help handling significantly
693 to do whats expected. exim_usage() emits usage/help information.
695 SC/01 Added the -bylocaldomain option to eximstats.
697 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
699 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
701 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
707 PH/01 Another patch from the Sieve maintainer.
709 PH/02 When an IPv6 address is converted to a string for single-key lookup
710 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
711 dots are used instead of colons so that keys in lsearch files need not
712 contain colons. This was done some time before quoting was made available
713 in lsearch files. However, iplsearch files do require colons in IPv6 keys
714 (notated using the quote facility) so as to distinguish them from IPv4
715 keys. This meant that lookups for IP addresses in host lists did not work
716 for iplsearch lookups.
718 This has been fixed by arranging for IPv6 addresses to be expressed with
719 colons if the lookup type is iplsearch. This is not incompatible, because
720 previously such lookups could never work.
722 The situation is now rather anomolous, since one *can* have colons in
723 ordinary lsearch keys. However, making the change in all cases is
724 incompatible and would probably break a number of configurations.
726 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
729 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
730 conversion specification without a maximum field width, thereby enabling
731 a rogue spamd server to cause a buffer overflow. While nobody in their
732 right mind would setup Exim to query an untrusted spamd server, an
733 attacker that gains access to a server running spamd could potentially
734 exploit this vulnerability to run arbitrary code as the Exim user.
736 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
737 $primary_hostname instead of what libspf2 thinks the hosts name is.
739 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
740 a directory entry by the name of the lookup key. Previously, if a
741 symlink pointed to a non-existing file or a file in a directory that
742 Exim lacked permissions to read, a lookup for a key matching that
743 symlink would fail. Now it is enough that a matching directory entry
744 exists, symlink or not. (Bugzilla 503.)
746 PH/03 The body_linecount and body_zerocount variables are now exported in the
749 PH/04 Added the $dnslist_matched variable.
751 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
752 This means they are set thereafter only if the connection becomes
755 PH/06 Added the client_condition to authenticators so that some can be skipped
756 by clients under certain conditions.
758 PH/07 The error message for a badly-placed control=no_multiline_responses left
759 "_responses" off the end of the name.
761 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
763 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
764 (without spaces) instead of just copying the configuration text.
766 PH/10 Added the /noupdate option to the ratelimit ACL condition.
768 PH/11 Added $max_received_linelength.
770 PH/12 Added +ignore_defer and +include_defer to host lists.
772 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
773 way in which PCRE > 7.0 is built.
775 PH/14 Implemented queue_only_load_latch.
777 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
778 MAIL command. The effect was to mangle the value on 64-bit systems.
780 PH/16 Another patch from the Sieve maintainer.
782 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
784 PH/18 If a system quota error occurred while trying to create the file for
785 a maildir delivery, the message "Mailbox is full" was not appended to the
786 bounce if the delivery eventually timed out. Change 4.67/27 below applied
787 only to a quota excession during the actual writing of the file.
789 PH/19 It seems that peer DN values may contain newlines (and other non-printing
790 characters?) which causes problems in log lines. The DN values are now
791 passed through string_printing() before being added to log lines.
793 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
794 and InterBase are left for another time.)
796 PH/21 Added message_body_newlines option.
798 PH/22 Guard against possible overflow in moan_check_errorcopy().
800 PH/23 POSIX allows open() to be a macro; guard against that.
802 PH/24 If the recipient of an error message contained an @ in the local part
803 (suitably quoted, of course), incorrect values were put in $domain and
804 $local_part during the evaluation of errors_copy.
810 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
811 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
814 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
815 issue a MAIL command.
817 PH/02 In an ACL statement such as
819 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
821 if a client was not listed at all, or was listed with a value other than
822 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
823 the condition was not true (as it should be), so access was not denied.
824 The bug was that the ! inversion was incorrectly passed on to the second
825 item. This has been fixed.
827 PH/03 Added additional dnslists conditions == and =& which are different from
828 = and & when the dns lookup returns more than one IP address.
830 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
831 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
833 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
834 FSYNC, which compiles an option called disable_fsync that allows for
835 bypassing fsync(). The documentation is heavily laced with warnings.
837 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
839 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
840 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
841 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
842 including adding "make clean"; (3) Added -fPIC when compiling the test
843 dynamically loaded module, to get rid of a warning.
845 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
846 message fails, move_frozen_messages = true and ignore_bounce_errors_after
847 = 0s. The bug is otherwise harmless.
849 PH/07 There was a bug in the dovecot authenticator such that the value of
850 $auth1 could be overwritten, and so not correctly preserved, after a
851 successful authentication. This usually meant that the value preserved by
852 the server_setid option was incorrect.
854 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
856 PH/09 Installed PCRE release 7.0.
858 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
859 run for batched SMTP input. It is now run at the start of every message
860 in the batch. While fixing this I discovered that the process information
861 (output by running exiwhat) was not always getting set for -bs and -bS
862 input. This is fixed, and it now also says "batched" for BSMTP.
864 PH/11 Added control=no_pipelining.
866 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
867 patch, slightly modified), and move the expansion of helo_data till after
868 the connection is made in the smtp transport (so it can use these
871 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
873 PH/14 Added log_selector = +pid.
875 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
877 PH/16 Add ${if forany and ${if forall.
879 PH/17 Added dsn_from option to vary the From: line in DSNs.
881 PH/18 Flush SMTP output before performing a callout, unless control =
882 no_callout_flush is set.
884 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
885 was true (the default) a successful delivery failed to delete the retry
886 item, thus causing premature timeout of the address. The bug is now
889 PH/20 Added hosts_avoid_pipelining to the smtp transport.
891 PH/21 Long custom messages for fakedefer and fakereject are now split up
892 into multiline reponses in the same way that messages for "deny" and
893 other ACL rejections are.
895 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
896 with slight modification.
898 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
899 draft, changing the syntax and factoring some duplicate code".
901 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
902 for deliveries of the second and subsequent messages over the same SMTP
905 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
906 ${reduce, with only minor "tidies".
908 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
910 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
911 expansion side effects.
913 PH/27 When a message times out after an over-quota error from an Exim-imposed
914 quota, the bounce message says "mailbox is full". This message was not
915 being given when it was a system quota that was exceeded. It now should
918 MH/03 Made $recipients available in local_scan(). local_scan() already has
919 better access to the recipient list through recipients_list[], but
920 $recipients can be useful in postmaster-provided expansion strings.
922 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
923 in the case of a MAIL command with additional options following the
924 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
925 were accidentally chopped off.
927 PH/29 SMTP synchronization checks are implemented when a command is read -
928 there is a check that no more input is waiting when there shouldn't be
929 any. However, for some commands, a delay in an ACL can mean that it is
930 some time before the response is written. In this time, more input might
931 arrive, invalidly. So now there are extra checks after an ACL has run for
932 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
933 pipelining has not been advertised.
935 PH/30 MH's patch to allow iscntrl() characters to be list separators.
937 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
938 returned in the SMTP response when smtp_return_error_details was false.
941 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
942 instead of the C I/O that was originally supplied, because problems were
945 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
946 Exim which did not show up earlier: it was assuming that a call to
947 SSL_CTX_set_info_callback() might give an error value. In fact, there is
948 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
949 was a macro that became an assignment, so it seemed to work. This has
950 changed to a proper function call with a void return, hence the compile
951 error. Exim's code has been fixed.
953 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
956 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
958 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
960 PH/37 If a message is not accepted after it has had an id assigned (e.g.
961 because it turns out to be too big or there is a timeout) there is no
962 "Completed" line in the log. When some messages of this type were
963 selected by exigrep, they were listed as "not completed". Others were
964 picked up by some special patterns. I have improved the selection
965 criteria to be more general.
967 PH/38 The host_find_failed option in the manualroute router can now be set
968 to "ignore", to completely ignore a host whose IP address cannot be
969 found. If all hosts are ignored, the behaviour is controlled by the new
970 host_all_ignored option.
972 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
973 homing or because of multiple MX records with /mx) generated more than
974 one IP address, and the following item turned out to be the local host,
975 all the secondary addresses of the first item were incorrectly removed
976 from the list, along with the local host and any following hosts (which
977 is what is supposed to happen).
979 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
980 whoever called Exim into the -H file. In the case of the daemon it was
981 behaving confusingly. When first started, it used values for whoever
982 started the daemon, but after a SIGHUP it used the Exim user (because it
983 calls itself on a restart). I have changed the code so that it now always
986 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
987 message are rejected with the same error (e.g. no authentication or bad
988 sender address), and a DATA command is nevertheless sent (as can happen
989 with PIPELINING or a stupid MUA), the error message that was given to the
990 RCPT commands is included in the rejection of the DATA command. This is
991 intended to be helpful for MUAs that show only the final error to their
994 PH/42 Another patch from the Sieve maintainer.
996 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
997 Eximstats - Fixed some broken HTML links and added missing column headers
999 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1000 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1002 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1004 PH/43 Yet another patch from the Sieve maintainer.
1006 PH/44 I found a way to check for a TCP/IP connection going away before sending
1007 the response to the final '.' that terminates a message, but only in the
1008 case where the client has not sent further data following the '.'
1009 (unfortunately, this is allowed). However, in many cases there won't be
1010 any further data because there won't be any more messages to send. A call
1011 to select() can be used: if it shows that the input is "ready", there is
1012 either input waiting, or the socket has been closed. An attempt to read
1013 the next input character can distinguish the two cases. Previously, Exim
1014 would have sent an OK response which the client would never have see.
1015 This could lead to message repetition. This fix should cure that, at
1016 least in a lot of common cases.
1018 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1019 advertised in response to EHLO.
1025 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1026 fixed by 4.65/MH/01 (is this a record?) are fixed:
1028 (i) An empty string was always treated as zero by the numeric comparison
1029 operators. This behaviour has been restored.
1031 (ii) It is documented that the numeric comparison operators always treat
1032 their arguments as decimal numbers. This was broken in that numbers
1033 starting with 0 were being interpreted as octal.
1035 While fixing these problems I realized that there was another issue that
1036 hadn't been noticed. Values of message_size_limit (both the global option
1037 and the transport option) were treated as octal if they started with 0.
1038 The documentation was vague. These values are now always treated as
1039 decimal, and I will make that clear in the documentation.
1045 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1046 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1049 MH/01 Don't check that the operands of numeric comparison operators are
1050 integers when their expansion is in "skipping" mode (fixes bug
1051 introduced by 4.64-PH/07).
1053 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1054 child addresses, Exim now panics and dies. Previously, because the count
1055 is held in a short int, deliveries were likely to be lost. As such a
1056 large number of recipients for a single message is ridiculous
1057 (performance will be very, very poor), I have chosen to impose a limit
1058 rather than extend the field.
1064 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1065 leftover -K file (the existence of which was triggered by #402).
1066 While we were at it, introduced process PID as part of the -K
1067 filename. This should rule out race conditions when creating
1070 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1071 processing considerably. Previous code took too long for large mails,
1072 triggering a timeout which in turn triggers #401.
1074 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1075 in the DK code in transports.c. sendfile() is not really portable,
1076 hence the _LINUX specificness.
1078 TF/01 In the add_headers option to the mail command in an Exim filter,
1079 there was a bug that Exim would claim a syntax error in any
1080 header after the first one which had an odd number of characters
1083 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1084 callout verification, Exim cached a "reject" for the entire domain. This
1085 is correct for most verifications, but it is not correct for a recipient
1086 verification with use_sender or use_postmaster set, because in that case
1087 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1088 case of MAIL FROM:<> rejection from other early rejections (e.g.
1089 rejection of HELO). When verifying a recipient using a non-null MAIL
1090 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1091 Whatever the result of the callout, the value of the domain cache is
1092 left unchanged (for any other kind of callout, getting as far as trying
1093 RCPT means that the domain itself is ok).
1095 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1098 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1099 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1100 session, the daemon ignores ECONNECT errors and logs others; it now
1101 ignores EPIPE as well.
1103 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1104 (quoted-printable decoding).
1106 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1107 later the small subsequent patch to fix an introduced bug.
1109 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1111 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1113 PH/08 An error is now given if message_size_limit is specified negative.
1115 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1116 to be given (somewhat) arbitrary names.
1118 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1121 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1122 miscellaneous code fixes
1124 PH/10 Added the log_reject_target ACL modifier to specify where to log
1127 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1128 hostname. This is wrong, because it relates to the incoming message (and
1129 probably the interface on which it is arriving) and not to the outgoing
1130 callout (which could be using a different interface). This has been
1131 changed to use the value of the helo_data option from the smtp transport
1132 instead - this is what is used when a message is actually being sent. If
1133 there is no remote transport (possible with a router that sets up host
1134 addresses), $smtp_active_hostname is used.
1136 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1137 tweaks were necessary in order to get it to work (see also 21 below):
1138 (a) The code assumed that strncpy() returns a negative number on buffer
1139 overflow, which isn't the case. Replaced with Exim's string_format()
1141 (b) There were several signed/unsigned issues. I just did the minimum
1142 hacking in of casts. There is scope for a larger refactoring.
1143 (c) The code used strcasecmp() which is not a standard C function.
1144 Replaced with Exim's strcmpic() function.
1145 (d) The code set only $1; it now sets $auth1 as well.
1146 (e) A simple test gave the error "authentication client didn't specify
1147 service in request". It would seem that Dovecot has changed its
1148 interface. Fortunately there's a specification; I followed it and
1149 changed what the client sends and it appears to be working now.
1151 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1154 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1155 address A is aliased to B and C, where B exists and C does not. Without
1156 -v the output is "A verified" because verification stops after a
1157 successful redirection if more than one address is generated. However,
1158 with -v the child addresses are also verified. Exim was outputting "A
1159 failed to verify" and then showing the successful verification for C,
1160 with its parentage. It now outputs "B failed to verify", showing B's
1161 parentage before showing the successful verification of C.
1163 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1164 look up a TXT record in a specific list after matching in a combined
1167 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1168 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1169 they consult the DNS. I had assumed they would set it the way they
1170 wanted; and indeed my experiments on Linux seem to show that in some
1171 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1172 To be on the safe side, however, I have now made the interface to
1173 host_find_byname() similar to host_find_bydns(), with an argument
1174 containing the DNS resolver options. The host_find_byname() function now
1175 sets these options at its start, just as host_find_bydns() does. The smtp
1176 transport options dns_qualify_single and dns_search_parents are passed to
1177 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1178 of host_find_byname() use the default settings of RES_DEFNAMES
1179 (qualify_single) but not RES_DNSRCH (search_parents).
1181 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1182 spool_read_header() do less string testing, by means of a preliminary
1183 switch on the second character of optional "-foo" lines. (This is
1184 overdue, caused by the large number of possibilities that now exist.
1185 Originally there were few.) While I was there, I also converted the
1186 str(n)cmp tests so they don't re-test the leading "-" and the first
1187 character, in the hope this might squeeze out yet more improvement.
1189 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1190 flag allowing group syntax was set by the header_syntax check but not
1191 turned off, possible causing trouble later; (2) The flag was not being
1192 set at all for the header_verify test, causing "group"-style headers to
1193 be rejected. I have now set it in this case, and also caused header_
1194 verify to ignore an empty address taken from a group. While doing this, I
1195 came across some other cases where the code for allowing group syntax
1196 while scanning a header line wasn't quite right (mostly, not resetting
1197 the flag correctly in the right place). These bugs could have caused
1198 trouble for malformed header lines. I hope it is now all correct.
1200 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1201 with the "reply" argument non-NULL. The code, however (which originally
1202 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1203 but it didn't always do it. This confused somebody who was copying the
1204 code for some other use. I have removed all the tests.
1206 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1207 feature that was used to support insecure browsers during the U.S. crypto
1208 embargo. It requires special client support, and Exim is probably the
1209 only MTA that supported it -- and would never use it because real RSA is
1210 always available. This code has been removed, because it had the bad
1211 effect of slowing Exim down by computing (never used) parameters for the
1212 RSA_EXPORT functionality.
1214 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1215 authenticator to fail if there's a tab character in the incoming data
1216 (there should never be unless someone is messing about, as it's supposed
1217 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1218 if the connection is using TLS or if the remote IP is the same as the
1219 local IP, and the "valid-client-cert option" if a client certificate has
1222 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1223 authenticators. This can be used for authorization after authentication
1224 succeeds. (In the case of plaintext, it servers for both authentication
1227 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1228 if any retry times were supplied.
1230 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1231 connection, where there is no client IP address to check. In this
1232 situation, the verify now always succeeds.
1234 PH/25 Applied John Jetmore's -Mset patch.
1236 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1238 PH/27 In a string expansion for a processed (not raw) header when multiple
1239 headers of the same name were present, leading whitespace was being
1240 removed from all of them, but trailing whitespace was being removed only
1241 from the last one. Now trailing whitespace is removed from each header
1242 before concatenation. Completely empty headers in a concatenation (as
1243 before) are ignored.
1245 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1246 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1248 PH/29 [Removed. This was a change that I later backed out, and forgot to
1249 correct the ChangeLog entry (that I had efficiently created) before
1250 committing the later change.]
1252 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1253 address errors (4xx response to RCPT) over the same connection as other
1254 messages routed to the same hosts. Such deliveries are always "forced",
1255 so retry times are not inspected. This resulted in far too many retries
1256 for the affected addresses. The effect occurred only when there were more
1257 hosts than the hosts_max_try setting in the smtp transport when it had
1258 the 4xx errors. Those hosts that it had tried were not added to the list
1259 of hosts for which the message was waiting, so if all were tried, there
1260 was no problem. Two fixes have been applied:
1262 (i) If there are any address or message errors in an SMTP delivery, none
1263 of the hosts (tried or untried) are now added to the list of hosts
1264 for which the message is waiting, so the message should not be a
1265 candidate for sending over the same connection that was used for a
1266 successful delivery of some other message. This seems entirely
1267 reasonable: after all the message is NOT "waiting for some host".
1268 This is so "obvious" that I'm not sure why it wasn't done
1269 previously. Hope I haven't missed anything, but it can't do any
1270 harm, as the worst effect is to miss an optimization.
1272 (ii) If, despite (i), such a delivery is accidentally attempted, the
1273 routing retry time is respected, so at least it doesn't keep
1274 hammering the server.
1276 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1277 in ${readsocket because some servers need this prod.
1279 PH/32 Added some extra debug output when updating a wait-xxx database.
1281 PH/33 The hint "could be header name not terminated by colon", which has been
1282 given for certain expansion errors for a long time, was not being given
1283 for the ${if def:h_colon_omitted{... case.
1285 PH/34 The spec says: "With one important exception, whenever a domain list is
1286 being scanned, $domain contains the subject domain." There was at least
1287 one case where this was not true.
1289 PH/35 The error "getsockname() failed: connection reset by peer" was being
1290 written to the panic log as well as the main log, but it isn't really
1291 panic-worthy as it just means the connection died rather early on. I have
1292 removed the panic log writing for the ECONNRESET error when getsockname()
1295 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1296 runs only) independently of the message's sender address. This meant
1297 that, if the 4xx error was in fact related to the sender, a different
1298 message to the same recipient with a different sender could confuse
1299 things. In particualar, this can happen when sending to a greylisting
1300 server, but other circumstances could also provoke similar problems.
1301 I have changed the default so that the retry time for these errors is now
1302 based a combination of the sender and recipient addresses. This change
1303 can be overridden by setting address_retry_include_sender=false in the
1306 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1307 remote server are returned as part of bounce messages. This was not
1308 happening for LMTP over a pipe (the lmtp transport), but now it is the
1309 same for both kinds of LMTP.
1311 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1312 in header lines that were in fact CNAMEs. This is no longer the case.
1314 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1315 and queue runs started by the daemon processed all messages. This has
1316 been fixed so that -R and -S can now usefully be given with -q<time>.
1318 PH/40 Import PCRE release 6.7 (fixes some bugs).
1320 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1322 PH/42 Give an error if -q is specified more than once.
1324 PH/43 Renamed the variables $interface_address and $interface_port as
1325 $received_ip_address and $received_port, to make it clear that these
1326 values apply to message reception, and not to the outgoing interface when
1327 a message is delivered. (The old names remain recognized, of course.)
1329 PH/44 There was no timeout on the connect() call when using a Unix domain
1330 socket in the ${readsocket expansion. There now is.
1332 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1333 be meaningful with "accept".
1335 SC/01 Eximstats V1.43
1336 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1338 SC/02 Eximstats V1.44
1339 Use a glob alias rather than an array ref in the generated
1340 parser. This improves both readability and performance.
1342 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1343 Collect SpamAssassin and rejection statistics.
1344 Don't display local sender or destination tables unless
1345 there is data to show.
1346 Added average volumes into the top table text output.
1348 SC/04 Eximstats V1.46
1349 Collect data on the number of addresses (recipients)
1350 as well as the number of messages.
1352 SC/05 Eximstats V1.47
1353 Added 'Message too big' to the list of mail rejection
1354 reasons (thanks to Marco Gaiarin).
1356 SC/06 Eximstats V1.48
1357 Mainlog lines which have GMT offsets and are too short to
1358 have a flag are now skipped.
1360 SC/07 Eximstats V1.49 (Alain Williams)
1361 Added the -emptyok flag.
1363 SC/08 Eximstats V1.50
1364 Fixes for obtaining the IP address from reject messages.
1366 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1367 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1368 whitesspace changes from 4.64-PH/27
1370 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1373 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1374 are found, allow negative numbers in numeric criteria)
1376 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1378 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1379 to match changes made in 4.64-PH/43
1381 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1383 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1384 "domain" is actually an IP address, and force a failure. This locks out
1385 those revolvers/nameservers that support "A-for-A" lookups, in
1386 contravention of the specifications.
1388 PH/48 When a host name was looked up from an IP address, and the subsequent
1389 forward lookup of the name timed out, the host name was left in
1390 $sender_host_name, contrary to the specification.
1392 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1393 restricted to single-key lookups, Exim was not diagnosing an error if
1394 * or *@ was used with a query-style lookup.
1396 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1398 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1399 long ago, but noone interested enough thought of it. Let's just say that
1400 the "1.1" means that there are some new functions that weren't there at
1401 some point in the past.
1403 PH/51 Error processing for expansion failure of helo_data from an smtp
1404 transport during callout processing was broken.
1406 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1407 tested/used via the -bh/-bhc/-bs options.
1409 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1410 bug, fixed in subsequent PCRE releases).
1412 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1413 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1415 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1421 SC/01 Use a glob alias rather than an array ref in eximstats generated
1422 parser. This improves both readability and performance.
1424 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1425 Don't display local sender or destination tables in eximstats unless
1426 there is data to show.
1427 Added average volumes into the eximstats top table text output.
1429 SC/03 Collect data on the number of addresses (recipients) as well
1430 as the number of messages in eximstats.
1432 TF/01 Correct an error in the documentation for the redirect router. Exim
1433 does (usually) call initgroups() when daemonizing.
1435 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1436 with consistent privilege compared to when running as a daemon.
1438 TF/03 Note in the spec that $authenticated_id is not set for local
1439 submissions from trusted users.
1441 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1442 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1444 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1445 by adding some example configuration directives to the default
1446 configuration file. A little bit of work is required to uncomment the
1447 directives and define how usernames and passwords are checked, but
1448 there is now a framework to start from.
1450 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1451 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1452 without this. I don't know how relevant this is to other LDAP libraries.
1454 PH/02 Add the verb name to the "unknown ACL verb" error.
1456 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1458 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1460 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1461 directory not expanded when it should be if an expanded home directory
1462 was set for the address (which is overridden by the transport).
1464 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1467 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1468 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1469 because it is too late at that time, and has no effect.
1471 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1472 security issue with \' (bugzilla #107). I could not use the
1473 PQescapeStringConn() function, because it needs a PGconn value as one of
1476 PH/08 When testing addresses using -bt, indicate those final addresses that
1477 are duplicates that would not cause an additional delivery. At least one
1478 person was confused, thinking that -bt output corresponded to deliveries.
1479 (Suppressing duplicates isn't a good idea as you lose the information
1480 about possibly different redirections that led to the duplicates.)
1482 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1483 systems where poll() doesn't work, in particular OS X.
1485 PH/10 Added more information to debugging output for retry time not reached.
1487 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1488 operations in malware.c.
1490 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1493 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1494 syslog_duplication set false, log lines that would normally be written
1495 both the the main log and to the reject log were not written to syslog at
1498 PH/14 In the default configuration, change the use of "message" in ACL warn
1499 statements to "add_header".
1501 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1502 not followed by a command (e.g. "seen endif").
1504 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1505 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1508 PH/17 Added extra conditions to the default value of delay_warning_condition
1512 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1513 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1514 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1517 The Auto-Submitted: and various List- headers are standardised, whereas I
1518 don't think Precedence: ever was.
1520 PH/18 Refactored debugging code in route_finduser() to show more information,
1521 in particular, the error code if getpwnam() issues one.
1523 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1524 This is apparently needed in addition to the PH/07 change above to avoid
1525 any possible encoding problems.
1527 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1528 but not after initializing Perl.
1530 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1531 output them only if debugging. By default they are written stderr,
1532 apparently, which is not desirable.
1534 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1537 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1540 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1542 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1543 authenticated or an ident call has been made. Suppress the default
1544 values for $authenticated_id and $authenticated_sender (but permit -oMai
1545 and -oMas) when testing with -bh.
1547 PH/24 Re-jigged the order of the tests in the default configuration so that the
1548 tests for valid domains and recipients precede the DNS black list and CSA
1549 tests, on the grounds that those ones are more expensive.
1551 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1552 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1553 This bug exists in every version of Exim that I still have, right back to
1556 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1557 However, an attempt to turn on case-sensitivity in a regex key by
1558 including (?-i) didn't work because the subject string was already
1559 lowercased, and the effects were non-intuitive. It turns out that a
1560 one-line patch can be used to allow (?-i) to work as expected.
1566 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1567 other effects) broke the use of negated acl sub-conditions.
1569 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1572 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1573 "Deny" causes Exim to reject the incoming connection with a 554 error.
1574 Unfortunately, if there is a major crisis, such as a disk failure,
1575 tcp-wrappers gives "deny", whereas what one would like would be some
1576 kind of temporary error. A kludge has been added to help with this.
1577 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1578 554 error is used if errno is still zero or contains ENOENT (which occurs
1579 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1582 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1584 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1585 errors. Otherwise a message that provokes a temporary error (when other
1586 messages do not) can cause a whole host to time out.
1588 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1589 addresses were routed directly to files or pipes from a redirect router.
1590 File deliveries just didn't batch; pipe deliveries might have suffered
1593 PH/06 A failure to get a lock for a hints database would erroneously always say
1594 "Failed to get write lock", even when it was really a read lock.
1596 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1597 of 0600. This has been changed to use the value of the lockfile_mode
1598 option (which defaults to 0600).
1600 PH/08 Applied small patch from the Sieve maintainer.
1602 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1603 folder from quota calculations, a direct delivery into this folder messed
1604 up the contents of the maildirsize file. This was because the regex was
1605 used only to exclude .Trash (or whatever) when the size of the mailbox
1606 was calculated. There was no check that a delivery was happening into an
1607 excluded directory. This bug has been fixed by ignoring all quota
1608 processing for deliveries into excluded directories.
1610 PH/10 Added the maildirfolder_create_regex option to appendfile.
1616 PH/01 The code for finding all the local interface addresses on a FreeBSD
1617 system running IPv6 was broken. This may well have applied to all BSD
1618 systems, as well as to others that have similar system calls. The broken
1619 code found IPv4 interfaces correctly, but gave incorrect values for the
1620 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1621 that it would not match correctly against @[] and not recognize the IPv6
1624 PH/02 The ipliteral router was not recognizing addresses of the form user@
1625 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1627 PH/03 Added disable_ipv6.
1629 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1630 decoded form, because it is most often used to construct To: headers
1631 lines in autoreplies, and the decoded form may well be syntactically
1632 invalid. However, $reply_address has leading white space removed, and all
1633 newlines turned into spaces so that the autoreply transport does not
1636 PH/05 If group was specified without a user on a router, and no group or user
1637 was specified on a transport, the group from the router was ignored.
1639 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1640 for visible compile-time settings that can be used to change these
1641 numbers, for those that want even more. Backwards compatibility with old
1642 spool files has been maintained. However, going back to a previous Exim
1643 release will lost any variables that are in spool files.
1645 PH/07 Two small changes when running in the test harness: increase delay when
1646 passing a TCP/IP connection to a new process, in case the original
1647 process has to generate a bounce, and remove special handling of
1648 127.0.0.2 (sic), which is no longer necessary.
1650 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1651 be the same on different OS.
1653 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1656 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1657 whether --show-vars was specified or not
1659 JJ/02 exipick: Added support for new ACL variable spool format introduced
1662 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1663 syntactically invalid From: or Reply-to: line, and a filter used this to
1664 generate an autoreply, and therefore failed to obtain an address for the
1665 autoreply, Exim could try to deliver to a non-existent relative file
1666 name, causing unrelated and misleading errors. What now happens is that
1667 it logs this as a hard delivery error, but does not attempt to create a
1670 PH/11 The exinext utility has a -C option for testing purposes, but although
1671 the given file was scanned by exinext itself; it wasn't being passed on
1672 when Exim was called.
1674 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1675 an end-of-file indication when reading a command response.
1677 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1678 compiled. In many other places in Exim, IPv6 addresses are always
1679 recognized, so I have changed this. It also means that IPv4 domain
1680 literals of the form [IPV4:n.n.n.n] are now always recognized.
1682 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1683 used if the router is not running as root, for example, when verifying at
1684 ACL time, or when using -bh. The debugging output from this situation was
1685 non-existent - all you got was a failure to exec. I have made two
1688 (a) Failures to set uid/gid, the current directory, or a process leader
1689 in a subprocess such as that created by queryprogram now generate
1690 suitable debugging ouput when -d is set.
1692 (b) The queryprogram router detects when it is not running as root,
1693 outputs suitable debugging information if -d is set, and then runs
1694 the subprocess without attempting to change uid/gid.
1696 PH/15 Minor change to Makefile for building test_host (undocumented testing
1699 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1700 additional section of a DNS packet that returns MX or SRV records.
1701 Instead, it always explicitly searches for A/AAAA records. This avoids
1702 major problems that occur when a DNS server includes only records of one
1703 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1704 fixed another bug: if SRV records were looked up and the corresponding
1705 address records were *not* found in the additional section, the port
1706 values from the SRV records were lost.
1708 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1709 using the correct key (the original address) when searching the retry
1710 rules in order to find which one to use for generating the retry hint.
1712 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1713 adding the default one. Similarly, if it contains a Reply-To: header, the
1714 errors_reply_to option, if set, is not used.
1716 PH/19 When calculating a retry time, Exim used to measure the "time since
1717 failure" by looking at the "first failed" field in the retry record. Now
1718 it does not use this if it is later than than the arrival time of the
1719 message. Instead it uses the arrival time. This makes for better
1720 behaviour in cases where some deliveries succeed, thus re-setting the
1721 "first failed" field. An example is a quota failure for a huge message
1722 when small messages continue to be delivered. Without this change, the
1723 "time since failure" will always be short, possible causing more frequent
1724 delivery attempts for the huge message than are intended.
1725 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1727 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1728 $1, $2, $3) because the numerical variables can be reset during some
1729 expansion items (e.g. "match"), thereby losing the authentication data.
1731 PH/21 Make -bV show the size of off_t variables so that the test suite can
1732 decide whether to run tests for quotas > 2G.
1734 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1735 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1736 filecount value is greater than 2G or if a quota value is greater than 2G
1737 on a system where the size of off_t is not greater than 4, a panic error
1740 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1741 never match. The debug and -bh output now contains an explicit error
1742 message indicating a malformed IPv4 address or mask.
1744 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1745 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1746 PH/23 above applies.
1748 PH/25 Do not write to syslog when running in the test harness. The only
1749 occasion when this arises is a failure to open the main or panic logs
1750 (for which there is an explicit test).
1752 PH/26 Added the /no_tell option to "control=freeze".
1754 PH/27 If a host name lookup failed very early in a connection, for example, if
1755 the IP address matched host_lookup and the reverse lookup yielded a name
1756 that did not have a forward lookup, an error message of the form "no IP
1757 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1758 could be logged. Now it outputs the IP address instead of "NULL".
1760 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1761 allows the sender and the authenticated sender to be set when
1762 submitting a message from within Exim. Since child_open_exim() is
1763 documented for local_scan(), the new function should be too.
1765 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1766 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1767 results in an empty string is now treated as unset.
1769 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1771 PH/31 Added sender_verify_fail logging option.
1773 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1774 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1775 tidied the source and removed it altogether.
1777 PH/33 When a queue run was abandoned because the load average was too high, a
1778 log line was always written; now it is written only if the queue_run log
1779 selector is set. In addition, the log line for abandonment now contains
1780 information about the queue run such as the pid. This is always present
1781 in "start" and "stop" lines but was omitted from the "abandon" line.
1783 PH/34 Omit spaces between a header name and the colon in the error message that
1784 is given when verify = headers_syntax fails (if there are lots of them,
1785 the message gets confusing).
1787 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1788 names, as there are now some PTR records that contain slashes. This check
1789 is only to protect against broken name servers that fall over on strange
1790 characters, so the fact that it applies to all lookups doesn't matter.
1792 PH/36 Now that the new test suite is complete, we can remove some of the
1793 special code in Exim that was needed for the old test suite. For example,
1794 sorting DNS records because real resolvers return them in an arbitrary
1795 order. The new test suite's fake resolver always returns records in the
1798 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1799 bounces) except when queue_only is set, to avoid logging races between
1800 the different processes.
1802 PH/38 Panic-die if .include specifies a non-absolute path.
1804 PH/39 A tweak to the "H" retry rule from its user.
1806 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1807 a label. They prevented compilation on older perls.
1809 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1810 a warning to be raised on newish perls.
1812 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1813 on queue. Changes to match documented behaviour of showing count of
1814 messages matching specified criteria.
1816 PH/40 Changed the default ident timeout from 30s to 5s.
1818 PH/41 Added support for the use of login_cap features, on those BSD systems
1819 that have them, for controlling the resources used by pipe deliveries.
1821 PH/42 The content-scanning code uses fopen() to create files in which to put
1822 message data. Previously it was not paying any attention to the mode of
1823 the files. Exim runs with umask(0) because the rest of the code creates
1824 files with open(), and sets the required mode explicitly. Thus, these
1825 files were ending up world-writeable. This was not a big issue, because,
1826 being within the spool directory, they were not world-accessible. I have
1827 created a function called modefopen, which takes an additional mode
1828 argument. It sets umask(777), creates the file, chmods it to the required
1829 mode, then resets the umask. All the relevant calls to fopen() in the
1830 content scanning code have been changed to use this function.
1832 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1833 to 24 hours. This avoids potential overflow problems when processing G
1834 and H retry rules. I suspect nobody ever tinkers with this value.
1836 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1838 PH/45 When the plaintext authenticator is running as a client, the server's
1839 challenges are checked to ensure they are valid base64 strings. By
1840 default, the authentication attempt is cancelled if an invalid string is
1841 received. Setting client_ignore_invalid_base64 true ignores these errors.
1842 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1843 they are received. Thus, the responses can be made to depend on the
1844 challenges. If an invalid string is ignored, an empty string is placed in
1847 PH/46 Messages that are created by the autoreply transport now contains a
1848 References: header, in accordance with RFCs 2822 and 3834.
1850 PH/47 Added authenticated_sender_force to the smtp transport.
1852 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1854 PH/49 Installed latest patch from the Sieve maintainer.
1856 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1857 was also set, the appendfile transport was unnecessarily scanning a
1858 directory of message files (e.g. for maildir delivery) to find the count
1859 of files (along with the size), even though it did not need this
1860 information. It now does the scan only if it needs to find either the
1861 size of the count of files.
1863 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1865 PH/52 Two bugs concerned with error handling when the smtp transport is
1868 (i) Exim was not creating retry information for temporary errors given
1869 for individual recipients after the DATA command when the smtp transport
1870 was used in LMTP mode. This meant that they could be retried too
1871 frequently, and not timed out correctly.
1873 (ii) Exim was setting the flag that allows error details to be returned
1874 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1875 recipients that were returned after the DATA command.
1877 PH/53 This is related to PH/52, but is more general: for any failing address,
1878 when detailed error information was permitted to be returned to the
1879 sender, but the error was temporary, then after the final timeout, only
1880 "retry timeout exceeded" was returned. Now it returns the full error as
1881 well as "retry timeout exceeded".
1883 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1884 do this, and (what is worse) MTAs that accept it.
1886 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1887 will now be deprecated.
1889 PH/56 New os.c-cygwin from the Cygwin maintainer.
1891 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1892 formats (previously only available in exim formats via -bpr, -bpru,
1893 and -bpra. Now also available in native and exiqgrep formats)
1895 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1896 with very large, slow to parse queues
1898 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1900 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1902 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1903 responses to authentication challenges, though it was showing the
1904 challenges; (ii) I've removed the CR characters from the debug output for
1907 PH/58 Allow for the insertion of a newline as well as a space when a string
1908 is turned into more than one encoded-word during RFC 2047 encoding. The
1909 Sieve code now uses this.
1911 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1912 data_4xx, lost_connection, tls_required.
1914 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1915 message was being sent as an SMTP response.
1917 PH/61 Add -l and -k options to exicyclog.
1919 PH/62 When verifying, if an address was redirected to one new address, so that
1920 verification continued, and the new address failed or deferred after
1921 having set something in $address_data, the value of $address_data was not
1922 passed back to the ACL. This was different to the case when no
1923 redirection occurred. The value is now passed back in both cases.
1925 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1926 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1927 use login_cap.h, so on its own it isn't the distinguishing feature. The
1928 new name refers directly to the setclassresources() function.
1930 PH/65 Added configuration files for NetBSD3.
1932 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1934 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1935 is preferred over IPv4.
1937 PH/68 The bounce_return_message and bounce_return_body options were not being
1938 honoured for bounces generated during the reception of non-SMTP messages.
1939 In particular, this applied to messages rejected by the ACL. This bug has
1940 been fixed. However, if bounce_return_message is true and bounce_return_
1941 body is false, the headers that are returned for a non-SMTP message
1942 include only those that have been read before the error was detected.
1943 (In the case of an ACL rejection, they have all been read.)
1945 PH/69 The HTML version of the specification is now built in a directory called
1946 spec_html instead of spec.html, because the latter looks like a path with
1947 a MIME-type, and this confuses some software.
1949 PH/70 Catch two compiler warnings in sieve.c.
1951 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1952 function verify_get_ident() calls ip_connect() to connect a socket, but
1953 if the "connect()" function timed out, ip_connect() used to close the
1954 socket. However, verify_get_ident() also closes the socket later, and in
1955 between Exim writes to the log, which may get opened at this point. When
1956 the socket was closed in ip_connect(), the log could get the same file
1957 descriptor number as the socket. This naturally causes chaos. The fix is
1958 not to close the socket in ip_connect(); the socket should be closed by
1959 the function that creates it. There was only one place in the code where
1960 this was missing, in the iplookup router, which I don't think anybody now
1961 uses, but I've fixed it anyway.
1963 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1964 well as to direct DNS lookups. Otherwise the handling of names in host
1965 lists is inconsistent and therefore confusing.
1971 PH/01 Two changes to the default runtime configuration:
1973 (1) Move the checks for relay_from_hosts and authenticated clients from
1974 after to before the (commented out) DNS black list checks.
1976 (2) Add control=submission to the relay_from_hosts and authenticated
1977 clients checks, on the grounds that messages accepted by these
1978 statements are most likely to be submissions.
1980 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1982 (1) Generate an error if the third argument for the ${prvs expansion is
1985 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1988 (3) Reset the variables that are obtained from the first argument of
1989 ${prvscheck and used in the second argument before leaving the code,
1990 because their memory is reclaimed, so using them afterwards may do
1993 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1994 one (it's much easier than Tom thought :-).
1996 (5) Because of (4), we can now allow for the use of $prvscheck_result
1997 inside the third argument.
1999 PH/03 For some reason, the default setting of PATH when running a command from
2000 a pipe transport was just "/usr/bin". I have changed it to
2003 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2004 anything to be listed in the output from -bV.
2006 PH/05 When a filter generated an autoreply, the entire To: header line was
2007 quoted in the delivery log line, like this:
2009 => >A.N.Other <ano@some.domain> <original@ddress> ...
2011 This has been changed so that it extracts the operative address. There
2012 may be more than one such address. If so, they are comma-separated, like
2015 => >ano@some.domain,ona@other.domain <original@ddress> ...
2017 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2018 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2019 not being looked up in the rDNS to get a host name, Exim was showing the
2020 IP address twice in Received: lines, even though the IP addresses were
2021 identical. For example:
2023 Received: from [1.2.3.4] (helo=[1.2.3.4])
2025 However, if the real host name was known, it was omitting the HELO data
2026 if it matched the actual IP address. This has been tidied up so that it
2027 doesn't show the same IP address twice.
2029 PH/07 When both +timestamp and +memory debugging was on, the value given by
2030 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2031 called by the debug printing, thereby overwriting the timestamp buffer.
2032 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2034 PH/08 When the original message was included in an autoreply transport, it
2035 always said "this is a copy of the message, including all the headers",
2036 even if body_only or headers_only was set. It now gives an appropriate
2039 PH/09 Applied a patch from the Sieve maintainer which:
2041 o fixes some comments
2042 o adds the (disabled) notify extension core
2043 o adds some debug output for the result of if/elsif tests
2044 o points to the current vacation draft in the documentation
2045 and documents the missing references header update
2049 o fixes a bug in processing the envelope test (when testing
2050 multiple envelope elements, the last element determinted the
2053 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2054 Electronic Mail") by including:
2056 Auto-submitted: auto-generated
2058 in the messages that it generates (bounce messages and others, such as
2059 warnings). In the case of bounce messages for non-SMTP mesages, there was
2060 also a typo: it was using "Auto_submitted" (underscore instead of
2061 hyphen). Since every message generated by Exim is necessarily in response
2062 to another message, thes have all been changed to:
2064 Auto-Submitted: auto-replied
2066 in accordance with these statements in the RFC:
2068 The auto-replied keyword:
2070 - SHOULD be used on messages sent in direct response to another
2071 message by an automatic process,
2073 - MUST NOT be used on manually-generated messages,
2075 - MAY be used on Delivery Status Notifications (DSNs) and Message
2076 Disposition Notifications (MDNs),
2078 - MUST NOT be used on messages generated by automatic or periodic
2079 processes, except for messages which are automatic responses to
2082 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2083 to the default Received: header definition.
2085 PH/12 Added log selector acl_warn_skipped (default on).
2087 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2088 variables because (a) they are in the wrong storage pool and (b) even if
2089 they were copied, it wouldn't work properly because of the caching.
2091 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2092 checking when decoding. Apparently there are clients that generate
2093 overlong encoded strings. Why am I not surprised?
2095 PH/15 If the first argument of "${if match_address" was not empty, but did not
2096 contain an "@" character, Exim crashed. Now it writes a panic log message
2097 and treats the condition as false.
2099 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2101 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2102 extension "envelope-auth". The code is finished and in agreement with
2103 other implementations, but there is no documentation so far and in fact,
2104 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2105 not changing the active code.
2107 Print executed "if" and "elsif" statements when debugging is used. This
2108 helps a great deal to understand what a filter does.
2110 Document more things not specified clearly in RFC3028. I had all this
2111 sorted out, when out of a sudden new issues came to my mind. Oops."
2113 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2116 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2117 canonical form (as documented). However, after a host name lookup from
2118 the IP address, check_host() was doing a simple string comparison with
2119 addresses acquired from the DNS when checking that the found name did
2120 have the original IP as one of its addresses. Since any found IPv6
2121 addresses are likely to be in abbreviated form, the comparison could
2122 fail. Luckily, there already exists a function for doing the comparison
2123 by converting both addresses to binary, so now that is used instead of
2124 the text comparison.
2126 PH/20 There was another similar case to PH/19, when a complete host name was
2127 given in a host list; looking up its IP address could give an abbreviated
2128 form, whereas the current host's name might or might not be abbreviated.
2129 The same fix has been applied.
2135 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2136 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2139 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2140 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2142 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2144 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2145 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2146 Also, for "Auto-Submitted", treat anything other than "no" as
2147 non-personal, in accordance with RFC 3834. (Previously it treated
2148 anything starting "auto-" as non-personal.)
2150 TF/01 The control=submission/name=... option had a problem with syntax
2151 errors if the name included a slash character. The /name= option
2152 now slurps the rest of the string, so it can include any characters
2153 but it must come last in the list of options (after /sender_retain
2156 PH/05 Some modifications to the interface to the fake nameserver for the new
2164 TK/01 Added the "success_on_redirect" address verification option. See
2165 NewStuff for rationale and an example.
2167 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2169 PH/02 Patch to exigrep to allow it to work on syslog lines.
2171 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2172 fread() to skip over the body file's header line, because in Cygwin the
2173 header line is locked and is inaccessible.
2175 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2176 co-exist for some time) to make it clear that it is the Exim ID that is
2177 referenced, not the Message-ID: header line.
2179 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2180 string_format() function, because snprintf() does not exist on all
2183 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2184 ${stat: expansion item.
2186 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2187 protocol synchronization error", to keep the pedants happy.
2189 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2190 well as for IRIX systems, when gcc is being used. See the host.c source
2193 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2195 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2198 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2199 transport and to the smtp transport in LMTP mode.
2201 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2203 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2204 run a filter in a subprocess. This could lead to confusion in subsequent
2205 lookups in the parent process. There should also be a search_tidyup() at
2206 the end of the subprocess.
2208 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2209 only if the host matched helo_try_verify_hosts, which caused the
2210 verification to occur when the EHLO/HELO command was issued. The ACL just
2211 tested the remembered result. Now, if a previous verification attempt has
2212 not happened, "verify = helo" does it there and then.
2214 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2216 TK/03 Fix log output including CR from clamd.
2218 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2219 exist provoked a memory error which could cause a segfault.
2221 PH/15 Installed PCRE 6.2
2223 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2225 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2226 of the problem. Specifically, suggested +O2 rather than +O1 for the
2229 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2231 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2232 record, the port was not correctly logged when the outgoing_port log
2233 selector was set (it logged the transort's default port).
2235 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2236 fallback_hosts, and "hosts" in the smtp transport.
2238 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2239 host errors such as "Connection refused".
2241 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2242 authentication with radiusclient 0.4.9:
2244 - Error returned from rc_read_config was caught wrongly
2245 - Username/password not passed on to radius server due to wrong length.
2247 The presumption is that some radiusclient API changes for 4.51/PH/17
2248 were not taken care of correctly. The code is still untested by me (my
2249 Linux distribution still has 0.3.2 of radiusclient), but it was
2250 contributed by a Radius user.
2252 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2253 expanding the "port" option of the smtp transport.
2255 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2256 while reading a MIME header. Thanks to Tom Hughes for a patch.
2258 PH/24 Include config.h inside local_scan.h so that configuration settings are
2261 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2262 that in an ACL for RCPT (for example), you can examine exactly what was
2265 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2266 commands, but it was not correctly comparing the address with the actual
2267 client host address. Thus, it would show the EHLO address in Received:
2268 header lines when this was not necessary.
2270 PH/27 Added the % operator to ${eval:}.
2272 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2273 it should be ignoring failures (because with -C, for example, it has lost
2274 privilege). It wasn't ignoring creation failures other than "already
2277 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2280 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2281 because an input line was too long, either on its own, or by virtue of
2282 too many continuations, the temporary file was not being removed, and the
2283 return code was incorrect.
2285 PH/31 Missing "BOOL" in function definition in filtertest.c.
2287 PH/32 Applied Sieve patches from the maintainer.
2289 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2291 PH/33 Added "verify = not_blind".
2293 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2294 Local/Makefile (with some defaults set). These are used in built scripts
2295 such as exicyclog, but they have never been used in the exim_install
2296 script (though there are many overriding facilities there). I have
2297 arranged that the exim_install script now takes note of these two
2300 PH/35 Installed configuration files for Dragonfly.
2302 PH/36 When a locally submitted message by a trusted user did not contain a
2303 From: header, and the sender address was obtained from -f or from an SMTP
2304 MAIL command, and the trusted user did not use -F to supply a sender
2305 name, $originator_name was incorrectly used when constructing a From:
2306 header. Furthermore, $originator_name was used for submission mode
2307 messages from external hosts without From: headers in a similar way,
2308 which is clearly wrong.
2310 PH/37 Added control=suppress_local_fixups.
2312 PH/38 When log_selector = +received_sender was set, and the addition of the
2313 sender made the log line's construction buffer exactly full, or one byte
2314 less than full, an overflow happened when the terminating "\n" was
2317 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2318 when the result of a list match is failure because a DNS lookup failed.
2320 PH/40 RM_COMMAND is now used in the building process.
2322 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2323 the "build-* directories that it finds.
2325 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2326 address in a domain literal was a prefix of an interface address.
2328 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2329 when verifying a sender address, unless rewrite_headers is false.
2331 PH/44 Wrote a long comment about why errors_to addresses are verified as
2332 recipients, not senders.
2334 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2335 the ratelimit ACL was added.
2337 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2339 PH/46 Added extra information about PostgreSQL errors to the error string.
2341 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2342 avoiding the need to install special zones in a real server. This is
2343 backwards compatible; if it can't find the fake resolver, it drops back.
2344 Thus, both old and new test suites can be run.
2346 TF/02 Added util/ratelimit.pl
2348 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2349 clock is set back in time.
2351 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2352 Candler <B.Candler@pobox.com>.
2354 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2355 OK for addresses that are the result of redirections.
2357 PH/48 A number of further additions for the benefit of the new test suite,
2358 including a fake gethostbyname() that interfaces to the fake DNS resolver
2361 TF/06 The fix for widen_domains has also been applied to qualify_single and
2362 search_parents which are the other dnslookup options that can cause
2365 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2368 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2370 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2371 service" if it encountered an MX record with an empty target hostname.
2372 The message is now "an MX or SRV record indicated no SMTP service".
2374 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2375 if the DNS of the sending site is misconfigured. This is quite a
2376 common situation. This change restores the behaviour of treating a
2377 helo verification defer as a failure.
2379 PH/51 If self=fail was set on a router, the bounce message did not include the
2380 actual error message.
2386 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2388 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2389 command itself ended in error, the underlying message about the transport
2390 filter timeout was being overwritten with the pipe command error. Now the
2391 underlying error message should be appended to the second error message.
2393 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2395 PH/02 Reduce the amount of output that "make" produces by default. Full output
2396 can still be requested.
2398 PH/03 The warning log line about a condition test deferring for a "warn" verb
2399 was being output only once per connection, rather than after each
2400 occurrence (because it was using the same function as for successful
2401 "warn" verbs). This seems wrong, so I have changed it.
2403 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2404 it should not have, which might have caused a crash in the right
2405 circumstances, but probably never did.
2407 PH/04 Installed a modified version of Tony Finch's patch to make submission
2408 mode fix the return path as well as the Sender: header line, and to
2409 add a /name= option so that you can make the user's friendly name appear
2412 TF/03 Added the control = fakedefer ACL modifier.
2414 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2415 Mark Lowes for thorough testing.
2417 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2419 TK/03 Merged latest SRS patch from Miles Wilton.
2421 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2422 with the definition in sysexits.h (which is #included earlier).
2423 Fortunately, Exim does not actually use EX_OK. The code used to try to
2424 preserve the sysexits.h value, by assumimg that macro definitions were
2425 scanned for macro replacements. I have been disabused of this notion,
2426 so now the code just undefines EX_OK before #including unistd.h.
2428 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2429 in the smtp transport. When a block could not be written in a single
2430 write() function, the timeout was being re-applied to each part-write.
2431 This seems wrong - if the receiver was accepting one byte at a time it
2432 would take for ever. The timeout is now adjusted when this happens. It
2433 doesn't have to be particularly precise.
2435 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2436 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2438 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2439 without a domain if the check to <postmaster@domain> fails.
2441 SC/01 Eximstats: added -xls and the ability to specify output files
2442 (patch written by Frank Heydlauf).
2444 SC/02 Eximstats: use FileHandles for outputing results.
2446 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2448 SC/04 Eximstats: fixed display of large numbers with -nvr option
2450 SC/05 Eximstats: fixed merging of reports with empty tables.
2452 SC/06 Eximstats: added the -include_original_destination flag
2454 SC/07 Eximstats: removed tabs and trailing whitespace.
2456 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2458 TK/06 MBOX spool code: Add real "From " MBOX separator line
2459 so the .eml file is really in mbox format (even though
2460 most programs do not really care). Patch from Alex Miller.
2462 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2463 The latter is generated from $received_to and is only set if the
2464 message has one envelope recipient. SA can use these headers,
2465 obviously out-of-the-box. Patch from Alex Miller.
2467 PH/08 The ${def test on a variable was returning false if the variable's
2468 value was "0", contrary to what the specification has always said!
2469 The result should be true unless the variable is empty.
2471 PH/09 The syntax error of a character other than { following "${if
2472 def:variable_name" (after optional whitespace) was not being diagnosed.
2473 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2474 accidental colon was present, for example, could give incorrect results.
2476 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2477 result is used (not including appendfile, where other changes are about
2480 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2481 This involved changing a lot of size variables from int to off_t. It
2482 should work with maildirs and everything.
2484 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2485 spamd dying while we are connected to it.
2487 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2490 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2491 function for BDB 4.3.
2493 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2495 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2496 the sources for PCRE have been re-arranged and more files are now
2499 PH/15 The code I had for printing potentially long long variables in PH/11
2500 above was not the best (it lost precision). The length of off_t variables
2501 is now inspected at build time, and an appropriate printing format (%ld
2502 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2503 to be "long long int" or "long int". This is needed for the internal
2504 formatting function string_vformat().
2506 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2507 the configuration file to be ":syslog", then the script "guesses" where
2508 the logs files are, rather than using the compiled in default. In our
2509 case the guess is not the same as the compiled default, so the script
2510 suddenly stopped working when I started to use syslog. The patch checks
2511 to see if log_file_path is "". If so, it attempts to read it from exim
2512 with no configuration file to get the compiled in version, before it
2513 falls back to the previous guessing code."
2515 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2516 implementing BATV in an Exim configuration. See NewStuff for the gory
2519 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2520 Makefile that are specific to HP-UX.
2522 PH/18 If the "use_postmaster" option was set for a recipient callout together
2523 with the "random" option, the postmaster address was used as the MAIL
2524 FROM address for the random test, but not for the subsequent recipient
2525 test. It is now used for both.
2527 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2528 patch removes a few documentation additions to RFC 3028, because the
2529 latest draft now contains them. It adds the new en;ascii-case comparator
2530 and a new error check for 8bit text in MIME parts. Comparator and
2531 require names are now matched exactly. I enabled the subaddress
2532 extension, but it is not well tested yet (read: it works for me)."
2534 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2535 rework some of the code of TK/09 above to avoid the hardwired use of
2536 "%lld" and "long long". Replaced the call to snprintf() with a call to
2539 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2540 records point to non-existent hosts", "retry timeout exceeded", and
2541 "retry time not reached for any host after a long failure period".
2543 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2544 experimental DomainKeys support:
2546 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2547 (2) On an error such as an illegally used "control", the wrong name for
2548 the control was given.
2550 These problems did NOT occur unless DomainKeys support was compiled.
2552 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2554 PH/24 Added ${if match_ip condition.
2556 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2557 databases so that it will be absolutely obvious if a crash occurs in the
2558 DB library. This is a regular occurrence (often caused by mis-matched
2561 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2562 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2563 picked up on a user's system that detects such things. There doesn't seem
2564 to be a gcc warning option for this - only an attribute that has to be
2565 put on the function's prototype. It seems that in Fedora Core 4 they have
2566 set this on a number of new functions. No doubt there will be more in due
2569 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2570 specify a transport. However, if an address that was verified by such a
2571 router was the subject of a callout, Exim crashed because it tried to
2572 read the rcpt_include_affixes from the non-existent transport. Now it
2573 just assumes that the setting of that option is false. This bug was
2574 introduced by 4.51/PH/31.
2576 PH/28 Changed -d+all to exclude +memory, because that information is very
2577 rarely of interest, but it makes the output a lot bigger. People tend to
2578 do -d+all out of habit.
2580 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2581 code in os-type was giving problems when libc.so lives in lib64, like on
2584 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2585 aren't the modern standard, and it seems that some systems' include files
2586 don't always have them. Exim was already checking for some of the newer
2587 ones like T_AAAA, and defining it itself. I've added checks for all the
2588 record types that Exim uses.
2590 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2591 not automatically generating a new one, as it is supposed to. This
2592 prevented TLS from working. If the file did exist, but contained invalid
2593 data, a new version was generated, as expected. It was only the case of a
2594 non-existent file that was broken.
2596 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2597 with a change in libdomainkeys > 0.64.
2599 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2600 from DNS. If the selector record carries the flag, it now has
2601 precedence over the domain-wide flag.
2603 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2605 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2606 the use of a port name that isn't defined in /etc/services) occurred, the
2607 message was deferred as in a normal delivery, and thus remained on the
2608 spool, instead of being failed because of the mua_wrapper setting. This
2609 is now fixed, and I tidied up some of the mua_wrapper messages at the
2612 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2613 the messages in a hash of arrays rather than using individual hashes.
2614 This is a bit cleaner and results in dramatic memory savings, albeit
2615 at a slight CPU cost.
2617 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2618 as requested by Marc Sherman.
2620 SC/10 Eximstats: added histograms for user specified patterns as requested
2623 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2625 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2626 fopen() in the content-scanning modules that did not already have it.
2632 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2633 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2635 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2637 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2639 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2640 to test Sieve filters that use "vacation".
2642 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2643 that changes the way the GnuTLS parameters are stored in the cache file.
2644 The new format can be generated externally. For backward compatibility,
2645 if the data in the cache doesn't make sense, Exim assumes it has read an
2646 old-format file, and it generates new data and writes a new file. This
2647 means that you can't go back to an older release without removing the
2650 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2651 work if there are any delivery delays because "one_time" forces the
2652 parent to be marked "delivered", so its unseen clone is never tried
2653 again. For this reason, Exim now forbids the simultaneous setting of
2656 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2657 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2658 entry if you want to know the details. The fix, however, neglected to
2659 consider the case when local delivery batching is involved. The test for
2660 "previously delivered" was not happening when checking to see if an
2661 address could be batched with a previous (undelivered) one; under
2662 certain circumstances this could lead to multiple deliveries to the same
2665 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2666 in its include files, and this causes problems building Exim.
2668 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2669 header_syntax) but Exim was just ignoring anything given after a slash.
2670 In particular, this caused confusion with an attempt to use "verify =
2671 reverse_host_lookup/defer_ok". An error is now given when options are
2672 supplied for verify items that do not have them. (Maybe reverse_host_
2673 lookup should have a defer_ok option, but that's a different point.)
2675 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2676 defined by RFC 821) to 2048, because there were problems with some AUTH
2677 commands, and RFC 1869 says the size should be increased for extended
2678 SMTP commands that take arguments.
2680 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2683 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2684 "unknown" error; now it says that the functionality isn't in the binary.
2686 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2687 an address' error message when a string expansion fails (syntax or
2688 whatever). Otherwise the password may appear in the log. Following change
2689 PH/42 below, there is no longer a chance of it appearing in a bounce
2692 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2694 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2695 of its addresses was ever tried. (Bugzilla bug #2.)
2697 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2698 the result incorrectly in the debug output. (It correctly added a newline
2699 to what was transported.)
2701 TF/01 Added $received_time.
2703 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2704 commented out examples of how to interface to a virus scanner and to
2705 SpamAssassin. Also added commented examples of av_scanner and
2706 spamd_address settings.
2708 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2709 and controls are allowed in which ACLs. There were a couple of minor
2710 errors. Some of the entries in the conditions table (which is a table of
2711 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2712 negation of where the condition IS allowed.
2714 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2716 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2717 header file does not have a version number, so I've had to invent a new
2718 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2719 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2720 radiusclient), but it was contributed by a Radius user.
2722 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2723 files or directories, for OpenSSL.
2725 PH/19 When an Exim process that is running as root has to create an Exim log
2726 file, it does so in a subprocess that runs as exim:exim so as to get the
2727 ownership right at creation (otherwise, other Exim processes might see
2728 the file with the wrong ownership). There was no test for failure of this
2729 fork() call, which would lead to the process getting stuck as it waited
2730 for a non-existent subprocess. Forks do occasionally fail when resources
2731 run out. I reviewed all the other calls to fork(); they all seem to check
2734 PH/20 When checking for unexpected SMTP input at connect time (before writing
2735 the banner), Exim was not dealing correctly with a non-positive return
2736 from the read() function. If the client had disconnected by this time,
2737 the result was a log entry for a synchronization error with an empty
2738 string after "input=" when read() returned zero. If read() returned -1
2739 (an event I could not check), uninitialized data bytes were printed.
2740 There were reports of junk text (parts of files, etc) appearing after
2743 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2745 PH/22 Added support for macro redefinition, and (re)definition in between
2746 driver and ACL definitions.
2748 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2749 forgetting to use the resulting value; it was using the unexpanded value.
2751 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2752 hadn't been configured. The fix is from Juergen Kreileder, who
2753 understands it better than I do:
2755 "Here's what I see happening with three configured cyrus_sasl
2756 authenticators configured (plain, login, cram-md5):
2758 On startup auth_cyrus_sasl_init() gets called for each of these.
2759 This means three calls to sasl_listmech() without a specified mech_list.
2760 => SASL tests which mechs of all available mechs actually work
2761 => three warnings about OTP not working
2762 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2764 With the patch, sasl_listmech() also gets called three times. But now
2765 SASL's mech_list option is set to the server_mech specified in the the
2766 authenticator. Or in other words, the answer from sasl_listmech()
2767 gets limited to just the mech you're testing for (which is different
2769 => the return list contains just 'plain' or 'login', 'cram-md5' or
2770 nothing depending on the value of ob->server_mech.
2772 I've just tested the patch: Authentication still works fine,
2773 unavailable mechs specified in the exim configuration are still
2774 caught, and the auth.log warnings about OTP are gone."
2776 PH/25 When debugging is enabled, the contents of the command line are added
2777 to the debugging output, even when log_selector=+arguments is not
2780 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2781 answer is "GNU", and only if the return is "GNU/something" is the answer
2784 PH/27 $acl_verify_message is now set immediately after the failure of a
2785 verification in an ACL, and so is available in subsequent modifiers. In
2786 particular, the message can be preserved by coding like this:
2788 warn !verify = sender
2789 set acl_m0 = $acl_verify_message
2791 Previously, $acl_verify_message was set only while expanding "message"
2792 and "log_message" when a very denied access.
2794 PH/28 Modified OS/os.c-Linux with
2796 -#ifndef OS_LOAD_AVERAGE
2797 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2799 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2800 nomenclature these days.)
2802 PH/29 Installed patch from the Sieve maintainer that adds the options
2803 sieve_useraddress and sieve_subaddress to the redirect router.
2805 PH/30 In these circumstances:
2806 . Two addresses routed to the same list of hosts;
2807 . First host does not offer TLS;
2808 . First host accepts first address;
2809 . First host gives temporary error to second address;
2810 . Second host offers TLS and a TLS session is established;
2811 . Second host accepts second address.
2812 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2813 and peerdn, if requested) that were in fact used only for the second
2816 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2817 attention to any local part prefix or suffix that was matched by the
2818 router that accepted the address. It now behaves in the same way as it
2819 does for delivery: the affixes are removed from the local part unless
2820 rcpt_include_affixes is set on the transport.
2822 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2823 timeout during the DATA phase of an incoming message.
2825 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2826 applied a patch sanctioned by the Sieve maintainer.
2828 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2829 the uid or gid is negative. A case of a negative gid caused this to be
2830 noticed. The fix allows for either to be negative.
2832 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2833 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2836 PH/36 The change PH/12 above was broken. Fixed it.
2838 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2839 the grounds that routing the same address twice would always produce the
2840 same answer. This might have been true once, but it is certainly no
2841 longer true now. Routing a child address may depend on the previous
2842 routing that produced that child. Some complicated redirection strategies
2843 went wrong when messages had multiple recipients, and made Exim's
2844 behaviour dependent on the order in which the addresses were given.
2846 I have moved the duplicate checking until after the routing is complete.
2847 Exim scans the addresses that are assigned to local and remote
2848 transports, and removes any duplicates. This means that more work will be
2849 done, as duplicates will always all be routed, but duplicates are
2850 presumably rare, so I don't expect this is of any significance.
2852 For deliveries to pipes, files, and autoreplies, the duplicate checking
2853 still happens during the routing process, since they are not going to be
2856 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2857 It corrects a timeout issue with spamd. This is Ian's comment: "The
2858 background is that sometimes spamd either never reads data from a
2859 connection it has accepted, or it never writes response data. The exiscan
2860 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2861 blindly assumes that writes won't block so it may never time out."
2863 PH/39 Allow G after quota size as well as K and M.
2865 PH/40 The value set for $authenticated_id in an authenticator may not contain
2866 binary zeroes or newlines because the value is written to log lines and
2867 to spool files. There was no check on this. Now the value is run through
2868 the string_printing() function so that such characters are converted to
2869 printable escape sequences.
2871 PH/41 $message_linecount is a new variable that contains the total number of
2872 lines in the message. Compare $body_linecount, which is the count for the
2875 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2876 bounce and delay warning messages, except in certain special cases, which
2879 (a) An SMTP error message from a remote host;
2880 (b) A message specified in a :fail: redirection;
2881 (c) A message specified in a "fail" command in a system filter;
2882 (d) A message specified in a FAIL return from the queryprogram router;
2883 (e) A message specified by the cannot_route_message router option.
2885 In these cases only, Exim does include the error details in bounce and
2886 warning messages. There are also a few cases where bland messages such
2887 as "unrouteable address" or "local delivery error" are given.
2889 PH/43 $value is now also set for the "else" part of a ${run expansion.
2891 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2892 being worked on, but at least Exim now implements the latest version to
2895 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2896 process to complete was treated as a delivery failure, a timeout while
2897 writing the message to the pipe was logged, but erroneously treated as a
2898 successful delivery. Such timeouts include transport filter timeouts. For
2899 consistency with the overall process timeout, these timeouts are now
2900 treated as errors, giving rise to delivery failures by default. However,
2901 there is now a new Boolean option for the pipe transport called
2902 timeout_defer, which, if set TRUE, converts the failures into defers for
2903 both kinds of timeout. A transport filter timeout is now identified in
2906 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2907 systems where "make" and "gmake" are different, calling "gmake" at top
2908 level broke things. I've arranged for the value of $(MAKE) to be passed
2909 from the Makefile to this script so that it can call the same version of
2913 A note about Exim versions 4.44 and 4.50
2914 ----------------------------------------
2916 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2917 changes of various kinds. As a consequence, a big documentation update was
2918 needed. This delayed the release for rather longer than seemed good, especially
2919 in the light of a couple of (minor) security issues. Therefore, the changes
2920 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2921 release. So 4.44 and 4.50 are in effect two different branches that both start
2924 I have left the 4.50 change log unchanged; it contains all the changes since
2925 4.43. The change log for 4.44 is below; many of its items are identical to
2926 those for 4.50. This seems to be the most sensible way to preserve the
2927 historical information.
2933 1. Minor wording change to the doc/README.SIEVE file.
2935 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2936 computation of the current number of files was incorrect.
2938 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2939 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2942 4. Give more explanation in the error message when the command for a transport
2943 filter fails to execute.
2945 5. There are several places where Exim runs a non-Exim command in a
2946 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2947 was being done only for the command run by the queryprogram router. It is
2948 now done for all such subprocesses. The other cases are: ${run, transport
2949 filters, and the commands run by the lmtp and pipe transports.
2951 6. Added CONFIGURE_GROUP build-time option.
2953 7. Some older OS have a limit of 256 on the maximum number of file
2954 descriptors. Exim was using setrlimit() to set 1000 as a large value
2955 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2956 systems. I've change it so that if it can't get 1000, it tries for 256.
2958 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2959 was an oversight, and furthermore, ever since the addition of extra
2960 controls (e.g. 4.43/32), the checks on when to allow different forms of
2961 "control" were broken. There should now be diagnostics for all cases when a
2962 control that does not make sense is encountered.
2964 9. Added the /retain_sender option to "control=submission".
2966 10. $recipients is now available in the predata ACL (oversight).
2968 11. Tidy the search cache before the fork to do a delivery from a message
2969 received from the command line. Otherwise the child will trigger a lookup
2970 failure and thereby defer the delivery if it tries to use (for example) a
2971 cached ldap connection that the parent has called unbind on.
2973 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2974 of $address_data from the recipient verification was clobbered by the
2975 sender verification.
2977 13. The value of address_data from a sender verification is now available in
2978 $sender_address_data in subsequent conditions in the ACL statement.
2980 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2982 15. Added a new option "connect=<time>" to callout options, to set a different
2985 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2986 was its contents. (It was OK if the option was not defined at all.)
2988 17. A "Completed" log line is now written for messages that are removed from
2989 the spool by the -Mrm option.
2991 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2992 information about exactly what failed.
2994 19. Added -dd to debug only the daemon process.
2996 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2997 handles timeouts, both on the server side and network timeouts. Renamed the
2998 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3000 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3001 It is now set to "smtps".
3003 22. $host_address is now set to the target address during the checking of
3004 ignore_target_hosts.
3006 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3007 being passed; this would have caused $sender_host_name to have been used if
3008 matching the list had actually called for a host name (not very likely,
3009 since this list is usually IP addresses). A host name is now passed as
3012 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3013 code that specifies a non-restarting handler (typically sigaction() in
3014 modern systems) in an attempt to fix a rare and obscure crash bug.
3016 25. Narrowed the window for a race in the daemon that could cause it to ignore
3017 SIGCHLD signals. This is not a major problem, because they are used only to
3018 wake it up if nothing else does.
3020 26. A malformed maildirsize file could cause Exim to calculate negative values
3021 for the mailbox size or file count. Odd effects could occur as a result.
3022 The maildirsize information is now recalculated if the size or filecount
3025 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3026 support for a long time. Removed HAVE_SYS_VFS_H.
3028 28. Installed the latest version of exipick from John Jetmore.
3030 29. In an address list, if the pattern was not a regular expression, an empty
3031 subject address (from a bounce message) matched only if the pattern was an
3032 empty string. Non-empty patterns were not even tested. This was the wrong
3033 because it is perfectly reasonable to use an empty address as part of a
3034 database query. An empty address is now tested by patterns that are
3035 lookups. However, all the other forms of pattern expect the subject to
3036 contain a local part and a domain, and therefore, for them, an empty
3037 address still always fails if the pattern is not itself empty.
3039 30. Exim went into a mad DNS loop when attempting to do a callout where the
3040 host was specified on an smtp transport, and looking it up yielded more
3041 than one IP address.
3043 31. Re-factored the code for checking spool and log partition space into a
3044 function that finds that data and another that does the check. The former
3045 is then used to implement four new variables: $spool_space, $log_space,
3046 $spool_inodes, and $log_inodes.
3048 32. The RFC2047 encoding function was originally intended for short strings
3049 such as real names; it was not keeping to the 75-character limit for
3050 encoded words that the RFC imposes. It now respects the limit, and
3051 generates multiple encoded words if necessary. To be on the safe side, I
3052 have increased the buffer size for the ${rfc2047: expansion operator from
3055 33. It is now permitted to omit both strings after an "if" condition; if the
3056 condition is true, the result is "true". As before, when the second string
3057 is omitted, a false condition yields an empty string. This makes it less
3058 cumbersome to write custom ACL and router conditions.
3060 34. Failure to deliver a bounce message always caused it to be frozen, even if
3061 there was an errors_to setting on the router. The errors_to setting is now
3064 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3065 canonical form (fully expanded) before being placed in
3066 $sender_host_address.
3068 36. The table in the code that translates DNS record types into text (T_A to
3069 "A" for instance) was missing entries for NS and CNAME. It is just possible
3070 that this could have caused confusion if both these types were looked up
3071 for the same domain, because the text type is used as part of Exim's
3072 per-process caching. But the chance of anyone hitting this buglet seems
3075 37. The dnsdb lookup has been extended in a number of ways.
3077 (1) There is a new type, "zns", which walks up the domain tree until it
3078 finds some nameserver records. It should be used with care.
3080 (2) There is a new type, "mxh", which is like "mx" except that it returns
3081 just the host names, not the priorities.
3083 (3) It is now possible to give a list of domains (or IP addresses) to be
3084 looked up. The behaviour when one of the lookups defers can be
3085 controlled by a keyword.
3087 (4) It is now possible to specify the separator character for use when
3088 multiple records are returned.
3090 38. The dnslists ACL condition has been extended: it is now possible to supply
3091 a list of IP addresses and/or domains to be looked up in a particular DNS
3094 39. Added log_selector=+queue_time_overall.
3096 40. When running the queue in the test harness, wait just a tad after forking a
3097 delivery process, to get repeatability of debugging output.
3099 41. Include certificate and key file names in error message when GnuTLS fails
3100 to set them up, because the GnuTLS error message doesn't include the name
3101 of the failing file when there is a problem reading it.
3103 42. Allow both -bf and -bF in the same test run.
3105 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3107 44. The "Exiscan patch" is now merged into the mainline Exim source.
3109 45. Sometimes the final signoff response after QUIT could fail to get
3110 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3111 < 0 before doing a fflush(). This bug looks as though it goes back to the
3112 introduction of TLS in release 3.20, but "sometimes" must have been rare
3113 because the tests only now provoked it.
3115 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3116 (this can affect the format of dates).
3118 47. exim_tidydb, when checking for the continued existence of a message for
3119 which it has found a message-specific retry record, was not finding
3120 messages that were in split spool directories. Consequently, it was
3121 deleting retry records that should have stayed in existence.
3123 48. Steve fixed some bugs in eximstats.
3125 49. The SPA authentication driver was not abandoning authentication and moving
3126 on to the next authenticator when an expansion was forced to fail,
3127 contradicting the general specification for all authenticators. Instead it
3128 was generating a temporary error. It now behaves as specified.
3130 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3131 (the order specifies the preference for clients). The order is now AES256,
3132 AES128, 3DES, ARCFOUR128.
3134 51. Small patch to Sieve code - explicitly set From: when generating an
3137 52. Exim crashed if a remote delivery caused a very long error message to be
3138 recorded - for instance if somebody sent an entire SpamAssassin report back
3139 as a large number of 550 error lines. This bug was coincidentally fixed by
3140 increasing the size of one of Exim's internal buffers (big_buffer) that
3141 happened as part of the Exiscan merge. However, to be on the safe side, I
3142 have made the code more robust (and fixed the comments that describe what
3145 53. Now that there can be additional text after "Completed" in log lines (if
3146 the queue_time_overall log selector is set), a one-byte patch to exigrep
3147 was needed to allow it to recognize "Completed" as not the last thing in
3150 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3151 patch that reportedly fixes this has been added. I am not expert enough to
3152 create a test for it. This is what the patch creator wrote:
3154 "I found a little strange behaviour of ldap code when working with
3155 Windows 2003 AD Domain, where users was placed in more than one
3156 Organization Units. When I tried to give exim partial DN, the exit code
3157 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3158 But simultaneously result of request was absolutely normal ldap result,
3159 so I produce this patch..."
3161 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3162 REFERENCE, so I have modified the code to exclude the patch when that macro
3165 55. Some experimental protocols are using DNS PTR records for new purposes. The
3166 keys for these records are domain names, not reversed IP addresses. The
3167 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3168 leaves it alone. Component reversal etc. now happens only for IP addresses.
3171 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3173 57. Double the size of the debug message buffer (to 2048) so that more of very
3174 long debug lines gets shown.
3176 58. The exicyclog utility now does better if the number of log files to keep
3177 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3179 59. Two changes related to the smtp_active_hostname option:
3181 (1) $smtp_active_hostname is now available as a variable.
3182 (2) The default for smtp_banner uses $smtp_active_hostname instead
3183 of $primary_hostname.
3185 60. The host_aton() function is supposed to be passed a string that is known
3186 to be a valid IP address. However, in the case of IPv6 addresses, it was
3187 not checking this. This is a hostage to fortune. Exim now panics and dies
3188 if the condition is not met. A case was found where this could be provoked
3189 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3190 components; fortuitously, this particular loophole had already been fixed
3191 by change 4.50/55 above.
3193 If there are any other similar loopholes, the new check in host_aton()
3194 itself should stop them being exploited. The report I received stated that
3195 data on the command line could provoke the exploit when Exim was running as
3196 exim, but did not say which command line option was involved. All I could
3197 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3198 running as the user.
3201 61. There was a buffer overflow vulnerability in the SPA authentication code
3202 (which came originally from the Samba project). I have added a test to the
3203 spa_base64_to_bits() function which I hope fixes it.
3206 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3207 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3209 63. The daemon start-up calls getloadavg() while still root for those OS that
3210 need the first call to be done as root, but it missed one case: when
3211 deliver_queue_load_max is set with deliver_drop_privilege. This is
3212 necessary for the benefit of the queue runner, because there is no re-exec
3213 when deliver_drop_privilege is set.
3215 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3216 This has been fixed.
3218 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3219 was in use, was not putting the data itself into the right store pool;
3220 consequently, it could be overwritten for a subsequent message in the same
3221 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3224 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3226 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3227 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3228 the function were treating the return as a boolean value, which happened to
3229 work because 0=false and not-0=true, but is not correct code.
3231 68. The host_aton() function was not handling scoped IPv6 addresses (those
3232 with, for example, "%eth0" on the end) correctly.
3234 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3235 negated items (that is, ~something) in unsigned ints. Some compilers
3236 apparently mutter when there is no cast.
3238 70. If an address verification called from an ACL failed, and did not produce a
3239 user-specific message (i.e. there was only a "system" message), nothing was
3240 put in $acl_verify_message. In this situation, it now puts the system
3243 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3244 session; change 4.31/43 added the unwanted input to the log line - except
3245 that it did not do this in the start of session case. It now does.
3247 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3248 This is wrong and can cause the other end to generate a synchronization
3249 error if it is another Exim or anything else that does the synchronization
3250 check. A QUIT command is no longer sent after a timeout.
3252 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3253 during host lookups.
3255 74. The defer_ok option of callout verification was not working if it was used
3256 when verifying addresses in header lines, that is, for this case:
3258 verify = header_sender/callout=defer_ok
3260 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3261 those file descriptors could be used for SMTP connections. If anything
3262 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3263 could be sent to the SMTP client, causing chaos. The daemon now opens
3264 stdin, stdout, and stderr to /dev/null when it puts itself into the
3267 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3268 log by default. The user can override this with suitable Perl magic.
3270 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3271 the log message when discard triggers, was not working for the DATA ACL or
3272 for the non-SMTP ACL.
3274 78. Error message wording change in sieve.c.
3276 79. If smtp_accept_max_per_host was set, the number of connections could be
3277 restricted to fewer than expected, because the daemon was trying to set up
3278 a new connection before checking whether the processes handling previous
3279 connections had finished. The check for completed processes is now done
3280 earlier. On busy systems, this bug wouldn't be noticed because something
3281 else would have woken the daemon, and it would have reaped the completed
3284 80. If a message was submitted locally by a user whose login name contained one
3285 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3286 It caused a spool format error. I have fixed the spool reading code. A
3287 related problem was that the "from" clause in the Received: line became
3288 illegal because of the space(s). It is now covered by ${quote_local_part.
3290 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3293 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3295 83. Patch from Sieve maintainer to fix unterminated string problem in
3296 "vacation" handling.
3298 84. Some minor changes to the Linux configuration files to help with other
3299 OS variants using glibc.
3301 85. One more patch for Sieve to update vacation handling to latest spec.
3304 ----------------------------------------------------
3305 See the note above about the 4.44 and 4.50 releases.
3306 ----------------------------------------------------
3312 1. Change 4.43/35 introduced a bug that caused file counts to be
3313 incorrectly computed when quota_filecount was set in an appendfile
3316 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3317 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3320 3. Give more explanation in the error message when the command for a transport
3321 filter fails to execute.
3323 4. There are several places where Exim runs a non-Exim command in a
3324 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3325 was being done only for the command run by the queryprogram router. It is
3326 now done for all such subprocesses. The other cases are: ${run, transport
3327 filters, and the commands run by the lmtp and pipe transports.
3329 5. Some older OS have a limit of 256 on the maximum number of file
3330 descriptors. Exim was using setrlimit() to set 1000 as a large value
3331 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3332 systems. I've change it so that if it can't get 1000, it tries for 256.
3334 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3335 was an oversight, and furthermore, ever since the addition of extra
3336 controls (e.g. 4.43/32), the checks on when to allow different forms of
3337 "control" were broken. There should now be diagnostics for all cases when a
3338 control that does not make sense is encountered.
3340 7. $recipients is now available in the predata ACL (oversight).
3342 8. Tidy the search cache before the fork to do a delivery from a message
3343 received from the command line. Otherwise the child will trigger a lookup
3344 failure and thereby defer the delivery if it tries to use (for example) a
3345 cached ldap connection that the parent has called unbind on.
3347 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3348 of $address_data from the recipient verification was clobbered by the
3349 sender verification.
3351 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3352 was its contents. (It was OK if the option was not defined at all.)
3354 11. A "Completed" log line is now written for messages that are removed from
3355 the spool by the -Mrm option.
3357 12. $host_address is now set to the target address during the checking of
3358 ignore_target_hosts.
3360 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3361 being passed; this would have caused $sender_host_name to have been used if
3362 matching the list had actually called for a host name (not very likely,
3363 since this list is usually IP addresses). A host name is now passed as
3366 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3367 code that specifies a non-restarting handler (typically sigaction() in
3368 modern systems) in an attempt to fix a rare and obscure crash bug.
3370 15. Narrowed the window for a race in the daemon that could cause it to ignore
3371 SIGCHLD signals. This is not a major problem, because they are used only to
3372 wake it up if nothing else does.
3374 16. A malformed maildirsize file could cause Exim to calculate negative values
3375 for the mailbox size or file count. Odd effects could occur as a result.
3376 The maildirsize information is now recalculated if the size or filecount
3379 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3380 support for a long time. Removed HAVE_SYS_VFS_H.
3382 18. Updated exipick to current release from John Jetmore.
3384 19. Allow an empty sender to be matched against a lookup in an address list.
3385 Previously the only cases considered were a regular expression, or an
3388 20. Exim went into a mad DNS lookup loop when doing a callout where the
3389 host was specified on the transport, if the DNS lookup yielded more than
3392 21. The RFC2047 encoding function was originally intended for short strings
3393 such as real names; it was not keeping to the 75-character limit for
3394 encoded words that the RFC imposes. It now respects the limit, and
3395 generates multiple encoded words if necessary. To be on the safe side, I
3396 have increased the buffer size for the ${rfc2047: expansion operator from
3399 22. Failure to deliver a bounce message always caused it to be frozen, even if
3400 there was an errors_to setting on the router. The errors_to setting is now
3403 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3404 canonical form (fully expanded) before being placed in
3405 $sender_host_address.
3407 24. Updated eximstats to version 1.33
3409 25. Include certificate and key file names in error message when GnuTLS fails
3410 to set them up, because the GnuTLS error message doesn't include the name
3411 of the failing file when there is a problem reading it.
3413 26. Expand error message when OpenSSL has problems setting up cert/key files.
3416 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3417 (this can affect the format of dates).
3419 28. exim_tidydb, when checking for the continued existence of a message for
3420 which it has found a message-specific retry record, was not finding
3421 messages that were in split spool directories. Consequently, it was
3422 deleting retry records that should have stayed in existence.
3424 29. eximstats updated to version 1.35
3425 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3426 1.35 - bugfix such that pie charts by volume are generated correctly
3428 30. The SPA authentication driver was not abandoning authentication and moving
3429 on to the next authenticator when an expansion was forced to fail,
3430 contradicting the general specification for all authenticators. Instead it
3431 was generating a temporary error. It now behaves as specified.
3433 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3434 (the order specifies the preference for clients). The order is now AES256,
3435 AES128, 3DES, ARCFOUR128.
3437 31. Small patch to Sieve code - explicitly set From: when generating an
3440 32. Exim crashed if a remote delivery caused a very long error message to be
3441 recorded - for instance if somebody sent an entire SpamAssassin report back
3442 as a large number of 550 error lines. This bug was coincidentally fixed by
3443 increasing the size of one of Exim's internal buffers (big_buffer) that
3444 happened as part of the Exiscan merge. However, to be on the safe side, I
3445 have made the code more robust (and fixed the comments that describe what
3448 33. Some experimental protocols are using DNS PTR records for new purposes. The
3449 keys for these records are domain names, not reversed IP addresses. The
3450 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3451 leaves it alone. Component reversal etc. now happens only for IP addresses.
3454 34. The host_aton() function is supposed to be passed a string that is known
3455 to be a valid IP address. However, in the case of IPv6 addresses, it was
3456 not checking this. This is a hostage to fortune. Exim now panics and dies
3457 if the condition is not met. A case was found where this could be provoked
3458 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3459 components; fortuitously, this particular loophole had already been fixed
3460 by change 4.50/55 or 4.44/33 above.
3462 If there are any other similar loopholes, the new check in host_aton()
3463 itself should stop them being exploited. The report I received stated that
3464 data on the command line could provoke the exploit when Exim was running as
3465 exim, but did not say which command line option was involved. All I could
3466 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3467 running as the user.
3470 35. There was a buffer overflow vulnerability in the SPA authentication code
3471 (which came originally from the Samba project). I have added a test to the
3472 spa_base64_to_bits() function which I hope fixes it.
3475 36. The daemon start-up calls getloadavg() while still root for those OS that
3476 need the first call to be done as root, but it missed one case: when
3477 deliver_queue_load_max is set with deliver_drop_privilege. This is
3478 necessary for the benefit of the queue runner, because there is no re-exec
3479 when deliver_drop_privilege is set.
3481 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3482 was in use, was not putting the data itself into the right store pool;
3483 consequently, it could be overwritten for a subsequent message in the same
3484 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3487 38. Sometimes the final signoff response after QUIT could fail to get
3488 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3489 < 0 before doing a fflush(). This bug looks as though it goes back to the
3490 introduction of TLS in release 3.20, but "sometimes" must have been rare
3491 because the tests only now provoked it.
3497 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3498 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3499 Now it returns an integer. A number of calls were still expecting a T/F
3500 return. Fortuitously, in all cases, the tests worked in OK situations,
3501 which is the norm. However, things would have gone wrong on any write
3502 failures on the smtp file descriptor. This function is used when sending
3503 messages over SMTP and also when doing verify callouts.
3505 2. When Exim is called to do synchronous delivery of a locally submitted
3506 message (the -odf or -odi options), it no longer closes stderr before doing
3509 3. Implemented the mua_wrapper option.
3511 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3513 5. Implemented the functions header_remove(), header_testname(),
3514 header_add_at_position(), and receive_remove_recipient(), and exported them
3517 6. If an ACL "warn" statement specified the addition of headers, Exim already
3518 inserted X-ACL-Warn: at the start if there was no header name. However, it
3519 was not making this test for the second and subsequent header lines if
3520 there were newlines in the string. This meant that an invalid header could
3521 be inserted if Exim was badly configured.
3523 7. Allow an ACL "warn" statement to add header lines at the start or after all
3524 the Received: headers, as well as at the end.
3526 8. Added the rcpt_4xx retry error code.
3528 9. Added postmaster_mailfrom=xxx to callout verification option.
3530 10. Added mailfrom=xxxx to the callout verification option, for verify=
3533 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3534 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3536 12. Inserted some casts to stop certain compilers warning when using pointer
3537 differences as field lengths or precisions in printf-type calls (mostly
3538 affecting debugging statements).
3540 13. Added optional readline() support for -be (dynamically loaded).
3542 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3543 same clock tick as a message's arrival, so that its received time was the
3544 same as the "first fail" time on the retry record, and that message
3545 remained on the queue past the ultimate address timeout, every queue runner
3546 would try a delivery (because it was past the ultimate address timeout) but
3547 after another failure, the ultimate address timeout, which should have then
3548 bounced the address, did not kick in. This was a "< instead of <=" error;
3549 in most cases the first failure would have been in the next clock tick
3550 after the received time, and all would be well.
3552 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3553 being recognized when the domain list was tested by the match_domain
3554 condition in an expansion string.
3556 16. Added the ${str2b64: operator.
3558 17. Exim was always calling setrlimit() to set a large limit for the number of
3559 processes, without checking whether the existing limit was already
3560 adequate. (It did check for the limit on file descriptors.) Furthermore,
3561 errors from getrlimit() and setrlimit() were being ignored. Now they are
3562 logged to the main and panic logs, but Exim does carry on, to try to do its
3563 job under whatever limits there are.
3565 18. Imported PCRE 5.0.
3567 19. Trivial typo in log message " temporarily refused connection" (the leading
3570 20. If the log selector return_path_on_delivery was set and an address was
3571 redirected to /dev/null, the delivery process crashed because it assumed
3572 that a return path would always be set for a "successful" delivery. In this
3573 case, the whole delivery is bypassed as an optimization, and therefore no
3576 21. Internal re-arrangement: the function for sending a challenge and reading
3577 a response while authentication was assuming a zero-terminated challenge
3578 string. It's now changed to take a pointer and a length, to allow for
3579 binary data in such strings.
3581 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3583 23. Exim was not respecting finduser_retries when seeking the login of the
3584 uid under which it was called; it was always trying 10 times. (The default
3585 setting of finduser_retries is zero.) Also, it was sleeping after the final
3586 failure, which is pointless.
3588 24. Implemented tls_on_connect_ports.
3590 25. Implemented acl_smtp_predata.
3592 26. If the domain in control=submission is set empty, Exim assumes that the
3593 authenticated id is a complete email address when it generates From: or
3594 Sender: header lines.
3596 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3597 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3598 chown and chgrp in /bin and hostname in /usr/bin.
3600 28. Exim was keeping the "process log" file open after each use, just as it
3601 does for the main log. This opens the possibility of it remaining open for
3602 long periods when the USR1 signal hits a daemon. Occasional processlog
3603 errors were reported, that could have been caused by this. Anyway, it seems
3604 much more sensible not to leave this file open at all, so that is what now
3607 29. The long-running daemon process does not normally write to the log once it
3608 has entered its main loop, and it closes the log before doing so. This is
3609 so that log files can straightforwardly be renamed and moved. However,
3610 there are a couple of unusual error situations where the daemon does write
3611 log entries, and I had neglected to close the log afterwards.
3613 30. The text of an SMTP error response that was received during a remote
3614 delivery was being truncated at 512 bytes. This is too short for some of
3615 the long messages that one sometimes sees. I've increased the limit to
3618 31. It is now possible to make retry rules that apply only when a message has a
3619 specific sender, in particular, an empty sender.
3621 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3622 it possible to be selective about when SMTP synchronization is enforced.
3624 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3626 32. Implemented hosts_connection_nolog.
3628 33. Added an ACL for QUIT.
3630 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3633 35. Added mailbox_size and mailbox_filecount to appendfile.
3635 36. Added control = no_multiline_responses to ACLs.
3637 37. There was a bug in the logic of the code that waits for the clock to tick
3638 in the case where the clock went backwards by a substantial amount such
3639 that the microsecond fraction of "now" was more than the microsecond
3640 fraction of "then" (but the whole seconds number was less).
3642 38. Added support for the libradius Radius client library this is found on
3643 FreeBSD (previously only the radiusclient library was supported).
3649 1. When certain lookups returned multiple values in the form name=value, the
3650 quoting of the values was not always being done properly. Specifically:
3651 (a) If the value started with a double quote, but contained no whitespace,
3653 (b) If the value contained whitespace other than a space character (i.e.
3654 tabs or newlines or carriage returns) it was not quoted.
3655 This fix has been applied to the mysql and pgsql lookups by writing a
3656 separate quoting function and calling it from the lookup code. The fix
3657 should probably also be applied to nisplus, ibase and oracle lookups, but
3658 since I cannot test any of those, I have not disturbed their existing code.
3660 2. A hit in the callout cache for a specific address caused a log line with no
3661 reason for rejecting RCPT. Now it says "Previous (cached) callout
3662 verification failure".
3664 3. There was an off-by-one bug in the queryprogram router. An over-long
3665 return line was truncated at 256 instead of 255 characters, thereby
3666 overflowing its buffer with the terminating zero. As well as fixing this, I
3667 have increased the buffer size to 1024 (and made a note to document this).
3669 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3670 when Exim is waiting for an SMTP response from a remote server, Exim
3671 restarts its select() call on the socket, thereby resetting its timeout.
3672 This is not a problem when such interrupts are rare. Somebody set up a cron
3673 job to run exiwhat every 2 minutes, which is less than the normal select()
3674 timeout (5 or 10 minutes). This meant that the select() timeout never
3675 kicked in because it was always reset. I have fixed this by comparing the
3676 time when an interrupt arrives with the time at the start of the first call
3677 to select(). If more time than the timeout has elapsed, the interrupt is
3678 treated as a timeout.
3680 5. Some internal re-factoring in preparation for the addition of Sieve
3681 extensions (by MH). In particular, the "personal" test is moved to a
3682 separate function, and given an option for scanning Cc: and Bcc: (which is
3683 not set for Exim filters).
3685 6. When Exim created an email address using the login of the caller as the
3686 local part (e.g. when creating a From: or Sender: header line), it was not
3687 quoting the local part when it contained special characters such as @.
3689 7. Installed new OpenBSD configuration files.
3691 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3692 try to make them clearer.
3694 9. Callout options, other than the timeout value, were being ignored when
3695 verifying sender addresses in header lines. For example, when using
3697 verify = header_sender/callout=no_cache
3699 the cache was (incorrectly) being used.
3701 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3702 only the Cygwin environment.
3704 11. When return_path_on_delivery was set as a log selector, if different remote
3705 addresses in the same message used different return paths and parallel
3706 remote delivery occurred, the wrong values would sometimes be logged.
3707 (Whenever a remote delivery process finished, the return path value from
3708 the most recently started remote delivery process was logged.)
3710 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3711 lines when AUTH and/or TLS are in use. This is the "received protocol"
3712 field. Exim used to use "asmtp" for authenticated SMTP, without any
3713 indication (in the protocol name) for TLS use. Now it follows the RFC and
3714 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3715 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3716 names appear in log lines as well as in Received: header lines.
3718 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3719 extensions, and comparison tests, and to fix some bugs.
3721 14. Changes to the "personal" filter test:
3723 (1) The test was buggy in that it was just doing the equivalent of
3724 "contains" tests on header lines. For example, if a user's address was
3725 anne@some.where, the "personal" test would incorrectly be true for
3727 To: susanne@some.where
3729 This test is now done by extracting each address from the header in turn,
3730 and checking the entire address. Other tests that are part of "personal"
3731 are now done using regular expressions (for example, to check local parts
3732 of addresses in From: header lines).
3734 (2) The list of non-personal local parts in From: addresses has been
3735 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3736 taken from the Sieve specification recommendations.
3738 (3) If the message contains any header line starting with "List-" it is
3739 treated as non-personal.
3741 (4) The test for "circular" in the Subject: header line has been removed
3742 because it now seems ill-conceived.
3744 15. Minor typos in src/EDITME comments corrected.
3746 16. Installed latest exipick from John Jetmore.
3748 17. If headers_add on a router specified a text string that was too long for
3749 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3750 of string_sprintf() is now avoided.
3752 18. $message_body_size was not set (it was always zero) when running the DATA
3753 ACL and the local_scan() function.
3755 19. For the "mail" command in an Exim filter, no default was being set for
3756 the once_repeat time, causing a random time value to be used if "once" was
3757 specified. (If the value happened to be <= 0, no repeat happened.) The
3758 default is now 0s, meaning "never repeat". The "vacation" command was OK
3759 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3760 (I found it when inspecting the code).
3762 20. There is now an overall timeout for performing a callout verification. It
3763 defaults to 4 times the callout timeout, which applies to individual SMTP
3764 commands during the callout. The overall timeout applies when there is more
3765 than one host that can be tried. The timeout is checked before trying the
3766 next host. This prevents very long delays if there are a large number of
3767 hosts and all are timing out (e.g. when the network connections are timing
3768 out). The value of the overall timeout can be changed by specifying an
3769 additional sub-option for "callout", called "maxwait". For example:
3771 verify = sender/callout=5s,maxwait=20s
3773 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3774 to the end before writing, but this should make it even safer).
3776 22. Exim was forgetting that it had advertised PIPELINING for the second and
3777 subsequent messages on an SMTP connection. It was also not resetting its
3778 memory on STARTTLS and an internal HELO.
3780 23. When Exim logs an SMTP synchronization error within a session, it now
3781 records whether PIPELINING has been advertised or not.
3783 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3784 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3785 rather than long int.
3787 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3789 26. Added the never_mail option to autoreply.
3795 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3796 crash if the getsockname() call failed; this can happen if a connection is
3797 closed very soon after it is established. The problem was simply in the
3798 order in which certain operations were done, causing Exim to try to write
3799 to the SMTP stream before it had set up the file descriptor. The bug has
3800 been fixed by making things happen in the correct order.
3806 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3807 before the connection was closed, thus losing the rejection response.
3809 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3810 some early Solaris releases, but causes trouble in current releases where
3811 socklen_t is defined.
3813 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3816 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3819 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3820 that releasing the top of it at the end releases what was used for sub-
3821 expansions (unless the block got too big). However, discard this block if
3822 the first thing is a variable or header, so that we can use its block when
3823 it is dynamic (useful for very large $message_headers, for example).
3825 6. Lookups now cache *every* query, not just the most recent. A new, separate
3826 store pool is used for this. It can be recovered when all lookup caches are
3827 flushed. Lookups now release memory at the end of their result strings.
3828 This has involved some general refactoring of the lookup sources.
3830 7. Some code has been added to the store_xxx() functions to reduce the amount
3831 of flapping under certain conditions.
3833 8. log_incoming_interface used to affect only the <= reception log lines. Now
3834 it causes the local interface and port to be added to several more SMTP log
3835 lines, for example "SMTP connection from", and rejection lines.
3837 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3839 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3841 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3842 could be overwritten at the end of the current message (or the start of a
3843 new message if it was set in a HELO ACL). The value is now preserved for
3844 the duration of the SMTP connection.
3846 12. If a transport had a headers_rewrite setting, and a matching header line
3847 contained an unqualified address, that address was qualified, even if it
3848 did not match any rewriting rules. The underlying bug was that the values
3849 of the flags that permit the existence of unqualified sender and recipient
3850 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3851 non-local messages, and by -bnq for local messages) were not being
3852 preserved with the message after it was received.
3854 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3855 "next input=" as part of the text comprising the host identity instead of
3856 the correct text. The code was using the same buffer for two different
3857 strings. However, depending on which order the printing function evaluated
3858 its arguments, the bug did not always show up. Under Linux, for example, my
3859 test suite worked just fine.
3861 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3862 doesn't work with some older versions of Perl. It has been changed to "my",
3863 which in any case is probably the better facility to use.
3865 15. A really picky compiler found some instances of statements for creating
3866 error messages that either had too many or two few arguments for the format
3869 16. The size of the buffer for calls to the DNS resolver has been increased
3870 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3871 for addresses that have a lot of PTR records. This alleviates a problem; it
3872 does not fully solve it.
3874 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3875 buffer now truncates the list and logs the incident, which is the same
3876 action as happens when Exim is looking up a host name and its aliases.
3877 Previously in this situation something unpredictable would happen;
3878 sometimes it was "internal error: store_reset failed".
3880 18. If a server dropped the connection unexpectedly when an Exim client was
3881 using GnuTLS and trying to read a response, the client delivery process
3882 crashed while trying to generate an error log message.
3884 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3885 string, for example:
3887 warn message = H1: something\nH2: something
3889 the text was added as a single header line from Exim's point of view
3890 though it ended up OK in the delivered message. However, searching for the
3891 second and subsequent header lines using $h_h2: did not work. This has been
3892 fixed. Similarly, if a system filter added multiple headers in this way,
3893 the routers could not see them.
3895 20. Expanded the error message when iplsearch is called with an invalid key to
3896 suggest using net-iplsearch in a host list.
3898 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3899 ACLs are no longer actually imposed (and a message to that effect is
3902 22. If a "gecos" field in a passwd entry contained escaped characters, in
3903 particular, if it contained a \" sequence, Exim got it wrong when building
3904 a From: or a Sender: header from that name. A second bug also caused
3905 incorrect handling when an unquoted " was present following a character
3906 that needed quoting.
3908 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3909 was not being matched caselessly.
3911 24. Arranged for all hyphens in the exim.8 source to be escaped with
3914 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3915 properly. Recipient callout cache records were still being keyed to include
3916 the sender, even when use_sender was set false. This led to far more
3917 callouts that were necessary. The sender is no longer included in the key
3918 when use_sender is false.
3920 26. Added "control = submission" modifier to ACLs.
3922 27. Added the ${base62d: operator to decode base 62 numbers.
3924 28. dnsdb lookups can now access SRV records.
3926 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3927 the configuration file.
3929 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3930 (-v) mode. This makes the output for a verbose queue run more intelligible.
3932 31. Added a use_postmaster feature to recipient callouts.
3934 32. Added the $body_zerocount variable, containing the number of binary zero
3935 bytes in the message body.
3937 33. The time of last modification of the "new" subdirectory is now used as the
3938 "mailbox time last read" when there is a quota error for a maildir
3941 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3943 35. Added +ignore_unknown as a special item in host lists.
3945 36. Code for decoding IPv6 addresses in host lists is now included, even if
3946 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3947 address was recognized as an IP address, but was then not correctly decoded
3948 into binary, causing unexpected and incorrect effects when compared with
3955 1. Very minor rewording of debugging text in manualroute to say "list of
3956 hosts" instead of "hostlist".
3958 2. If verify=header_syntax was set, and a header line with an unqualified
3959 address (no domain) and a large number of spaces between the end of the
3960 name and the colon was received, the reception process suffered a buffer
3961 overflow, and (when I tested it) crashed. This was caused by some obsolete
3962 code that should have been removed. The fix is to remove it!
3964 3. When running in the test harness, delay a bit after writing a bounce
3965 message to get a bit more predictability in the log output.
3967 4. Added a call to search_tidyup() just before forking a reception process. In
3968 theory, someone could use a lookup in the expansion of smtp_accept_max_
3969 per_host which, without the tidyup, could leave open a database connection.
3971 5. Added the variables $recipient_data and $sender_data which get set from a
3972 lookup success in an ACL "recipients" or "senders" condition, or a router
3973 "senders" option, similar to $domain_data and $local_part_data.
3975 6. Moved the writing of debug_print from before to after the "senders" test
3978 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3979 problems for message scanning, either using a data ACL, or using
3980 local_scan() because the Received: header was not generated till after they
3981 were called (in order to set the time as the time of reception completion).
3982 I have revised the way this works. The header is now generated after the
3983 body is received, but before the ACL or local_scan() are called. After they
3984 are run, the timestamp in the header is updated.
3990 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3991 before starting a queue runner without re-exec. This happened only when
3992 deliver_drop_privilege was set or when the Exim user was set to root. The
3993 effect of the bug was that timeouts during subsequent deliveries caused
3994 crashes instead of being properly handled. The handler is now left at its
3995 default (and expected) setting.
3997 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3998 message, again when deliver_drop_privilege is set or Exim is run as root.
3999 The bug described in (1) was not present in this case, but the tidying up
4000 of the other signals was missing. I have made the two cases consistent.
4002 3. The ignore_target_hosts setting on a manualroute router was being ignored
4003 for hosts that were looked up using the /MX notation.
4005 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4008 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4009 operated on the sender address. After changing the $sender_address to <>
4010 for the sender address verify, Exim was re-instated it as the original
4011 (before rewriting) address, but remembering that it had rewritten it, so it
4012 wasn't rewriting it again. This bug also had the effect of breaking the
4013 sender address verification caching when the sender address was rewritten.
4015 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4016 This has been changed so that if the ip literal address matches
4017 ignore_target_hosts, the router declines.
4019 7. Added expansion conditions match_domain, match_address, and match_local_
4020 part (NOT match_host).
4022 8. The placeholder for the Received: header didn't have a length field set.
4024 9. Added code to Exim itself and to exim_lock to test for a specific race
4025 condition that could lead to file corruption when using MBX delivery. The
4026 issue is with the lockfile that is created in /tmp. If this file is removed
4027 after a process has opened it but before that process has acquired a lock,
4028 there is the potential for a second process to recreate the file and also
4029 acquire a lock. This could lead to two Exim processes writing to the file
4030 at the same time. The added code performs the same test as UW imapd; it
4031 checks after acquiring the lock that its file descriptor still refers to
4032 the same named file.
4034 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4035 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4038 11. Added the smtp_active_hostname option. If used, this will typically be made
4039 to depend on the incoming interface address. Because $interface_address is
4040 not set up until the daemon has forked a reception process, error responses
4041 that can happen earlier (such as "too many connections") no longer contain
4044 12. If an expansion in a condition on a "warn" statement fails because a lookup
4045 defers, the "warn" statement is abandoned, and the next ACL statement is
4046 processed. Previously this caused the whole ACL to be aborted.
4048 13. Added the iplsearch lookup type.
4050 14. Added ident_timeout as a log selector.
4052 15. Added tls_certificate_verified as a log selector.
4054 16. Added a global option tls_require_ciphers (compare the smtp transport
4055 option of the same name). This controls incoming TLS connections.
4057 17. I finally figured out how to make tls_require_ciphers do a similar thing
4058 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4059 before starting the TLS session.
4061 18. Tabs are now shown as \t in -bP output.
4063 19. If the log selector return_path_on_delivery was set, Exim crashed when
4064 bouncing a message because it had too many Received: header lines.
4066 20. If two routers both had headers_remove settings, and the first one included
4067 a superfluous trailing colon, the final name in the first list and the
4068 first name in the second list were incorrectly joined into one item (with a
4069 colon in the middle).
4075 1. Added -C and -D options to the exinext utility, mainly to make it easier
4076 to include in the automated testing, but these could be helpful when
4077 multiple configurations are in use.
4079 2. The exinext utility was not formatting the output nicely when there was
4080 an alternate port involved in the retry record key, nor when there was a
4081 message id as well (for retries that were specific to a specific message
4082 and a specific host). It was also confused by IPv6 addresses, because of
4083 the additional colons they contain. I have fixed the IPv4 problem, and
4084 patched it up to do a reasonable job for IPv6.
4086 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4087 delivery, the log line now contains "pipelined" if PIPELINING was used.
4089 4. An SMTP transport process used to panic and die if the bind() call to set
4090 an explicit outgoing interface failed. This has been changed; it is now
4091 treated in the same way as a connect() failure.
4093 5. A reference to $sender_host_name in the part of a conditional expansion
4094 that was being skipped was still causing a DNS lookup. This no longer
4097 6. The def: expansion condition was not recognizing references to header lines
4098 that used bh_ and bheader_.
4100 7. Added the _cache feature to named lists.
4102 8. The code for checking quota_filecount in the appendfile transport was
4103 allowing one more file than it should have been.
4105 9. For compatibility with Sendmail, the command line option
4113 and sets the incoming protocol and host name (for trusted callers). The
4114 host name and its colon can be omitted when only the protocol is to be set.
4115 Note the Exim already has two private options, -pd and -ps, that refer to
4116 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4117 "s", but I don't think that's a major issue.
4119 10. A number of refactoring changes to the code, none of which should affect
4122 (a) The number of logging options was getting close to filling up the
4123 32-bit word that was used as a bit map. I have split them into two classes:
4124 those that are passed in the argument to log_write(), and those that are
4125 only ever tested independently outside of that function. These are now in
4126 separate 32-bit words, so there is plenty of room for expansion again.
4127 There is no change in the user interface or the logging behaviour.
4129 (b) When building, for example, log lines, the code previously used a
4130 macro that called string_cat() twice, in order to add two strings. This is
4131 not really sufficiently general. Furthermore, there was one instance where
4132 it was actually wrong because one of the argument was used twice, and in
4133 one call a function was used. (As it happened, calling the function twice
4134 did not affect the overall behaviour.) The macro has been replaced by a
4135 function that can join an arbitrary number of extra strings onto a growing
4138 (c) The code for expansion conditions now uses a table and a binary chop
4139 instead of a serial search (which was left over from when there were very
4140 few conditions). Also, it now recognizes conditions like "pam" even when
4141 the relevant support is not compiled in: a suitably worded error message is
4142 given if an attempt is made to use such a condition.
4144 11. Added ${time_interval:xxxxx}.
4146 12. A bug was causing one of the ddress fields not to be passed back correctly
4147 from remote delivery subprocesses. The field in question was not being
4148 subsequently used, so this caused to problems in practice.
4150 13. Added new log selectors queue_time and deliver_time.
4152 14. Might have fixed a bug in maildirsizefile handling that threw up
4153 "unexpected character" debug warnings, and recalculated the data
4154 unnecessarily. In any case, I expanded the warning message to give more
4157 15. Added the message "Restricted characters in address" to the statements in
4158 the default ACL that block characters like @ and % in local parts.
4160 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4161 Three changes have been made:
4163 (a) There was a serious bug; a negative response to MAIL caused the whole
4164 recipient domain to be cached as invalid, thereby blocking all messages
4165 to all local parts at the same domain, from all senders. This bug has
4166 been fixed. The domain is no longer cached after a negative response to
4167 MAIL if the sender used is not empty.
4169 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4172 (c) A new callout option, "use_sender" has been added for people who want
4173 the modified behaviour.
4179 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4182 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4183 indeed breaks things for older releases.
4185 3. Added additional logging to the case where there is a problem reading data
4186 from a filter that is running in a subprocess using a pipe, in order to
4187 try to track down a specific problem.
4189 4. Testing facility fudge: when running in the test harness and attempting
4190 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4191 getting "No route to host". Convert this to a timeout.
4193 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4196 6. Some OS don't have socklen_t but use size_t instead. This affects the
4197 fifth argument of getsockopt() amongst other things. This is now
4198 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4199 can be set for individual OS. I have set it for SunOS5, OSF1, and
4200 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4201 some earlier ones do not.
4203 7. Change 4.30/15 was not doing the test caselessly.
4205 8. The standard form for an IPv6 address literal was being rejected by address
4206 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4207 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4208 this, as well as the form without the "IPv6" on the front (but only when
4209 address literals are enabled, of course).
4211 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4213 10. Exim crashed if a message with an empty sender address specified by -f
4214 encountered a router with an errors_to setting. This could be provoked only
4215 by a command such as
4219 where an empty string was supplied; "<>" did not hit this bug.
4221 11. Installed PCRE release 4.5.
4223 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4224 remained set. It is now erased.
4226 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4227 times from message ids (which are base 36 rather than the normal 62).
4229 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4230 were being counted as actual protocol errors, and logged if the log
4231 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4232 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4233 and DATA following a set of rejected RCPTs do not count as protocol errors.
4234 In other words, Exim assumes they were pipelined, though this may not
4235 actually be the case. Of course, in all cases the client gets an
4236 appropriate error code.
4238 15. If a lookup fails in an ACL condition, a message about the failure may
4239 be available; it is used if testing the ACL cannot continue, because most
4240 such messages specify what the cause of the deferral is. However, some
4241 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4242 that caused an old message to be retained and used if a later statement
4243 caused a defer, replacing the real cause of the deferral.
4245 16. If an IP address had so many PTR records that the DNS lookup buffer
4246 was not large enough to hold them, Exim could crash while trying to process
4247 the truncated data. It now detects and logs this case.
4249 17. Further to 4.21/58, another change has been made: if (and only if) the
4250 first line of a message (the first header line) ends with CRLF, a bare LF
4251 in a subsequent header line has a space inserted after it, so as not to
4252 terminate the header.
4254 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4255 unnecessarily, used atoi() instead of strtol(), and didn't check the
4256 termination when getting file sizes from file names by regex.
4258 19. Completely re-implemented the support for maildirsize files, in the light
4259 of a number of problems with the previous contributed implementation
4260 (4.30/29). In particular:
4262 . If the quota is zero, the maildirsize file is maintained, but no quota is
4265 . If the maildir directory does not exist, it is created before any attempt
4266 to write a maildirsize file.
4268 . The quota value in the file is just a cache; if the quota is changed in
4269 the transport, the new value overrides.
4271 . A regular expression is available for excluding directories from the
4274 20. The autoreply transport checks the characters in options that define the
4275 message's headers; it allows continued headers, but it was checking with
4276 isspace() after an embedded newline instead of explicitly looking for a
4279 21. If all the "regular" hosts to which an address was routed had passed their
4280 expiry times, and had not reached their retry times, the address was
4281 bounced, even if fallback hosts were defined. Now Exim should go on to try
4284 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4285 equivalent code in the SMTP transport. Some hosts send humungous responses
4286 to HELO/EHLO, more than 1024 it seems.
4288 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4289 gives compiler warnings in some environments. I've now done it "properly",
4292 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4293 (because of problems with the built-in one) was declared to return uschar *
4294 instead of char *, causing compiler failure.
4296 25. Fixed a file descriptor leak when processing alias/forward files.
4298 26. Fixed a minor format string issue in dbfn.c.
4300 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4302 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4303 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4306 29. When a sender address is verified, it is cached, to save repeating the test
4307 when there is more than one recipient in a message. However, when the
4308 verification involves a callout, it is possible for different callout
4309 options to be set for different recipients. It is too complicated to keep
4310 track of this in the cache, so now Exim always runs a verification when a
4311 callout is required, relying on the callout cache for the optimization.
4312 The overhead is duplication of the address routing, but this should not be
4315 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4316 to be verified with callout=postmaster, and the main callout worked but the
4317 postmaster check failed, the verification correctly failed. However, if a
4318 subsequent RCPT command asked for sender verification *without* the
4319 postmaster check, incorrect caching caused this verification also to fail,
4322 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4323 it was not caching the DNS options (qualify_single, search_parents) that
4324 were used when the lookup failed. A subsequent lookup with different
4325 options therefore always gave the same answer, though there were cases
4326 where it should not have. (Example: a "domains = !$mx_any" option on a
4327 dnslookup router: the "domains" option is always processed without any
4328 widening, but the router might have qualify_single set.) Now Exim uses the
4329 cached value only when the same options are set.
4331 32. Added John Jetmore's "exipick" utility to the distribution.
4333 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4334 than a timeout (e.g. a certificate is required, and is not provided), an
4335 Exim server now closes the connection immediately. Previously it waited for
4336 the client to close - but if the client is SSL, it seems that they each
4337 wait for each other, leading to a delay before one of them times out.
4339 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4340 maintained 0.8.x compatibility because I don't think many are using it, and
4341 it is clearly obsolete.
4343 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4346 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4347 client certificate was expired. A simple patch fixes this, though I don't
4348 understand the full logic of why the verify callback is called multiple
4351 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4352 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4353 which causes problems with some clients (such as the Certicom SSL Plus
4354 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4355 disables the coutermeasure allowing Eudora to connect."
4357 38. Exim was not checking that a write() to a log file succeeded. This could
4358 lead to Bad Things if a log got too big, in particular if it hit a file
4359 size limit. Exim now panics and dies if it cannot write to a log file, just
4360 as it does if it cannot open a log file.
4362 39. Modified OS/Makefile-Linux so that it now contains
4364 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4366 The two -D definitions ensure that Exim is compiled with large file
4367 support, which makes it possible to handle log files that are bigger than
4370 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4371 instance) a domain was checked against a named list that involved a lookup,
4372 causing $domain_data to be set, then another domain was checked against the
4373 same list, then the first domain was re-checked, the value of $domain_data
4374 after the final check could be wrong. In particular, if the second check
4375 failed, it could be set empty. This bug probably also applied to
4378 41. The strip_trailing_dot option was not being applied to the address given
4379 with the -f command-line option.
4381 42. The code for reading a message's header from the spool was incrementing
4382 $received_count, but never initializing it. This meant that the value was
4383 incorrect (doubled) while delivering a message in the same process in which
4384 it was received. In the most common configuration of Exim, this never
4385 happens - a fresh exec is done - but it can happen when
4386 deliver_drop_privilege is set.
4388 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4389 it now includes up to 150 characters of the unexpected data in the log
4392 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4393 and building data strings. The size of both of these buffers was 10 000
4394 bytes - far larger than anybody would *ever* want, thought I. Needless to
4395 say, somebody hit the limit. I have increased the maximum line length to
4396 20 000 and the maximum data length of concatenated lines to 100 000. I have
4397 also fixed two bugs, because there was no checking on these buffers. Tsk,
4398 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4399 buffer is too small.
4401 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4402 lsearch lookups. Now it does.
4404 46. When parsing a route_list item in a manualroute router, a fixed-length
4405 buffer was used for the list of hosts. I made this 1024 bytes long,
4406 thinking that nobody would ever have a list of hosts that long. Wrong.
4407 Somebody had a whole pile of complicated expansion conditions, and the
4408 string was silently truncated, leading to an expansion error. It turns out
4409 that it is easier to change to an unlimited length (owing to other changes
4410 that have happened since this code was originally written) than to build
4411 structure for giving a limitation error. The length of the item that
4412 expands into the list of hosts is now unlimited.
4414 47. The lsearch lookup could not handle data where the length of text line was
4415 more than 4095 characters. Such lines were truncated, leading to shortened
4416 data being returned. It should now handle lines of any length.
4418 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4419 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4422 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4423 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4424 no longer happens in comment lines. A list of replacements is now placed
4425 at the head of all of the source files, except those whose only change is
4426 to replace PERL_COMMAND in the very first #! line.
4428 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4429 messages on the queue, with a bottom-up merge sort, using code contributed
4430 by Michael Haardt. This should make operations like -bp somewhat faster on
4431 large queues. It won't affect queue runners, except when queue_run_in_order
4434 51. Installed eximstats 1.31 in the distribution.
4436 52. Added support for SRV lookups to the dnslookup router.
4438 53. If an ACL referred to $message_body or $message_body_end, the value was not
4439 reset for any messages that followed in the same SMTP session.
4441 54. The store-handling optimization for building very long strings was not
4442 differentiating between the different store pools. I don't think this
4443 actually made any difference in practice, but I've tidied it.
4445 55. While running the routers to verify a sender address, $sender_address
4446 was still set to the sender address. This is wrong, because when routing to
4447 send a bounce to the sender, it would be empty. Therefore, I have changed
4448 it so that, while verifying a sender address, $sender_address is set to <>.
4449 (There is no change to what happens when verifying a recipient address.)
4451 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4452 target A or AAAA records (if not already returned) without resetting the
4453 qualify_single or search_parents options of the DNS resolver. These are
4454 inappropriate in this case because the targets of MX and SRV records must
4455 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4456 target that, when qualified, matched something in the local domain. These
4457 two options are now turned off when doing these lookups.
4459 57. It seems that at least some releases of Reiserfs (which does not have the
4460 concept of a fixed number of inodes) returns zero and not -1 for the
4461 number of available inodes. This interacted badly with check_spool_inodes,
4462 which assumed that -1 was the "no such thing" setting. What I have done is
4463 to check that the total number of inodes is greater than zero before doing
4464 the test of how many are available.
4466 58. When a "warn" ACL statement has a log_message modifier, the message is
4467 remembered, and not repeated. This is to avoid a lot of repetition when a
4468 message has many recipients that cause the same warning to be written.
4469 Howewer, Exim was preserving the list of already written lines for an
4470 entire SMTP session, which doesn't seem right. The memory is now reset if a
4471 new message is started.
4473 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4474 debugging output unless log_rewrite was also set.
4476 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4477 of (int)(handle) when we know that handle contains (void *)(-1).
4479 61. The Exim daemon panic-logs an error return when it closes the incoming
4480 connection. However "connection reset by peer" seems to be common, and
4481 isn't really an error worthy of noting specially, so that particular error
4484 62. When Exim is trying to find all the local interfaces, it used to panic and
4485 die if the ioctl to get the interface flags failed. However, it seems that
4486 on at least one OS (Solaris 9) it is possible to have an interface that is
4487 included in the list of interfaces, but for which you get a failure error
4488 for this call. This happens when the interface is not "plumbed" into a
4489 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4490 failure of the "get flags" call assumes that the interface is down.
4492 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4493 makes life easier for people who are doing arithmetic on fields extracted
4494 from dates, where you often get leading zeros that should not be
4495 interpreted as octal.
4497 64. Added qualify_domain to the redirect router, to override the global
4500 65. If a pathologically long header line contained very many addresses (the
4501 report of this problem mentioned 10 000) and each of them was rewritten,
4502 Exim could use up a very large amount of memory. (It kept on making new
4503 copies of the header line as it rewrote, and never released the old ones.)
4504 At the expense of a bit more processing, the header rewriting function has
4505 been changed so that it no longer eats memory in this way.
4507 66. The generation of the Received: header has been moved from the time that a
4508 message starts to be received, to the time that it finishes. The timestamp
4509 in the Received: header should now be very close to that of the <= log
4510 line. There are two side-effects of this change:
4512 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4513 logged header lines no longer include the local Received: line, because
4514 it has not yet been created. The same applies to a copy of the message
4515 that is returned to a non-SMTP sender when a message is rejected.
4517 (b) When a filter file is tested using -bf, no additional Received: header
4518 is added to the test message. After some thought, I decided that this
4521 This change does not affect the value of $received_for. It is still set
4522 after address rewriting, but before local_scan() is called.
4524 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4526 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4527 gave an unhelpful panic error message, and a defer error. I have managed to
4528 change this behaviour so that it now rejects any supplied certificate,
4529 which seems right, as the list of acceptable certificates is empty.
4531 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4532 gave an unhelpful defer error. I have not managed to make this reject any
4533 supplied certificates, but the error message it gives is "no certificate
4534 supplied", which is not helpful.
4536 70. exigrep's output now also includes lines that are not associated with any
4537 message, but which match the given pattern. Implemented by a patch from
4538 Martin Sluka, which also tidied up the Perl a bit.
4540 71. Recipient callout verification, like sender verification, was using <> in
4541 the MAIL FROM command. This isn't really the right thing, since the actual
4542 sender may affect whether the remote host accepts the recipient or not. I
4543 have changed it to use the actual sender in the callout; this means that
4544 the cache record is now keyed on a recipient/sender pair, not just the
4545 recipient address. There doesn't seem to be a real danger of callout loops,
4546 since a callout by the remote host to check the sender would use <>.
4547 [SEE ABOVE: changed after hitting problems.]
4549 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4550 temporary errors. However, in the case of such a code being given after
4551 the end of a data transmission (i.e. after ".") Exim was failing to write
4552 a retry record for the message. (Yes, there was some broken host that was
4553 actually sending 8xx at this point.)
4555 73. An unknown lookup type in a host list could cause Exim to panic-die when
4556 the list was checked. (An example that provoked this was putting <; in the
4557 middle of a list instead of at the start.) If this happened during a DATA
4558 ACL check, a -D file could be left lying around. This kind of configuration
4559 error no longer causes Exim to die; instead it causes a defer errror. The
4560 incident is still logged to the main and panic logs.
4562 74. Buglet left over from Exim 3 conversion. The message "too many messages
4563 in one connection" was written to the rejectlog but not the mainlog, except
4564 when address rewriting (yes!) was being logged.
4566 75. Added write_rejectlog option.
4568 76. When a system filter was run not as root (that is, when system_filter_user
4569 was set), the values of the $n variables were not being returned to the
4570 main process; thus, they were not subsequently available in the $sn
4573 77. Added +return_path_on_delivery log selector.
4575 78. A connection timeout was being treated differently from recipients deferred
4576 when testing hosts_max_try with a message that was older than the host's
4577 retry timeout. (The host should not be counted, thus allowing all hosts to
4578 be tried at least once before bouncing.) This may have been the cause of an
4579 occasionally reported bug whereby a message would remain on the queue
4580 longer than the retry timeout, but would be bounced if a delivery was
4581 forced. I say "may" because I never totally pinned down the problem;
4582 setting up timeout/retry tests is difficult. See also the next item.
4584 79. The ultimate address timeout was not being applied to errors that involved
4585 a combination of host plus message (for example, a timeout on a MAIL
4586 command). When an address resolved to a number of possible hosts, and they
4587 were not all tried for each delivery (e.g. because of hosts_max_try), a
4588 message could remain on the queue longer than the retry timeout.
4590 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4593 81. Fixed an obscure SMTP outgoing bug which required at least the following
4594 conditions: (a) there was another message waiting for the same server;
4595 (b) the server returned 5xx to all RCPT commands in the first message so
4596 that the message was not completed; (c) the server dropped the connection
4597 or gave a negative response to the RSET that Exim sends to abort the
4598 transaction. The observed case was a dropped connection after DATA that had
4599 been sent in pipelining mode. That is, the server had advertised PIPELINING
4600 but was not implementing it correctly. The effect of the bug was incorrect
4601 behaviour, such as trying another host, and this could lead to a crash.
4607 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4608 and daemon.c were passed as pointers to ints; they should have been
4609 pointers to socklen_t variables (which are typically unsigned ints).
4611 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4614 3. Fixed a really odd bug that affected only the testing scheme; patching a
4615 certain fixed string in the binary changed the value of another string that
4616 happened to be identical to the end of the original first string.
4618 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4619 name", it returns that address as the IP address. On some operating
4620 systems (e.g. Solaris), it also passes back the IP address string as the
4621 "host name". However, on others (e.g. Linux), it passes back an empty
4622 string. Exim wasn't checking for this, and was changing the host name to an
4623 empty string, assuming it had been canonicized.
4625 5. Although rare, it is permitted to have more than one PTR record for a given
4626 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4627 all the names associated with an address, because they do in Solaris.
4628 However, it seems that they do not in Linux for data that comes from the
4629 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4630 I found this out when I moved to a new Linux workstation and tried to run
4631 the Exim test suite.
4633 To get round this problem I have changed the code so that it now does its
4634 own call to the DNS to look up PTR records when searching for a host name.
4635 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4636 addresses that are only in /etc/hosts are still found.
4638 This behaviour is, however, controlled by an option called host_lookup_
4639 order, which defaults to "bydns:byaddr". If people want to use the other
4640 order, or indeed, just use one or the other means of lookup, they can
4641 specify it in this variable.
4643 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4644 some operating systems, this comes back from gethostbyaddr() as an empty
4645 string, and this is what Exim used to test for. However, it seems that in
4646 other systems, "." is yielded. Exim now tests for this case too.
4648 7. The values of check_spool_space and check_log_space are now held internally
4649 as a number of kilobytes instead of an absolute number of bytes. If a
4650 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4651 kilobyte. This means that much larger values can be stored.
4653 8. Exim monitor: an attempt to get the action menu when not actually pointing
4654 at a message produces an empty menu entitled "No message selected". This
4655 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4656 no entries in it ("Shell widget menu has zero width and/or height"). So I
4657 have added a single, blank menu entry in this case.
4659 9. Added ${quote_local_part.
4661 10. MIME decoding is now applied to the contents of Subject: header lines when
4664 11. Now that a reference to $sender_host_address automatically causes a reverse
4665 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4666 host lookup before query-style lookups in lists that might use this
4667 variable. This has therefore been abolished, and the "net-" prefix is no
4668 longer necessary for query-style lookups.
4670 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4671 have been a typo for LFLAGS, so it has been changed.
4673 13. The install script calls Exim with "-C /dev/null" in order to find the
4674 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4675 to be output. Howeve, since Exim outputs its version number before the
4676 error, it didn't break the script. It just looked ugly. I fixed this by
4677 always allowing "-C /dev/null" if the caller is root.
4679 14. Ignore overlarge ACL variable number when reading spool file - insurance
4680 against a later release with more variables having written the file.
4682 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4683 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4684 this, as well as the form without the "IPv6" on the front.
4686 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4687 OS/Makefile-Darwin file.
4689 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4690 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4692 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4693 variable. However, some OS use ints for inodes. I've added cast to long int
4694 to get rid of the compiler warning.
4696 19. I had forgotten to lock out "/../" in configuration file names when
4697 ALT_CONFIG_PREFIX was set.
4699 20. Routers used for verification do not need to specify transports. However,
4700 if such a router generated a host list, and callout was configured, Exim
4701 crashed, because it could not find a port number from the (non-existent)
4702 transport. It now assumes port 25 in this circumstance.
4704 21. Added the -t option to exigrep.
4706 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4707 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4708 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4709 them is set, LOOKUP_LSEARCH is forced.
4711 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4713 24. Added sender and host information to the "rejected by local_scan()" log
4714 line; previously there was no indication of these.
4716 25. Added .include_if_exists.
4718 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4719 Linux. It turns out that not all file systems support this. Apparently some
4720 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4721 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4722 not supported on the file descriptor, is now ignored when Exim is trying to
4723 sync a directory. This applies only to Linux.
4725 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4727 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4728 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4731 29. Added support for maildirsize files from supplied patch (modified a bit).
4733 30. The use of :fail: followed by an empty string could lead Exim to respond to
4734 sender verification failures with (e.g.):
4736 550 Verification failed for <xxx>
4737 550 Sender verify failed
4739 where the first response line was missing the '-' that indicates it is not
4740 the final line of the response.
4742 31. The loop for finding the name of the user that called Exim had a hardwired
4743 limit of 10; it now uses the value of finduser_retries, which is used for
4744 all other user lookups.
4746 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4749 33. Exim was neglecting to zero errno before one call of strtol() when
4750 expanding a string and expecting an integer value. On some systems this
4751 resulted in spurious "integer overflow" errors. Also, it was casting the
4752 result into an int without checking.
4754 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4755 did not work. The code looks as if it has *never* worked, though it appears
4756 to have been documented since at least releast 1.62. I have made it work.
4758 35. The "timeout_DNS" error in retry rules, also documented since at least
4759 1.62, also never worked. As it isn't clear exactly what this means, and
4760 clearly it isn't a major issue, I have abolished the feature by treating it
4761 as "timeout", and writing a warning to the main and panic logs.
4763 36. The display of retry rules for -brt wasn't always showing the error code
4766 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4767 timeout_connect_A, timeout_connect_MX.
4769 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4770 to the empty sender.
4772 39. The daemon was not analysing the content of -oX till after it had closed
4773 stderr and disconnected from the controlling terminal. This meant that any
4774 syntax errors were only noted on the panic log, and the return code from
4775 the command was 0. By re-arranging the code a little, I've made the
4776 decoding happen first, so such errors now appear on stderr, and the return
4777 code is 1. However, the actual setting up of the sockets still happens in
4778 the disconnected process, so errors there are still only recorded on the
4781 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4782 connections (as happens on some IP stacks) was logged at start up time as
4783 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4784 it from "IPv6 and IPv4", which means that two separate sockets are being
4787 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4788 says whether AF_INET or AF_INET6 was passed as an argument.
4790 42. Exiwhat output was messed up when time zones were included in log
4793 43. Exiwhat now gives more information about the daemon's listening ports,
4794 and whether -tls-on-connect was used.
4796 44. The "port" option of the smtp transport is now expanded.
4798 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4799 silently ignored. Now an error message is written to the main and panic
4802 46. There's a new ACL modifier called "logwrite" which writes to a log file
4803 as soon as it is encountered.
4805 47. Added $local_user_uid and $local_user_gid at routing time.
4807 48. Exim crashed when trying to verify a sender address that was being
4810 49. Exim was recognizing only a space character after ".include". It now also
4811 recognizes a tab character.
4813 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4814 extracting the relevant information from the specification. The man page no
4815 longer contains scrambled data for the -d option, and I've added a section
4816 at the front about calling Exim under different names.
4818 51. Added "extra_headers" argument to the "mail" command in filter files.
4820 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4823 53. Installed eximstats 1.29.
4825 54. Added transport_filter_timeout as a generic transport option.
4827 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4828 Cc: header lines. This was required by RFC 822, but it not required by RFC
4831 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4832 incoming messages that did not have them. Now it does so only if the
4833 message originates locally, that is, if there is no associated remote host
4834 address. When Resent- header lines are present, this applies to the Resent-
4835 lines rather than the non-Resent- lines.
4837 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4838 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4840 58. Messages for configuration errors now include the name of the main
4841 configuration file - useful now that there may be more than one file in a
4842 list (.included file names were always shown).
4844 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4845 for those rare installations that do not start the daemon as root or run it
4846 setuid root. I've cut out the call to initgroups() if the daemon is not
4849 60. The Exim user and group can now be bound into the binary as text strings
4850 that are looked up at the start of Exim's processing.
4852 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4854 62. Added $mailstore_basename variable.
4856 63. Installed patch to sieve.c from Michael Haardt.
4858 64. When Exim failed to open the panic log after failing to open the main log,
4859 the original message it was trying to log was written to stderr and debug
4860 output, but if they were not available (the usual case in production), it
4861 was lost. Now it is written to syslog before the two lines that record the
4862 failures to open the logs.
4864 65. Users' Exim filters run in subprocesses under the user's uid. It is
4865 possible for a "deliver" command or an alias in a "personal" command to
4866 provoke an address rewrite. If logging of address rewriting is configured,
4867 this fails because the process is not running as root or exim. There may be
4868 a better way of dealing with this, but for the moment (because 4.30 needs
4869 to be released), I have disabled address rewrite logging when running a
4870 filter in a non-root, non-exim process.
4876 1. The buildconfig auxiliary program wasn't quoting the value set for
4877 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4878 not defined. This bug was masked in 4.22 by the effect that was fixed in
4881 2. Some messages that were rejected after a message id was allocated were
4882 shown as "incomplete" by exigrep. It no longer does this for messages that
4883 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4885 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4886 have allow_domain_literals set, the ID did not get logged in the <= line.
4887 Domain literals are now always recognized in Message-ID: header lines.
4889 4. The first argument for a ${extract expansion item is the key name or field
4890 number. Leading and trailing spaces in this item were not being ignored,
4891 causing some misleading effects.
4893 5. When deliver_drop_privilege was set, single queue runner processes started
4894 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4895 same command in the process it spins off) were not dropping privilege.
4897 6. When the daemon running as "exim" started a queue runner, it always
4898 re-executed Exim in the spun-off process. This is a waste of effort when
4899 deliver_drop_privilege is set. The new process now just calls the
4900 queue-runner function directly.
4906 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4909 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4910 ignored. Though the use of -oP was forcing the writing of a pid file, it
4911 was always written to the default place.
4913 3. If the message "no IP address found for host xxxx" is generated during
4914 incoming verification, it is now followed by identification of the incoming
4915 connection (so you can more easily find what provoked it).
4917 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4919 5. Added some features to "harden" Exim a bit more against certain attacks:
4921 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4922 be put in Local/Makefile. This is like the never_users runtime option,
4923 but it cannot be overridden. The default setting is "root".
4925 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4926 prefix string with which any file named in a -C command line option
4929 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4930 is retained for -C and -D only if the caller of Exim is root. Without
4931 it, the exim user may also use -C and -D and retain privilege.
4933 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4934 command line option is disabled.
4936 6. Macro names set by the -D option must start with an upper case letter, just
4937 like macro names defined in the configuration file.
4939 7. Added "dereference=" facility to LDAP.
4941 8. Two instances of the typo "uknown" in the source files are fixed.
4943 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4944 the Configure-Makefile script screwed up while processing it.
4946 10. Incorporated PCRE 4.4.
4948 11. The SMTP synchronization check was not operating right at the start of an
4949 SMTP session. For example, it could not catch a HELO sent before the client
4950 waited for the greeting. There is now a check for outstanding input at the
4951 point when the greeting is written. Because of the duplex, asynchronous
4952 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4953 way, but not yet received, when the check is performed.
4955 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4956 on TCP/IP sockets, because this apparently causes some broken clients to
4959 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4960 unchanged) from the Cygwin maintainer.
4962 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4963 was supported instead of testing for mbx. Effectively a typo.
4965 15. The spa authenticator server code was not checking that the input it
4966 received was valid base64.
4968 16. The debug output line for the "set" modifier in ACLs was not showing the
4969 name of the variable that was being set.
4971 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4973 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4974 lookup on its own. Something else was needed to trigger the lookup. For
4975 example, a match in host_lookup or the need for a host name in a host list.
4976 Now, if $sender_host_name is referenced and the host name has not yet been
4977 looked up, a lookup is performed. If the lookup fails, the variable remains
4978 empty, and $host_lookup_failed is set to "1".
4980 19. Added "eqi" as a case-independent comparison operator.
4982 20. The saslauthd authentication condition could segfault if neither service
4983 nor realm was specified.
4985 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4986 error message that was logged was misleading, and incoming SMTP
4987 connections were dropped. The message is now more accurate, and temporary
4988 errors are given to SMTP connections.
4990 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4991 (except RSET) with a 421 error, until QUIT is received. However, it was
4992 failing to send a response to QUIT.
4994 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4995 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4996 ACL is now run after the helo_try_verify_hosts code.
4998 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4999 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5000 case-independent, so other case variants are also recognized). Apparently
5001 some systems use these upper case variants.
5003 25. If more than two messages were waiting for the same host, and a transport
5004 filter was specified for the transport, Exim sent two messages over the
5005 same TCP/IP connection, and then failed with "socket operation on non-
5006 socket" when it tried to send the third.
5008 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5010 27. The extern definition of crypt16() in expand.c was not being excluded when
5011 the OS had its own crypt16() function.
5013 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5014 as a preferred synonym for return_size_limit, both as an option and as an
5017 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5019 30. Changed the default configuration ACL to relax the local part checking rule
5020 for addresses that are not in any local domains. For these addresses,
5021 slashes and pipe symbols are allowed within local parts, but the sequence
5022 /../ is explicitly forbidden.
5024 31. SPA server authentication was not clearing the challenge buffer before
5027 32. log_message in a "warn" ACL statement was writing to the reject log as
5028 well as to the main log, which contradicts the documentation and doesn't
5029 seem right (because no rejection is happening). So I have stopped it.
5031 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5032 However, I am unable to do any testing of this.
5034 34. Fixed an infelicity in the appendfile transport. When checking directories
5035 for a mailbox, to see if any needed to be created, it was accidentally
5036 using path names with one or more superfluous leading slashes; tracing
5037 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5039 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5040 contents are added to the log line that is written for every discarded
5041 recipient. (Previously a log_message setting was ignored.)
5043 36. The ${quote: operator now quotes the string if it is empty.
5045 37. The install script runs exim in order to find its version number. If for
5046 some reason other than non-existence or emptiness, which it checks, it
5047 could not run './exim', it was installing it with an empty version number,
5048 i.e. as "exim-". This error state is now caught, and the installation is
5051 38. An argument was missing from the function that creates an error message
5052 when Exim fails to connect to the socket for saslauthd authentication.
5053 This could cause Exim to crash, or give a corrupted message.
5055 39. Added isip, isip4, and isip6 to ${if conditions.
5057 40. The ACL variables $acl_xx are now saved with the message, and can be
5058 accessed later in routers, transports, and filters.
5060 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5061 strings in the file are not string-expanded.
5063 42. If a MAIL command specified a SIZE value that was too large to fit into an
5064 int variable, the check against message_size_limit failed. Such values are
5065 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5066 day this will have to be increased, but I don't think I want to be around
5067 when emails are that large.
5074 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5075 iconv() is not standard in FreeBSD.
5077 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5078 IPv6 enabled. The observed symptom was a segmentation fault on return from
5079 the function os_common_find_running_interfaces() in src/os.c.
5081 3. In the check_special_case() function in daemon.c I had used "errno" as an
5082 argument name, which causes warnings on some systems. This was basically a
5083 typo, since it was named "eno" in the comments!
5085 4. The code that waits for the clock to tick (at a resolution of some fraction
5086 of a second) so as to ensure message-id uniqueness was always waiting for
5087 at least one whole tick, when it could have waited for less. [This is
5088 almost certainly not relevant at current processor speeds, where it is
5089 unlikely to ever wait at all. But we try to future-proof.]
5091 5. The function that sleeps for a time interval that includes fractions of a
5092 second contained a race. It did not block SIGALRM between setting the
5093 timer, and suspending (a couple of lines later). If the interval was short
5094 and the sigsuspend() was delayed until after it had expired, the suspension
5095 never ended. On busy systems this could lead to processes getting stuck for
5098 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5099 process, before it forks any delivery processes. The open lookup caching
5100 mechanism meant that the open file or database connection was passed into
5101 the delivery process. The problem was that delivery processes always tidy
5102 up cached lookup data. This could cause a problem for the next delivery
5103 process started by the queue runner, because the external queue runner
5104 process does not know about the closure. So the next delivery process
5105 still has data in the lookup cache. In the case of a file lookup, there was
5106 no problem because closing a file descriptor in a subprocess doesn't affect
5107 the parent. However, if the lookup was caching a connection to a database,
5108 the connection was closed, and the second delivery process was likely to
5109 see errors such as "PGSQL: query failed: server closed the connection
5110 unexpectedly". The problem has been fixed by closing all cached lookups
5111 in a queue runner before running a delivery process.
5113 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5114 seem to have the "const" qualifier which it has on other OS. I've
5117 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5118 *of the same type* (client or server) with the same public name that an
5119 error should be diagnosed.
5121 9. When Exim looked up a host name for an IP address, but failed to find the
5122 original IP address when looking up the host name (a safety check), it
5123 output the message "<ip address> does not match any IP for NULL", which was
5124 confusing, to say the least. The bug was that the host name should have
5125 appeared instead of "NULL".
5127 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5128 user that is built into the binary, and the -C or -D options is used, root
5129 privilege is dropped before the configuration file is read. In addition,
5130 logging is switched to stderr instead of the normal log files. If the
5131 configuration then re-defines the Exim user, the unprivileged environment
5132 is probably not what is expected, so Exim logs a panic warning message (but
5135 However, if deliver_drop_privilege is set, the unprivileged state may well
5136 be exactly what is intended, so the warning has been cut out in that case,
5137 and Exim is allowed to try to write to its normal log files.
5143 1. smtp_return_error_details was not giving details for temporary sender
5144 or receiver verification errors.
5146 2. Diagnose a configuration error if two authenticators have the same public
5149 3. Exim used not to create the message log file for a message until the first
5150 delivery attempt. This could be confusing when incoming messages were held
5151 for policy or load reasons. The message log file is now created at the time
5152 the message is received, and an initial "Received" line is written to it.
5154 4. The automatically generated man page for command line options had a minor
5155 bug that caused no ill effects; however, a more serious problem was that
5156 the procedure for building the man page automatically didn't always
5157 operate. Consequently, release 4.20 contains an out-of-date version. This
5158 shouldn't happen again.
5160 5. When building Exim with embedded Perl support, the script that builds the
5161 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5162 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5164 6. The freeze_tell option was not being used for messages that were frozen on
5165 arrival, either by an ACL or by local_scan().
5167 7. Added the smtp_incomplete_transaction log selector.
5169 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5170 was accepting AUTH without a new EHLO.
5172 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5173 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5176 10. Logging of TCP/IP connections (when configured) now happens in the main
5177 daemon process instead of the child process, so that the TCP/IP connection
5178 count is more accurate (but it can never be perfect).
5180 11. The use of "drop" in a nested ACL was not being handled correctly in the
5181 outer ACL. Now, if condition failure induced by the nested "drop" causes
5182 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5183 or "require"), the connection is dropped.
5185 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5186 that yield "discard" can now be used with an "accept" or a "discard" verb,
5187 but an error is generated for any others (because I can't see a useful way
5188 to define what should happen).
5190 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5191 are now processed in the same way as lines in the Exim configuration file.
5192 In particular, continuation lines are supported.
5194 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5196 15. Added -ti meaning -t -i.
5198 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5199 domains, and warn by logging if others are found.
5201 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5202 returned from GIFCONF when Exim is trying to find the list of interfaces on
5203 a host. The code in os.c has been modified to copy each ifreq to an aligned
5204 structure in all cases.
5206 Also, in some cases, the returned ifreq's were being copied to a 'struct
5207 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5208 means the last couple of bytes of an IPv6 address could be chopped if the
5209 ifreq contained only a normal sockaddr (14 bytes storage).
5211 18. Named domain lists were not supported in the hosts_treat_as_local option.
5212 An entry such as +xxxx was not recognized, and was treated as a literal
5215 19. Ensure that header lines added by a DATA ACL are included in the reject log
5216 if the ACL subsequently rejects the message.
5218 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5219 MD5 (which is deprecated).
5221 21. When testing a filter file using -bf, Exim was writing a message when it
5222 took the sender from a "From " line in the message, but it was not doing so
5223 when it took $return_path from a Return-Path: header line. It now does.
5225 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5226 with a valid header line field name (a series of printing characters
5227 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5229 23. Changed "disc" in the source to "disk" to conform to the documentation and
5230 the book and for uniformity.
5232 24. Ignore Sendmail's -Ooption=value command line item.
5234 25. When execve() failed while trying to run a command in a pipe transport,
5235 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5236 could be confused with a return value of 69 from the command itself. This
5237 has been changed to 127, the value the shell returns if it is asked to run
5238 a non-existent command. The wording for the related log line suggests a
5239 non-existent command as the problem.
5241 26. If received_header_text expands to an empty string, do not add a Received:
5242 header line to the message. (Well, it adds a token one on the spool, but
5243 marks it "old" so that it doesn't get used or transmitted.)
5245 27. Installed eximstats 1.28 (addition of -nt option).
5247 28. There was no check for failure on the call to getsockname() in the daemon
5248 code. This can fail if there is a shortage of resources on the system, with
5249 ENOMEM, for example. A temporary error is now given on failure.
5251 29. Contrary to the C standard, it seems that in some environments, the
5252 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5253 program. Exim now does this explicitly; it affects the formatting of
5254 timestamps using strftime().
5256 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5257 complaints. I've now initialized all the variables, to avoid this.
5259 32. Header lines added by a system filter were not being "seen" during
5260 transport-time rewrites.
5262 33. The info_callback() function passed to OpenSSL is set up with type void
5263 (*)(SSL *, int, int), as described somewhere. However, when calling the
5264 function (actually a macro) that sets it up, the type void(*)() is
5265 expected. I've put in a cast to prevent warnings from picky compilers.
5267 34. If a DNS black list lookup found a CNAME record, but there were no A
5268 records associated with the domain it pointed at, Exim crashed.
5270 35. If a DNS black list lookup returned more than one A record, Exim ignored
5271 all but the first. It now scans all returned addresses if a particular IP
5272 value is being sought. In this situation, the contents of the
5273 $dnslist_value variable are a list of all the addresses, separated by a
5276 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5277 to accept a host name and all its aliases if the forward lookup for any of
5278 them yielded the IP address of the incoming connection. Now it accepts only
5279 those names whose forward lookup yields the correct IP address. Any other
5280 names are discarded. This closes a loophole whereby a rogue DNS
5281 administrator could create reverse DNS records to break through a
5282 wildcarded host restriction in an ACL.
5284 37. If a user filter or a system filter that ran in a subprocess used any of
5285 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5286 the wrong values were passed to the pipe command ($thisaddress had the
5287 value of $0, $0 had the value of $1, etc). This bug was introduced by
5288 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5290 38. Improved the line breaking for long SMTP error messages from ACLs.
5291 Previously, if there was no break point between 40 and 75 characters, Exim
5292 left the rest of the message alone. Two changes have been made: (a) I've
5293 reduced the minimum length to 35 characters; (b) if it can't find a break
5294 point between 35 and 75 characters, it looks ahead and uses the first one
5295 that it finds. This may give the occasional overlong line, but at least the
5296 remaining text gets split now.
5298 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5299 file descriptors might be low, and that setting 1000 would always raise it.
5300 It turns out that in some environments, the limit is already over 1000 and
5301 that lowering it causes trouble. So now Exim takes care not to decrease it.
5303 40. When delivering a message, the value of $return_path is set to $sender_
5304 address at the start of routing (routers may change the value). By an
5305 oversight, this default was not being set up when an address was tested by
5306 -bt or -bv, which affected the outcome if any router or filter referred to
5309 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5310 log only when "message" or "log_message" are set. However, if one of the
5311 conditions was an address verification, or a call to a nested ACL, the
5312 messages generated by the underlying test were being passed through. This
5313 no longer happens. The underlying message is available in $acl_verify_
5314 message for both "message" and "log_message" expansions, so it can be
5315 passed through if needed.
5317 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5318 new expansion $bh_ to give the encoded byte string without charset
5319 translation. Translation happens only if iconv() is available; HAVE_ICONV
5320 indicates this at build time. HEADERS_CHARSET gives the charset to
5321 translate to; headers_charset can change it in the configuration, and
5322 "headers charset" can change it in an individual filter file.
5324 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5325 that creates RFC 2047 encoded "words" labels them as that charset instead
5326 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5327 expansion operator; (ii) when Exim creates a From: line for a local
5328 message; (iii) when a header line is rewritten to include a "phrase" part.
5330 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5331 buggy, causing it to skip the first lines of messages whose message ID
5332 ended in 'D'. This would not have bitten before Exim release 4.14, because
5333 message IDs were unlikely to end in 'D' before then. The effect was to have
5334 incorrect size information for certain domains.
5336 45. #include "config.h" was missing at the start of the crypt16.c module. This
5337 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5340 46. If there was a timeout during a "random" callout check, Exim treated it as
5341 a failure of the random address, and carried on sending RSET and the real
5342 address. If the delay was just some slowness somewhere, the response to the
5343 original RCPT would be taken as a response to RSET and so on, causing
5344 mayhem of various kinds.
5346 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5347 when I implemented it. It didn't allow for the fact that some option values
5348 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5349 the right test for positive values.
5351 48. Domain names in DNS records are case-independent. Exim always looks them up
5352 in lower case. Some resolvers return domain names in exactly the case they
5353 appear in the zone file, that is, they may contain uppercase letters. Not
5354 all resolvers do this - some return always lower case. Exim was treating a
5355 change of case by a resolver as a change of domain, similar to a widening
5356 of a domain abbreviation. This triggered its re-routing code and so it was
5357 trying to route what was effectively the same domain again. This normally
5358 caused routing to fail (because the router wouldn't handle the domain
5359 twice). Now Exim checks for this case specially, and just changes the
5360 casing of the domain that it ultimately uses when it transmits the message
5363 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5366 50. If a filter generated a file delivery with a non-absolute name (possible if
5367 no home directory exists for the router), the forbid_file option was not
5370 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5371 the existing equality matching.
5373 52. Exim was using ints instead of ino_t variables in some places where it was
5374 dealing with inode numbers.
5376 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5377 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5378 called TMPDIR, and if it finds it is different, it changes its value.
5380 54. The smtp_printf() function is now made available to local_scan() so
5381 additional output lines can be written before returning. There is also an
5382 smtp_fflush() function to enable the detection of a dropped connection.
5383 The variables smtp_input and smtp_batched_input are exported to
5386 55. Changed the default runtime configuration: the message "Unknown user"
5387 has been removed from the ACL, and instead placed on the localuser router,
5388 using the cannot_route_message feature. This means that any verification
5389 failures that generate their own messages won't get overridden. Similarly,
5390 the "Unrouteable address" message that was in the ACL for unverifiable
5391 relay addresses has also been removed.
5393 56. Added hosts_avoid_esmtp to the smtp transport.
5395 57. The exicyclog script was not checking for the esoteric option
5396 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5397 will work only if exicyclog is run under the appropriate euid.
5399 58. Following a discussion on the list, the rules by which Exim recognises line
5400 endings on incoming messages have been changed. The -dropcr and drop_cr
5401 options are now no-ops, retained only for backwards compatibility. The
5402 following line terminators are recognized: LF CRLF CR. However, special
5403 processing applies to CR:
5405 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5406 nor a local message in the state where . is a terminator.
5408 (ii) If a bare CR is encountered in a header line, an extra space is added
5409 after the line terminator so as not to end the header. The reasoning
5410 behind this is that bare CRs in header lines are most likely either
5411 to be mistakes, or people trying to play silly games.
5413 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5414 was being incorrectly given as 18 bytes larger than it should have been.
5415 This is a VOB (very old bug).
5417 60. This may never have affected anything current, but just in case it has:
5418 When the local host is found other than at the start of a list of hosts,
5419 the local host, those with the same MX, and any that follow, are discarded.
5420 When the list in question was part of a longer list of hosts, the following
5421 hosts (not currently being processed) were also being discarded. This no
5422 longer happens. I'm not sure if this situation could ever has previously
5425 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5428 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5429 header. This is something that is recommended in a new Internet Draft, and
5430 is something that is documented as being done by Sendmail. There are two
5431 possible values. For messages generated by the autoreply transport, Exim
5434 Auto-Submitted: auto-replied
5436 whereas for all other generated messages (e.g. bounces) it adds
5438 Auto-Submitted: auto-generated
5440 63. The "personal" condition in filters now includes a test for the
5441 Auto-Submitted: header. If it contains the string "auto-" the message it
5442 not considered personal.
5444 64. Added rcpt_include_affixes as a generic transport option.
5446 65. Added queue_only_override (default true).
5448 66. Added the syslog_duplication option.
5450 67. If what should have been the first header line of a message consisted of
5451 a space followed by a colon, Exim was mis-interpreting it as a header line.
5452 It isn't of course - it is syntactically invalid and should therefore be
5453 treated as the start of the message body. The misbehaviour could have
5454 caused a number of strange effects, including loss of data in subsequent
5455 header lines, and spool format errors.
5457 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5458 client host had authenticated. This control can now be exercised by an ACL
5459 for more flexibility.
5461 69. By default, callouts do not happen when testing with -bh. There is now a
5462 variant, -bhc, which does actually run the callout code, including
5463 consulting and updating the callout cache.
5465 70. Added support for saslauthd authentication, courtesy of Alexander
5468 71. If statvfs() failed on the spool or log directories while checking their
5469 size for availability, Exim confusingly gave the error "space shortage".
5470 Furthermore, in debugging mode it crashed with a floating point exception.
5471 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5472 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5473 serious problem, Exim now writes to the main and panic logs when this
5474 happens, with details of the failure. It then refuses to accept the
5475 incoming message, giving the message "spool directory problem" or "log
5476 directory problem" with a 421 code for SMTP messages.
5478 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5479 0, 1, and 2 exist, because some OS complain for execs without them (see
5480 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5481 descriptors. However, the code omitted to check that the open succeeded,
5482 causing mysterious errors if for some reason the permissions on /dev/null
5483 got screwed. Now Exim writes a message to the main and panic logs, and
5484 bombs out if it can't open /dev/null.
5486 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5487 interact so that it is all more flexible. It is supposed to remain
5488 backwards compatible. Also added extra_local_interfaces.
5490 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5491 to bomb out with an assertion failure - to the client this appears as a
5492 connection drop. This problem occurs in the part of the code that was taken
5493 from the Samba project. Fortunately, the assertion is in a very simple
5494 function, so I have fixed this by reproducing the function inline in the
5495 one place where it is called, and arranging for authentication to fail
5496 instead of killing the process with assert().
5498 75. The SPA client code was not working when the server requested OEM rather
5499 than Unicode encoding.
5501 76. Added code to make require_files with a specific uid setting more usable in
5502 the case where statting the file as root fails - usually a non-root-mounted
5503 NFS file system. When this happens and the failure is EACCES, Exim now
5504 forks a subprocess and does the per-uid checking as the relevant uid.
5506 77. Added process_log_path.
5508 78. If log_file_path was not explicitly set, a setting of check_log_space or
5509 check_log_inodes was ignored.
5511 79. If a space check for the spool or log partitions fails, the incident is now
5512 logged. Of course, in the latter case the data may get lost...
5514 80. Added the %p formatting code to string_format() so that it can be used to
5515 print addresses in debug_print(). Adjusted all the address printing in the
5516 debugging in store.c to use %p rather than %d.
5518 81. There was a concern that a line of code in smtp_in.c could overflow a
5519 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5520 initially expressed, the concern was not well-founded, because trailing
5521 spaces are removed early. However, if the trailing spaces were followed by
5522 a NULL, they did not get removed, so the overflow was possible. Two fixes
5525 (a) I re-wrote the offending code in a cleaner fashion.
5526 (b) If an incoming SMTP command contains a NULL character, it is rejected
5529 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5530 runs initgroups(), so that if the Exim user is in any additional groups,
5531 they will be used during message reception.
5537 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob