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>
56 PP/01 Solaris build fix for Oracle's LDAP libraries.
57 Bugzilla 1109, patch from Stephen Usher.
59 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
61 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
64 TF/02 Fix a couple more cases where we did not log the error message
65 when unlink() failed. See also change 4.74-TF/03.
67 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
68 lock up or crash if it happened to be inside a call to libc when it
69 got a SIGUSR1 from exiwhat.
71 The SIGUSR1 handler appends the current process status to the process
72 log which is later printed by exiwhat. It used to use the general
73 purpose logging code to do this, but several functions it calls are
76 The new output code in the SIGUSR1 handler is specific to the process
77 log, and simple enough that it's easy to inspect for signal safety.
78 Removing some special cases also simplifies the general logging code.
79 Removing the spurious timestamps from the process log simplifies
82 TF/04 Improved ratelimit ACL condition.
84 The /noupdate option has been deprecated in favour of /readonly which
85 has clearer semantics. The /leaky, /strict, and /readonly update modes
86 are mutually exclusive. The update mode is no longer included in the
87 database key; it just determines when the database is updated. (This
88 means that when you upgrde Exim will forget old rate measurements.)
90 Exim now checks that the per_* options are used with an update mode that
91 makes sense for the current ACL. For example, when Exim is processing a
92 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
93 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
94 must specify per_mail/readonly. If you omit the update mode it defaults to
95 /leaky where that makes sense (as before) or /readonly where required.
97 The /noupdate option is now undocumented but still supported for
98 backwards compatibility. It is equivalent to /readonly except that in
99 ACLs where /readonly is required you may specify /leaky/noupdate or
100 /strict/noupdate which are treated the same as /readonly.
102 A useful new feature is the /count= option. This is a generalization
103 of the per_byte option, so that you can measure the throughput of other
104 aggregate values. For example, the per_byte option is now equivalent
105 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
107 The per_rcpt option has been generalized using the /count= mechanism
108 (though it's more complicated than the per_byte equivalence). When it is
109 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
110 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
111 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
112 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
113 Note that using per_rcpt with a non-readonly update mode in more than
114 one ACL will cause the recipients to be double-counted. (The per_mail
115 and per_byte options don't have this problem.)
117 The handling of very low rates has changed slightly. If the computed rate
118 is less than the event's count (usually one) then this event is the first
119 after a long gap. In this case the rate is set to the same as this event's
120 count, so that the first message of a spam run is counted properly.
122 The major new feature is a mechanism for counting the rate of unique
123 events. The new per_addr option counts the number of different
124 recipients that someone has sent messages to in the last time period. It
125 behaves like per_rcpt if all the recipient addresses are different, but
126 duplicate recipient addresses do not increase the measured rate. Like
127 the /count= option this is a general mechanism, so the per_addr option
128 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
129 example, measure the rate that a client uses different sender addresses
130 with the options per_mail/unique=$sender_address. There are further
131 details in the main documentation.
133 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
135 TF/06 Removed a few PCRE remnants.
137 TF/07 Automatically extract Exim's version number from tags in the git
138 repository when doing development or release builds.
140 PP/02 Raise smtp_cmd_buffer_size to 16kB.
141 Bugzilla 879. Patch from Paul Fisher.
143 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
144 Heavily based on revision 40f9a89a from Simon Arlott's tree.
147 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
149 PP/05 Variable $av_failed, true if the AV scanner deferred.
150 Bugzilla 1078. Patch from John Horne.
152 PP/06 Stop make process more reliably on build failure.
153 Bugzilla 1087. Patch from Heiko Schlittermann.
155 PP/07 Make maildir_use_size_file an _expandable_ boolean.
156 Bugzilla 1089. Patch from Heiko Schlittermann.
158 PP/08 Handle ${run} returning more data than OS pipe buffer size.
159 Bugzilla 1131. Patch from Holger Weiß.
161 PP/09 Handle IPv6 addresses with SPF.
162 Bugzilla 860. Patch from Wolfgang Breyha.
164 PP/10 GnuTLS: support TLS 1.2 & 1.1.
166 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
169 PP/11 match_* no longer expand right-hand-side by default.
170 New compile-time build option, EXPAND_LISTMATCH_RHS.
171 New expansion conditions, "inlist", "inlisti".
173 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
175 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
177 PP/14 fix log_write() format string regression from TF/03.
178 Bugzilla 1152. Patch from Dmitry Isaikin.
184 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
186 PP/02 Harmonised TLS library version reporting; only show if debugging.
187 Layout now matches that introduced for other libraries in 4.74 PP/03.
189 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
191 PP/04 New "dns_use_edns0" global option.
193 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
196 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
197 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
199 TK/01 Updated PolarSSL code to 0.14.2.
200 Bugzilla 1097. Patch from Andreas Metzler.
202 PP/07 Catch divide-by-zero in ${eval:...}.
205 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
208 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
209 format-string attack -- SECURITY: remote arbitrary code execution.
211 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
212 time unintentionally subject to list matching rules, letting the header
213 cause arbitrary Exim lookups (of items which can occur in lists, *not*
214 arbitrary string expansion). This allowed for information disclosure.
216 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
217 INT_MIN/-1 -- value coerced to INT_MAX.
223 NM/01 Workround for PCRE version dependency in version reporting
226 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
227 This fixes portability to compilers other than gcc, notably
228 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
230 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
231 makefiles for portability to HP-UX and POSIX correctness.
233 PP/01 Permit LOOKUP_foo enabling on the make command-line.
234 Also via indented variable definition in the Makefile.
235 (Debugging by Oliver Heesakkers).
237 PP/02 Restore caching of spamd results with expanded spamd_address.
238 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
240 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
241 Improves build reliability. Fix from: Frank Elsner
243 NM/02 Fix wide character breakage in the rfc2047 coding
244 Fixes bug 1064. Patch from Andrey N. Oktyabrski
246 NM/03 Allow underscore in dnslist lookups
247 Fixes bug 1026. Patch from Graeme Fowler
249 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
250 Code patches from Adam Ciarcinski of NetBSD.
252 NM/04 Fixed exiqgrep to cope with mailq missing size issue
255 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
256 is logged, to avoid truncation. Patch from John Horne.
258 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
259 Patch from Jakob Hirsch.
261 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
262 SQL string expansion failure details.
263 Patch from Andrey Oktyabrski.
265 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
266 Patch from Simon Arlott.
268 PP/09 New lookups functionality failed to compile on old gcc which rejects
269 extern declarations in function scope.
270 Patch from Oliver Fleischmann
272 PP/10 Use sig_atomic_t for flags set from signal handlers.
273 Check getgroups() return and improve debugging.
274 Fixed developed for diagnosis in bug 927 (which turned out to be
277 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
278 Patch from Mark Zealey.
280 PP/12 Bugzilla 1056: Improved spamd server selection.
281 Patch from Mark Zealey.
283 PP/13 Bugzilla 1086: Deal with maildir quota file races.
284 Based on patch from Heiko Schlittermann.
286 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
287 Patch from Uwe Doering, sign-off by Michael Haardt.
289 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
290 variable declaration deep within a block. Bug and patch from
293 PP/15 lookups-Makefile IRIX compatibilty coercion.
295 PP/16 Make DISABLE_DKIM build knob functional.
297 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
298 Patch by Simon Arlott
300 TF/03 Fix valgrind.h portability to C89 compilers that do not support
301 variable argument macros. Our copy now differs from upstream.
307 TF/01 Failure to get a lock on a hints database can have serious
308 consequences so log it to the panic log.
310 TF/02 Log LMTP confirmation messages in the same way as SMTP,
311 controlled using the smtp_confirmation log selector.
313 TF/03 Include the error message when we fail to unlink a spool file.
315 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
316 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
317 for maintaining out-of-tree patches for some time.
319 PP/01 Bugzilla 139: Documentation and portability issues.
320 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
321 Handle per-OS dynamic-module compilation flags.
323 PP/02 Let /dev/null have normal permissions.
324 The 4.73 fixes were a little too stringent and complained about the
325 permissions on /dev/null. Exempt it from some checks.
326 Reported by Andreas M. Kirchwitz.
328 PP/03 Report version information for many libraries, including
329 Exim version information for dynamically loaded libraries. Created
330 version.h, now support a version extension string for distributors
331 who patch heavily. Dynamic module ABI change.
333 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
334 privilege escalation vulnerability whereby the Exim run-time user
335 can cause root to append content of the attacker's choosing to
338 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
341 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
342 If dropping privileges for untrusted macros, we disabled normal logging
343 on the basis that it would fail; for the Exim run-time user, this is not
344 the case, and it resulted in successful deliveries going unlogged.
345 Fixed. Reported by Andreas Metzler.
351 PP/01 Date: & Message-Id: revert to normally being appended to a message,
352 only prepend for the Resent-* case. Fixes regression introduced in
353 Exim 4.70 by NM/22 for Bugzilla 607.
355 PP/02 Include check_rfc2047_length in configure.default because we're seeing
356 increasing numbers of administrators be bitten by this.
358 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
360 PP/03 Bugzilla 994: added openssl_options main configuration option.
362 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
364 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
366 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
368 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
369 without a peer certificate, leading to a segfault because of an
370 assumption that peers always have certificates. Be a little more
371 paranoid. Problem reported by Martin Tscholak.
373 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
374 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
375 NB: ClamAV planning to remove STREAM in "middle of 2010".
376 CL also introduces -bmalware, various -d+acl logging additions and
377 more caution in buffer sizes.
379 PP/09 Implemented reverse_ip expansion operator.
381 PP/10 Bugzilla 937: provide a "debug" ACL control.
383 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
385 PP/12 Bugzilla 973: Implement --version.
387 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
389 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
391 PP/15 Bugzilla 816: support multiple condition rules on Routers.
393 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
394 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
395 ignore trailing whitespace.
397 JJ/02 prevent non-panic DKIM error from being sent to paniclog
399 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
402 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
403 Notification from Dr Andrew Aitchison.
405 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
406 ExtendedDetectionInfo response format.
407 Notification from John Horne.
409 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
412 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
413 XSL and documented dependency on system catalogs, with examples of how
416 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
419 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
420 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
421 configuration file which is writeable by the Exim user or group.
423 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
424 of configuration files to cover files specified with the -C option if
425 they are going to be used with root privileges, not just the default
428 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
429 option (effectively making it always true).
431 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
432 files to be used while preserving root privileges.
434 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
435 that rogue child processes cannot use them.
437 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
438 run-time user, instead of root.
440 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
441 Exim run-time user without dropping privileges.
443 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
444 result string, instead of calling string_vformat() twice with the same
447 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
448 for other users. Others should always drop root privileges if they use
449 -C on the command line, even for a whitelisted configure file.
451 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
453 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
459 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
460 $data_path, and $header_path variables; fixed documentation bugs and
463 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
464 exipick to access non-standard spools, including the "frozen" queue
467 NM/01 Bugzilla 965: Support mysql stored procedures.
468 Patch from Alain Williams
470 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
472 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
473 Patch from Andreas Metzler
475 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
476 Patch from Kirill Miazine
478 NM/05 Bugzilla 671: Added umask to procmail example.
480 JJ/03 installed exipick 20100323.0, fixing doc bug
482 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
483 directory. Notification and patch from Dan Rosenberg.
485 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
487 TK/02 Improve log output when DKIM signing operation fails.
489 MH/01 Treat the transport option dkim_domain as a colon separated
490 list, not as a single string, and sign the message with each element,
491 omitting multiple occurences of the same signer.
493 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
494 Bugzilla 985, 986. Patch by Simon Arlott
496 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
497 Patch by Simon Arlott
499 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
500 MBX locking. Notification from Dan Rosenberg.
506 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
508 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
510 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
512 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
514 NM/04 Bugzilla 915: Flag broken perl installation during build.
520 TK/01 Added patch by Johannes Berg that expands the main option
521 "spamd_address" if it starts with a dollar sign.
523 TK/02 Write list of recipients to X-Envelope-Sender header when building
524 the mbox-format spool file for content scanning (suggested by Jakob
527 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
528 (http://www.dcc-servers.net/) support via dccifd. Activated by
529 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
531 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
532 by Mark Daniel Reidel <mr@df.eu>.
534 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
535 When building exim an external PCRE library is now needed -
536 PCRE is a system library on the majority of modern systems.
537 See entry on PCRE_LIBS in EDITME file.
539 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
540 conversation. Added nologin parameter to request.
541 Patch contributed by Kirill Miazine.
543 TF/01 Do not log submission mode rewrites if they do not change the address.
545 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
547 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
548 log files in place. Contributed by Roberto Lima.
550 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
552 TF/03 Bugzilla 615: When checking the local_parts router precondition
553 after a local_part_suffix or local_part_prefix option, Exim now
554 does not use the address's named list lookup cache, since this
555 contains cached lookups for the whole local part.
557 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
558 Robert Millan. Documentation is in experimental-spec.txt.
560 TF/04 Bugzilla 668: Fix parallel build (make -j).
562 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
564 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
565 Patch provided by Jan Srzednicki.
567 TF/05 Leading white space used to be stripped from $spam_report which
568 wrecked the formatting. Now it is preserved.
570 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
571 that they are available at delivery time.
573 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
575 TF/08 TLS error reporting now respects the incoming_interface and
576 incoming_port log selectors.
578 TF/09 Produce a more useful error message if an SMTP transport's hosts
579 setting expands to an empty string.
581 NM/06 Bugzilla 744: EXPN did not work under TLS.
582 Patch provided by Phil Pennock.
584 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
585 Patch provided by Richard Godbee.
587 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
588 acl_smtp_notquit, added index entry.
590 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
591 Patch provided by Eugene Bujak.
593 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
594 accept(). Patch provided by Maxim Dounin.
596 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
597 Patch provided by Phil Pennock.
599 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
601 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
602 Patch provided by Brad "anomie" Jorsch.
604 NM/14 Bugzilla 622: Added timeout setting to transport filter.
605 Patch provided by Dean Brooks.
607 TK/05 Add native DKIM support (does not depend on external libraries).
609 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
610 Patch provided by Graeme Fowler.
612 NM/16 Bugzilla 851: Documentation example syntax fix.
614 NM/17 Changed NOTICE file to remove references to embedded PCRE.
616 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
619 NM/19 Bugzilla 745: TLS version reporting.
620 Patch provided by Phil Pennock.
622 NM/20 Bugzilla 167: bool: condition support.
623 Patch provided by Phil Pennock.
625 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
626 clients. Patch provided by Phil Pennock.
628 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
629 Patch provided by Brad "anomie" Jorsch.
631 NM/23 Bugzilla 687: Fix misparses in eximstats.
632 Patch provided by Heiko Schlittermann.
634 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
635 Patch provided by Heiko Schlittermann.
637 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
638 plus update to original patch.
640 NM/26 Bugzilla 799: Documentation correction for ratelimit.
642 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
643 Patch provided by David Brownlee.
645 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
647 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
649 NM/30 Bugzilla 888: TLS documentation bugfixes.
651 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
653 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
654 Unlike the original bugzilla I have changed all shell scripts in src tree.
656 NM/33 Bugzilla 898: Transport filter timeout fix.
657 Patch by Todd Rinaldo.
659 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
660 Patch by Serge Demonchaux.
662 NM/35 Bugzilla 39: Base64 decode bug fixes.
663 Patch by Jakob Hirsch.
665 NM/36 Bugzilla 909: Correct connect() call in dcc code.
667 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
669 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
671 NM/39 Bugzilla 911: Fixed MakeLinks build script.
677 TK/01 Add preliminary DKIM support. Currently requires a forked version of
678 ALT-N's libdkim that I have put here:
679 http://duncanthrax.net/exim-experimental/
681 Note to Michael Haardt: I had to rename some vars in sieve.c. They
682 were called 'true' and it seems that C99 defines that as a reserved
683 keyword to be used with 'bool' variable types. That means you could
684 not include C99-style headers which use bools without triggering
685 build errors in sieve.c.
687 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
688 as mailq or other aliases. Changed the --help handling significantly
689 to do whats expected. exim_usage() emits usage/help information.
691 SC/01 Added the -bylocaldomain option to eximstats.
693 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
695 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
697 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
703 PH/01 Another patch from the Sieve maintainer.
705 PH/02 When an IPv6 address is converted to a string for single-key lookup
706 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
707 dots are used instead of colons so that keys in lsearch files need not
708 contain colons. This was done some time before quoting was made available
709 in lsearch files. However, iplsearch files do require colons in IPv6 keys
710 (notated using the quote facility) so as to distinguish them from IPv4
711 keys. This meant that lookups for IP addresses in host lists did not work
712 for iplsearch lookups.
714 This has been fixed by arranging for IPv6 addresses to be expressed with
715 colons if the lookup type is iplsearch. This is not incompatible, because
716 previously such lookups could never work.
718 The situation is now rather anomolous, since one *can* have colons in
719 ordinary lsearch keys. However, making the change in all cases is
720 incompatible and would probably break a number of configurations.
722 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
725 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
726 conversion specification without a maximum field width, thereby enabling
727 a rogue spamd server to cause a buffer overflow. While nobody in their
728 right mind would setup Exim to query an untrusted spamd server, an
729 attacker that gains access to a server running spamd could potentially
730 exploit this vulnerability to run arbitrary code as the Exim user.
732 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
733 $primary_hostname instead of what libspf2 thinks the hosts name is.
735 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
736 a directory entry by the name of the lookup key. Previously, if a
737 symlink pointed to a non-existing file or a file in a directory that
738 Exim lacked permissions to read, a lookup for a key matching that
739 symlink would fail. Now it is enough that a matching directory entry
740 exists, symlink or not. (Bugzilla 503.)
742 PH/03 The body_linecount and body_zerocount variables are now exported in the
745 PH/04 Added the $dnslist_matched variable.
747 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
748 This means they are set thereafter only if the connection becomes
751 PH/06 Added the client_condition to authenticators so that some can be skipped
752 by clients under certain conditions.
754 PH/07 The error message for a badly-placed control=no_multiline_responses left
755 "_responses" off the end of the name.
757 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
759 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
760 (without spaces) instead of just copying the configuration text.
762 PH/10 Added the /noupdate option to the ratelimit ACL condition.
764 PH/11 Added $max_received_linelength.
766 PH/12 Added +ignore_defer and +include_defer to host lists.
768 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
769 way in which PCRE > 7.0 is built.
771 PH/14 Implemented queue_only_load_latch.
773 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
774 MAIL command. The effect was to mangle the value on 64-bit systems.
776 PH/16 Another patch from the Sieve maintainer.
778 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
780 PH/18 If a system quota error occurred while trying to create the file for
781 a maildir delivery, the message "Mailbox is full" was not appended to the
782 bounce if the delivery eventually timed out. Change 4.67/27 below applied
783 only to a quota excession during the actual writing of the file.
785 PH/19 It seems that peer DN values may contain newlines (and other non-printing
786 characters?) which causes problems in log lines. The DN values are now
787 passed through string_printing() before being added to log lines.
789 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
790 and InterBase are left for another time.)
792 PH/21 Added message_body_newlines option.
794 PH/22 Guard against possible overflow in moan_check_errorcopy().
796 PH/23 POSIX allows open() to be a macro; guard against that.
798 PH/24 If the recipient of an error message contained an @ in the local part
799 (suitably quoted, of course), incorrect values were put in $domain and
800 $local_part during the evaluation of errors_copy.
806 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
807 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
810 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
811 issue a MAIL command.
813 PH/02 In an ACL statement such as
815 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
817 if a client was not listed at all, or was listed with a value other than
818 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
819 the condition was not true (as it should be), so access was not denied.
820 The bug was that the ! inversion was incorrectly passed on to the second
821 item. This has been fixed.
823 PH/03 Added additional dnslists conditions == and =& which are different from
824 = and & when the dns lookup returns more than one IP address.
826 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
827 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
829 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
830 FSYNC, which compiles an option called disable_fsync that allows for
831 bypassing fsync(). The documentation is heavily laced with warnings.
833 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
835 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
836 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
837 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
838 including adding "make clean"; (3) Added -fPIC when compiling the test
839 dynamically loaded module, to get rid of a warning.
841 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
842 message fails, move_frozen_messages = true and ignore_bounce_errors_after
843 = 0s. The bug is otherwise harmless.
845 PH/07 There was a bug in the dovecot authenticator such that the value of
846 $auth1 could be overwritten, and so not correctly preserved, after a
847 successful authentication. This usually meant that the value preserved by
848 the server_setid option was incorrect.
850 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
852 PH/09 Installed PCRE release 7.0.
854 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
855 run for batched SMTP input. It is now run at the start of every message
856 in the batch. While fixing this I discovered that the process information
857 (output by running exiwhat) was not always getting set for -bs and -bS
858 input. This is fixed, and it now also says "batched" for BSMTP.
860 PH/11 Added control=no_pipelining.
862 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
863 patch, slightly modified), and move the expansion of helo_data till after
864 the connection is made in the smtp transport (so it can use these
867 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
869 PH/14 Added log_selector = +pid.
871 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
873 PH/16 Add ${if forany and ${if forall.
875 PH/17 Added dsn_from option to vary the From: line in DSNs.
877 PH/18 Flush SMTP output before performing a callout, unless control =
878 no_callout_flush is set.
880 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
881 was true (the default) a successful delivery failed to delete the retry
882 item, thus causing premature timeout of the address. The bug is now
885 PH/20 Added hosts_avoid_pipelining to the smtp transport.
887 PH/21 Long custom messages for fakedefer and fakereject are now split up
888 into multiline reponses in the same way that messages for "deny" and
889 other ACL rejections are.
891 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
892 with slight modification.
894 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
895 draft, changing the syntax and factoring some duplicate code".
897 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
898 for deliveries of the second and subsequent messages over the same SMTP
901 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
902 ${reduce, with only minor "tidies".
904 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
906 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
907 expansion side effects.
909 PH/27 When a message times out after an over-quota error from an Exim-imposed
910 quota, the bounce message says "mailbox is full". This message was not
911 being given when it was a system quota that was exceeded. It now should
914 MH/03 Made $recipients available in local_scan(). local_scan() already has
915 better access to the recipient list through recipients_list[], but
916 $recipients can be useful in postmaster-provided expansion strings.
918 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
919 in the case of a MAIL command with additional options following the
920 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
921 were accidentally chopped off.
923 PH/29 SMTP synchronization checks are implemented when a command is read -
924 there is a check that no more input is waiting when there shouldn't be
925 any. However, for some commands, a delay in an ACL can mean that it is
926 some time before the response is written. In this time, more input might
927 arrive, invalidly. So now there are extra checks after an ACL has run for
928 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
929 pipelining has not been advertised.
931 PH/30 MH's patch to allow iscntrl() characters to be list separators.
933 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
934 returned in the SMTP response when smtp_return_error_details was false.
937 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
938 instead of the C I/O that was originally supplied, because problems were
941 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
942 Exim which did not show up earlier: it was assuming that a call to
943 SSL_CTX_set_info_callback() might give an error value. In fact, there is
944 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
945 was a macro that became an assignment, so it seemed to work. This has
946 changed to a proper function call with a void return, hence the compile
947 error. Exim's code has been fixed.
949 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
952 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
954 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
956 PH/37 If a message is not accepted after it has had an id assigned (e.g.
957 because it turns out to be too big or there is a timeout) there is no
958 "Completed" line in the log. When some messages of this type were
959 selected by exigrep, they were listed as "not completed". Others were
960 picked up by some special patterns. I have improved the selection
961 criteria to be more general.
963 PH/38 The host_find_failed option in the manualroute router can now be set
964 to "ignore", to completely ignore a host whose IP address cannot be
965 found. If all hosts are ignored, the behaviour is controlled by the new
966 host_all_ignored option.
968 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
969 homing or because of multiple MX records with /mx) generated more than
970 one IP address, and the following item turned out to be the local host,
971 all the secondary addresses of the first item were incorrectly removed
972 from the list, along with the local host and any following hosts (which
973 is what is supposed to happen).
975 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
976 whoever called Exim into the -H file. In the case of the daemon it was
977 behaving confusingly. When first started, it used values for whoever
978 started the daemon, but after a SIGHUP it used the Exim user (because it
979 calls itself on a restart). I have changed the code so that it now always
982 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
983 message are rejected with the same error (e.g. no authentication or bad
984 sender address), and a DATA command is nevertheless sent (as can happen
985 with PIPELINING or a stupid MUA), the error message that was given to the
986 RCPT commands is included in the rejection of the DATA command. This is
987 intended to be helpful for MUAs that show only the final error to their
990 PH/42 Another patch from the Sieve maintainer.
992 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
993 Eximstats - Fixed some broken HTML links and added missing column headers
995 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
996 columns for Rejects, Temp Rejects, Ham, and Spam rows.
998 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1000 PH/43 Yet another patch from the Sieve maintainer.
1002 PH/44 I found a way to check for a TCP/IP connection going away before sending
1003 the response to the final '.' that terminates a message, but only in the
1004 case where the client has not sent further data following the '.'
1005 (unfortunately, this is allowed). However, in many cases there won't be
1006 any further data because there won't be any more messages to send. A call
1007 to select() can be used: if it shows that the input is "ready", there is
1008 either input waiting, or the socket has been closed. An attempt to read
1009 the next input character can distinguish the two cases. Previously, Exim
1010 would have sent an OK response which the client would never have see.
1011 This could lead to message repetition. This fix should cure that, at
1012 least in a lot of common cases.
1014 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1015 advertised in response to EHLO.
1021 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1022 fixed by 4.65/MH/01 (is this a record?) are fixed:
1024 (i) An empty string was always treated as zero by the numeric comparison
1025 operators. This behaviour has been restored.
1027 (ii) It is documented that the numeric comparison operators always treat
1028 their arguments as decimal numbers. This was broken in that numbers
1029 starting with 0 were being interpreted as octal.
1031 While fixing these problems I realized that there was another issue that
1032 hadn't been noticed. Values of message_size_limit (both the global option
1033 and the transport option) were treated as octal if they started with 0.
1034 The documentation was vague. These values are now always treated as
1035 decimal, and I will make that clear in the documentation.
1041 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1042 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1045 MH/01 Don't check that the operands of numeric comparison operators are
1046 integers when their expansion is in "skipping" mode (fixes bug
1047 introduced by 4.64-PH/07).
1049 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1050 child addresses, Exim now panics and dies. Previously, because the count
1051 is held in a short int, deliveries were likely to be lost. As such a
1052 large number of recipients for a single message is ridiculous
1053 (performance will be very, very poor), I have chosen to impose a limit
1054 rather than extend the field.
1060 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1061 leftover -K file (the existence of which was triggered by #402).
1062 While we were at it, introduced process PID as part of the -K
1063 filename. This should rule out race conditions when creating
1066 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1067 processing considerably. Previous code took too long for large mails,
1068 triggering a timeout which in turn triggers #401.
1070 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1071 in the DK code in transports.c. sendfile() is not really portable,
1072 hence the _LINUX specificness.
1074 TF/01 In the add_headers option to the mail command in an Exim filter,
1075 there was a bug that Exim would claim a syntax error in any
1076 header after the first one which had an odd number of characters
1079 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1080 callout verification, Exim cached a "reject" for the entire domain. This
1081 is correct for most verifications, but it is not correct for a recipient
1082 verification with use_sender or use_postmaster set, because in that case
1083 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1084 case of MAIL FROM:<> rejection from other early rejections (e.g.
1085 rejection of HELO). When verifying a recipient using a non-null MAIL
1086 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1087 Whatever the result of the callout, the value of the domain cache is
1088 left unchanged (for any other kind of callout, getting as far as trying
1089 RCPT means that the domain itself is ok).
1091 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1094 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1095 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1096 session, the daemon ignores ECONNECT errors and logs others; it now
1097 ignores EPIPE as well.
1099 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1100 (quoted-printable decoding).
1102 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1103 later the small subsequent patch to fix an introduced bug.
1105 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1107 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1109 PH/08 An error is now given if message_size_limit is specified negative.
1111 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1112 to be given (somewhat) arbitrary names.
1114 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1117 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1118 miscellaneous code fixes
1120 PH/10 Added the log_reject_target ACL modifier to specify where to log
1123 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1124 hostname. This is wrong, because it relates to the incoming message (and
1125 probably the interface on which it is arriving) and not to the outgoing
1126 callout (which could be using a different interface). This has been
1127 changed to use the value of the helo_data option from the smtp transport
1128 instead - this is what is used when a message is actually being sent. If
1129 there is no remote transport (possible with a router that sets up host
1130 addresses), $smtp_active_hostname is used.
1132 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1133 tweaks were necessary in order to get it to work (see also 21 below):
1134 (a) The code assumed that strncpy() returns a negative number on buffer
1135 overflow, which isn't the case. Replaced with Exim's string_format()
1137 (b) There were several signed/unsigned issues. I just did the minimum
1138 hacking in of casts. There is scope for a larger refactoring.
1139 (c) The code used strcasecmp() which is not a standard C function.
1140 Replaced with Exim's strcmpic() function.
1141 (d) The code set only $1; it now sets $auth1 as well.
1142 (e) A simple test gave the error "authentication client didn't specify
1143 service in request". It would seem that Dovecot has changed its
1144 interface. Fortunately there's a specification; I followed it and
1145 changed what the client sends and it appears to be working now.
1147 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1150 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1151 address A is aliased to B and C, where B exists and C does not. Without
1152 -v the output is "A verified" because verification stops after a
1153 successful redirection if more than one address is generated. However,
1154 with -v the child addresses are also verified. Exim was outputting "A
1155 failed to verify" and then showing the successful verification for C,
1156 with its parentage. It now outputs "B failed to verify", showing B's
1157 parentage before showing the successful verification of C.
1159 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1160 look up a TXT record in a specific list after matching in a combined
1163 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1164 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1165 they consult the DNS. I had assumed they would set it the way they
1166 wanted; and indeed my experiments on Linux seem to show that in some
1167 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1168 To be on the safe side, however, I have now made the interface to
1169 host_find_byname() similar to host_find_bydns(), with an argument
1170 containing the DNS resolver options. The host_find_byname() function now
1171 sets these options at its start, just as host_find_bydns() does. The smtp
1172 transport options dns_qualify_single and dns_search_parents are passed to
1173 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1174 of host_find_byname() use the default settings of RES_DEFNAMES
1175 (qualify_single) but not RES_DNSRCH (search_parents).
1177 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1178 spool_read_header() do less string testing, by means of a preliminary
1179 switch on the second character of optional "-foo" lines. (This is
1180 overdue, caused by the large number of possibilities that now exist.
1181 Originally there were few.) While I was there, I also converted the
1182 str(n)cmp tests so they don't re-test the leading "-" and the first
1183 character, in the hope this might squeeze out yet more improvement.
1185 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1186 flag allowing group syntax was set by the header_syntax check but not
1187 turned off, possible causing trouble later; (2) The flag was not being
1188 set at all for the header_verify test, causing "group"-style headers to
1189 be rejected. I have now set it in this case, and also caused header_
1190 verify to ignore an empty address taken from a group. While doing this, I
1191 came across some other cases where the code for allowing group syntax
1192 while scanning a header line wasn't quite right (mostly, not resetting
1193 the flag correctly in the right place). These bugs could have caused
1194 trouble for malformed header lines. I hope it is now all correct.
1196 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1197 with the "reply" argument non-NULL. The code, however (which originally
1198 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1199 but it didn't always do it. This confused somebody who was copying the
1200 code for some other use. I have removed all the tests.
1202 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1203 feature that was used to support insecure browsers during the U.S. crypto
1204 embargo. It requires special client support, and Exim is probably the
1205 only MTA that supported it -- and would never use it because real RSA is
1206 always available. This code has been removed, because it had the bad
1207 effect of slowing Exim down by computing (never used) parameters for the
1208 RSA_EXPORT functionality.
1210 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1211 authenticator to fail if there's a tab character in the incoming data
1212 (there should never be unless someone is messing about, as it's supposed
1213 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1214 if the connection is using TLS or if the remote IP is the same as the
1215 local IP, and the "valid-client-cert option" if a client certificate has
1218 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1219 authenticators. This can be used for authorization after authentication
1220 succeeds. (In the case of plaintext, it servers for both authentication
1223 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1224 if any retry times were supplied.
1226 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1227 connection, where there is no client IP address to check. In this
1228 situation, the verify now always succeeds.
1230 PH/25 Applied John Jetmore's -Mset patch.
1232 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1234 PH/27 In a string expansion for a processed (not raw) header when multiple
1235 headers of the same name were present, leading whitespace was being
1236 removed from all of them, but trailing whitespace was being removed only
1237 from the last one. Now trailing whitespace is removed from each header
1238 before concatenation. Completely empty headers in a concatenation (as
1239 before) are ignored.
1241 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1242 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1244 PH/29 [Removed. This was a change that I later backed out, and forgot to
1245 correct the ChangeLog entry (that I had efficiently created) before
1246 committing the later change.]
1248 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1249 address errors (4xx response to RCPT) over the same connection as other
1250 messages routed to the same hosts. Such deliveries are always "forced",
1251 so retry times are not inspected. This resulted in far too many retries
1252 for the affected addresses. The effect occurred only when there were more
1253 hosts than the hosts_max_try setting in the smtp transport when it had
1254 the 4xx errors. Those hosts that it had tried were not added to the list
1255 of hosts for which the message was waiting, so if all were tried, there
1256 was no problem. Two fixes have been applied:
1258 (i) If there are any address or message errors in an SMTP delivery, none
1259 of the hosts (tried or untried) are now added to the list of hosts
1260 for which the message is waiting, so the message should not be a
1261 candidate for sending over the same connection that was used for a
1262 successful delivery of some other message. This seems entirely
1263 reasonable: after all the message is NOT "waiting for some host".
1264 This is so "obvious" that I'm not sure why it wasn't done
1265 previously. Hope I haven't missed anything, but it can't do any
1266 harm, as the worst effect is to miss an optimization.
1268 (ii) If, despite (i), such a delivery is accidentally attempted, the
1269 routing retry time is respected, so at least it doesn't keep
1270 hammering the server.
1272 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1273 in ${readsocket because some servers need this prod.
1275 PH/32 Added some extra debug output when updating a wait-xxx database.
1277 PH/33 The hint "could be header name not terminated by colon", which has been
1278 given for certain expansion errors for a long time, was not being given
1279 for the ${if def:h_colon_omitted{... case.
1281 PH/34 The spec says: "With one important exception, whenever a domain list is
1282 being scanned, $domain contains the subject domain." There was at least
1283 one case where this was not true.
1285 PH/35 The error "getsockname() failed: connection reset by peer" was being
1286 written to the panic log as well as the main log, but it isn't really
1287 panic-worthy as it just means the connection died rather early on. I have
1288 removed the panic log writing for the ECONNRESET error when getsockname()
1291 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1292 runs only) independently of the message's sender address. This meant
1293 that, if the 4xx error was in fact related to the sender, a different
1294 message to the same recipient with a different sender could confuse
1295 things. In particualar, this can happen when sending to a greylisting
1296 server, but other circumstances could also provoke similar problems.
1297 I have changed the default so that the retry time for these errors is now
1298 based a combination of the sender and recipient addresses. This change
1299 can be overridden by setting address_retry_include_sender=false in the
1302 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1303 remote server are returned as part of bounce messages. This was not
1304 happening for LMTP over a pipe (the lmtp transport), but now it is the
1305 same for both kinds of LMTP.
1307 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1308 in header lines that were in fact CNAMEs. This is no longer the case.
1310 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1311 and queue runs started by the daemon processed all messages. This has
1312 been fixed so that -R and -S can now usefully be given with -q<time>.
1314 PH/40 Import PCRE release 6.7 (fixes some bugs).
1316 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1318 PH/42 Give an error if -q is specified more than once.
1320 PH/43 Renamed the variables $interface_address and $interface_port as
1321 $received_ip_address and $received_port, to make it clear that these
1322 values apply to message reception, and not to the outgoing interface when
1323 a message is delivered. (The old names remain recognized, of course.)
1325 PH/44 There was no timeout on the connect() call when using a Unix domain
1326 socket in the ${readsocket expansion. There now is.
1328 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1329 be meaningful with "accept".
1331 SC/01 Eximstats V1.43
1332 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1334 SC/02 Eximstats V1.44
1335 Use a glob alias rather than an array ref in the generated
1336 parser. This improves both readability and performance.
1338 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1339 Collect SpamAssassin and rejection statistics.
1340 Don't display local sender or destination tables unless
1341 there is data to show.
1342 Added average volumes into the top table text output.
1344 SC/04 Eximstats V1.46
1345 Collect data on the number of addresses (recipients)
1346 as well as the number of messages.
1348 SC/05 Eximstats V1.47
1349 Added 'Message too big' to the list of mail rejection
1350 reasons (thanks to Marco Gaiarin).
1352 SC/06 Eximstats V1.48
1353 Mainlog lines which have GMT offsets and are too short to
1354 have a flag are now skipped.
1356 SC/07 Eximstats V1.49 (Alain Williams)
1357 Added the -emptyok flag.
1359 SC/08 Eximstats V1.50
1360 Fixes for obtaining the IP address from reject messages.
1362 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1363 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1364 whitesspace changes from 4.64-PH/27
1366 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1369 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1370 are found, allow negative numbers in numeric criteria)
1372 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1374 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1375 to match changes made in 4.64-PH/43
1377 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1379 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1380 "domain" is actually an IP address, and force a failure. This locks out
1381 those revolvers/nameservers that support "A-for-A" lookups, in
1382 contravention of the specifications.
1384 PH/48 When a host name was looked up from an IP address, and the subsequent
1385 forward lookup of the name timed out, the host name was left in
1386 $sender_host_name, contrary to the specification.
1388 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1389 restricted to single-key lookups, Exim was not diagnosing an error if
1390 * or *@ was used with a query-style lookup.
1392 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1394 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1395 long ago, but noone interested enough thought of it. Let's just say that
1396 the "1.1" means that there are some new functions that weren't there at
1397 some point in the past.
1399 PH/51 Error processing for expansion failure of helo_data from an smtp
1400 transport during callout processing was broken.
1402 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1403 tested/used via the -bh/-bhc/-bs options.
1405 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1406 bug, fixed in subsequent PCRE releases).
1408 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1409 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1411 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1417 SC/01 Use a glob alias rather than an array ref in eximstats generated
1418 parser. This improves both readability and performance.
1420 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1421 Don't display local sender or destination tables in eximstats unless
1422 there is data to show.
1423 Added average volumes into the eximstats top table text output.
1425 SC/03 Collect data on the number of addresses (recipients) as well
1426 as the number of messages in eximstats.
1428 TF/01 Correct an error in the documentation for the redirect router. Exim
1429 does (usually) call initgroups() when daemonizing.
1431 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1432 with consistent privilege compared to when running as a daemon.
1434 TF/03 Note in the spec that $authenticated_id is not set for local
1435 submissions from trusted users.
1437 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1438 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1440 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1441 by adding some example configuration directives to the default
1442 configuration file. A little bit of work is required to uncomment the
1443 directives and define how usernames and passwords are checked, but
1444 there is now a framework to start from.
1446 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1447 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1448 without this. I don't know how relevant this is to other LDAP libraries.
1450 PH/02 Add the verb name to the "unknown ACL verb" error.
1452 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1454 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1456 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1457 directory not expanded when it should be if an expanded home directory
1458 was set for the address (which is overridden by the transport).
1460 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1463 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1464 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1465 because it is too late at that time, and has no effect.
1467 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1468 security issue with \' (bugzilla #107). I could not use the
1469 PQescapeStringConn() function, because it needs a PGconn value as one of
1472 PH/08 When testing addresses using -bt, indicate those final addresses that
1473 are duplicates that would not cause an additional delivery. At least one
1474 person was confused, thinking that -bt output corresponded to deliveries.
1475 (Suppressing duplicates isn't a good idea as you lose the information
1476 about possibly different redirections that led to the duplicates.)
1478 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1479 systems where poll() doesn't work, in particular OS X.
1481 PH/10 Added more information to debugging output for retry time not reached.
1483 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1484 operations in malware.c.
1486 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1489 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1490 syslog_duplication set false, log lines that would normally be written
1491 both the the main log and to the reject log were not written to syslog at
1494 PH/14 In the default configuration, change the use of "message" in ACL warn
1495 statements to "add_header".
1497 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1498 not followed by a command (e.g. "seen endif").
1500 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1501 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1504 PH/17 Added extra conditions to the default value of delay_warning_condition
1508 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1509 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1510 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1513 The Auto-Submitted: and various List- headers are standardised, whereas I
1514 don't think Precedence: ever was.
1516 PH/18 Refactored debugging code in route_finduser() to show more information,
1517 in particular, the error code if getpwnam() issues one.
1519 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1520 This is apparently needed in addition to the PH/07 change above to avoid
1521 any possible encoding problems.
1523 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1524 but not after initializing Perl.
1526 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1527 output them only if debugging. By default they are written stderr,
1528 apparently, which is not desirable.
1530 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1533 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1536 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1538 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1539 authenticated or an ident call has been made. Suppress the default
1540 values for $authenticated_id and $authenticated_sender (but permit -oMai
1541 and -oMas) when testing with -bh.
1543 PH/24 Re-jigged the order of the tests in the default configuration so that the
1544 tests for valid domains and recipients precede the DNS black list and CSA
1545 tests, on the grounds that those ones are more expensive.
1547 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1548 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1549 This bug exists in every version of Exim that I still have, right back to
1552 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1553 However, an attempt to turn on case-sensitivity in a regex key by
1554 including (?-i) didn't work because the subject string was already
1555 lowercased, and the effects were non-intuitive. It turns out that a
1556 one-line patch can be used to allow (?-i) to work as expected.
1562 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1563 other effects) broke the use of negated acl sub-conditions.
1565 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1568 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1569 "Deny" causes Exim to reject the incoming connection with a 554 error.
1570 Unfortunately, if there is a major crisis, such as a disk failure,
1571 tcp-wrappers gives "deny", whereas what one would like would be some
1572 kind of temporary error. A kludge has been added to help with this.
1573 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1574 554 error is used if errno is still zero or contains ENOENT (which occurs
1575 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1578 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1580 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1581 errors. Otherwise a message that provokes a temporary error (when other
1582 messages do not) can cause a whole host to time out.
1584 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1585 addresses were routed directly to files or pipes from a redirect router.
1586 File deliveries just didn't batch; pipe deliveries might have suffered
1589 PH/06 A failure to get a lock for a hints database would erroneously always say
1590 "Failed to get write lock", even when it was really a read lock.
1592 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1593 of 0600. This has been changed to use the value of the lockfile_mode
1594 option (which defaults to 0600).
1596 PH/08 Applied small patch from the Sieve maintainer.
1598 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1599 folder from quota calculations, a direct delivery into this folder messed
1600 up the contents of the maildirsize file. This was because the regex was
1601 used only to exclude .Trash (or whatever) when the size of the mailbox
1602 was calculated. There was no check that a delivery was happening into an
1603 excluded directory. This bug has been fixed by ignoring all quota
1604 processing for deliveries into excluded directories.
1606 PH/10 Added the maildirfolder_create_regex option to appendfile.
1612 PH/01 The code for finding all the local interface addresses on a FreeBSD
1613 system running IPv6 was broken. This may well have applied to all BSD
1614 systems, as well as to others that have similar system calls. The broken
1615 code found IPv4 interfaces correctly, but gave incorrect values for the
1616 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1617 that it would not match correctly against @[] and not recognize the IPv6
1620 PH/02 The ipliteral router was not recognizing addresses of the form user@
1621 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1623 PH/03 Added disable_ipv6.
1625 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1626 decoded form, because it is most often used to construct To: headers
1627 lines in autoreplies, and the decoded form may well be syntactically
1628 invalid. However, $reply_address has leading white space removed, and all
1629 newlines turned into spaces so that the autoreply transport does not
1632 PH/05 If group was specified without a user on a router, and no group or user
1633 was specified on a transport, the group from the router was ignored.
1635 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1636 for visible compile-time settings that can be used to change these
1637 numbers, for those that want even more. Backwards compatibility with old
1638 spool files has been maintained. However, going back to a previous Exim
1639 release will lost any variables that are in spool files.
1641 PH/07 Two small changes when running in the test harness: increase delay when
1642 passing a TCP/IP connection to a new process, in case the original
1643 process has to generate a bounce, and remove special handling of
1644 127.0.0.2 (sic), which is no longer necessary.
1646 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1647 be the same on different OS.
1649 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1652 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1653 whether --show-vars was specified or not
1655 JJ/02 exipick: Added support for new ACL variable spool format introduced
1658 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1659 syntactically invalid From: or Reply-to: line, and a filter used this to
1660 generate an autoreply, and therefore failed to obtain an address for the
1661 autoreply, Exim could try to deliver to a non-existent relative file
1662 name, causing unrelated and misleading errors. What now happens is that
1663 it logs this as a hard delivery error, but does not attempt to create a
1666 PH/11 The exinext utility has a -C option for testing purposes, but although
1667 the given file was scanned by exinext itself; it wasn't being passed on
1668 when Exim was called.
1670 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1671 an end-of-file indication when reading a command response.
1673 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1674 compiled. In many other places in Exim, IPv6 addresses are always
1675 recognized, so I have changed this. It also means that IPv4 domain
1676 literals of the form [IPV4:n.n.n.n] are now always recognized.
1678 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1679 used if the router is not running as root, for example, when verifying at
1680 ACL time, or when using -bh. The debugging output from this situation was
1681 non-existent - all you got was a failure to exec. I have made two
1684 (a) Failures to set uid/gid, the current directory, or a process leader
1685 in a subprocess such as that created by queryprogram now generate
1686 suitable debugging ouput when -d is set.
1688 (b) The queryprogram router detects when it is not running as root,
1689 outputs suitable debugging information if -d is set, and then runs
1690 the subprocess without attempting to change uid/gid.
1692 PH/15 Minor change to Makefile for building test_host (undocumented testing
1695 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1696 additional section of a DNS packet that returns MX or SRV records.
1697 Instead, it always explicitly searches for A/AAAA records. This avoids
1698 major problems that occur when a DNS server includes only records of one
1699 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1700 fixed another bug: if SRV records were looked up and the corresponding
1701 address records were *not* found in the additional section, the port
1702 values from the SRV records were lost.
1704 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1705 using the correct key (the original address) when searching the retry
1706 rules in order to find which one to use for generating the retry hint.
1708 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1709 adding the default one. Similarly, if it contains a Reply-To: header, the
1710 errors_reply_to option, if set, is not used.
1712 PH/19 When calculating a retry time, Exim used to measure the "time since
1713 failure" by looking at the "first failed" field in the retry record. Now
1714 it does not use this if it is later than than the arrival time of the
1715 message. Instead it uses the arrival time. This makes for better
1716 behaviour in cases where some deliveries succeed, thus re-setting the
1717 "first failed" field. An example is a quota failure for a huge message
1718 when small messages continue to be delivered. Without this change, the
1719 "time since failure" will always be short, possible causing more frequent
1720 delivery attempts for the huge message than are intended.
1721 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1723 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1724 $1, $2, $3) because the numerical variables can be reset during some
1725 expansion items (e.g. "match"), thereby losing the authentication data.
1727 PH/21 Make -bV show the size of off_t variables so that the test suite can
1728 decide whether to run tests for quotas > 2G.
1730 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1731 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1732 filecount value is greater than 2G or if a quota value is greater than 2G
1733 on a system where the size of off_t is not greater than 4, a panic error
1736 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1737 never match. The debug and -bh output now contains an explicit error
1738 message indicating a malformed IPv4 address or mask.
1740 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1741 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1742 PH/23 above applies.
1744 PH/25 Do not write to syslog when running in the test harness. The only
1745 occasion when this arises is a failure to open the main or panic logs
1746 (for which there is an explicit test).
1748 PH/26 Added the /no_tell option to "control=freeze".
1750 PH/27 If a host name lookup failed very early in a connection, for example, if
1751 the IP address matched host_lookup and the reverse lookup yielded a name
1752 that did not have a forward lookup, an error message of the form "no IP
1753 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1754 could be logged. Now it outputs the IP address instead of "NULL".
1756 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1757 allows the sender and the authenticated sender to be set when
1758 submitting a message from within Exim. Since child_open_exim() is
1759 documented for local_scan(), the new function should be too.
1761 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1762 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1763 results in an empty string is now treated as unset.
1765 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1767 PH/31 Added sender_verify_fail logging option.
1769 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1770 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1771 tidied the source and removed it altogether.
1773 PH/33 When a queue run was abandoned because the load average was too high, a
1774 log line was always written; now it is written only if the queue_run log
1775 selector is set. In addition, the log line for abandonment now contains
1776 information about the queue run such as the pid. This is always present
1777 in "start" and "stop" lines but was omitted from the "abandon" line.
1779 PH/34 Omit spaces between a header name and the colon in the error message that
1780 is given when verify = headers_syntax fails (if there are lots of them,
1781 the message gets confusing).
1783 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1784 names, as there are now some PTR records that contain slashes. This check
1785 is only to protect against broken name servers that fall over on strange
1786 characters, so the fact that it applies to all lookups doesn't matter.
1788 PH/36 Now that the new test suite is complete, we can remove some of the
1789 special code in Exim that was needed for the old test suite. For example,
1790 sorting DNS records because real resolvers return them in an arbitrary
1791 order. The new test suite's fake resolver always returns records in the
1794 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1795 bounces) except when queue_only is set, to avoid logging races between
1796 the different processes.
1798 PH/38 Panic-die if .include specifies a non-absolute path.
1800 PH/39 A tweak to the "H" retry rule from its user.
1802 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1803 a label. They prevented compilation on older perls.
1805 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1806 a warning to be raised on newish perls.
1808 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1809 on queue. Changes to match documented behaviour of showing count of
1810 messages matching specified criteria.
1812 PH/40 Changed the default ident timeout from 30s to 5s.
1814 PH/41 Added support for the use of login_cap features, on those BSD systems
1815 that have them, for controlling the resources used by pipe deliveries.
1817 PH/42 The content-scanning code uses fopen() to create files in which to put
1818 message data. Previously it was not paying any attention to the mode of
1819 the files. Exim runs with umask(0) because the rest of the code creates
1820 files with open(), and sets the required mode explicitly. Thus, these
1821 files were ending up world-writeable. This was not a big issue, because,
1822 being within the spool directory, they were not world-accessible. I have
1823 created a function called modefopen, which takes an additional mode
1824 argument. It sets umask(777), creates the file, chmods it to the required
1825 mode, then resets the umask. All the relevant calls to fopen() in the
1826 content scanning code have been changed to use this function.
1828 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1829 to 24 hours. This avoids potential overflow problems when processing G
1830 and H retry rules. I suspect nobody ever tinkers with this value.
1832 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1834 PH/45 When the plaintext authenticator is running as a client, the server's
1835 challenges are checked to ensure they are valid base64 strings. By
1836 default, the authentication attempt is cancelled if an invalid string is
1837 received. Setting client_ignore_invalid_base64 true ignores these errors.
1838 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1839 they are received. Thus, the responses can be made to depend on the
1840 challenges. If an invalid string is ignored, an empty string is placed in
1843 PH/46 Messages that are created by the autoreply transport now contains a
1844 References: header, in accordance with RFCs 2822 and 3834.
1846 PH/47 Added authenticated_sender_force to the smtp transport.
1848 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1850 PH/49 Installed latest patch from the Sieve maintainer.
1852 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1853 was also set, the appendfile transport was unnecessarily scanning a
1854 directory of message files (e.g. for maildir delivery) to find the count
1855 of files (along with the size), even though it did not need this
1856 information. It now does the scan only if it needs to find either the
1857 size of the count of files.
1859 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1861 PH/52 Two bugs concerned with error handling when the smtp transport is
1864 (i) Exim was not creating retry information for temporary errors given
1865 for individual recipients after the DATA command when the smtp transport
1866 was used in LMTP mode. This meant that they could be retried too
1867 frequently, and not timed out correctly.
1869 (ii) Exim was setting the flag that allows error details to be returned
1870 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1871 recipients that were returned after the DATA command.
1873 PH/53 This is related to PH/52, but is more general: for any failing address,
1874 when detailed error information was permitted to be returned to the
1875 sender, but the error was temporary, then after the final timeout, only
1876 "retry timeout exceeded" was returned. Now it returns the full error as
1877 well as "retry timeout exceeded".
1879 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1880 do this, and (what is worse) MTAs that accept it.
1882 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1883 will now be deprecated.
1885 PH/56 New os.c-cygwin from the Cygwin maintainer.
1887 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1888 formats (previously only available in exim formats via -bpr, -bpru,
1889 and -bpra. Now also available in native and exiqgrep formats)
1891 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1892 with very large, slow to parse queues
1894 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1896 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1898 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1899 responses to authentication challenges, though it was showing the
1900 challenges; (ii) I've removed the CR characters from the debug output for
1903 PH/58 Allow for the insertion of a newline as well as a space when a string
1904 is turned into more than one encoded-word during RFC 2047 encoding. The
1905 Sieve code now uses this.
1907 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1908 data_4xx, lost_connection, tls_required.
1910 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1911 message was being sent as an SMTP response.
1913 PH/61 Add -l and -k options to exicyclog.
1915 PH/62 When verifying, if an address was redirected to one new address, so that
1916 verification continued, and the new address failed or deferred after
1917 having set something in $address_data, the value of $address_data was not
1918 passed back to the ACL. This was different to the case when no
1919 redirection occurred. The value is now passed back in both cases.
1921 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1922 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1923 use login_cap.h, so on its own it isn't the distinguishing feature. The
1924 new name refers directly to the setclassresources() function.
1926 PH/65 Added configuration files for NetBSD3.
1928 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1930 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1931 is preferred over IPv4.
1933 PH/68 The bounce_return_message and bounce_return_body options were not being
1934 honoured for bounces generated during the reception of non-SMTP messages.
1935 In particular, this applied to messages rejected by the ACL. This bug has
1936 been fixed. However, if bounce_return_message is true and bounce_return_
1937 body is false, the headers that are returned for a non-SMTP message
1938 include only those that have been read before the error was detected.
1939 (In the case of an ACL rejection, they have all been read.)
1941 PH/69 The HTML version of the specification is now built in a directory called
1942 spec_html instead of spec.html, because the latter looks like a path with
1943 a MIME-type, and this confuses some software.
1945 PH/70 Catch two compiler warnings in sieve.c.
1947 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1948 function verify_get_ident() calls ip_connect() to connect a socket, but
1949 if the "connect()" function timed out, ip_connect() used to close the
1950 socket. However, verify_get_ident() also closes the socket later, and in
1951 between Exim writes to the log, which may get opened at this point. When
1952 the socket was closed in ip_connect(), the log could get the same file
1953 descriptor number as the socket. This naturally causes chaos. The fix is
1954 not to close the socket in ip_connect(); the socket should be closed by
1955 the function that creates it. There was only one place in the code where
1956 this was missing, in the iplookup router, which I don't think anybody now
1957 uses, but I've fixed it anyway.
1959 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1960 well as to direct DNS lookups. Otherwise the handling of names in host
1961 lists is inconsistent and therefore confusing.
1967 PH/01 Two changes to the default runtime configuration:
1969 (1) Move the checks for relay_from_hosts and authenticated clients from
1970 after to before the (commented out) DNS black list checks.
1972 (2) Add control=submission to the relay_from_hosts and authenticated
1973 clients checks, on the grounds that messages accepted by these
1974 statements are most likely to be submissions.
1976 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1978 (1) Generate an error if the third argument for the ${prvs expansion is
1981 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1984 (3) Reset the variables that are obtained from the first argument of
1985 ${prvscheck and used in the second argument before leaving the code,
1986 because their memory is reclaimed, so using them afterwards may do
1989 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1990 one (it's much easier than Tom thought :-).
1992 (5) Because of (4), we can now allow for the use of $prvscheck_result
1993 inside the third argument.
1995 PH/03 For some reason, the default setting of PATH when running a command from
1996 a pipe transport was just "/usr/bin". I have changed it to
1999 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2000 anything to be listed in the output from -bV.
2002 PH/05 When a filter generated an autoreply, the entire To: header line was
2003 quoted in the delivery log line, like this:
2005 => >A.N.Other <ano@some.domain> <original@ddress> ...
2007 This has been changed so that it extracts the operative address. There
2008 may be more than one such address. If so, they are comma-separated, like
2011 => >ano@some.domain,ona@other.domain <original@ddress> ...
2013 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2014 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2015 not being looked up in the rDNS to get a host name, Exim was showing the
2016 IP address twice in Received: lines, even though the IP addresses were
2017 identical. For example:
2019 Received: from [1.2.3.4] (helo=[1.2.3.4])
2021 However, if the real host name was known, it was omitting the HELO data
2022 if it matched the actual IP address. This has been tidied up so that it
2023 doesn't show the same IP address twice.
2025 PH/07 When both +timestamp and +memory debugging was on, the value given by
2026 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2027 called by the debug printing, thereby overwriting the timestamp buffer.
2028 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2030 PH/08 When the original message was included in an autoreply transport, it
2031 always said "this is a copy of the message, including all the headers",
2032 even if body_only or headers_only was set. It now gives an appropriate
2035 PH/09 Applied a patch from the Sieve maintainer which:
2037 o fixes some comments
2038 o adds the (disabled) notify extension core
2039 o adds some debug output for the result of if/elsif tests
2040 o points to the current vacation draft in the documentation
2041 and documents the missing references header update
2045 o fixes a bug in processing the envelope test (when testing
2046 multiple envelope elements, the last element determinted the
2049 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2050 Electronic Mail") by including:
2052 Auto-submitted: auto-generated
2054 in the messages that it generates (bounce messages and others, such as
2055 warnings). In the case of bounce messages for non-SMTP mesages, there was
2056 also a typo: it was using "Auto_submitted" (underscore instead of
2057 hyphen). Since every message generated by Exim is necessarily in response
2058 to another message, thes have all been changed to:
2060 Auto-Submitted: auto-replied
2062 in accordance with these statements in the RFC:
2064 The auto-replied keyword:
2066 - SHOULD be used on messages sent in direct response to another
2067 message by an automatic process,
2069 - MUST NOT be used on manually-generated messages,
2071 - MAY be used on Delivery Status Notifications (DSNs) and Message
2072 Disposition Notifications (MDNs),
2074 - MUST NOT be used on messages generated by automatic or periodic
2075 processes, except for messages which are automatic responses to
2078 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2079 to the default Received: header definition.
2081 PH/12 Added log selector acl_warn_skipped (default on).
2083 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2084 variables because (a) they are in the wrong storage pool and (b) even if
2085 they were copied, it wouldn't work properly because of the caching.
2087 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2088 checking when decoding. Apparently there are clients that generate
2089 overlong encoded strings. Why am I not surprised?
2091 PH/15 If the first argument of "${if match_address" was not empty, but did not
2092 contain an "@" character, Exim crashed. Now it writes a panic log message
2093 and treats the condition as false.
2095 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2097 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2098 extension "envelope-auth". The code is finished and in agreement with
2099 other implementations, but there is no documentation so far and in fact,
2100 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2101 not changing the active code.
2103 Print executed "if" and "elsif" statements when debugging is used. This
2104 helps a great deal to understand what a filter does.
2106 Document more things not specified clearly in RFC3028. I had all this
2107 sorted out, when out of a sudden new issues came to my mind. Oops."
2109 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2112 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2113 canonical form (as documented). However, after a host name lookup from
2114 the IP address, check_host() was doing a simple string comparison with
2115 addresses acquired from the DNS when checking that the found name did
2116 have the original IP as one of its addresses. Since any found IPv6
2117 addresses are likely to be in abbreviated form, the comparison could
2118 fail. Luckily, there already exists a function for doing the comparison
2119 by converting both addresses to binary, so now that is used instead of
2120 the text comparison.
2122 PH/20 There was another similar case to PH/19, when a complete host name was
2123 given in a host list; looking up its IP address could give an abbreviated
2124 form, whereas the current host's name might or might not be abbreviated.
2125 The same fix has been applied.
2131 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2132 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2135 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2136 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2138 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2140 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2141 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2142 Also, for "Auto-Submitted", treat anything other than "no" as
2143 non-personal, in accordance with RFC 3834. (Previously it treated
2144 anything starting "auto-" as non-personal.)
2146 TF/01 The control=submission/name=... option had a problem with syntax
2147 errors if the name included a slash character. The /name= option
2148 now slurps the rest of the string, so it can include any characters
2149 but it must come last in the list of options (after /sender_retain
2152 PH/05 Some modifications to the interface to the fake nameserver for the new
2160 TK/01 Added the "success_on_redirect" address verification option. See
2161 NewStuff for rationale and an example.
2163 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2165 PH/02 Patch to exigrep to allow it to work on syslog lines.
2167 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2168 fread() to skip over the body file's header line, because in Cygwin the
2169 header line is locked and is inaccessible.
2171 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2172 co-exist for some time) to make it clear that it is the Exim ID that is
2173 referenced, not the Message-ID: header line.
2175 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2176 string_format() function, because snprintf() does not exist on all
2179 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2180 ${stat: expansion item.
2182 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2183 protocol synchronization error", to keep the pedants happy.
2185 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2186 well as for IRIX systems, when gcc is being used. See the host.c source
2189 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2191 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2194 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2195 transport and to the smtp transport in LMTP mode.
2197 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2199 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2200 run a filter in a subprocess. This could lead to confusion in subsequent
2201 lookups in the parent process. There should also be a search_tidyup() at
2202 the end of the subprocess.
2204 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2205 only if the host matched helo_try_verify_hosts, which caused the
2206 verification to occur when the EHLO/HELO command was issued. The ACL just
2207 tested the remembered result. Now, if a previous verification attempt has
2208 not happened, "verify = helo" does it there and then.
2210 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2212 TK/03 Fix log output including CR from clamd.
2214 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2215 exist provoked a memory error which could cause a segfault.
2217 PH/15 Installed PCRE 6.2
2219 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2221 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2222 of the problem. Specifically, suggested +O2 rather than +O1 for the
2225 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2227 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2228 record, the port was not correctly logged when the outgoing_port log
2229 selector was set (it logged the transort's default port).
2231 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2232 fallback_hosts, and "hosts" in the smtp transport.
2234 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2235 host errors such as "Connection refused".
2237 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2238 authentication with radiusclient 0.4.9:
2240 - Error returned from rc_read_config was caught wrongly
2241 - Username/password not passed on to radius server due to wrong length.
2243 The presumption is that some radiusclient API changes for 4.51/PH/17
2244 were not taken care of correctly. The code is still untested by me (my
2245 Linux distribution still has 0.3.2 of radiusclient), but it was
2246 contributed by a Radius user.
2248 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2249 expanding the "port" option of the smtp transport.
2251 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2252 while reading a MIME header. Thanks to Tom Hughes for a patch.
2254 PH/24 Include config.h inside local_scan.h so that configuration settings are
2257 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2258 that in an ACL for RCPT (for example), you can examine exactly what was
2261 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2262 commands, but it was not correctly comparing the address with the actual
2263 client host address. Thus, it would show the EHLO address in Received:
2264 header lines when this was not necessary.
2266 PH/27 Added the % operator to ${eval:}.
2268 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2269 it should be ignoring failures (because with -C, for example, it has lost
2270 privilege). It wasn't ignoring creation failures other than "already
2273 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2276 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2277 because an input line was too long, either on its own, or by virtue of
2278 too many continuations, the temporary file was not being removed, and the
2279 return code was incorrect.
2281 PH/31 Missing "BOOL" in function definition in filtertest.c.
2283 PH/32 Applied Sieve patches from the maintainer.
2285 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2287 PH/33 Added "verify = not_blind".
2289 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2290 Local/Makefile (with some defaults set). These are used in built scripts
2291 such as exicyclog, but they have never been used in the exim_install
2292 script (though there are many overriding facilities there). I have
2293 arranged that the exim_install script now takes note of these two
2296 PH/35 Installed configuration files for Dragonfly.
2298 PH/36 When a locally submitted message by a trusted user did not contain a
2299 From: header, and the sender address was obtained from -f or from an SMTP
2300 MAIL command, and the trusted user did not use -F to supply a sender
2301 name, $originator_name was incorrectly used when constructing a From:
2302 header. Furthermore, $originator_name was used for submission mode
2303 messages from external hosts without From: headers in a similar way,
2304 which is clearly wrong.
2306 PH/37 Added control=suppress_local_fixups.
2308 PH/38 When log_selector = +received_sender was set, and the addition of the
2309 sender made the log line's construction buffer exactly full, or one byte
2310 less than full, an overflow happened when the terminating "\n" was
2313 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2314 when the result of a list match is failure because a DNS lookup failed.
2316 PH/40 RM_COMMAND is now used in the building process.
2318 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2319 the "build-* directories that it finds.
2321 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2322 address in a domain literal was a prefix of an interface address.
2324 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2325 when verifying a sender address, unless rewrite_headers is false.
2327 PH/44 Wrote a long comment about why errors_to addresses are verified as
2328 recipients, not senders.
2330 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2331 the ratelimit ACL was added.
2333 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2335 PH/46 Added extra information about PostgreSQL errors to the error string.
2337 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2338 avoiding the need to install special zones in a real server. This is
2339 backwards compatible; if it can't find the fake resolver, it drops back.
2340 Thus, both old and new test suites can be run.
2342 TF/02 Added util/ratelimit.pl
2344 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2345 clock is set back in time.
2347 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2348 Candler <B.Candler@pobox.com>.
2350 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2351 OK for addresses that are the result of redirections.
2353 PH/48 A number of further additions for the benefit of the new test suite,
2354 including a fake gethostbyname() that interfaces to the fake DNS resolver
2357 TF/06 The fix for widen_domains has also been applied to qualify_single and
2358 search_parents which are the other dnslookup options that can cause
2361 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2364 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2366 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2367 service" if it encountered an MX record with an empty target hostname.
2368 The message is now "an MX or SRV record indicated no SMTP service".
2370 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2371 if the DNS of the sending site is misconfigured. This is quite a
2372 common situation. This change restores the behaviour of treating a
2373 helo verification defer as a failure.
2375 PH/51 If self=fail was set on a router, the bounce message did not include the
2376 actual error message.
2382 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2384 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2385 command itself ended in error, the underlying message about the transport
2386 filter timeout was being overwritten with the pipe command error. Now the
2387 underlying error message should be appended to the second error message.
2389 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2391 PH/02 Reduce the amount of output that "make" produces by default. Full output
2392 can still be requested.
2394 PH/03 The warning log line about a condition test deferring for a "warn" verb
2395 was being output only once per connection, rather than after each
2396 occurrence (because it was using the same function as for successful
2397 "warn" verbs). This seems wrong, so I have changed it.
2399 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2400 it should not have, which might have caused a crash in the right
2401 circumstances, but probably never did.
2403 PH/04 Installed a modified version of Tony Finch's patch to make submission
2404 mode fix the return path as well as the Sender: header line, and to
2405 add a /name= option so that you can make the user's friendly name appear
2408 TF/03 Added the control = fakedefer ACL modifier.
2410 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2411 Mark Lowes for thorough testing.
2413 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2415 TK/03 Merged latest SRS patch from Miles Wilton.
2417 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2418 with the definition in sysexits.h (which is #included earlier).
2419 Fortunately, Exim does not actually use EX_OK. The code used to try to
2420 preserve the sysexits.h value, by assumimg that macro definitions were
2421 scanned for macro replacements. I have been disabused of this notion,
2422 so now the code just undefines EX_OK before #including unistd.h.
2424 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2425 in the smtp transport. When a block could not be written in a single
2426 write() function, the timeout was being re-applied to each part-write.
2427 This seems wrong - if the receiver was accepting one byte at a time it
2428 would take for ever. The timeout is now adjusted when this happens. It
2429 doesn't have to be particularly precise.
2431 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2432 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2434 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2435 without a domain if the check to <postmaster@domain> fails.
2437 SC/01 Eximstats: added -xls and the ability to specify output files
2438 (patch written by Frank Heydlauf).
2440 SC/02 Eximstats: use FileHandles for outputing results.
2442 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2444 SC/04 Eximstats: fixed display of large numbers with -nvr option
2446 SC/05 Eximstats: fixed merging of reports with empty tables.
2448 SC/06 Eximstats: added the -include_original_destination flag
2450 SC/07 Eximstats: removed tabs and trailing whitespace.
2452 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2454 TK/06 MBOX spool code: Add real "From " MBOX separator line
2455 so the .eml file is really in mbox format (even though
2456 most programs do not really care). Patch from Alex Miller.
2458 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2459 The latter is generated from $received_to and is only set if the
2460 message has one envelope recipient. SA can use these headers,
2461 obviously out-of-the-box. Patch from Alex Miller.
2463 PH/08 The ${def test on a variable was returning false if the variable's
2464 value was "0", contrary to what the specification has always said!
2465 The result should be true unless the variable is empty.
2467 PH/09 The syntax error of a character other than { following "${if
2468 def:variable_name" (after optional whitespace) was not being diagnosed.
2469 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2470 accidental colon was present, for example, could give incorrect results.
2472 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2473 result is used (not including appendfile, where other changes are about
2476 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2477 This involved changing a lot of size variables from int to off_t. It
2478 should work with maildirs and everything.
2480 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2481 spamd dying while we are connected to it.
2483 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2486 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2487 function for BDB 4.3.
2489 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2491 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2492 the sources for PCRE have been re-arranged and more files are now
2495 PH/15 The code I had for printing potentially long long variables in PH/11
2496 above was not the best (it lost precision). The length of off_t variables
2497 is now inspected at build time, and an appropriate printing format (%ld
2498 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2499 to be "long long int" or "long int". This is needed for the internal
2500 formatting function string_vformat().
2502 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2503 the configuration file to be ":syslog", then the script "guesses" where
2504 the logs files are, rather than using the compiled in default. In our
2505 case the guess is not the same as the compiled default, so the script
2506 suddenly stopped working when I started to use syslog. The patch checks
2507 to see if log_file_path is "". If so, it attempts to read it from exim
2508 with no configuration file to get the compiled in version, before it
2509 falls back to the previous guessing code."
2511 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2512 implementing BATV in an Exim configuration. See NewStuff for the gory
2515 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2516 Makefile that are specific to HP-UX.
2518 PH/18 If the "use_postmaster" option was set for a recipient callout together
2519 with the "random" option, the postmaster address was used as the MAIL
2520 FROM address for the random test, but not for the subsequent recipient
2521 test. It is now used for both.
2523 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2524 patch removes a few documentation additions to RFC 3028, because the
2525 latest draft now contains them. It adds the new en;ascii-case comparator
2526 and a new error check for 8bit text in MIME parts. Comparator and
2527 require names are now matched exactly. I enabled the subaddress
2528 extension, but it is not well tested yet (read: it works for me)."
2530 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2531 rework some of the code of TK/09 above to avoid the hardwired use of
2532 "%lld" and "long long". Replaced the call to snprintf() with a call to
2535 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2536 records point to non-existent hosts", "retry timeout exceeded", and
2537 "retry time not reached for any host after a long failure period".
2539 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2540 experimental DomainKeys support:
2542 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2543 (2) On an error such as an illegally used "control", the wrong name for
2544 the control was given.
2546 These problems did NOT occur unless DomainKeys support was compiled.
2548 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2550 PH/24 Added ${if match_ip condition.
2552 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2553 databases so that it will be absolutely obvious if a crash occurs in the
2554 DB library. This is a regular occurrence (often caused by mis-matched
2557 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2558 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2559 picked up on a user's system that detects such things. There doesn't seem
2560 to be a gcc warning option for this - only an attribute that has to be
2561 put on the function's prototype. It seems that in Fedora Core 4 they have
2562 set this on a number of new functions. No doubt there will be more in due
2565 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2566 specify a transport. However, if an address that was verified by such a
2567 router was the subject of a callout, Exim crashed because it tried to
2568 read the rcpt_include_affixes from the non-existent transport. Now it
2569 just assumes that the setting of that option is false. This bug was
2570 introduced by 4.51/PH/31.
2572 PH/28 Changed -d+all to exclude +memory, because that information is very
2573 rarely of interest, but it makes the output a lot bigger. People tend to
2574 do -d+all out of habit.
2576 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2577 code in os-type was giving problems when libc.so lives in lib64, like on
2580 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2581 aren't the modern standard, and it seems that some systems' include files
2582 don't always have them. Exim was already checking for some of the newer
2583 ones like T_AAAA, and defining it itself. I've added checks for all the
2584 record types that Exim uses.
2586 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2587 not automatically generating a new one, as it is supposed to. This
2588 prevented TLS from working. If the file did exist, but contained invalid
2589 data, a new version was generated, as expected. It was only the case of a
2590 non-existent file that was broken.
2592 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2593 with a change in libdomainkeys > 0.64.
2595 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2596 from DNS. If the selector record carries the flag, it now has
2597 precedence over the domain-wide flag.
2599 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2601 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2602 the use of a port name that isn't defined in /etc/services) occurred, the
2603 message was deferred as in a normal delivery, and thus remained on the
2604 spool, instead of being failed because of the mua_wrapper setting. This
2605 is now fixed, and I tidied up some of the mua_wrapper messages at the
2608 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2609 the messages in a hash of arrays rather than using individual hashes.
2610 This is a bit cleaner and results in dramatic memory savings, albeit
2611 at a slight CPU cost.
2613 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2614 as requested by Marc Sherman.
2616 SC/10 Eximstats: added histograms for user specified patterns as requested
2619 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2621 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2622 fopen() in the content-scanning modules that did not already have it.
2628 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2629 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2631 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2633 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2635 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2636 to test Sieve filters that use "vacation".
2638 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2639 that changes the way the GnuTLS parameters are stored in the cache file.
2640 The new format can be generated externally. For backward compatibility,
2641 if the data in the cache doesn't make sense, Exim assumes it has read an
2642 old-format file, and it generates new data and writes a new file. This
2643 means that you can't go back to an older release without removing the
2646 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2647 work if there are any delivery delays because "one_time" forces the
2648 parent to be marked "delivered", so its unseen clone is never tried
2649 again. For this reason, Exim now forbids the simultaneous setting of
2652 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2653 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2654 entry if you want to know the details. The fix, however, neglected to
2655 consider the case when local delivery batching is involved. The test for
2656 "previously delivered" was not happening when checking to see if an
2657 address could be batched with a previous (undelivered) one; under
2658 certain circumstances this could lead to multiple deliveries to the same
2661 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2662 in its include files, and this causes problems building Exim.
2664 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2665 header_syntax) but Exim was just ignoring anything given after a slash.
2666 In particular, this caused confusion with an attempt to use "verify =
2667 reverse_host_lookup/defer_ok". An error is now given when options are
2668 supplied for verify items that do not have them. (Maybe reverse_host_
2669 lookup should have a defer_ok option, but that's a different point.)
2671 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2672 defined by RFC 821) to 2048, because there were problems with some AUTH
2673 commands, and RFC 1869 says the size should be increased for extended
2674 SMTP commands that take arguments.
2676 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2679 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2680 "unknown" error; now it says that the functionality isn't in the binary.
2682 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2683 an address' error message when a string expansion fails (syntax or
2684 whatever). Otherwise the password may appear in the log. Following change
2685 PH/42 below, there is no longer a chance of it appearing in a bounce
2688 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2690 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2691 of its addresses was ever tried. (Bugzilla bug #2.)
2693 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2694 the result incorrectly in the debug output. (It correctly added a newline
2695 to what was transported.)
2697 TF/01 Added $received_time.
2699 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2700 commented out examples of how to interface to a virus scanner and to
2701 SpamAssassin. Also added commented examples of av_scanner and
2702 spamd_address settings.
2704 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2705 and controls are allowed in which ACLs. There were a couple of minor
2706 errors. Some of the entries in the conditions table (which is a table of
2707 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2708 negation of where the condition IS allowed.
2710 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2712 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2713 header file does not have a version number, so I've had to invent a new
2714 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2715 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2716 radiusclient), but it was contributed by a Radius user.
2718 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2719 files or directories, for OpenSSL.
2721 PH/19 When an Exim process that is running as root has to create an Exim log
2722 file, it does so in a subprocess that runs as exim:exim so as to get the
2723 ownership right at creation (otherwise, other Exim processes might see
2724 the file with the wrong ownership). There was no test for failure of this
2725 fork() call, which would lead to the process getting stuck as it waited
2726 for a non-existent subprocess. Forks do occasionally fail when resources
2727 run out. I reviewed all the other calls to fork(); they all seem to check
2730 PH/20 When checking for unexpected SMTP input at connect time (before writing
2731 the banner), Exim was not dealing correctly with a non-positive return
2732 from the read() function. If the client had disconnected by this time,
2733 the result was a log entry for a synchronization error with an empty
2734 string after "input=" when read() returned zero. If read() returned -1
2735 (an event I could not check), uninitialized data bytes were printed.
2736 There were reports of junk text (parts of files, etc) appearing after
2739 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2741 PH/22 Added support for macro redefinition, and (re)definition in between
2742 driver and ACL definitions.
2744 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2745 forgetting to use the resulting value; it was using the unexpanded value.
2747 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2748 hadn't been configured. The fix is from Juergen Kreileder, who
2749 understands it better than I do:
2751 "Here's what I see happening with three configured cyrus_sasl
2752 authenticators configured (plain, login, cram-md5):
2754 On startup auth_cyrus_sasl_init() gets called for each of these.
2755 This means three calls to sasl_listmech() without a specified mech_list.
2756 => SASL tests which mechs of all available mechs actually work
2757 => three warnings about OTP not working
2758 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2760 With the patch, sasl_listmech() also gets called three times. But now
2761 SASL's mech_list option is set to the server_mech specified in the the
2762 authenticator. Or in other words, the answer from sasl_listmech()
2763 gets limited to just the mech you're testing for (which is different
2765 => the return list contains just 'plain' or 'login', 'cram-md5' or
2766 nothing depending on the value of ob->server_mech.
2768 I've just tested the patch: Authentication still works fine,
2769 unavailable mechs specified in the exim configuration are still
2770 caught, and the auth.log warnings about OTP are gone."
2772 PH/25 When debugging is enabled, the contents of the command line are added
2773 to the debugging output, even when log_selector=+arguments is not
2776 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2777 answer is "GNU", and only if the return is "GNU/something" is the answer
2780 PH/27 $acl_verify_message is now set immediately after the failure of a
2781 verification in an ACL, and so is available in subsequent modifiers. In
2782 particular, the message can be preserved by coding like this:
2784 warn !verify = sender
2785 set acl_m0 = $acl_verify_message
2787 Previously, $acl_verify_message was set only while expanding "message"
2788 and "log_message" when a very denied access.
2790 PH/28 Modified OS/os.c-Linux with
2792 -#ifndef OS_LOAD_AVERAGE
2793 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2795 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2796 nomenclature these days.)
2798 PH/29 Installed patch from the Sieve maintainer that adds the options
2799 sieve_useraddress and sieve_subaddress to the redirect router.
2801 PH/30 In these circumstances:
2802 . Two addresses routed to the same list of hosts;
2803 . First host does not offer TLS;
2804 . First host accepts first address;
2805 . First host gives temporary error to second address;
2806 . Second host offers TLS and a TLS session is established;
2807 . Second host accepts second address.
2808 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2809 and peerdn, if requested) that were in fact used only for the second
2812 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2813 attention to any local part prefix or suffix that was matched by the
2814 router that accepted the address. It now behaves in the same way as it
2815 does for delivery: the affixes are removed from the local part unless
2816 rcpt_include_affixes is set on the transport.
2818 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2819 timeout during the DATA phase of an incoming message.
2821 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2822 applied a patch sanctioned by the Sieve maintainer.
2824 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2825 the uid or gid is negative. A case of a negative gid caused this to be
2826 noticed. The fix allows for either to be negative.
2828 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2829 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2832 PH/36 The change PH/12 above was broken. Fixed it.
2834 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2835 the grounds that routing the same address twice would always produce the
2836 same answer. This might have been true once, but it is certainly no
2837 longer true now. Routing a child address may depend on the previous
2838 routing that produced that child. Some complicated redirection strategies
2839 went wrong when messages had multiple recipients, and made Exim's
2840 behaviour dependent on the order in which the addresses were given.
2842 I have moved the duplicate checking until after the routing is complete.
2843 Exim scans the addresses that are assigned to local and remote
2844 transports, and removes any duplicates. This means that more work will be
2845 done, as duplicates will always all be routed, but duplicates are
2846 presumably rare, so I don't expect this is of any significance.
2848 For deliveries to pipes, files, and autoreplies, the duplicate checking
2849 still happens during the routing process, since they are not going to be
2852 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2853 It corrects a timeout issue with spamd. This is Ian's comment: "The
2854 background is that sometimes spamd either never reads data from a
2855 connection it has accepted, or it never writes response data. The exiscan
2856 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2857 blindly assumes that writes won't block so it may never time out."
2859 PH/39 Allow G after quota size as well as K and M.
2861 PH/40 The value set for $authenticated_id in an authenticator may not contain
2862 binary zeroes or newlines because the value is written to log lines and
2863 to spool files. There was no check on this. Now the value is run through
2864 the string_printing() function so that such characters are converted to
2865 printable escape sequences.
2867 PH/41 $message_linecount is a new variable that contains the total number of
2868 lines in the message. Compare $body_linecount, which is the count for the
2871 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2872 bounce and delay warning messages, except in certain special cases, which
2875 (a) An SMTP error message from a remote host;
2876 (b) A message specified in a :fail: redirection;
2877 (c) A message specified in a "fail" command in a system filter;
2878 (d) A message specified in a FAIL return from the queryprogram router;
2879 (e) A message specified by the cannot_route_message router option.
2881 In these cases only, Exim does include the error details in bounce and
2882 warning messages. There are also a few cases where bland messages such
2883 as "unrouteable address" or "local delivery error" are given.
2885 PH/43 $value is now also set for the "else" part of a ${run expansion.
2887 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2888 being worked on, but at least Exim now implements the latest version to
2891 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2892 process to complete was treated as a delivery failure, a timeout while
2893 writing the message to the pipe was logged, but erroneously treated as a
2894 successful delivery. Such timeouts include transport filter timeouts. For
2895 consistency with the overall process timeout, these timeouts are now
2896 treated as errors, giving rise to delivery failures by default. However,
2897 there is now a new Boolean option for the pipe transport called
2898 timeout_defer, which, if set TRUE, converts the failures into defers for
2899 both kinds of timeout. A transport filter timeout is now identified in
2902 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2903 systems where "make" and "gmake" are different, calling "gmake" at top
2904 level broke things. I've arranged for the value of $(MAKE) to be passed
2905 from the Makefile to this script so that it can call the same version of
2909 A note about Exim versions 4.44 and 4.50
2910 ----------------------------------------
2912 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2913 changes of various kinds. As a consequence, a big documentation update was
2914 needed. This delayed the release for rather longer than seemed good, especially
2915 in the light of a couple of (minor) security issues. Therefore, the changes
2916 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2917 release. So 4.44 and 4.50 are in effect two different branches that both start
2920 I have left the 4.50 change log unchanged; it contains all the changes since
2921 4.43. The change log for 4.44 is below; many of its items are identical to
2922 those for 4.50. This seems to be the most sensible way to preserve the
2923 historical information.
2929 1. Minor wording change to the doc/README.SIEVE file.
2931 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2932 computation of the current number of files was incorrect.
2934 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2935 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2938 4. Give more explanation in the error message when the command for a transport
2939 filter fails to execute.
2941 5. There are several places where Exim runs a non-Exim command in a
2942 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2943 was being done only for the command run by the queryprogram router. It is
2944 now done for all such subprocesses. The other cases are: ${run, transport
2945 filters, and the commands run by the lmtp and pipe transports.
2947 6. Added CONFIGURE_GROUP build-time option.
2949 7. Some older OS have a limit of 256 on the maximum number of file
2950 descriptors. Exim was using setrlimit() to set 1000 as a large value
2951 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2952 systems. I've change it so that if it can't get 1000, it tries for 256.
2954 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2955 was an oversight, and furthermore, ever since the addition of extra
2956 controls (e.g. 4.43/32), the checks on when to allow different forms of
2957 "control" were broken. There should now be diagnostics for all cases when a
2958 control that does not make sense is encountered.
2960 9. Added the /retain_sender option to "control=submission".
2962 10. $recipients is now available in the predata ACL (oversight).
2964 11. Tidy the search cache before the fork to do a delivery from a message
2965 received from the command line. Otherwise the child will trigger a lookup
2966 failure and thereby defer the delivery if it tries to use (for example) a
2967 cached ldap connection that the parent has called unbind on.
2969 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2970 of $address_data from the recipient verification was clobbered by the
2971 sender verification.
2973 13. The value of address_data from a sender verification is now available in
2974 $sender_address_data in subsequent conditions in the ACL statement.
2976 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2978 15. Added a new option "connect=<time>" to callout options, to set a different
2981 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2982 was its contents. (It was OK if the option was not defined at all.)
2984 17. A "Completed" log line is now written for messages that are removed from
2985 the spool by the -Mrm option.
2987 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2988 information about exactly what failed.
2990 19. Added -dd to debug only the daemon process.
2992 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2993 handles timeouts, both on the server side and network timeouts. Renamed the
2994 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2996 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2997 It is now set to "smtps".
2999 22. $host_address is now set to the target address during the checking of
3000 ignore_target_hosts.
3002 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3003 being passed; this would have caused $sender_host_name to have been used if
3004 matching the list had actually called for a host name (not very likely,
3005 since this list is usually IP addresses). A host name is now passed as
3008 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3009 code that specifies a non-restarting handler (typically sigaction() in
3010 modern systems) in an attempt to fix a rare and obscure crash bug.
3012 25. Narrowed the window for a race in the daemon that could cause it to ignore
3013 SIGCHLD signals. This is not a major problem, because they are used only to
3014 wake it up if nothing else does.
3016 26. A malformed maildirsize file could cause Exim to calculate negative values
3017 for the mailbox size or file count. Odd effects could occur as a result.
3018 The maildirsize information is now recalculated if the size or filecount
3021 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3022 support for a long time. Removed HAVE_SYS_VFS_H.
3024 28. Installed the latest version of exipick from John Jetmore.
3026 29. In an address list, if the pattern was not a regular expression, an empty
3027 subject address (from a bounce message) matched only if the pattern was an
3028 empty string. Non-empty patterns were not even tested. This was the wrong
3029 because it is perfectly reasonable to use an empty address as part of a
3030 database query. An empty address is now tested by patterns that are
3031 lookups. However, all the other forms of pattern expect the subject to
3032 contain a local part and a domain, and therefore, for them, an empty
3033 address still always fails if the pattern is not itself empty.
3035 30. Exim went into a mad DNS loop when attempting to do a callout where the
3036 host was specified on an smtp transport, and looking it up yielded more
3037 than one IP address.
3039 31. Re-factored the code for checking spool and log partition space into a
3040 function that finds that data and another that does the check. The former
3041 is then used to implement four new variables: $spool_space, $log_space,
3042 $spool_inodes, and $log_inodes.
3044 32. The RFC2047 encoding function was originally intended for short strings
3045 such as real names; it was not keeping to the 75-character limit for
3046 encoded words that the RFC imposes. It now respects the limit, and
3047 generates multiple encoded words if necessary. To be on the safe side, I
3048 have increased the buffer size for the ${rfc2047: expansion operator from
3051 33. It is now permitted to omit both strings after an "if" condition; if the
3052 condition is true, the result is "true". As before, when the second string
3053 is omitted, a false condition yields an empty string. This makes it less
3054 cumbersome to write custom ACL and router conditions.
3056 34. Failure to deliver a bounce message always caused it to be frozen, even if
3057 there was an errors_to setting on the router. The errors_to setting is now
3060 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3061 canonical form (fully expanded) before being placed in
3062 $sender_host_address.
3064 36. The table in the code that translates DNS record types into text (T_A to
3065 "A" for instance) was missing entries for NS and CNAME. It is just possible
3066 that this could have caused confusion if both these types were looked up
3067 for the same domain, because the text type is used as part of Exim's
3068 per-process caching. But the chance of anyone hitting this buglet seems
3071 37. The dnsdb lookup has been extended in a number of ways.
3073 (1) There is a new type, "zns", which walks up the domain tree until it
3074 finds some nameserver records. It should be used with care.
3076 (2) There is a new type, "mxh", which is like "mx" except that it returns
3077 just the host names, not the priorities.
3079 (3) It is now possible to give a list of domains (or IP addresses) to be
3080 looked up. The behaviour when one of the lookups defers can be
3081 controlled by a keyword.
3083 (4) It is now possible to specify the separator character for use when
3084 multiple records are returned.
3086 38. The dnslists ACL condition has been extended: it is now possible to supply
3087 a list of IP addresses and/or domains to be looked up in a particular DNS
3090 39. Added log_selector=+queue_time_overall.
3092 40. When running the queue in the test harness, wait just a tad after forking a
3093 delivery process, to get repeatability of debugging output.
3095 41. Include certificate and key file names in error message when GnuTLS fails
3096 to set them up, because the GnuTLS error message doesn't include the name
3097 of the failing file when there is a problem reading it.
3099 42. Allow both -bf and -bF in the same test run.
3101 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3103 44. The "Exiscan patch" is now merged into the mainline Exim source.
3105 45. Sometimes the final signoff response after QUIT could fail to get
3106 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3107 < 0 before doing a fflush(). This bug looks as though it goes back to the
3108 introduction of TLS in release 3.20, but "sometimes" must have been rare
3109 because the tests only now provoked it.
3111 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3112 (this can affect the format of dates).
3114 47. exim_tidydb, when checking for the continued existence of a message for
3115 which it has found a message-specific retry record, was not finding
3116 messages that were in split spool directories. Consequently, it was
3117 deleting retry records that should have stayed in existence.
3119 48. Steve fixed some bugs in eximstats.
3121 49. The SPA authentication driver was not abandoning authentication and moving
3122 on to the next authenticator when an expansion was forced to fail,
3123 contradicting the general specification for all authenticators. Instead it
3124 was generating a temporary error. It now behaves as specified.
3126 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3127 (the order specifies the preference for clients). The order is now AES256,
3128 AES128, 3DES, ARCFOUR128.
3130 51. Small patch to Sieve code - explicitly set From: when generating an
3133 52. Exim crashed if a remote delivery caused a very long error message to be
3134 recorded - for instance if somebody sent an entire SpamAssassin report back
3135 as a large number of 550 error lines. This bug was coincidentally fixed by
3136 increasing the size of one of Exim's internal buffers (big_buffer) that
3137 happened as part of the Exiscan merge. However, to be on the safe side, I
3138 have made the code more robust (and fixed the comments that describe what
3141 53. Now that there can be additional text after "Completed" in log lines (if
3142 the queue_time_overall log selector is set), a one-byte patch to exigrep
3143 was needed to allow it to recognize "Completed" as not the last thing in
3146 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3147 patch that reportedly fixes this has been added. I am not expert enough to
3148 create a test for it. This is what the patch creator wrote:
3150 "I found a little strange behaviour of ldap code when working with
3151 Windows 2003 AD Domain, where users was placed in more than one
3152 Organization Units. When I tried to give exim partial DN, the exit code
3153 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3154 But simultaneously result of request was absolutely normal ldap result,
3155 so I produce this patch..."
3157 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3158 REFERENCE, so I have modified the code to exclude the patch when that macro
3161 55. Some experimental protocols are using DNS PTR records for new purposes. The
3162 keys for these records are domain names, not reversed IP addresses. The
3163 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3164 leaves it alone. Component reversal etc. now happens only for IP addresses.
3167 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3169 57. Double the size of the debug message buffer (to 2048) so that more of very
3170 long debug lines gets shown.
3172 58. The exicyclog utility now does better if the number of log files to keep
3173 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3175 59. Two changes related to the smtp_active_hostname option:
3177 (1) $smtp_active_hostname is now available as a variable.
3178 (2) The default for smtp_banner uses $smtp_active_hostname instead
3179 of $primary_hostname.
3181 60. The host_aton() function is supposed to be passed a string that is known
3182 to be a valid IP address. However, in the case of IPv6 addresses, it was
3183 not checking this. This is a hostage to fortune. Exim now panics and dies
3184 if the condition is not met. A case was found where this could be provoked
3185 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3186 components; fortuitously, this particular loophole had already been fixed
3187 by change 4.50/55 above.
3189 If there are any other similar loopholes, the new check in host_aton()
3190 itself should stop them being exploited. The report I received stated that
3191 data on the command line could provoke the exploit when Exim was running as
3192 exim, but did not say which command line option was involved. All I could
3193 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3194 running as the user.
3197 61. There was a buffer overflow vulnerability in the SPA authentication code
3198 (which came originally from the Samba project). I have added a test to the
3199 spa_base64_to_bits() function which I hope fixes it.
3202 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3203 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3205 63. The daemon start-up calls getloadavg() while still root for those OS that
3206 need the first call to be done as root, but it missed one case: when
3207 deliver_queue_load_max is set with deliver_drop_privilege. This is
3208 necessary for the benefit of the queue runner, because there is no re-exec
3209 when deliver_drop_privilege is set.
3211 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3212 This has been fixed.
3214 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3215 was in use, was not putting the data itself into the right store pool;
3216 consequently, it could be overwritten for a subsequent message in the same
3217 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3220 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3222 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3223 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3224 the function were treating the return as a boolean value, which happened to
3225 work because 0=false and not-0=true, but is not correct code.
3227 68. The host_aton() function was not handling scoped IPv6 addresses (those
3228 with, for example, "%eth0" on the end) correctly.
3230 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3231 negated items (that is, ~something) in unsigned ints. Some compilers
3232 apparently mutter when there is no cast.
3234 70. If an address verification called from an ACL failed, and did not produce a
3235 user-specific message (i.e. there was only a "system" message), nothing was
3236 put in $acl_verify_message. In this situation, it now puts the system
3239 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3240 session; change 4.31/43 added the unwanted input to the log line - except
3241 that it did not do this in the start of session case. It now does.
3243 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3244 This is wrong and can cause the other end to generate a synchronization
3245 error if it is another Exim or anything else that does the synchronization
3246 check. A QUIT command is no longer sent after a timeout.
3248 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3249 during host lookups.
3251 74. The defer_ok option of callout verification was not working if it was used
3252 when verifying addresses in header lines, that is, for this case:
3254 verify = header_sender/callout=defer_ok
3256 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3257 those file descriptors could be used for SMTP connections. If anything
3258 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3259 could be sent to the SMTP client, causing chaos. The daemon now opens
3260 stdin, stdout, and stderr to /dev/null when it puts itself into the
3263 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3264 log by default. The user can override this with suitable Perl magic.
3266 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3267 the log message when discard triggers, was not working for the DATA ACL or
3268 for the non-SMTP ACL.
3270 78. Error message wording change in sieve.c.
3272 79. If smtp_accept_max_per_host was set, the number of connections could be
3273 restricted to fewer than expected, because the daemon was trying to set up
3274 a new connection before checking whether the processes handling previous
3275 connections had finished. The check for completed processes is now done
3276 earlier. On busy systems, this bug wouldn't be noticed because something
3277 else would have woken the daemon, and it would have reaped the completed
3280 80. If a message was submitted locally by a user whose login name contained one
3281 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3282 It caused a spool format error. I have fixed the spool reading code. A
3283 related problem was that the "from" clause in the Received: line became
3284 illegal because of the space(s). It is now covered by ${quote_local_part.
3286 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3289 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3291 83. Patch from Sieve maintainer to fix unterminated string problem in
3292 "vacation" handling.
3294 84. Some minor changes to the Linux configuration files to help with other
3295 OS variants using glibc.
3297 85. One more patch for Sieve to update vacation handling to latest spec.
3300 ----------------------------------------------------
3301 See the note above about the 4.44 and 4.50 releases.
3302 ----------------------------------------------------
3308 1. Change 4.43/35 introduced a bug that caused file counts to be
3309 incorrectly computed when quota_filecount was set in an appendfile
3312 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3313 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3316 3. Give more explanation in the error message when the command for a transport
3317 filter fails to execute.
3319 4. There are several places where Exim runs a non-Exim command in a
3320 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3321 was being done only for the command run by the queryprogram router. It is
3322 now done for all such subprocesses. The other cases are: ${run, transport
3323 filters, and the commands run by the lmtp and pipe transports.
3325 5. Some older OS have a limit of 256 on the maximum number of file
3326 descriptors. Exim was using setrlimit() to set 1000 as a large value
3327 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3328 systems. I've change it so that if it can't get 1000, it tries for 256.
3330 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3331 was an oversight, and furthermore, ever since the addition of extra
3332 controls (e.g. 4.43/32), the checks on when to allow different forms of
3333 "control" were broken. There should now be diagnostics for all cases when a
3334 control that does not make sense is encountered.
3336 7. $recipients is now available in the predata ACL (oversight).
3338 8. Tidy the search cache before the fork to do a delivery from a message
3339 received from the command line. Otherwise the child will trigger a lookup
3340 failure and thereby defer the delivery if it tries to use (for example) a
3341 cached ldap connection that the parent has called unbind on.
3343 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3344 of $address_data from the recipient verification was clobbered by the
3345 sender verification.
3347 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3348 was its contents. (It was OK if the option was not defined at all.)
3350 11. A "Completed" log line is now written for messages that are removed from
3351 the spool by the -Mrm option.
3353 12. $host_address is now set to the target address during the checking of
3354 ignore_target_hosts.
3356 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3357 being passed; this would have caused $sender_host_name to have been used if
3358 matching the list had actually called for a host name (not very likely,
3359 since this list is usually IP addresses). A host name is now passed as
3362 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3363 code that specifies a non-restarting handler (typically sigaction() in
3364 modern systems) in an attempt to fix a rare and obscure crash bug.
3366 15. Narrowed the window for a race in the daemon that could cause it to ignore
3367 SIGCHLD signals. This is not a major problem, because they are used only to
3368 wake it up if nothing else does.
3370 16. A malformed maildirsize file could cause Exim to calculate negative values
3371 for the mailbox size or file count. Odd effects could occur as a result.
3372 The maildirsize information is now recalculated if the size or filecount
3375 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3376 support for a long time. Removed HAVE_SYS_VFS_H.
3378 18. Updated exipick to current release from John Jetmore.
3380 19. Allow an empty sender to be matched against a lookup in an address list.
3381 Previously the only cases considered were a regular expression, or an
3384 20. Exim went into a mad DNS lookup loop when doing a callout where the
3385 host was specified on the transport, if the DNS lookup yielded more than
3388 21. The RFC2047 encoding function was originally intended for short strings
3389 such as real names; it was not keeping to the 75-character limit for
3390 encoded words that the RFC imposes. It now respects the limit, and
3391 generates multiple encoded words if necessary. To be on the safe side, I
3392 have increased the buffer size for the ${rfc2047: expansion operator from
3395 22. Failure to deliver a bounce message always caused it to be frozen, even if
3396 there was an errors_to setting on the router. The errors_to setting is now
3399 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3400 canonical form (fully expanded) before being placed in
3401 $sender_host_address.
3403 24. Updated eximstats to version 1.33
3405 25. Include certificate and key file names in error message when GnuTLS fails
3406 to set them up, because the GnuTLS error message doesn't include the name
3407 of the failing file when there is a problem reading it.
3409 26. Expand error message when OpenSSL has problems setting up cert/key files.
3412 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3413 (this can affect the format of dates).
3415 28. exim_tidydb, when checking for the continued existence of a message for
3416 which it has found a message-specific retry record, was not finding
3417 messages that were in split spool directories. Consequently, it was
3418 deleting retry records that should have stayed in existence.
3420 29. eximstats updated to version 1.35
3421 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3422 1.35 - bugfix such that pie charts by volume are generated correctly
3424 30. The SPA authentication driver was not abandoning authentication and moving
3425 on to the next authenticator when an expansion was forced to fail,
3426 contradicting the general specification for all authenticators. Instead it
3427 was generating a temporary error. It now behaves as specified.
3429 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3430 (the order specifies the preference for clients). The order is now AES256,
3431 AES128, 3DES, ARCFOUR128.
3433 31. Small patch to Sieve code - explicitly set From: when generating an
3436 32. Exim crashed if a remote delivery caused a very long error message to be
3437 recorded - for instance if somebody sent an entire SpamAssassin report back
3438 as a large number of 550 error lines. This bug was coincidentally fixed by
3439 increasing the size of one of Exim's internal buffers (big_buffer) that
3440 happened as part of the Exiscan merge. However, to be on the safe side, I
3441 have made the code more robust (and fixed the comments that describe what
3444 33. Some experimental protocols are using DNS PTR records for new purposes. The
3445 keys for these records are domain names, not reversed IP addresses. The
3446 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3447 leaves it alone. Component reversal etc. now happens only for IP addresses.
3450 34. The host_aton() function is supposed to be passed a string that is known
3451 to be a valid IP address. However, in the case of IPv6 addresses, it was
3452 not checking this. This is a hostage to fortune. Exim now panics and dies
3453 if the condition is not met. A case was found where this could be provoked
3454 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3455 components; fortuitously, this particular loophole had already been fixed
3456 by change 4.50/55 or 4.44/33 above.
3458 If there are any other similar loopholes, the new check in host_aton()
3459 itself should stop them being exploited. The report I received stated that
3460 data on the command line could provoke the exploit when Exim was running as
3461 exim, but did not say which command line option was involved. All I could
3462 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3463 running as the user.
3466 35. There was a buffer overflow vulnerability in the SPA authentication code
3467 (which came originally from the Samba project). I have added a test to the
3468 spa_base64_to_bits() function which I hope fixes it.
3471 36. The daemon start-up calls getloadavg() while still root for those OS that
3472 need the first call to be done as root, but it missed one case: when
3473 deliver_queue_load_max is set with deliver_drop_privilege. This is
3474 necessary for the benefit of the queue runner, because there is no re-exec
3475 when deliver_drop_privilege is set.
3477 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3478 was in use, was not putting the data itself into the right store pool;
3479 consequently, it could be overwritten for a subsequent message in the same
3480 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3483 38. Sometimes the final signoff response after QUIT could fail to get
3484 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3485 < 0 before doing a fflush(). This bug looks as though it goes back to the
3486 introduction of TLS in release 3.20, but "sometimes" must have been rare
3487 because the tests only now provoked it.
3493 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3494 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3495 Now it returns an integer. A number of calls were still expecting a T/F
3496 return. Fortuitously, in all cases, the tests worked in OK situations,
3497 which is the norm. However, things would have gone wrong on any write
3498 failures on the smtp file descriptor. This function is used when sending
3499 messages over SMTP and also when doing verify callouts.
3501 2. When Exim is called to do synchronous delivery of a locally submitted
3502 message (the -odf or -odi options), it no longer closes stderr before doing
3505 3. Implemented the mua_wrapper option.
3507 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3509 5. Implemented the functions header_remove(), header_testname(),
3510 header_add_at_position(), and receive_remove_recipient(), and exported them
3513 6. If an ACL "warn" statement specified the addition of headers, Exim already
3514 inserted X-ACL-Warn: at the start if there was no header name. However, it
3515 was not making this test for the second and subsequent header lines if
3516 there were newlines in the string. This meant that an invalid header could
3517 be inserted if Exim was badly configured.
3519 7. Allow an ACL "warn" statement to add header lines at the start or after all
3520 the Received: headers, as well as at the end.
3522 8. Added the rcpt_4xx retry error code.
3524 9. Added postmaster_mailfrom=xxx to callout verification option.
3526 10. Added mailfrom=xxxx to the callout verification option, for verify=
3529 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3530 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3532 12. Inserted some casts to stop certain compilers warning when using pointer
3533 differences as field lengths or precisions in printf-type calls (mostly
3534 affecting debugging statements).
3536 13. Added optional readline() support for -be (dynamically loaded).
3538 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3539 same clock tick as a message's arrival, so that its received time was the
3540 same as the "first fail" time on the retry record, and that message
3541 remained on the queue past the ultimate address timeout, every queue runner
3542 would try a delivery (because it was past the ultimate address timeout) but
3543 after another failure, the ultimate address timeout, which should have then
3544 bounced the address, did not kick in. This was a "< instead of <=" error;
3545 in most cases the first failure would have been in the next clock tick
3546 after the received time, and all would be well.
3548 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3549 being recognized when the domain list was tested by the match_domain
3550 condition in an expansion string.
3552 16. Added the ${str2b64: operator.
3554 17. Exim was always calling setrlimit() to set a large limit for the number of
3555 processes, without checking whether the existing limit was already
3556 adequate. (It did check for the limit on file descriptors.) Furthermore,
3557 errors from getrlimit() and setrlimit() were being ignored. Now they are
3558 logged to the main and panic logs, but Exim does carry on, to try to do its
3559 job under whatever limits there are.
3561 18. Imported PCRE 5.0.
3563 19. Trivial typo in log message " temporarily refused connection" (the leading
3566 20. If the log selector return_path_on_delivery was set and an address was
3567 redirected to /dev/null, the delivery process crashed because it assumed
3568 that a return path would always be set for a "successful" delivery. In this
3569 case, the whole delivery is bypassed as an optimization, and therefore no
3572 21. Internal re-arrangement: the function for sending a challenge and reading
3573 a response while authentication was assuming a zero-terminated challenge
3574 string. It's now changed to take a pointer and a length, to allow for
3575 binary data in such strings.
3577 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3579 23. Exim was not respecting finduser_retries when seeking the login of the
3580 uid under which it was called; it was always trying 10 times. (The default
3581 setting of finduser_retries is zero.) Also, it was sleeping after the final
3582 failure, which is pointless.
3584 24. Implemented tls_on_connect_ports.
3586 25. Implemented acl_smtp_predata.
3588 26. If the domain in control=submission is set empty, Exim assumes that the
3589 authenticated id is a complete email address when it generates From: or
3590 Sender: header lines.
3592 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3593 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3594 chown and chgrp in /bin and hostname in /usr/bin.
3596 28. Exim was keeping the "process log" file open after each use, just as it
3597 does for the main log. This opens the possibility of it remaining open for
3598 long periods when the USR1 signal hits a daemon. Occasional processlog
3599 errors were reported, that could have been caused by this. Anyway, it seems
3600 much more sensible not to leave this file open at all, so that is what now
3603 29. The long-running daemon process does not normally write to the log once it
3604 has entered its main loop, and it closes the log before doing so. This is
3605 so that log files can straightforwardly be renamed and moved. However,
3606 there are a couple of unusual error situations where the daemon does write
3607 log entries, and I had neglected to close the log afterwards.
3609 30. The text of an SMTP error response that was received during a remote
3610 delivery was being truncated at 512 bytes. This is too short for some of
3611 the long messages that one sometimes sees. I've increased the limit to
3614 31. It is now possible to make retry rules that apply only when a message has a
3615 specific sender, in particular, an empty sender.
3617 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3618 it possible to be selective about when SMTP synchronization is enforced.
3620 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3622 32. Implemented hosts_connection_nolog.
3624 33. Added an ACL for QUIT.
3626 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3629 35. Added mailbox_size and mailbox_filecount to appendfile.
3631 36. Added control = no_multiline_responses to ACLs.
3633 37. There was a bug in the logic of the code that waits for the clock to tick
3634 in the case where the clock went backwards by a substantial amount such
3635 that the microsecond fraction of "now" was more than the microsecond
3636 fraction of "then" (but the whole seconds number was less).
3638 38. Added support for the libradius Radius client library this is found on
3639 FreeBSD (previously only the radiusclient library was supported).
3645 1. When certain lookups returned multiple values in the form name=value, the
3646 quoting of the values was not always being done properly. Specifically:
3647 (a) If the value started with a double quote, but contained no whitespace,
3649 (b) If the value contained whitespace other than a space character (i.e.
3650 tabs or newlines or carriage returns) it was not quoted.
3651 This fix has been applied to the mysql and pgsql lookups by writing a
3652 separate quoting function and calling it from the lookup code. The fix
3653 should probably also be applied to nisplus, ibase and oracle lookups, but
3654 since I cannot test any of those, I have not disturbed their existing code.
3656 2. A hit in the callout cache for a specific address caused a log line with no
3657 reason for rejecting RCPT. Now it says "Previous (cached) callout
3658 verification failure".
3660 3. There was an off-by-one bug in the queryprogram router. An over-long
3661 return line was truncated at 256 instead of 255 characters, thereby
3662 overflowing its buffer with the terminating zero. As well as fixing this, I
3663 have increased the buffer size to 1024 (and made a note to document this).
3665 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3666 when Exim is waiting for an SMTP response from a remote server, Exim
3667 restarts its select() call on the socket, thereby resetting its timeout.
3668 This is not a problem when such interrupts are rare. Somebody set up a cron
3669 job to run exiwhat every 2 minutes, which is less than the normal select()
3670 timeout (5 or 10 minutes). This meant that the select() timeout never
3671 kicked in because it was always reset. I have fixed this by comparing the
3672 time when an interrupt arrives with the time at the start of the first call
3673 to select(). If more time than the timeout has elapsed, the interrupt is
3674 treated as a timeout.
3676 5. Some internal re-factoring in preparation for the addition of Sieve
3677 extensions (by MH). In particular, the "personal" test is moved to a
3678 separate function, and given an option for scanning Cc: and Bcc: (which is
3679 not set for Exim filters).
3681 6. When Exim created an email address using the login of the caller as the
3682 local part (e.g. when creating a From: or Sender: header line), it was not
3683 quoting the local part when it contained special characters such as @.
3685 7. Installed new OpenBSD configuration files.
3687 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3688 try to make them clearer.
3690 9. Callout options, other than the timeout value, were being ignored when
3691 verifying sender addresses in header lines. For example, when using
3693 verify = header_sender/callout=no_cache
3695 the cache was (incorrectly) being used.
3697 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3698 only the Cygwin environment.
3700 11. When return_path_on_delivery was set as a log selector, if different remote
3701 addresses in the same message used different return paths and parallel
3702 remote delivery occurred, the wrong values would sometimes be logged.
3703 (Whenever a remote delivery process finished, the return path value from
3704 the most recently started remote delivery process was logged.)
3706 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3707 lines when AUTH and/or TLS are in use. This is the "received protocol"
3708 field. Exim used to use "asmtp" for authenticated SMTP, without any
3709 indication (in the protocol name) for TLS use. Now it follows the RFC and
3710 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3711 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3712 names appear in log lines as well as in Received: header lines.
3714 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3715 extensions, and comparison tests, and to fix some bugs.
3717 14. Changes to the "personal" filter test:
3719 (1) The test was buggy in that it was just doing the equivalent of
3720 "contains" tests on header lines. For example, if a user's address was
3721 anne@some.where, the "personal" test would incorrectly be true for
3723 To: susanne@some.where
3725 This test is now done by extracting each address from the header in turn,
3726 and checking the entire address. Other tests that are part of "personal"
3727 are now done using regular expressions (for example, to check local parts
3728 of addresses in From: header lines).
3730 (2) The list of non-personal local parts in From: addresses has been
3731 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3732 taken from the Sieve specification recommendations.
3734 (3) If the message contains any header line starting with "List-" it is
3735 treated as non-personal.
3737 (4) The test for "circular" in the Subject: header line has been removed
3738 because it now seems ill-conceived.
3740 15. Minor typos in src/EDITME comments corrected.
3742 16. Installed latest exipick from John Jetmore.
3744 17. If headers_add on a router specified a text string that was too long for
3745 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3746 of string_sprintf() is now avoided.
3748 18. $message_body_size was not set (it was always zero) when running the DATA
3749 ACL and the local_scan() function.
3751 19. For the "mail" command in an Exim filter, no default was being set for
3752 the once_repeat time, causing a random time value to be used if "once" was
3753 specified. (If the value happened to be <= 0, no repeat happened.) The
3754 default is now 0s, meaning "never repeat". The "vacation" command was OK
3755 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3756 (I found it when inspecting the code).
3758 20. There is now an overall timeout for performing a callout verification. It
3759 defaults to 4 times the callout timeout, which applies to individual SMTP
3760 commands during the callout. The overall timeout applies when there is more
3761 than one host that can be tried. The timeout is checked before trying the
3762 next host. This prevents very long delays if there are a large number of
3763 hosts and all are timing out (e.g. when the network connections are timing
3764 out). The value of the overall timeout can be changed by specifying an
3765 additional sub-option for "callout", called "maxwait". For example:
3767 verify = sender/callout=5s,maxwait=20s
3769 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3770 to the end before writing, but this should make it even safer).
3772 22. Exim was forgetting that it had advertised PIPELINING for the second and
3773 subsequent messages on an SMTP connection. It was also not resetting its
3774 memory on STARTTLS and an internal HELO.
3776 23. When Exim logs an SMTP synchronization error within a session, it now
3777 records whether PIPELINING has been advertised or not.
3779 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3780 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3781 rather than long int.
3783 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3785 26. Added the never_mail option to autoreply.
3791 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3792 crash if the getsockname() call failed; this can happen if a connection is
3793 closed very soon after it is established. The problem was simply in the
3794 order in which certain operations were done, causing Exim to try to write
3795 to the SMTP stream before it had set up the file descriptor. The bug has
3796 been fixed by making things happen in the correct order.
3802 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3803 before the connection was closed, thus losing the rejection response.
3805 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3806 some early Solaris releases, but causes trouble in current releases where
3807 socklen_t is defined.
3809 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3812 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3815 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3816 that releasing the top of it at the end releases what was used for sub-
3817 expansions (unless the block got too big). However, discard this block if
3818 the first thing is a variable or header, so that we can use its block when
3819 it is dynamic (useful for very large $message_headers, for example).
3821 6. Lookups now cache *every* query, not just the most recent. A new, separate
3822 store pool is used for this. It can be recovered when all lookup caches are
3823 flushed. Lookups now release memory at the end of their result strings.
3824 This has involved some general refactoring of the lookup sources.
3826 7. Some code has been added to the store_xxx() functions to reduce the amount
3827 of flapping under certain conditions.
3829 8. log_incoming_interface used to affect only the <= reception log lines. Now
3830 it causes the local interface and port to be added to several more SMTP log
3831 lines, for example "SMTP connection from", and rejection lines.
3833 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3835 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3837 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3838 could be overwritten at the end of the current message (or the start of a
3839 new message if it was set in a HELO ACL). The value is now preserved for
3840 the duration of the SMTP connection.
3842 12. If a transport had a headers_rewrite setting, and a matching header line
3843 contained an unqualified address, that address was qualified, even if it
3844 did not match any rewriting rules. The underlying bug was that the values
3845 of the flags that permit the existence of unqualified sender and recipient
3846 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3847 non-local messages, and by -bnq for local messages) were not being
3848 preserved with the message after it was received.
3850 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3851 "next input=" as part of the text comprising the host identity instead of
3852 the correct text. The code was using the same buffer for two different
3853 strings. However, depending on which order the printing function evaluated
3854 its arguments, the bug did not always show up. Under Linux, for example, my
3855 test suite worked just fine.
3857 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3858 doesn't work with some older versions of Perl. It has been changed to "my",
3859 which in any case is probably the better facility to use.
3861 15. A really picky compiler found some instances of statements for creating
3862 error messages that either had too many or two few arguments for the format
3865 16. The size of the buffer for calls to the DNS resolver has been increased
3866 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3867 for addresses that have a lot of PTR records. This alleviates a problem; it
3868 does not fully solve it.
3870 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3871 buffer now truncates the list and logs the incident, which is the same
3872 action as happens when Exim is looking up a host name and its aliases.
3873 Previously in this situation something unpredictable would happen;
3874 sometimes it was "internal error: store_reset failed".
3876 18. If a server dropped the connection unexpectedly when an Exim client was
3877 using GnuTLS and trying to read a response, the client delivery process
3878 crashed while trying to generate an error log message.
3880 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3881 string, for example:
3883 warn message = H1: something\nH2: something
3885 the text was added as a single header line from Exim's point of view
3886 though it ended up OK in the delivered message. However, searching for the
3887 second and subsequent header lines using $h_h2: did not work. This has been
3888 fixed. Similarly, if a system filter added multiple headers in this way,
3889 the routers could not see them.
3891 20. Expanded the error message when iplsearch is called with an invalid key to
3892 suggest using net-iplsearch in a host list.
3894 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3895 ACLs are no longer actually imposed (and a message to that effect is
3898 22. If a "gecos" field in a passwd entry contained escaped characters, in
3899 particular, if it contained a \" sequence, Exim got it wrong when building
3900 a From: or a Sender: header from that name. A second bug also caused
3901 incorrect handling when an unquoted " was present following a character
3902 that needed quoting.
3904 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3905 was not being matched caselessly.
3907 24. Arranged for all hyphens in the exim.8 source to be escaped with
3910 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3911 properly. Recipient callout cache records were still being keyed to include
3912 the sender, even when use_sender was set false. This led to far more
3913 callouts that were necessary. The sender is no longer included in the key
3914 when use_sender is false.
3916 26. Added "control = submission" modifier to ACLs.
3918 27. Added the ${base62d: operator to decode base 62 numbers.
3920 28. dnsdb lookups can now access SRV records.
3922 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3923 the configuration file.
3925 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3926 (-v) mode. This makes the output for a verbose queue run more intelligible.
3928 31. Added a use_postmaster feature to recipient callouts.
3930 32. Added the $body_zerocount variable, containing the number of binary zero
3931 bytes in the message body.
3933 33. The time of last modification of the "new" subdirectory is now used as the
3934 "mailbox time last read" when there is a quota error for a maildir
3937 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3939 35. Added +ignore_unknown as a special item in host lists.
3941 36. Code for decoding IPv6 addresses in host lists is now included, even if
3942 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3943 address was recognized as an IP address, but was then not correctly decoded
3944 into binary, causing unexpected and incorrect effects when compared with
3951 1. Very minor rewording of debugging text in manualroute to say "list of
3952 hosts" instead of "hostlist".
3954 2. If verify=header_syntax was set, and a header line with an unqualified
3955 address (no domain) and a large number of spaces between the end of the
3956 name and the colon was received, the reception process suffered a buffer
3957 overflow, and (when I tested it) crashed. This was caused by some obsolete
3958 code that should have been removed. The fix is to remove it!
3960 3. When running in the test harness, delay a bit after writing a bounce
3961 message to get a bit more predictability in the log output.
3963 4. Added a call to search_tidyup() just before forking a reception process. In
3964 theory, someone could use a lookup in the expansion of smtp_accept_max_
3965 per_host which, without the tidyup, could leave open a database connection.
3967 5. Added the variables $recipient_data and $sender_data which get set from a
3968 lookup success in an ACL "recipients" or "senders" condition, or a router
3969 "senders" option, similar to $domain_data and $local_part_data.
3971 6. Moved the writing of debug_print from before to after the "senders" test
3974 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3975 problems for message scanning, either using a data ACL, or using
3976 local_scan() because the Received: header was not generated till after they
3977 were called (in order to set the time as the time of reception completion).
3978 I have revised the way this works. The header is now generated after the
3979 body is received, but before the ACL or local_scan() are called. After they
3980 are run, the timestamp in the header is updated.
3986 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3987 before starting a queue runner without re-exec. This happened only when
3988 deliver_drop_privilege was set or when the Exim user was set to root. The
3989 effect of the bug was that timeouts during subsequent deliveries caused
3990 crashes instead of being properly handled. The handler is now left at its
3991 default (and expected) setting.
3993 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3994 message, again when deliver_drop_privilege is set or Exim is run as root.
3995 The bug described in (1) was not present in this case, but the tidying up
3996 of the other signals was missing. I have made the two cases consistent.
3998 3. The ignore_target_hosts setting on a manualroute router was being ignored
3999 for hosts that were looked up using the /MX notation.
4001 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4004 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4005 operated on the sender address. After changing the $sender_address to <>
4006 for the sender address verify, Exim was re-instated it as the original
4007 (before rewriting) address, but remembering that it had rewritten it, so it
4008 wasn't rewriting it again. This bug also had the effect of breaking the
4009 sender address verification caching when the sender address was rewritten.
4011 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4012 This has been changed so that if the ip literal address matches
4013 ignore_target_hosts, the router declines.
4015 7. Added expansion conditions match_domain, match_address, and match_local_
4016 part (NOT match_host).
4018 8. The placeholder for the Received: header didn't have a length field set.
4020 9. Added code to Exim itself and to exim_lock to test for a specific race
4021 condition that could lead to file corruption when using MBX delivery. The
4022 issue is with the lockfile that is created in /tmp. If this file is removed
4023 after a process has opened it but before that process has acquired a lock,
4024 there is the potential for a second process to recreate the file and also
4025 acquire a lock. This could lead to two Exim processes writing to the file
4026 at the same time. The added code performs the same test as UW imapd; it
4027 checks after acquiring the lock that its file descriptor still refers to
4028 the same named file.
4030 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4031 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4034 11. Added the smtp_active_hostname option. If used, this will typically be made
4035 to depend on the incoming interface address. Because $interface_address is
4036 not set up until the daemon has forked a reception process, error responses
4037 that can happen earlier (such as "too many connections") no longer contain
4040 12. If an expansion in a condition on a "warn" statement fails because a lookup
4041 defers, the "warn" statement is abandoned, and the next ACL statement is
4042 processed. Previously this caused the whole ACL to be aborted.
4044 13. Added the iplsearch lookup type.
4046 14. Added ident_timeout as a log selector.
4048 15. Added tls_certificate_verified as a log selector.
4050 16. Added a global option tls_require_ciphers (compare the smtp transport
4051 option of the same name). This controls incoming TLS connections.
4053 17. I finally figured out how to make tls_require_ciphers do a similar thing
4054 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4055 before starting the TLS session.
4057 18. Tabs are now shown as \t in -bP output.
4059 19. If the log selector return_path_on_delivery was set, Exim crashed when
4060 bouncing a message because it had too many Received: header lines.
4062 20. If two routers both had headers_remove settings, and the first one included
4063 a superfluous trailing colon, the final name in the first list and the
4064 first name in the second list were incorrectly joined into one item (with a
4065 colon in the middle).
4071 1. Added -C and -D options to the exinext utility, mainly to make it easier
4072 to include in the automated testing, but these could be helpful when
4073 multiple configurations are in use.
4075 2. The exinext utility was not formatting the output nicely when there was
4076 an alternate port involved in the retry record key, nor when there was a
4077 message id as well (for retries that were specific to a specific message
4078 and a specific host). It was also confused by IPv6 addresses, because of
4079 the additional colons they contain. I have fixed the IPv4 problem, and
4080 patched it up to do a reasonable job for IPv6.
4082 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4083 delivery, the log line now contains "pipelined" if PIPELINING was used.
4085 4. An SMTP transport process used to panic and die if the bind() call to set
4086 an explicit outgoing interface failed. This has been changed; it is now
4087 treated in the same way as a connect() failure.
4089 5. A reference to $sender_host_name in the part of a conditional expansion
4090 that was being skipped was still causing a DNS lookup. This no longer
4093 6. The def: expansion condition was not recognizing references to header lines
4094 that used bh_ and bheader_.
4096 7. Added the _cache feature to named lists.
4098 8. The code for checking quota_filecount in the appendfile transport was
4099 allowing one more file than it should have been.
4101 9. For compatibility with Sendmail, the command line option
4109 and sets the incoming protocol and host name (for trusted callers). The
4110 host name and its colon can be omitted when only the protocol is to be set.
4111 Note the Exim already has two private options, -pd and -ps, that refer to
4112 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4113 "s", but I don't think that's a major issue.
4115 10. A number of refactoring changes to the code, none of which should affect
4118 (a) The number of logging options was getting close to filling up the
4119 32-bit word that was used as a bit map. I have split them into two classes:
4120 those that are passed in the argument to log_write(), and those that are
4121 only ever tested independently outside of that function. These are now in
4122 separate 32-bit words, so there is plenty of room for expansion again.
4123 There is no change in the user interface or the logging behaviour.
4125 (b) When building, for example, log lines, the code previously used a
4126 macro that called string_cat() twice, in order to add two strings. This is
4127 not really sufficiently general. Furthermore, there was one instance where
4128 it was actually wrong because one of the argument was used twice, and in
4129 one call a function was used. (As it happened, calling the function twice
4130 did not affect the overall behaviour.) The macro has been replaced by a
4131 function that can join an arbitrary number of extra strings onto a growing
4134 (c) The code for expansion conditions now uses a table and a binary chop
4135 instead of a serial search (which was left over from when there were very
4136 few conditions). Also, it now recognizes conditions like "pam" even when
4137 the relevant support is not compiled in: a suitably worded error message is
4138 given if an attempt is made to use such a condition.
4140 11. Added ${time_interval:xxxxx}.
4142 12. A bug was causing one of the ddress fields not to be passed back correctly
4143 from remote delivery subprocesses. The field in question was not being
4144 subsequently used, so this caused to problems in practice.
4146 13. Added new log selectors queue_time and deliver_time.
4148 14. Might have fixed a bug in maildirsizefile handling that threw up
4149 "unexpected character" debug warnings, and recalculated the data
4150 unnecessarily. In any case, I expanded the warning message to give more
4153 15. Added the message "Restricted characters in address" to the statements in
4154 the default ACL that block characters like @ and % in local parts.
4156 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4157 Three changes have been made:
4159 (a) There was a serious bug; a negative response to MAIL caused the whole
4160 recipient domain to be cached as invalid, thereby blocking all messages
4161 to all local parts at the same domain, from all senders. This bug has
4162 been fixed. The domain is no longer cached after a negative response to
4163 MAIL if the sender used is not empty.
4165 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4168 (c) A new callout option, "use_sender" has been added for people who want
4169 the modified behaviour.
4175 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4178 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4179 indeed breaks things for older releases.
4181 3. Added additional logging to the case where there is a problem reading data
4182 from a filter that is running in a subprocess using a pipe, in order to
4183 try to track down a specific problem.
4185 4. Testing facility fudge: when running in the test harness and attempting
4186 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4187 getting "No route to host". Convert this to a timeout.
4189 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4192 6. Some OS don't have socklen_t but use size_t instead. This affects the
4193 fifth argument of getsockopt() amongst other things. This is now
4194 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4195 can be set for individual OS. I have set it for SunOS5, OSF1, and
4196 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4197 some earlier ones do not.
4199 7. Change 4.30/15 was not doing the test caselessly.
4201 8. The standard form for an IPv6 address literal was being rejected by address
4202 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4203 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4204 this, as well as the form without the "IPv6" on the front (but only when
4205 address literals are enabled, of course).
4207 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4209 10. Exim crashed if a message with an empty sender address specified by -f
4210 encountered a router with an errors_to setting. This could be provoked only
4211 by a command such as
4215 where an empty string was supplied; "<>" did not hit this bug.
4217 11. Installed PCRE release 4.5.
4219 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4220 remained set. It is now erased.
4222 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4223 times from message ids (which are base 36 rather than the normal 62).
4225 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4226 were being counted as actual protocol errors, and logged if the log
4227 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4228 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4229 and DATA following a set of rejected RCPTs do not count as protocol errors.
4230 In other words, Exim assumes they were pipelined, though this may not
4231 actually be the case. Of course, in all cases the client gets an
4232 appropriate error code.
4234 15. If a lookup fails in an ACL condition, a message about the failure may
4235 be available; it is used if testing the ACL cannot continue, because most
4236 such messages specify what the cause of the deferral is. However, some
4237 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4238 that caused an old message to be retained and used if a later statement
4239 caused a defer, replacing the real cause of the deferral.
4241 16. If an IP address had so many PTR records that the DNS lookup buffer
4242 was not large enough to hold them, Exim could crash while trying to process
4243 the truncated data. It now detects and logs this case.
4245 17. Further to 4.21/58, another change has been made: if (and only if) the
4246 first line of a message (the first header line) ends with CRLF, a bare LF
4247 in a subsequent header line has a space inserted after it, so as not to
4248 terminate the header.
4250 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4251 unnecessarily, used atoi() instead of strtol(), and didn't check the
4252 termination when getting file sizes from file names by regex.
4254 19. Completely re-implemented the support for maildirsize files, in the light
4255 of a number of problems with the previous contributed implementation
4256 (4.30/29). In particular:
4258 . If the quota is zero, the maildirsize file is maintained, but no quota is
4261 . If the maildir directory does not exist, it is created before any attempt
4262 to write a maildirsize file.
4264 . The quota value in the file is just a cache; if the quota is changed in
4265 the transport, the new value overrides.
4267 . A regular expression is available for excluding directories from the
4270 20. The autoreply transport checks the characters in options that define the
4271 message's headers; it allows continued headers, but it was checking with
4272 isspace() after an embedded newline instead of explicitly looking for a
4275 21. If all the "regular" hosts to which an address was routed had passed their
4276 expiry times, and had not reached their retry times, the address was
4277 bounced, even if fallback hosts were defined. Now Exim should go on to try
4280 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4281 equivalent code in the SMTP transport. Some hosts send humungous responses
4282 to HELO/EHLO, more than 1024 it seems.
4284 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4285 gives compiler warnings in some environments. I've now done it "properly",
4288 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4289 (because of problems with the built-in one) was declared to return uschar *
4290 instead of char *, causing compiler failure.
4292 25. Fixed a file descriptor leak when processing alias/forward files.
4294 26. Fixed a minor format string issue in dbfn.c.
4296 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4298 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4299 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4302 29. When a sender address is verified, it is cached, to save repeating the test
4303 when there is more than one recipient in a message. However, when the
4304 verification involves a callout, it is possible for different callout
4305 options to be set for different recipients. It is too complicated to keep
4306 track of this in the cache, so now Exim always runs a verification when a
4307 callout is required, relying on the callout cache for the optimization.
4308 The overhead is duplication of the address routing, but this should not be
4311 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4312 to be verified with callout=postmaster, and the main callout worked but the
4313 postmaster check failed, the verification correctly failed. However, if a
4314 subsequent RCPT command asked for sender verification *without* the
4315 postmaster check, incorrect caching caused this verification also to fail,
4318 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4319 it was not caching the DNS options (qualify_single, search_parents) that
4320 were used when the lookup failed. A subsequent lookup with different
4321 options therefore always gave the same answer, though there were cases
4322 where it should not have. (Example: a "domains = !$mx_any" option on a
4323 dnslookup router: the "domains" option is always processed without any
4324 widening, but the router might have qualify_single set.) Now Exim uses the
4325 cached value only when the same options are set.
4327 32. Added John Jetmore's "exipick" utility to the distribution.
4329 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4330 than a timeout (e.g. a certificate is required, and is not provided), an
4331 Exim server now closes the connection immediately. Previously it waited for
4332 the client to close - but if the client is SSL, it seems that they each
4333 wait for each other, leading to a delay before one of them times out.
4335 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4336 maintained 0.8.x compatibility because I don't think many are using it, and
4337 it is clearly obsolete.
4339 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4342 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4343 client certificate was expired. A simple patch fixes this, though I don't
4344 understand the full logic of why the verify callback is called multiple
4347 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4348 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4349 which causes problems with some clients (such as the Certicom SSL Plus
4350 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4351 disables the coutermeasure allowing Eudora to connect."
4353 38. Exim was not checking that a write() to a log file succeeded. This could
4354 lead to Bad Things if a log got too big, in particular if it hit a file
4355 size limit. Exim now panics and dies if it cannot write to a log file, just
4356 as it does if it cannot open a log file.
4358 39. Modified OS/Makefile-Linux so that it now contains
4360 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4362 The two -D definitions ensure that Exim is compiled with large file
4363 support, which makes it possible to handle log files that are bigger than
4366 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4367 instance) a domain was checked against a named list that involved a lookup,
4368 causing $domain_data to be set, then another domain was checked against the
4369 same list, then the first domain was re-checked, the value of $domain_data
4370 after the final check could be wrong. In particular, if the second check
4371 failed, it could be set empty. This bug probably also applied to
4374 41. The strip_trailing_dot option was not being applied to the address given
4375 with the -f command-line option.
4377 42. The code for reading a message's header from the spool was incrementing
4378 $received_count, but never initializing it. This meant that the value was
4379 incorrect (doubled) while delivering a message in the same process in which
4380 it was received. In the most common configuration of Exim, this never
4381 happens - a fresh exec is done - but it can happen when
4382 deliver_drop_privilege is set.
4384 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4385 it now includes up to 150 characters of the unexpected data in the log
4388 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4389 and building data strings. The size of both of these buffers was 10 000
4390 bytes - far larger than anybody would *ever* want, thought I. Needless to
4391 say, somebody hit the limit. I have increased the maximum line length to
4392 20 000 and the maximum data length of concatenated lines to 100 000. I have
4393 also fixed two bugs, because there was no checking on these buffers. Tsk,
4394 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4395 buffer is too small.
4397 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4398 lsearch lookups. Now it does.
4400 46. When parsing a route_list item in a manualroute router, a fixed-length
4401 buffer was used for the list of hosts. I made this 1024 bytes long,
4402 thinking that nobody would ever have a list of hosts that long. Wrong.
4403 Somebody had a whole pile of complicated expansion conditions, and the
4404 string was silently truncated, leading to an expansion error. It turns out
4405 that it is easier to change to an unlimited length (owing to other changes
4406 that have happened since this code was originally written) than to build
4407 structure for giving a limitation error. The length of the item that
4408 expands into the list of hosts is now unlimited.
4410 47. The lsearch lookup could not handle data where the length of text line was
4411 more than 4095 characters. Such lines were truncated, leading to shortened
4412 data being returned. It should now handle lines of any length.
4414 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4415 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4418 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4419 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4420 no longer happens in comment lines. A list of replacements is now placed
4421 at the head of all of the source files, except those whose only change is
4422 to replace PERL_COMMAND in the very first #! line.
4424 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4425 messages on the queue, with a bottom-up merge sort, using code contributed
4426 by Michael Haardt. This should make operations like -bp somewhat faster on
4427 large queues. It won't affect queue runners, except when queue_run_in_order
4430 51. Installed eximstats 1.31 in the distribution.
4432 52. Added support for SRV lookups to the dnslookup router.
4434 53. If an ACL referred to $message_body or $message_body_end, the value was not
4435 reset for any messages that followed in the same SMTP session.
4437 54. The store-handling optimization for building very long strings was not
4438 differentiating between the different store pools. I don't think this
4439 actually made any difference in practice, but I've tidied it.
4441 55. While running the routers to verify a sender address, $sender_address
4442 was still set to the sender address. This is wrong, because when routing to
4443 send a bounce to the sender, it would be empty. Therefore, I have changed
4444 it so that, while verifying a sender address, $sender_address is set to <>.
4445 (There is no change to what happens when verifying a recipient address.)
4447 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4448 target A or AAAA records (if not already returned) without resetting the
4449 qualify_single or search_parents options of the DNS resolver. These are
4450 inappropriate in this case because the targets of MX and SRV records must
4451 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4452 target that, when qualified, matched something in the local domain. These
4453 two options are now turned off when doing these lookups.
4455 57. It seems that at least some releases of Reiserfs (which does not have the
4456 concept of a fixed number of inodes) returns zero and not -1 for the
4457 number of available inodes. This interacted badly with check_spool_inodes,
4458 which assumed that -1 was the "no such thing" setting. What I have done is
4459 to check that the total number of inodes is greater than zero before doing
4460 the test of how many are available.
4462 58. When a "warn" ACL statement has a log_message modifier, the message is
4463 remembered, and not repeated. This is to avoid a lot of repetition when a
4464 message has many recipients that cause the same warning to be written.
4465 Howewer, Exim was preserving the list of already written lines for an
4466 entire SMTP session, which doesn't seem right. The memory is now reset if a
4467 new message is started.
4469 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4470 debugging output unless log_rewrite was also set.
4472 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4473 of (int)(handle) when we know that handle contains (void *)(-1).
4475 61. The Exim daemon panic-logs an error return when it closes the incoming
4476 connection. However "connection reset by peer" seems to be common, and
4477 isn't really an error worthy of noting specially, so that particular error
4480 62. When Exim is trying to find all the local interfaces, it used to panic and
4481 die if the ioctl to get the interface flags failed. However, it seems that
4482 on at least one OS (Solaris 9) it is possible to have an interface that is
4483 included in the list of interfaces, but for which you get a failure error
4484 for this call. This happens when the interface is not "plumbed" into a
4485 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4486 failure of the "get flags" call assumes that the interface is down.
4488 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4489 makes life easier for people who are doing arithmetic on fields extracted
4490 from dates, where you often get leading zeros that should not be
4491 interpreted as octal.
4493 64. Added qualify_domain to the redirect router, to override the global
4496 65. If a pathologically long header line contained very many addresses (the
4497 report of this problem mentioned 10 000) and each of them was rewritten,
4498 Exim could use up a very large amount of memory. (It kept on making new
4499 copies of the header line as it rewrote, and never released the old ones.)
4500 At the expense of a bit more processing, the header rewriting function has
4501 been changed so that it no longer eats memory in this way.
4503 66. The generation of the Received: header has been moved from the time that a
4504 message starts to be received, to the time that it finishes. The timestamp
4505 in the Received: header should now be very close to that of the <= log
4506 line. There are two side-effects of this change:
4508 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4509 logged header lines no longer include the local Received: line, because
4510 it has not yet been created. The same applies to a copy of the message
4511 that is returned to a non-SMTP sender when a message is rejected.
4513 (b) When a filter file is tested using -bf, no additional Received: header
4514 is added to the test message. After some thought, I decided that this
4517 This change does not affect the value of $received_for. It is still set
4518 after address rewriting, but before local_scan() is called.
4520 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4522 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4523 gave an unhelpful panic error message, and a defer error. I have managed to
4524 change this behaviour so that it now rejects any supplied certificate,
4525 which seems right, as the list of acceptable certificates is empty.
4527 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4528 gave an unhelpful defer error. I have not managed to make this reject any
4529 supplied certificates, but the error message it gives is "no certificate
4530 supplied", which is not helpful.
4532 70. exigrep's output now also includes lines that are not associated with any
4533 message, but which match the given pattern. Implemented by a patch from
4534 Martin Sluka, which also tidied up the Perl a bit.
4536 71. Recipient callout verification, like sender verification, was using <> in
4537 the MAIL FROM command. This isn't really the right thing, since the actual
4538 sender may affect whether the remote host accepts the recipient or not. I
4539 have changed it to use the actual sender in the callout; this means that
4540 the cache record is now keyed on a recipient/sender pair, not just the
4541 recipient address. There doesn't seem to be a real danger of callout loops,
4542 since a callout by the remote host to check the sender would use <>.
4543 [SEE ABOVE: changed after hitting problems.]
4545 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4546 temporary errors. However, in the case of such a code being given after
4547 the end of a data transmission (i.e. after ".") Exim was failing to write
4548 a retry record for the message. (Yes, there was some broken host that was
4549 actually sending 8xx at this point.)
4551 73. An unknown lookup type in a host list could cause Exim to panic-die when
4552 the list was checked. (An example that provoked this was putting <; in the
4553 middle of a list instead of at the start.) If this happened during a DATA
4554 ACL check, a -D file could be left lying around. This kind of configuration
4555 error no longer causes Exim to die; instead it causes a defer errror. The
4556 incident is still logged to the main and panic logs.
4558 74. Buglet left over from Exim 3 conversion. The message "too many messages
4559 in one connection" was written to the rejectlog but not the mainlog, except
4560 when address rewriting (yes!) was being logged.
4562 75. Added write_rejectlog option.
4564 76. When a system filter was run not as root (that is, when system_filter_user
4565 was set), the values of the $n variables were not being returned to the
4566 main process; thus, they were not subsequently available in the $sn
4569 77. Added +return_path_on_delivery log selector.
4571 78. A connection timeout was being treated differently from recipients deferred
4572 when testing hosts_max_try with a message that was older than the host's
4573 retry timeout. (The host should not be counted, thus allowing all hosts to
4574 be tried at least once before bouncing.) This may have been the cause of an
4575 occasionally reported bug whereby a message would remain on the queue
4576 longer than the retry timeout, but would be bounced if a delivery was
4577 forced. I say "may" because I never totally pinned down the problem;
4578 setting up timeout/retry tests is difficult. See also the next item.
4580 79. The ultimate address timeout was not being applied to errors that involved
4581 a combination of host plus message (for example, a timeout on a MAIL
4582 command). When an address resolved to a number of possible hosts, and they
4583 were not all tried for each delivery (e.g. because of hosts_max_try), a
4584 message could remain on the queue longer than the retry timeout.
4586 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4589 81. Fixed an obscure SMTP outgoing bug which required at least the following
4590 conditions: (a) there was another message waiting for the same server;
4591 (b) the server returned 5xx to all RCPT commands in the first message so
4592 that the message was not completed; (c) the server dropped the connection
4593 or gave a negative response to the RSET that Exim sends to abort the
4594 transaction. The observed case was a dropped connection after DATA that had
4595 been sent in pipelining mode. That is, the server had advertised PIPELINING
4596 but was not implementing it correctly. The effect of the bug was incorrect
4597 behaviour, such as trying another host, and this could lead to a crash.
4603 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4604 and daemon.c were passed as pointers to ints; they should have been
4605 pointers to socklen_t variables (which are typically unsigned ints).
4607 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4610 3. Fixed a really odd bug that affected only the testing scheme; patching a
4611 certain fixed string in the binary changed the value of another string that
4612 happened to be identical to the end of the original first string.
4614 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4615 name", it returns that address as the IP address. On some operating
4616 systems (e.g. Solaris), it also passes back the IP address string as the
4617 "host name". However, on others (e.g. Linux), it passes back an empty
4618 string. Exim wasn't checking for this, and was changing the host name to an
4619 empty string, assuming it had been canonicized.
4621 5. Although rare, it is permitted to have more than one PTR record for a given
4622 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4623 all the names associated with an address, because they do in Solaris.
4624 However, it seems that they do not in Linux for data that comes from the
4625 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4626 I found this out when I moved to a new Linux workstation and tried to run
4627 the Exim test suite.
4629 To get round this problem I have changed the code so that it now does its
4630 own call to the DNS to look up PTR records when searching for a host name.
4631 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4632 addresses that are only in /etc/hosts are still found.
4634 This behaviour is, however, controlled by an option called host_lookup_
4635 order, which defaults to "bydns:byaddr". If people want to use the other
4636 order, or indeed, just use one or the other means of lookup, they can
4637 specify it in this variable.
4639 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4640 some operating systems, this comes back from gethostbyaddr() as an empty
4641 string, and this is what Exim used to test for. However, it seems that in
4642 other systems, "." is yielded. Exim now tests for this case too.
4644 7. The values of check_spool_space and check_log_space are now held internally
4645 as a number of kilobytes instead of an absolute number of bytes. If a
4646 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4647 kilobyte. This means that much larger values can be stored.
4649 8. Exim monitor: an attempt to get the action menu when not actually pointing
4650 at a message produces an empty menu entitled "No message selected". This
4651 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4652 no entries in it ("Shell widget menu has zero width and/or height"). So I
4653 have added a single, blank menu entry in this case.
4655 9. Added ${quote_local_part.
4657 10. MIME decoding is now applied to the contents of Subject: header lines when
4660 11. Now that a reference to $sender_host_address automatically causes a reverse
4661 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4662 host lookup before query-style lookups in lists that might use this
4663 variable. This has therefore been abolished, and the "net-" prefix is no
4664 longer necessary for query-style lookups.
4666 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4667 have been a typo for LFLAGS, so it has been changed.
4669 13. The install script calls Exim with "-C /dev/null" in order to find the
4670 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4671 to be output. Howeve, since Exim outputs its version number before the
4672 error, it didn't break the script. It just looked ugly. I fixed this by
4673 always allowing "-C /dev/null" if the caller is root.
4675 14. Ignore overlarge ACL variable number when reading spool file - insurance
4676 against a later release with more variables having written the file.
4678 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4679 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4680 this, as well as the form without the "IPv6" on the front.
4682 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4683 OS/Makefile-Darwin file.
4685 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4686 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4688 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4689 variable. However, some OS use ints for inodes. I've added cast to long int
4690 to get rid of the compiler warning.
4692 19. I had forgotten to lock out "/../" in configuration file names when
4693 ALT_CONFIG_PREFIX was set.
4695 20. Routers used for verification do not need to specify transports. However,
4696 if such a router generated a host list, and callout was configured, Exim
4697 crashed, because it could not find a port number from the (non-existent)
4698 transport. It now assumes port 25 in this circumstance.
4700 21. Added the -t option to exigrep.
4702 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4703 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4704 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4705 them is set, LOOKUP_LSEARCH is forced.
4707 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4709 24. Added sender and host information to the "rejected by local_scan()" log
4710 line; previously there was no indication of these.
4712 25. Added .include_if_exists.
4714 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4715 Linux. It turns out that not all file systems support this. Apparently some
4716 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4717 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4718 not supported on the file descriptor, is now ignored when Exim is trying to
4719 sync a directory. This applies only to Linux.
4721 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4723 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4724 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4727 29. Added support for maildirsize files from supplied patch (modified a bit).
4729 30. The use of :fail: followed by an empty string could lead Exim to respond to
4730 sender verification failures with (e.g.):
4732 550 Verification failed for <xxx>
4733 550 Sender verify failed
4735 where the first response line was missing the '-' that indicates it is not
4736 the final line of the response.
4738 31. The loop for finding the name of the user that called Exim had a hardwired
4739 limit of 10; it now uses the value of finduser_retries, which is used for
4740 all other user lookups.
4742 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4745 33. Exim was neglecting to zero errno before one call of strtol() when
4746 expanding a string and expecting an integer value. On some systems this
4747 resulted in spurious "integer overflow" errors. Also, it was casting the
4748 result into an int without checking.
4750 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4751 did not work. The code looks as if it has *never* worked, though it appears
4752 to have been documented since at least releast 1.62. I have made it work.
4754 35. The "timeout_DNS" error in retry rules, also documented since at least
4755 1.62, also never worked. As it isn't clear exactly what this means, and
4756 clearly it isn't a major issue, I have abolished the feature by treating it
4757 as "timeout", and writing a warning to the main and panic logs.
4759 36. The display of retry rules for -brt wasn't always showing the error code
4762 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4763 timeout_connect_A, timeout_connect_MX.
4765 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4766 to the empty sender.
4768 39. The daemon was not analysing the content of -oX till after it had closed
4769 stderr and disconnected from the controlling terminal. This meant that any
4770 syntax errors were only noted on the panic log, and the return code from
4771 the command was 0. By re-arranging the code a little, I've made the
4772 decoding happen first, so such errors now appear on stderr, and the return
4773 code is 1. However, the actual setting up of the sockets still happens in
4774 the disconnected process, so errors there are still only recorded on the
4777 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4778 connections (as happens on some IP stacks) was logged at start up time as
4779 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4780 it from "IPv6 and IPv4", which means that two separate sockets are being
4783 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4784 says whether AF_INET or AF_INET6 was passed as an argument.
4786 42. Exiwhat output was messed up when time zones were included in log
4789 43. Exiwhat now gives more information about the daemon's listening ports,
4790 and whether -tls-on-connect was used.
4792 44. The "port" option of the smtp transport is now expanded.
4794 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4795 silently ignored. Now an error message is written to the main and panic
4798 46. There's a new ACL modifier called "logwrite" which writes to a log file
4799 as soon as it is encountered.
4801 47. Added $local_user_uid and $local_user_gid at routing time.
4803 48. Exim crashed when trying to verify a sender address that was being
4806 49. Exim was recognizing only a space character after ".include". It now also
4807 recognizes a tab character.
4809 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4810 extracting the relevant information from the specification. The man page no
4811 longer contains scrambled data for the -d option, and I've added a section
4812 at the front about calling Exim under different names.
4814 51. Added "extra_headers" argument to the "mail" command in filter files.
4816 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4819 53. Installed eximstats 1.29.
4821 54. Added transport_filter_timeout as a generic transport option.
4823 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4824 Cc: header lines. This was required by RFC 822, but it not required by RFC
4827 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4828 incoming messages that did not have them. Now it does so only if the
4829 message originates locally, that is, if there is no associated remote host
4830 address. When Resent- header lines are present, this applies to the Resent-
4831 lines rather than the non-Resent- lines.
4833 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4834 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4836 58. Messages for configuration errors now include the name of the main
4837 configuration file - useful now that there may be more than one file in a
4838 list (.included file names were always shown).
4840 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4841 for those rare installations that do not start the daemon as root or run it
4842 setuid root. I've cut out the call to initgroups() if the daemon is not
4845 60. The Exim user and group can now be bound into the binary as text strings
4846 that are looked up at the start of Exim's processing.
4848 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4850 62. Added $mailstore_basename variable.
4852 63. Installed patch to sieve.c from Michael Haardt.
4854 64. When Exim failed to open the panic log after failing to open the main log,
4855 the original message it was trying to log was written to stderr and debug
4856 output, but if they were not available (the usual case in production), it
4857 was lost. Now it is written to syslog before the two lines that record the
4858 failures to open the logs.
4860 65. Users' Exim filters run in subprocesses under the user's uid. It is
4861 possible for a "deliver" command or an alias in a "personal" command to
4862 provoke an address rewrite. If logging of address rewriting is configured,
4863 this fails because the process is not running as root or exim. There may be
4864 a better way of dealing with this, but for the moment (because 4.30 needs
4865 to be released), I have disabled address rewrite logging when running a
4866 filter in a non-root, non-exim process.
4872 1. The buildconfig auxiliary program wasn't quoting the value set for
4873 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4874 not defined. This bug was masked in 4.22 by the effect that was fixed in
4877 2. Some messages that were rejected after a message id was allocated were
4878 shown as "incomplete" by exigrep. It no longer does this for messages that
4879 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4881 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4882 have allow_domain_literals set, the ID did not get logged in the <= line.
4883 Domain literals are now always recognized in Message-ID: header lines.
4885 4. The first argument for a ${extract expansion item is the key name or field
4886 number. Leading and trailing spaces in this item were not being ignored,
4887 causing some misleading effects.
4889 5. When deliver_drop_privilege was set, single queue runner processes started
4890 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4891 same command in the process it spins off) were not dropping privilege.
4893 6. When the daemon running as "exim" started a queue runner, it always
4894 re-executed Exim in the spun-off process. This is a waste of effort when
4895 deliver_drop_privilege is set. The new process now just calls the
4896 queue-runner function directly.
4902 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4905 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4906 ignored. Though the use of -oP was forcing the writing of a pid file, it
4907 was always written to the default place.
4909 3. If the message "no IP address found for host xxxx" is generated during
4910 incoming verification, it is now followed by identification of the incoming
4911 connection (so you can more easily find what provoked it).
4913 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4915 5. Added some features to "harden" Exim a bit more against certain attacks:
4917 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4918 be put in Local/Makefile. This is like the never_users runtime option,
4919 but it cannot be overridden. The default setting is "root".
4921 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4922 prefix string with which any file named in a -C command line option
4925 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4926 is retained for -C and -D only if the caller of Exim is root. Without
4927 it, the exim user may also use -C and -D and retain privilege.
4929 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4930 command line option is disabled.
4932 6. Macro names set by the -D option must start with an upper case letter, just
4933 like macro names defined in the configuration file.
4935 7. Added "dereference=" facility to LDAP.
4937 8. Two instances of the typo "uknown" in the source files are fixed.
4939 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4940 the Configure-Makefile script screwed up while processing it.
4942 10. Incorporated PCRE 4.4.
4944 11. The SMTP synchronization check was not operating right at the start of an
4945 SMTP session. For example, it could not catch a HELO sent before the client
4946 waited for the greeting. There is now a check for outstanding input at the
4947 point when the greeting is written. Because of the duplex, asynchronous
4948 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4949 way, but not yet received, when the check is performed.
4951 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4952 on TCP/IP sockets, because this apparently causes some broken clients to
4955 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4956 unchanged) from the Cygwin maintainer.
4958 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4959 was supported instead of testing for mbx. Effectively a typo.
4961 15. The spa authenticator server code was not checking that the input it
4962 received was valid base64.
4964 16. The debug output line for the "set" modifier in ACLs was not showing the
4965 name of the variable that was being set.
4967 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4969 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4970 lookup on its own. Something else was needed to trigger the lookup. For
4971 example, a match in host_lookup or the need for a host name in a host list.
4972 Now, if $sender_host_name is referenced and the host name has not yet been
4973 looked up, a lookup is performed. If the lookup fails, the variable remains
4974 empty, and $host_lookup_failed is set to "1".
4976 19. Added "eqi" as a case-independent comparison operator.
4978 20. The saslauthd authentication condition could segfault if neither service
4979 nor realm was specified.
4981 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4982 error message that was logged was misleading, and incoming SMTP
4983 connections were dropped. The message is now more accurate, and temporary
4984 errors are given to SMTP connections.
4986 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4987 (except RSET) with a 421 error, until QUIT is received. However, it was
4988 failing to send a response to QUIT.
4990 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4991 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4992 ACL is now run after the helo_try_verify_hosts code.
4994 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4995 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4996 case-independent, so other case variants are also recognized). Apparently
4997 some systems use these upper case variants.
4999 25. If more than two messages were waiting for the same host, and a transport
5000 filter was specified for the transport, Exim sent two messages over the
5001 same TCP/IP connection, and then failed with "socket operation on non-
5002 socket" when it tried to send the third.
5004 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5006 27. The extern definition of crypt16() in expand.c was not being excluded when
5007 the OS had its own crypt16() function.
5009 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5010 as a preferred synonym for return_size_limit, both as an option and as an
5013 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5015 30. Changed the default configuration ACL to relax the local part checking rule
5016 for addresses that are not in any local domains. For these addresses,
5017 slashes and pipe symbols are allowed within local parts, but the sequence
5018 /../ is explicitly forbidden.
5020 31. SPA server authentication was not clearing the challenge buffer before
5023 32. log_message in a "warn" ACL statement was writing to the reject log as
5024 well as to the main log, which contradicts the documentation and doesn't
5025 seem right (because no rejection is happening). So I have stopped it.
5027 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5028 However, I am unable to do any testing of this.
5030 34. Fixed an infelicity in the appendfile transport. When checking directories
5031 for a mailbox, to see if any needed to be created, it was accidentally
5032 using path names with one or more superfluous leading slashes; tracing
5033 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5035 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5036 contents are added to the log line that is written for every discarded
5037 recipient. (Previously a log_message setting was ignored.)
5039 36. The ${quote: operator now quotes the string if it is empty.
5041 37. The install script runs exim in order to find its version number. If for
5042 some reason other than non-existence or emptiness, which it checks, it
5043 could not run './exim', it was installing it with an empty version number,
5044 i.e. as "exim-". This error state is now caught, and the installation is
5047 38. An argument was missing from the function that creates an error message
5048 when Exim fails to connect to the socket for saslauthd authentication.
5049 This could cause Exim to crash, or give a corrupted message.
5051 39. Added isip, isip4, and isip6 to ${if conditions.
5053 40. The ACL variables $acl_xx are now saved with the message, and can be
5054 accessed later in routers, transports, and filters.
5056 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5057 strings in the file are not string-expanded.
5059 42. If a MAIL command specified a SIZE value that was too large to fit into an
5060 int variable, the check against message_size_limit failed. Such values are
5061 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5062 day this will have to be increased, but I don't think I want to be around
5063 when emails are that large.
5070 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5071 iconv() is not standard in FreeBSD.
5073 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5074 IPv6 enabled. The observed symptom was a segmentation fault on return from
5075 the function os_common_find_running_interfaces() in src/os.c.
5077 3. In the check_special_case() function in daemon.c I had used "errno" as an
5078 argument name, which causes warnings on some systems. This was basically a
5079 typo, since it was named "eno" in the comments!
5081 4. The code that waits for the clock to tick (at a resolution of some fraction
5082 of a second) so as to ensure message-id uniqueness was always waiting for
5083 at least one whole tick, when it could have waited for less. [This is
5084 almost certainly not relevant at current processor speeds, where it is
5085 unlikely to ever wait at all. But we try to future-proof.]
5087 5. The function that sleeps for a time interval that includes fractions of a
5088 second contained a race. It did not block SIGALRM between setting the
5089 timer, and suspending (a couple of lines later). If the interval was short
5090 and the sigsuspend() was delayed until after it had expired, the suspension
5091 never ended. On busy systems this could lead to processes getting stuck for
5094 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5095 process, before it forks any delivery processes. The open lookup caching
5096 mechanism meant that the open file or database connection was passed into
5097 the delivery process. The problem was that delivery processes always tidy
5098 up cached lookup data. This could cause a problem for the next delivery
5099 process started by the queue runner, because the external queue runner
5100 process does not know about the closure. So the next delivery process
5101 still has data in the lookup cache. In the case of a file lookup, there was
5102 no problem because closing a file descriptor in a subprocess doesn't affect
5103 the parent. However, if the lookup was caching a connection to a database,
5104 the connection was closed, and the second delivery process was likely to
5105 see errors such as "PGSQL: query failed: server closed the connection
5106 unexpectedly". The problem has been fixed by closing all cached lookups
5107 in a queue runner before running a delivery process.
5109 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5110 seem to have the "const" qualifier which it has on other OS. I've
5113 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5114 *of the same type* (client or server) with the same public name that an
5115 error should be diagnosed.
5117 9. When Exim looked up a host name for an IP address, but failed to find the
5118 original IP address when looking up the host name (a safety check), it
5119 output the message "<ip address> does not match any IP for NULL", which was
5120 confusing, to say the least. The bug was that the host name should have
5121 appeared instead of "NULL".
5123 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5124 user that is built into the binary, and the -C or -D options is used, root
5125 privilege is dropped before the configuration file is read. In addition,
5126 logging is switched to stderr instead of the normal log files. If the
5127 configuration then re-defines the Exim user, the unprivileged environment
5128 is probably not what is expected, so Exim logs a panic warning message (but
5131 However, if deliver_drop_privilege is set, the unprivileged state may well
5132 be exactly what is intended, so the warning has been cut out in that case,
5133 and Exim is allowed to try to write to its normal log files.
5139 1. smtp_return_error_details was not giving details for temporary sender
5140 or receiver verification errors.
5142 2. Diagnose a configuration error if two authenticators have the same public
5145 3. Exim used not to create the message log file for a message until the first
5146 delivery attempt. This could be confusing when incoming messages were held
5147 for policy or load reasons. The message log file is now created at the time
5148 the message is received, and an initial "Received" line is written to it.
5150 4. The automatically generated man page for command line options had a minor
5151 bug that caused no ill effects; however, a more serious problem was that
5152 the procedure for building the man page automatically didn't always
5153 operate. Consequently, release 4.20 contains an out-of-date version. This
5154 shouldn't happen again.
5156 5. When building Exim with embedded Perl support, the script that builds the
5157 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5158 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5160 6. The freeze_tell option was not being used for messages that were frozen on
5161 arrival, either by an ACL or by local_scan().
5163 7. Added the smtp_incomplete_transaction log selector.
5165 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5166 was accepting AUTH without a new EHLO.
5168 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5169 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5172 10. Logging of TCP/IP connections (when configured) now happens in the main
5173 daemon process instead of the child process, so that the TCP/IP connection
5174 count is more accurate (but it can never be perfect).
5176 11. The use of "drop" in a nested ACL was not being handled correctly in the
5177 outer ACL. Now, if condition failure induced by the nested "drop" causes
5178 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5179 or "require"), the connection is dropped.
5181 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5182 that yield "discard" can now be used with an "accept" or a "discard" verb,
5183 but an error is generated for any others (because I can't see a useful way
5184 to define what should happen).
5186 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5187 are now processed in the same way as lines in the Exim configuration file.
5188 In particular, continuation lines are supported.
5190 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5192 15. Added -ti meaning -t -i.
5194 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5195 domains, and warn by logging if others are found.
5197 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5198 returned from GIFCONF when Exim is trying to find the list of interfaces on
5199 a host. The code in os.c has been modified to copy each ifreq to an aligned
5200 structure in all cases.
5202 Also, in some cases, the returned ifreq's were being copied to a 'struct
5203 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5204 means the last couple of bytes of an IPv6 address could be chopped if the
5205 ifreq contained only a normal sockaddr (14 bytes storage).
5207 18. Named domain lists were not supported in the hosts_treat_as_local option.
5208 An entry such as +xxxx was not recognized, and was treated as a literal
5211 19. Ensure that header lines added by a DATA ACL are included in the reject log
5212 if the ACL subsequently rejects the message.
5214 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5215 MD5 (which is deprecated).
5217 21. When testing a filter file using -bf, Exim was writing a message when it
5218 took the sender from a "From " line in the message, but it was not doing so
5219 when it took $return_path from a Return-Path: header line. It now does.
5221 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5222 with a valid header line field name (a series of printing characters
5223 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5225 23. Changed "disc" in the source to "disk" to conform to the documentation and
5226 the book and for uniformity.
5228 24. Ignore Sendmail's -Ooption=value command line item.
5230 25. When execve() failed while trying to run a command in a pipe transport,
5231 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5232 could be confused with a return value of 69 from the command itself. This
5233 has been changed to 127, the value the shell returns if it is asked to run
5234 a non-existent command. The wording for the related log line suggests a
5235 non-existent command as the problem.
5237 26. If received_header_text expands to an empty string, do not add a Received:
5238 header line to the message. (Well, it adds a token one on the spool, but
5239 marks it "old" so that it doesn't get used or transmitted.)
5241 27. Installed eximstats 1.28 (addition of -nt option).
5243 28. There was no check for failure on the call to getsockname() in the daemon
5244 code. This can fail if there is a shortage of resources on the system, with
5245 ENOMEM, for example. A temporary error is now given on failure.
5247 29. Contrary to the C standard, it seems that in some environments, the
5248 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5249 program. Exim now does this explicitly; it affects the formatting of
5250 timestamps using strftime().
5252 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5253 complaints. I've now initialized all the variables, to avoid this.
5255 32. Header lines added by a system filter were not being "seen" during
5256 transport-time rewrites.
5258 33. The info_callback() function passed to OpenSSL is set up with type void
5259 (*)(SSL *, int, int), as described somewhere. However, when calling the
5260 function (actually a macro) that sets it up, the type void(*)() is
5261 expected. I've put in a cast to prevent warnings from picky compilers.
5263 34. If a DNS black list lookup found a CNAME record, but there were no A
5264 records associated with the domain it pointed at, Exim crashed.
5266 35. If a DNS black list lookup returned more than one A record, Exim ignored
5267 all but the first. It now scans all returned addresses if a particular IP
5268 value is being sought. In this situation, the contents of the
5269 $dnslist_value variable are a list of all the addresses, separated by a
5272 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5273 to accept a host name and all its aliases if the forward lookup for any of
5274 them yielded the IP address of the incoming connection. Now it accepts only
5275 those names whose forward lookup yields the correct IP address. Any other
5276 names are discarded. This closes a loophole whereby a rogue DNS
5277 administrator could create reverse DNS records to break through a
5278 wildcarded host restriction in an ACL.
5280 37. If a user filter or a system filter that ran in a subprocess used any of
5281 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5282 the wrong values were passed to the pipe command ($thisaddress had the
5283 value of $0, $0 had the value of $1, etc). This bug was introduced by
5284 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5286 38. Improved the line breaking for long SMTP error messages from ACLs.
5287 Previously, if there was no break point between 40 and 75 characters, Exim
5288 left the rest of the message alone. Two changes have been made: (a) I've
5289 reduced the minimum length to 35 characters; (b) if it can't find a break
5290 point between 35 and 75 characters, it looks ahead and uses the first one
5291 that it finds. This may give the occasional overlong line, but at least the
5292 remaining text gets split now.
5294 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5295 file descriptors might be low, and that setting 1000 would always raise it.
5296 It turns out that in some environments, the limit is already over 1000 and
5297 that lowering it causes trouble. So now Exim takes care not to decrease it.
5299 40. When delivering a message, the value of $return_path is set to $sender_
5300 address at the start of routing (routers may change the value). By an
5301 oversight, this default was not being set up when an address was tested by
5302 -bt or -bv, which affected the outcome if any router or filter referred to
5305 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5306 log only when "message" or "log_message" are set. However, if one of the
5307 conditions was an address verification, or a call to a nested ACL, the
5308 messages generated by the underlying test were being passed through. This
5309 no longer happens. The underlying message is available in $acl_verify_
5310 message for both "message" and "log_message" expansions, so it can be
5311 passed through if needed.
5313 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5314 new expansion $bh_ to give the encoded byte string without charset
5315 translation. Translation happens only if iconv() is available; HAVE_ICONV
5316 indicates this at build time. HEADERS_CHARSET gives the charset to
5317 translate to; headers_charset can change it in the configuration, and
5318 "headers charset" can change it in an individual filter file.
5320 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5321 that creates RFC 2047 encoded "words" labels them as that charset instead
5322 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5323 expansion operator; (ii) when Exim creates a From: line for a local
5324 message; (iii) when a header line is rewritten to include a "phrase" part.
5326 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5327 buggy, causing it to skip the first lines of messages whose message ID
5328 ended in 'D'. This would not have bitten before Exim release 4.14, because
5329 message IDs were unlikely to end in 'D' before then. The effect was to have
5330 incorrect size information for certain domains.
5332 45. #include "config.h" was missing at the start of the crypt16.c module. This
5333 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5336 46. If there was a timeout during a "random" callout check, Exim treated it as
5337 a failure of the random address, and carried on sending RSET and the real
5338 address. If the delay was just some slowness somewhere, the response to the
5339 original RCPT would be taken as a response to RSET and so on, causing
5340 mayhem of various kinds.
5342 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5343 when I implemented it. It didn't allow for the fact that some option values
5344 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5345 the right test for positive values.
5347 48. Domain names in DNS records are case-independent. Exim always looks them up
5348 in lower case. Some resolvers return domain names in exactly the case they
5349 appear in the zone file, that is, they may contain uppercase letters. Not
5350 all resolvers do this - some return always lower case. Exim was treating a
5351 change of case by a resolver as a change of domain, similar to a widening
5352 of a domain abbreviation. This triggered its re-routing code and so it was
5353 trying to route what was effectively the same domain again. This normally
5354 caused routing to fail (because the router wouldn't handle the domain
5355 twice). Now Exim checks for this case specially, and just changes the
5356 casing of the domain that it ultimately uses when it transmits the message
5359 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5362 50. If a filter generated a file delivery with a non-absolute name (possible if
5363 no home directory exists for the router), the forbid_file option was not
5366 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5367 the existing equality matching.
5369 52. Exim was using ints instead of ino_t variables in some places where it was
5370 dealing with inode numbers.
5372 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5373 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5374 called TMPDIR, and if it finds it is different, it changes its value.
5376 54. The smtp_printf() function is now made available to local_scan() so
5377 additional output lines can be written before returning. There is also an
5378 smtp_fflush() function to enable the detection of a dropped connection.
5379 The variables smtp_input and smtp_batched_input are exported to
5382 55. Changed the default runtime configuration: the message "Unknown user"
5383 has been removed from the ACL, and instead placed on the localuser router,
5384 using the cannot_route_message feature. This means that any verification
5385 failures that generate their own messages won't get overridden. Similarly,
5386 the "Unrouteable address" message that was in the ACL for unverifiable
5387 relay addresses has also been removed.
5389 56. Added hosts_avoid_esmtp to the smtp transport.
5391 57. The exicyclog script was not checking for the esoteric option
5392 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5393 will work only if exicyclog is run under the appropriate euid.
5395 58. Following a discussion on the list, the rules by which Exim recognises line
5396 endings on incoming messages have been changed. The -dropcr and drop_cr
5397 options are now no-ops, retained only for backwards compatibility. The
5398 following line terminators are recognized: LF CRLF CR. However, special
5399 processing applies to CR:
5401 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5402 nor a local message in the state where . is a terminator.
5404 (ii) If a bare CR is encountered in a header line, an extra space is added
5405 after the line terminator so as not to end the header. The reasoning
5406 behind this is that bare CRs in header lines are most likely either
5407 to be mistakes, or people trying to play silly games.
5409 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5410 was being incorrectly given as 18 bytes larger than it should have been.
5411 This is a VOB (very old bug).
5413 60. This may never have affected anything current, but just in case it has:
5414 When the local host is found other than at the start of a list of hosts,
5415 the local host, those with the same MX, and any that follow, are discarded.
5416 When the list in question was part of a longer list of hosts, the following
5417 hosts (not currently being processed) were also being discarded. This no
5418 longer happens. I'm not sure if this situation could ever has previously
5421 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5424 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5425 header. This is something that is recommended in a new Internet Draft, and
5426 is something that is documented as being done by Sendmail. There are two
5427 possible values. For messages generated by the autoreply transport, Exim
5430 Auto-Submitted: auto-replied
5432 whereas for all other generated messages (e.g. bounces) it adds
5434 Auto-Submitted: auto-generated
5436 63. The "personal" condition in filters now includes a test for the
5437 Auto-Submitted: header. If it contains the string "auto-" the message it
5438 not considered personal.
5440 64. Added rcpt_include_affixes as a generic transport option.
5442 65. Added queue_only_override (default true).
5444 66. Added the syslog_duplication option.
5446 67. If what should have been the first header line of a message consisted of
5447 a space followed by a colon, Exim was mis-interpreting it as a header line.
5448 It isn't of course - it is syntactically invalid and should therefore be
5449 treated as the start of the message body. The misbehaviour could have
5450 caused a number of strange effects, including loss of data in subsequent
5451 header lines, and spool format errors.
5453 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5454 client host had authenticated. This control can now be exercised by an ACL
5455 for more flexibility.
5457 69. By default, callouts do not happen when testing with -bh. There is now a
5458 variant, -bhc, which does actually run the callout code, including
5459 consulting and updating the callout cache.
5461 70. Added support for saslauthd authentication, courtesy of Alexander
5464 71. If statvfs() failed on the spool or log directories while checking their
5465 size for availability, Exim confusingly gave the error "space shortage".
5466 Furthermore, in debugging mode it crashed with a floating point exception.
5467 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5468 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5469 serious problem, Exim now writes to the main and panic logs when this
5470 happens, with details of the failure. It then refuses to accept the
5471 incoming message, giving the message "spool directory problem" or "log
5472 directory problem" with a 421 code for SMTP messages.
5474 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5475 0, 1, and 2 exist, because some OS complain for execs without them (see
5476 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5477 descriptors. However, the code omitted to check that the open succeeded,
5478 causing mysterious errors if for some reason the permissions on /dev/null
5479 got screwed. Now Exim writes a message to the main and panic logs, and
5480 bombs out if it can't open /dev/null.
5482 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5483 interact so that it is all more flexible. It is supposed to remain
5484 backwards compatible. Also added extra_local_interfaces.
5486 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5487 to bomb out with an assertion failure - to the client this appears as a
5488 connection drop. This problem occurs in the part of the code that was taken
5489 from the Samba project. Fortunately, the assertion is in a very simple
5490 function, so I have fixed this by reproducing the function inline in the
5491 one place where it is called, and arranging for authentication to fail
5492 instead of killing the process with assert().
5494 75. The SPA client code was not working when the server requested OEM rather
5495 than Unicode encoding.
5497 76. Added code to make require_files with a specific uid setting more usable in
5498 the case where statting the file as root fails - usually a non-root-mounted
5499 NFS file system. When this happens and the failure is EACCES, Exim now
5500 forks a subprocess and does the per-uid checking as the relevant uid.
5502 77. Added process_log_path.
5504 78. If log_file_path was not explicitly set, a setting of check_log_space or
5505 check_log_inodes was ignored.
5507 79. If a space check for the spool or log partitions fails, the incident is now
5508 logged. Of course, in the latter case the data may get lost...
5510 80. Added the %p formatting code to string_format() so that it can be used to
5511 print addresses in debug_print(). Adjusted all the address printing in the
5512 debugging in store.c to use %p rather than %d.
5514 81. There was a concern that a line of code in smtp_in.c could overflow a
5515 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5516 initially expressed, the concern was not well-founded, because trailing
5517 spaces are removed early. However, if the trailing spaces were followed by
5518 a NULL, they did not get removed, so the overflow was possible. Two fixes
5521 (a) I re-wrote the offending code in a cleaner fashion.
5522 (b) If an incoming SMTP command contains a NULL character, it is rejected
5525 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5526 runs initgroups(), so that if the Exim user is in any additional groups,
5527 they will be used during message reception.
5533 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob