1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Solaris build fix for Oracle's LDAP libraries.
8 Bugzilla 1109, patch from Stephen Usher.
10 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
12 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
15 TF/02 Fix a couple more cases where we did not log the error message
16 when unlink() failed. See also change 4.74-TF/03.
18 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
19 lock up or crash if it happened to be inside a call to libc when it
20 got a SIGUSR1 from exiwhat.
22 The SIGUSR1 handler appends the current process status to the process
23 log which is later printed by exiwhat. It used to use the general
24 purpose logging code to do this, but several functions it calls are
27 The new output code in the SIGUSR1 handler is specific to the process
28 log, and simple enough that it's easy to inspect for signal safety.
29 Removing some special cases also simplifies the general logging code.
30 Removing the spurious timestamps from the process log simplifies
33 TF/04 Improved ratelimit ACL condition.
35 The /noupdate option has been deprecated in favour of /readonly which
36 has clearer semantics. The /leaky, /strict, and /readonly update modes
37 are mutually exclusive. The update mode is no longer included in the
38 database key; it just determines when the database is updated. (This
39 means that when you upgrde Exim will forget old rate measurements.)
41 Exim now checks that the per_* options are used with an update mode that
42 makes sense for the current ACL. For example, when Exim is processing a
43 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
44 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
45 must specify per_mail/readonly. If you omit the update mode it defaults to
46 /leaky where that makes sense (as before) or /readonly where required.
48 The /noupdate option is now undocumented but still supported for
49 backwards compatibility. It is equivalent to /readonly except that in
50 ACLs where /readonly is required you may specify /leaky/noupdate or
51 /strict/noupdate which are treated the same as /readonly.
53 A useful new feature is the /count= option. This is a generalization
54 of the per_byte option, so that you can measure the throughput of other
55 aggregate values. For example, the per_byte option is now equivalent
56 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
58 The per_rcpt option has been generalized using the /count= mechanism
59 (though it's more complicated than the per_byte equivalence). When it is
60 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
61 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
62 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
63 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
64 Note that using per_rcpt with a non-readonly update mode in more than
65 one ACL will cause the recipients to be double-counted. (The per_mail
66 and per_byte options don't have this problem.)
68 The handling of very low rates has changed slightly. If the computed rate
69 is less than the event's count (usually one) then this event is the first
70 after a long gap. In this case the rate is set to the same as this event's
71 count, so that the first message of a spam run is counted properly.
73 The major new feature is a mechanism for counting the rate of unique
74 events. The new per_addr option counts the number of different
75 recipients that someone has sent messages to in the last time period. It
76 behaves like per_rcpt if all the recipient addresses are different, but
77 duplicate recipient addresses do not increase the measured rate. Like
78 the /count= option this is a general mechanism, so the per_addr option
79 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
80 example, measure the rate that a client uses different sender addresses
81 with the options per_mail/unique=$sender_address. There are further
82 details in the main documentation.
84 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
86 TF/06 Removed a few PCRE remnants.
88 TF/07 Automatically extract Exim's version number from tags in the git
89 repository when doing development or release builds.
91 PP/02 Raise smtp_cmd_buffer_size to 16kB. Patch from Paul Fisher.
94 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
95 Heavily based on revision 40f9a89a from Simon Arlott's tree.
98 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
104 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
106 PP/02 Harmonised TLS library version reporting; only show if debugging.
107 Layout now matches that introduced for other libraries in 4.74 PP/03.
109 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
111 PP/04 New "dns_use_edns0" global option.
113 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
116 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
117 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
119 TK/01 Updated PolarSSL code to 0.14.2.
120 Bugzilla 1097. Patch from Andreas Metzler.
122 PP/07 Catch divide-by-zero in ${eval:...}.
125 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
128 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
129 format-string attack -- SECURITY: remote arbitrary code execution.
131 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
132 time unintentionally subject to list matching rules, letting the header
133 cause arbitrary Exim lookups (of items which can occur in lists, *not*
134 arbitrary string expansion). This allowed for information disclosure.
136 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
137 INT_MIN/-1 -- value coerced to INT_MAX.
143 NM/01 Workround for PCRE version dependency in version reporting
146 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
147 This fixes portability to compilers other than gcc, notably
148 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
150 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
151 makefiles for portability to HP-UX and POSIX correctness.
153 PP/01 Permit LOOKUP_foo enabling on the make command-line.
154 Also via indented variable definition in the Makefile.
155 (Debugging by Oliver Heesakkers).
157 PP/02 Restore caching of spamd results with expanded spamd_address.
158 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
160 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
161 Improves build reliability. Fix from: Frank Elsner
163 NM/02 Fix wide character breakage in the rfc2047 coding
164 Fixes bug 1064. Patch from Andrey N. Oktyabrski
166 NM/03 Allow underscore in dnslist lookups
167 Fixes bug 1026. Patch from Graeme Fowler
169 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
170 Code patches from Adam Ciarcinski of NetBSD.
172 NM/04 Fixed exiqgrep to cope with mailq missing size issue
175 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
176 is logged, to avoid truncation. Patch from John Horne.
178 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
179 Patch from Jakob Hirsch.
181 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
182 SQL string expansion failure details.
183 Patch from Andrey Oktyabrski.
185 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
186 Patch from Simon Arlott.
188 PP/09 New lookups functionality failed to compile on old gcc which rejects
189 extern declarations in function scope.
190 Patch from Oliver Fleischmann
192 PP/10 Use sig_atomic_t for flags set from signal handlers.
193 Check getgroups() return and improve debugging.
194 Fixed developed for diagnosis in bug 927 (which turned out to be
197 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
198 Patch from Mark Zealey.
200 PP/12 Bugzilla 1056: Improved spamd server selection.
201 Patch from Mark Zealey.
203 PP/13 Bugzilla 1086: Deal with maildir quota file races.
204 Based on patch from Heiko Schlittermann.
206 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
207 Patch from Uwe Doering, sign-off by Michael Haardt.
209 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
210 variable declaration deep within a block. Bug and patch from
213 PP/15 lookups-Makefile IRIX compatibilty coercion.
215 PP/16 Make DISABLE_DKIM build knob functional.
217 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
218 Patch by Simon Arlott
220 TF/03 Fix valgrind.h portability to C89 compilers that do not support
221 variable argument macros. Our copy now differs from upstream.
227 TF/01 Failure to get a lock on a hints database can have serious
228 consequences so log it to the panic log.
230 TF/02 Log LMTP confirmation messages in the same way as SMTP,
231 controlled using the smtp_confirmation log selector.
233 TF/03 Include the error message when we fail to unlink a spool file.
235 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
236 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
237 for maintaining out-of-tree patches for some time.
239 PP/01 Bugzilla 139: Documentation and portability issues.
240 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
241 Handle per-OS dynamic-module compilation flags.
243 PP/02 Let /dev/null have normal permissions.
244 The 4.73 fixes were a little too stringent and complained about the
245 permissions on /dev/null. Exempt it from some checks.
246 Reported by Andreas M. Kirchwitz.
248 PP/03 Report version information for many libraries, including
249 Exim version information for dynamically loaded libraries. Created
250 version.h, now support a version extension string for distributors
251 who patch heavily. Dynamic module ABI change.
253 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
254 privilege escalation vulnerability whereby the Exim run-time user
255 can cause root to append content of the attacker's choosing to
258 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
261 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
262 If dropping privileges for untrusted macros, we disabled normal logging
263 on the basis that it would fail; for the Exim run-time user, this is not
264 the case, and it resulted in successful deliveries going unlogged.
265 Fixed. Reported by Andreas Metzler.
271 PP/01 Date: & Message-Id: revert to normally being appended to a message,
272 only prepend for the Resent-* case. Fixes regression introduced in
273 Exim 4.70 by NM/22 for Bugzilla 607.
275 PP/02 Include check_rfc2047_length in configure.default because we're seeing
276 increasing numbers of administrators be bitten by this.
278 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
280 PP/03 Bugzilla 994: added openssl_options main configuration option.
282 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
284 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
286 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
288 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
289 without a peer certificate, leading to a segfault because of an
290 assumption that peers always have certificates. Be a little more
291 paranoid. Problem reported by Martin Tscholak.
293 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
294 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
295 NB: ClamAV planning to remove STREAM in "middle of 2010".
296 CL also introduces -bmalware, various -d+acl logging additions and
297 more caution in buffer sizes.
299 PP/09 Implemented reverse_ip expansion operator.
301 PP/10 Bugzilla 937: provide a "debug" ACL control.
303 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
305 PP/12 Bugzilla 973: Implement --version.
307 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
309 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
311 PP/15 Bugzilla 816: support multiple condition rules on Routers.
313 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
314 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
315 ignore trailing whitespace.
317 JJ/02 prevent non-panic DKIM error from being sent to paniclog
319 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
322 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
323 Notification from Dr Andrew Aitchison.
325 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
326 ExtendedDetectionInfo response format.
327 Notification from John Horne.
329 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
332 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
333 XSL and documented dependency on system catalogs, with examples of how
336 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
339 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
340 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
341 configuration file which is writeable by the Exim user or group.
343 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
344 of configuration files to cover files specified with the -C option if
345 they are going to be used with root privileges, not just the default
348 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
349 option (effectively making it always true).
351 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
352 files to be used while preserving root privileges.
354 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
355 that rogue child processes cannot use them.
357 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
358 run-time user, instead of root.
360 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
361 Exim run-time user without dropping privileges.
363 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
364 result string, instead of calling string_vformat() twice with the same
367 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
368 for other users. Others should always drop root privileges if they use
369 -C on the command line, even for a whitelisted configure file.
371 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
373 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
379 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
380 $data_path, and $header_path variables; fixed documentation bugs and
383 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
384 exipick to access non-standard spools, including the "frozen" queue
387 NM/01 Bugzilla 965: Support mysql stored procedures.
388 Patch from Alain Williams
390 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
392 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
393 Patch from Andreas Metzler
395 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
396 Patch from Kirill Miazine
398 NM/05 Bugzilla 671: Added umask to procmail example.
400 JJ/03 installed exipick 20100323.0, fixing doc bug
402 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
403 directory. Notification and patch from Dan Rosenberg.
405 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
407 TK/02 Improve log output when DKIM signing operation fails.
409 MH/01 Treat the transport option dkim_domain as a colon separated
410 list, not as a single string, and sign the message with each element,
411 omitting multiple occurences of the same signer.
413 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
414 Bugzilla 985, 986. Patch by Simon Arlott
416 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
417 Patch by Simon Arlott
419 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
420 MBX locking. Notification from Dan Rosenberg.
426 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
428 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
430 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
432 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
434 NM/04 Bugzilla 915: Flag broken perl installation during build.
440 TK/01 Added patch by Johannes Berg that expands the main option
441 "spamd_address" if it starts with a dollar sign.
443 TK/02 Write list of recipients to X-Envelope-Sender header when building
444 the mbox-format spool file for content scanning (suggested by Jakob
447 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
448 (http://www.dcc-servers.net/) support via dccifd. Activated by
449 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
451 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
452 by Mark Daniel Reidel <mr@df.eu>.
454 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
455 When building exim an external PCRE library is now needed -
456 PCRE is a system library on the majority of modern systems.
457 See entry on PCRE_LIBS in EDITME file.
459 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
460 conversation. Added nologin parameter to request.
461 Patch contributed by Kirill Miazine.
463 TF/01 Do not log submission mode rewrites if they do not change the address.
465 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
467 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
468 log files in place. Contributed by Roberto Lima.
470 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
472 TF/03 Bugzilla 615: When checking the local_parts router precondition
473 after a local_part_suffix or local_part_prefix option, Exim now
474 does not use the address's named list lookup cache, since this
475 contains cached lookups for the whole local part.
477 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
478 Robert Millan. Documentation is in experimental-spec.txt.
480 TF/04 Bugzilla 668: Fix parallel build (make -j).
482 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
484 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
485 Patch provided by Jan Srzednicki.
487 TF/05 Leading white space used to be stripped from $spam_report which
488 wrecked the formatting. Now it is preserved.
490 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
491 that they are available at delivery time.
493 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
495 TF/08 TLS error reporting now respects the incoming_interface and
496 incoming_port log selectors.
498 TF/09 Produce a more useful error message if an SMTP transport's hosts
499 setting expands to an empty string.
501 NM/06 Bugzilla 744: EXPN did not work under TLS.
502 Patch provided by Phil Pennock.
504 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
505 Patch provided by Richard Godbee.
507 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
508 acl_smtp_notquit, added index entry.
510 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
511 Patch provided by Eugene Bujak.
513 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
514 accept(). Patch provided by Maxim Dounin.
516 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
517 Patch provided by Phil Pennock.
519 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
521 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
522 Patch provided by Brad "anomie" Jorsch.
524 NM/14 Bugzilla 622: Added timeout setting to transport filter.
525 Patch provided by Dean Brooks.
527 TK/05 Add native DKIM support (does not depend on external libraries).
529 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
530 Patch provided by Graeme Fowler.
532 NM/16 Bugzilla 851: Documentation example syntax fix.
534 NM/17 Changed NOTICE file to remove references to embedded PCRE.
536 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
539 NM/19 Bugzilla 745: TLS version reporting.
540 Patch provided by Phil Pennock.
542 NM/20 Bugzilla 167: bool: condition support.
543 Patch provided by Phil Pennock.
545 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
546 clients. Patch provided by Phil Pennock.
548 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
549 Patch provided by Brad "anomie" Jorsch.
551 NM/23 Bugzilla 687: Fix misparses in eximstats.
552 Patch provided by Heiko Schlittermann.
554 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
555 Patch provided by Heiko Schlittermann.
557 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
558 plus update to original patch.
560 NM/26 Bugzilla 799: Documentation correction for ratelimit.
562 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
563 Patch provided by David Brownlee.
565 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
567 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
569 NM/30 Bugzilla 888: TLS documentation bugfixes.
571 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
573 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
574 Unlike the original bugzilla I have changed all shell scripts in src tree.
576 NM/33 Bugzilla 898: Transport filter timeout fix.
577 Patch by Todd Rinaldo.
579 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
580 Patch by Serge Demonchaux.
582 NM/35 Bugzilla 39: Base64 decode bug fixes.
583 Patch by Jakob Hirsch.
585 NM/36 Bugzilla 909: Correct connect() call in dcc code.
587 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
589 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
591 NM/39 Bugzilla 911: Fixed MakeLinks build script.
597 TK/01 Add preliminary DKIM support. Currently requires a forked version of
598 ALT-N's libdkim that I have put here:
599 http://duncanthrax.net/exim-experimental/
601 Note to Michael Haardt: I had to rename some vars in sieve.c. They
602 were called 'true' and it seems that C99 defines that as a reserved
603 keyword to be used with 'bool' variable types. That means you could
604 not include C99-style headers which use bools without triggering
605 build errors in sieve.c.
607 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
608 as mailq or other aliases. Changed the --help handling significantly
609 to do whats expected. exim_usage() emits usage/help information.
611 SC/01 Added the -bylocaldomain option to eximstats.
613 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
615 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
617 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
623 PH/01 Another patch from the Sieve maintainer.
625 PH/02 When an IPv6 address is converted to a string for single-key lookup
626 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
627 dots are used instead of colons so that keys in lsearch files need not
628 contain colons. This was done some time before quoting was made available
629 in lsearch files. However, iplsearch files do require colons in IPv6 keys
630 (notated using the quote facility) so as to distinguish them from IPv4
631 keys. This meant that lookups for IP addresses in host lists did not work
632 for iplsearch lookups.
634 This has been fixed by arranging for IPv6 addresses to be expressed with
635 colons if the lookup type is iplsearch. This is not incompatible, because
636 previously such lookups could never work.
638 The situation is now rather anomolous, since one *can* have colons in
639 ordinary lsearch keys. However, making the change in all cases is
640 incompatible and would probably break a number of configurations.
642 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
645 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
646 conversion specification without a maximum field width, thereby enabling
647 a rogue spamd server to cause a buffer overflow. While nobody in their
648 right mind would setup Exim to query an untrusted spamd server, an
649 attacker that gains access to a server running spamd could potentially
650 exploit this vulnerability to run arbitrary code as the Exim user.
652 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
653 $primary_hostname instead of what libspf2 thinks the hosts name is.
655 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
656 a directory entry by the name of the lookup key. Previously, if a
657 symlink pointed to a non-existing file or a file in a directory that
658 Exim lacked permissions to read, a lookup for a key matching that
659 symlink would fail. Now it is enough that a matching directory entry
660 exists, symlink or not. (Bugzilla 503.)
662 PH/03 The body_linecount and body_zerocount variables are now exported in the
665 PH/04 Added the $dnslist_matched variable.
667 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
668 This means they are set thereafter only if the connection becomes
671 PH/06 Added the client_condition to authenticators so that some can be skipped
672 by clients under certain conditions.
674 PH/07 The error message for a badly-placed control=no_multiline_responses left
675 "_responses" off the end of the name.
677 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
679 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
680 (without spaces) instead of just copying the configuration text.
682 PH/10 Added the /noupdate option to the ratelimit ACL condition.
684 PH/11 Added $max_received_linelength.
686 PH/12 Added +ignore_defer and +include_defer to host lists.
688 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
689 way in which PCRE > 7.0 is built.
691 PH/14 Implemented queue_only_load_latch.
693 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
694 MAIL command. The effect was to mangle the value on 64-bit systems.
696 PH/16 Another patch from the Sieve maintainer.
698 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
700 PH/18 If a system quota error occurred while trying to create the file for
701 a maildir delivery, the message "Mailbox is full" was not appended to the
702 bounce if the delivery eventually timed out. Change 4.67/27 below applied
703 only to a quota excession during the actual writing of the file.
705 PH/19 It seems that peer DN values may contain newlines (and other non-printing
706 characters?) which causes problems in log lines. The DN values are now
707 passed through string_printing() before being added to log lines.
709 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
710 and InterBase are left for another time.)
712 PH/21 Added message_body_newlines option.
714 PH/22 Guard against possible overflow in moan_check_errorcopy().
716 PH/23 POSIX allows open() to be a macro; guard against that.
718 PH/24 If the recipient of an error message contained an @ in the local part
719 (suitably quoted, of course), incorrect values were put in $domain and
720 $local_part during the evaluation of errors_copy.
726 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
727 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
730 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
731 issue a MAIL command.
733 PH/02 In an ACL statement such as
735 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
737 if a client was not listed at all, or was listed with a value other than
738 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
739 the condition was not true (as it should be), so access was not denied.
740 The bug was that the ! inversion was incorrectly passed on to the second
741 item. This has been fixed.
743 PH/03 Added additional dnslists conditions == and =& which are different from
744 = and & when the dns lookup returns more than one IP address.
746 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
747 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
749 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
750 FSYNC, which compiles an option called disable_fsync that allows for
751 bypassing fsync(). The documentation is heavily laced with warnings.
753 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
755 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
756 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
757 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
758 including adding "make clean"; (3) Added -fPIC when compiling the test
759 dynamically loaded module, to get rid of a warning.
761 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
762 message fails, move_frozen_messages = true and ignore_bounce_errors_after
763 = 0s. The bug is otherwise harmless.
765 PH/07 There was a bug in the dovecot authenticator such that the value of
766 $auth1 could be overwritten, and so not correctly preserved, after a
767 successful authentication. This usually meant that the value preserved by
768 the server_setid option was incorrect.
770 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
772 PH/09 Installed PCRE release 7.0.
774 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
775 run for batched SMTP input. It is now run at the start of every message
776 in the batch. While fixing this I discovered that the process information
777 (output by running exiwhat) was not always getting set for -bs and -bS
778 input. This is fixed, and it now also says "batched" for BSMTP.
780 PH/11 Added control=no_pipelining.
782 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
783 patch, slightly modified), and move the expansion of helo_data till after
784 the connection is made in the smtp transport (so it can use these
787 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
789 PH/14 Added log_selector = +pid.
791 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
793 PH/16 Add ${if forany and ${if forall.
795 PH/17 Added dsn_from option to vary the From: line in DSNs.
797 PH/18 Flush SMTP output before performing a callout, unless control =
798 no_callout_flush is set.
800 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
801 was true (the default) a successful delivery failed to delete the retry
802 item, thus causing premature timeout of the address. The bug is now
805 PH/20 Added hosts_avoid_pipelining to the smtp transport.
807 PH/21 Long custom messages for fakedefer and fakereject are now split up
808 into multiline reponses in the same way that messages for "deny" and
809 other ACL rejections are.
811 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
812 with slight modification.
814 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
815 draft, changing the syntax and factoring some duplicate code".
817 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
818 for deliveries of the second and subsequent messages over the same SMTP
821 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
822 ${reduce, with only minor "tidies".
824 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
826 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
827 expansion side effects.
829 PH/27 When a message times out after an over-quota error from an Exim-imposed
830 quota, the bounce message says "mailbox is full". This message was not
831 being given when it was a system quota that was exceeded. It now should
834 MH/03 Made $recipients available in local_scan(). local_scan() already has
835 better access to the recipient list through recipients_list[], but
836 $recipients can be useful in postmaster-provided expansion strings.
838 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
839 in the case of a MAIL command with additional options following the
840 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
841 were accidentally chopped off.
843 PH/29 SMTP synchronization checks are implemented when a command is read -
844 there is a check that no more input is waiting when there shouldn't be
845 any. However, for some commands, a delay in an ACL can mean that it is
846 some time before the response is written. In this time, more input might
847 arrive, invalidly. So now there are extra checks after an ACL has run for
848 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
849 pipelining has not been advertised.
851 PH/30 MH's patch to allow iscntrl() characters to be list separators.
853 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
854 returned in the SMTP response when smtp_return_error_details was false.
857 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
858 instead of the C I/O that was originally supplied, because problems were
861 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
862 Exim which did not show up earlier: it was assuming that a call to
863 SSL_CTX_set_info_callback() might give an error value. In fact, there is
864 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
865 was a macro that became an assignment, so it seemed to work. This has
866 changed to a proper function call with a void return, hence the compile
867 error. Exim's code has been fixed.
869 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
872 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
874 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
876 PH/37 If a message is not accepted after it has had an id assigned (e.g.
877 because it turns out to be too big or there is a timeout) there is no
878 "Completed" line in the log. When some messages of this type were
879 selected by exigrep, they were listed as "not completed". Others were
880 picked up by some special patterns. I have improved the selection
881 criteria to be more general.
883 PH/38 The host_find_failed option in the manualroute router can now be set
884 to "ignore", to completely ignore a host whose IP address cannot be
885 found. If all hosts are ignored, the behaviour is controlled by the new
886 host_all_ignored option.
888 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
889 homing or because of multiple MX records with /mx) generated more than
890 one IP address, and the following item turned out to be the local host,
891 all the secondary addresses of the first item were incorrectly removed
892 from the list, along with the local host and any following hosts (which
893 is what is supposed to happen).
895 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
896 whoever called Exim into the -H file. In the case of the daemon it was
897 behaving confusingly. When first started, it used values for whoever
898 started the daemon, but after a SIGHUP it used the Exim user (because it
899 calls itself on a restart). I have changed the code so that it now always
902 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
903 message are rejected with the same error (e.g. no authentication or bad
904 sender address), and a DATA command is nevertheless sent (as can happen
905 with PIPELINING or a stupid MUA), the error message that was given to the
906 RCPT commands is included in the rejection of the DATA command. This is
907 intended to be helpful for MUAs that show only the final error to their
910 PH/42 Another patch from the Sieve maintainer.
912 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
913 Eximstats - Fixed some broken HTML links and added missing column headers
915 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
916 columns for Rejects, Temp Rejects, Ham, and Spam rows.
918 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
920 PH/43 Yet another patch from the Sieve maintainer.
922 PH/44 I found a way to check for a TCP/IP connection going away before sending
923 the response to the final '.' that terminates a message, but only in the
924 case where the client has not sent further data following the '.'
925 (unfortunately, this is allowed). However, in many cases there won't be
926 any further data because there won't be any more messages to send. A call
927 to select() can be used: if it shows that the input is "ready", there is
928 either input waiting, or the socket has been closed. An attempt to read
929 the next input character can distinguish the two cases. Previously, Exim
930 would have sent an OK response which the client would never have see.
931 This could lead to message repetition. This fix should cure that, at
932 least in a lot of common cases.
934 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
935 advertised in response to EHLO.
941 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
942 fixed by 4.65/MH/01 (is this a record?) are fixed:
944 (i) An empty string was always treated as zero by the numeric comparison
945 operators. This behaviour has been restored.
947 (ii) It is documented that the numeric comparison operators always treat
948 their arguments as decimal numbers. This was broken in that numbers
949 starting with 0 were being interpreted as octal.
951 While fixing these problems I realized that there was another issue that
952 hadn't been noticed. Values of message_size_limit (both the global option
953 and the transport option) were treated as octal if they started with 0.
954 The documentation was vague. These values are now always treated as
955 decimal, and I will make that clear in the documentation.
961 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
962 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
965 MH/01 Don't check that the operands of numeric comparison operators are
966 integers when their expansion is in "skipping" mode (fixes bug
967 introduced by 4.64-PH/07).
969 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
970 child addresses, Exim now panics and dies. Previously, because the count
971 is held in a short int, deliveries were likely to be lost. As such a
972 large number of recipients for a single message is ridiculous
973 (performance will be very, very poor), I have chosen to impose a limit
974 rather than extend the field.
980 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
981 leftover -K file (the existence of which was triggered by #402).
982 While we were at it, introduced process PID as part of the -K
983 filename. This should rule out race conditions when creating
986 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
987 processing considerably. Previous code took too long for large mails,
988 triggering a timeout which in turn triggers #401.
990 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
991 in the DK code in transports.c. sendfile() is not really portable,
992 hence the _LINUX specificness.
994 TF/01 In the add_headers option to the mail command in an Exim filter,
995 there was a bug that Exim would claim a syntax error in any
996 header after the first one which had an odd number of characters
999 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1000 callout verification, Exim cached a "reject" for the entire domain. This
1001 is correct for most verifications, but it is not correct for a recipient
1002 verification with use_sender or use_postmaster set, because in that case
1003 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1004 case of MAIL FROM:<> rejection from other early rejections (e.g.
1005 rejection of HELO). When verifying a recipient using a non-null MAIL
1006 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1007 Whatever the result of the callout, the value of the domain cache is
1008 left unchanged (for any other kind of callout, getting as far as trying
1009 RCPT means that the domain itself is ok).
1011 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1014 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1015 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1016 session, the daemon ignores ECONNECT errors and logs others; it now
1017 ignores EPIPE as well.
1019 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1020 (quoted-printable decoding).
1022 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1023 later the small subsequent patch to fix an introduced bug.
1025 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1027 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1029 PH/08 An error is now given if message_size_limit is specified negative.
1031 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1032 to be given (somewhat) arbitrary names.
1034 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1037 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1038 miscellaneous code fixes
1040 PH/10 Added the log_reject_target ACL modifier to specify where to log
1043 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1044 hostname. This is wrong, because it relates to the incoming message (and
1045 probably the interface on which it is arriving) and not to the outgoing
1046 callout (which could be using a different interface). This has been
1047 changed to use the value of the helo_data option from the smtp transport
1048 instead - this is what is used when a message is actually being sent. If
1049 there is no remote transport (possible with a router that sets up host
1050 addresses), $smtp_active_hostname is used.
1052 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1053 tweaks were necessary in order to get it to work (see also 21 below):
1054 (a) The code assumed that strncpy() returns a negative number on buffer
1055 overflow, which isn't the case. Replaced with Exim's string_format()
1057 (b) There were several signed/unsigned issues. I just did the minimum
1058 hacking in of casts. There is scope for a larger refactoring.
1059 (c) The code used strcasecmp() which is not a standard C function.
1060 Replaced with Exim's strcmpic() function.
1061 (d) The code set only $1; it now sets $auth1 as well.
1062 (e) A simple test gave the error "authentication client didn't specify
1063 service in request". It would seem that Dovecot has changed its
1064 interface. Fortunately there's a specification; I followed it and
1065 changed what the client sends and it appears to be working now.
1067 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1070 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1071 address A is aliased to B and C, where B exists and C does not. Without
1072 -v the output is "A verified" because verification stops after a
1073 successful redirection if more than one address is generated. However,
1074 with -v the child addresses are also verified. Exim was outputting "A
1075 failed to verify" and then showing the successful verification for C,
1076 with its parentage. It now outputs "B failed to verify", showing B's
1077 parentage before showing the successful verification of C.
1079 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1080 look up a TXT record in a specific list after matching in a combined
1083 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1084 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1085 they consult the DNS. I had assumed they would set it the way they
1086 wanted; and indeed my experiments on Linux seem to show that in some
1087 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1088 To be on the safe side, however, I have now made the interface to
1089 host_find_byname() similar to host_find_bydns(), with an argument
1090 containing the DNS resolver options. The host_find_byname() function now
1091 sets these options at its start, just as host_find_bydns() does. The smtp
1092 transport options dns_qualify_single and dns_search_parents are passed to
1093 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1094 of host_find_byname() use the default settings of RES_DEFNAMES
1095 (qualify_single) but not RES_DNSRCH (search_parents).
1097 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1098 spool_read_header() do less string testing, by means of a preliminary
1099 switch on the second character of optional "-foo" lines. (This is
1100 overdue, caused by the large number of possibilities that now exist.
1101 Originally there were few.) While I was there, I also converted the
1102 str(n)cmp tests so they don't re-test the leading "-" and the first
1103 character, in the hope this might squeeze out yet more improvement.
1105 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1106 flag allowing group syntax was set by the header_syntax check but not
1107 turned off, possible causing trouble later; (2) The flag was not being
1108 set at all for the header_verify test, causing "group"-style headers to
1109 be rejected. I have now set it in this case, and also caused header_
1110 verify to ignore an empty address taken from a group. While doing this, I
1111 came across some other cases where the code for allowing group syntax
1112 while scanning a header line wasn't quite right (mostly, not resetting
1113 the flag correctly in the right place). These bugs could have caused
1114 trouble for malformed header lines. I hope it is now all correct.
1116 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1117 with the "reply" argument non-NULL. The code, however (which originally
1118 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1119 but it didn't always do it. This confused somebody who was copying the
1120 code for some other use. I have removed all the tests.
1122 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1123 feature that was used to support insecure browsers during the U.S. crypto
1124 embargo. It requires special client support, and Exim is probably the
1125 only MTA that supported it -- and would never use it because real RSA is
1126 always available. This code has been removed, because it had the bad
1127 effect of slowing Exim down by computing (never used) parameters for the
1128 RSA_EXPORT functionality.
1130 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1131 authenticator to fail if there's a tab character in the incoming data
1132 (there should never be unless someone is messing about, as it's supposed
1133 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1134 if the connection is using TLS or if the remote IP is the same as the
1135 local IP, and the "valid-client-cert option" if a client certificate has
1138 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1139 authenticators. This can be used for authorization after authentication
1140 succeeds. (In the case of plaintext, it servers for both authentication
1143 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1144 if any retry times were supplied.
1146 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1147 connection, where there is no client IP address to check. In this
1148 situation, the verify now always succeeds.
1150 PH/25 Applied John Jetmore's -Mset patch.
1152 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1154 PH/27 In a string expansion for a processed (not raw) header when multiple
1155 headers of the same name were present, leading whitespace was being
1156 removed from all of them, but trailing whitespace was being removed only
1157 from the last one. Now trailing whitespace is removed from each header
1158 before concatenation. Completely empty headers in a concatenation (as
1159 before) are ignored.
1161 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1162 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1164 PH/29 [Removed. This was a change that I later backed out, and forgot to
1165 correct the ChangeLog entry (that I had efficiently created) before
1166 committing the later change.]
1168 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1169 address errors (4xx response to RCPT) over the same connection as other
1170 messages routed to the same hosts. Such deliveries are always "forced",
1171 so retry times are not inspected. This resulted in far too many retries
1172 for the affected addresses. The effect occurred only when there were more
1173 hosts than the hosts_max_try setting in the smtp transport when it had
1174 the 4xx errors. Those hosts that it had tried were not added to the list
1175 of hosts for which the message was waiting, so if all were tried, there
1176 was no problem. Two fixes have been applied:
1178 (i) If there are any address or message errors in an SMTP delivery, none
1179 of the hosts (tried or untried) are now added to the list of hosts
1180 for which the message is waiting, so the message should not be a
1181 candidate for sending over the same connection that was used for a
1182 successful delivery of some other message. This seems entirely
1183 reasonable: after all the message is NOT "waiting for some host".
1184 This is so "obvious" that I'm not sure why it wasn't done
1185 previously. Hope I haven't missed anything, but it can't do any
1186 harm, as the worst effect is to miss an optimization.
1188 (ii) If, despite (i), such a delivery is accidentally attempted, the
1189 routing retry time is respected, so at least it doesn't keep
1190 hammering the server.
1192 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1193 in ${readsocket because some servers need this prod.
1195 PH/32 Added some extra debug output when updating a wait-xxx database.
1197 PH/33 The hint "could be header name not terminated by colon", which has been
1198 given for certain expansion errors for a long time, was not being given
1199 for the ${if def:h_colon_omitted{... case.
1201 PH/34 The spec says: "With one important exception, whenever a domain list is
1202 being scanned, $domain contains the subject domain." There was at least
1203 one case where this was not true.
1205 PH/35 The error "getsockname() failed: connection reset by peer" was being
1206 written to the panic log as well as the main log, but it isn't really
1207 panic-worthy as it just means the connection died rather early on. I have
1208 removed the panic log writing for the ECONNRESET error when getsockname()
1211 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1212 runs only) independently of the message's sender address. This meant
1213 that, if the 4xx error was in fact related to the sender, a different
1214 message to the same recipient with a different sender could confuse
1215 things. In particualar, this can happen when sending to a greylisting
1216 server, but other circumstances could also provoke similar problems.
1217 I have changed the default so that the retry time for these errors is now
1218 based a combination of the sender and recipient addresses. This change
1219 can be overridden by setting address_retry_include_sender=false in the
1222 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1223 remote server are returned as part of bounce messages. This was not
1224 happening for LMTP over a pipe (the lmtp transport), but now it is the
1225 same for both kinds of LMTP.
1227 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1228 in header lines that were in fact CNAMEs. This is no longer the case.
1230 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1231 and queue runs started by the daemon processed all messages. This has
1232 been fixed so that -R and -S can now usefully be given with -q<time>.
1234 PH/40 Import PCRE release 6.7 (fixes some bugs).
1236 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1238 PH/42 Give an error if -q is specified more than once.
1240 PH/43 Renamed the variables $interface_address and $interface_port as
1241 $received_ip_address and $received_port, to make it clear that these
1242 values apply to message reception, and not to the outgoing interface when
1243 a message is delivered. (The old names remain recognized, of course.)
1245 PH/44 There was no timeout on the connect() call when using a Unix domain
1246 socket in the ${readsocket expansion. There now is.
1248 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1249 be meaningful with "accept".
1251 SC/01 Eximstats V1.43
1252 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1254 SC/02 Eximstats V1.44
1255 Use a glob alias rather than an array ref in the generated
1256 parser. This improves both readability and performance.
1258 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1259 Collect SpamAssassin and rejection statistics.
1260 Don't display local sender or destination tables unless
1261 there is data to show.
1262 Added average volumes into the top table text output.
1264 SC/04 Eximstats V1.46
1265 Collect data on the number of addresses (recipients)
1266 as well as the number of messages.
1268 SC/05 Eximstats V1.47
1269 Added 'Message too big' to the list of mail rejection
1270 reasons (thanks to Marco Gaiarin).
1272 SC/06 Eximstats V1.48
1273 Mainlog lines which have GMT offsets and are too short to
1274 have a flag are now skipped.
1276 SC/07 Eximstats V1.49 (Alain Williams)
1277 Added the -emptyok flag.
1279 SC/08 Eximstats V1.50
1280 Fixes for obtaining the IP address from reject messages.
1282 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1283 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1284 whitesspace changes from 4.64-PH/27
1286 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1289 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1290 are found, allow negative numbers in numeric criteria)
1292 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1294 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1295 to match changes made in 4.64-PH/43
1297 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1299 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1300 "domain" is actually an IP address, and force a failure. This locks out
1301 those revolvers/nameservers that support "A-for-A" lookups, in
1302 contravention of the specifications.
1304 PH/48 When a host name was looked up from an IP address, and the subsequent
1305 forward lookup of the name timed out, the host name was left in
1306 $sender_host_name, contrary to the specification.
1308 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1309 restricted to single-key lookups, Exim was not diagnosing an error if
1310 * or *@ was used with a query-style lookup.
1312 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1314 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1315 long ago, but noone interested enough thought of it. Let's just say that
1316 the "1.1" means that there are some new functions that weren't there at
1317 some point in the past.
1319 PH/51 Error processing for expansion failure of helo_data from an smtp
1320 transport during callout processing was broken.
1322 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1323 tested/used via the -bh/-bhc/-bs options.
1325 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1326 bug, fixed in subsequent PCRE releases).
1328 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1329 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1331 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1337 SC/01 Use a glob alias rather than an array ref in eximstats generated
1338 parser. This improves both readability and performance.
1340 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1341 Don't display local sender or destination tables in eximstats unless
1342 there is data to show.
1343 Added average volumes into the eximstats top table text output.
1345 SC/03 Collect data on the number of addresses (recipients) as well
1346 as the number of messages in eximstats.
1348 TF/01 Correct an error in the documentation for the redirect router. Exim
1349 does (usually) call initgroups() when daemonizing.
1351 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1352 with consistent privilege compared to when running as a daemon.
1354 TF/03 Note in the spec that $authenticated_id is not set for local
1355 submissions from trusted users.
1357 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1358 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1360 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1361 by adding some example configuration directives to the default
1362 configuration file. A little bit of work is required to uncomment the
1363 directives and define how usernames and passwords are checked, but
1364 there is now a framework to start from.
1366 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1367 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1368 without this. I don't know how relevant this is to other LDAP libraries.
1370 PH/02 Add the verb name to the "unknown ACL verb" error.
1372 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1374 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1376 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1377 directory not expanded when it should be if an expanded home directory
1378 was set for the address (which is overridden by the transport).
1380 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1383 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1384 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1385 because it is too late at that time, and has no effect.
1387 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1388 security issue with \' (bugzilla #107). I could not use the
1389 PQescapeStringConn() function, because it needs a PGconn value as one of
1392 PH/08 When testing addresses using -bt, indicate those final addresses that
1393 are duplicates that would not cause an additional delivery. At least one
1394 person was confused, thinking that -bt output corresponded to deliveries.
1395 (Suppressing duplicates isn't a good idea as you lose the information
1396 about possibly different redirections that led to the duplicates.)
1398 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1399 systems where poll() doesn't work, in particular OS X.
1401 PH/10 Added more information to debugging output for retry time not reached.
1403 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1404 operations in malware.c.
1406 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1409 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1410 syslog_duplication set false, log lines that would normally be written
1411 both the the main log and to the reject log were not written to syslog at
1414 PH/14 In the default configuration, change the use of "message" in ACL warn
1415 statements to "add_header".
1417 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1418 not followed by a command (e.g. "seen endif").
1420 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1421 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1424 PH/17 Added extra conditions to the default value of delay_warning_condition
1428 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1429 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1430 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1433 The Auto-Submitted: and various List- headers are standardised, whereas I
1434 don't think Precedence: ever was.
1436 PH/18 Refactored debugging code in route_finduser() to show more information,
1437 in particular, the error code if getpwnam() issues one.
1439 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1440 This is apparently needed in addition to the PH/07 change above to avoid
1441 any possible encoding problems.
1443 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1444 but not after initializing Perl.
1446 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1447 output them only if debugging. By default they are written stderr,
1448 apparently, which is not desirable.
1450 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1453 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1456 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1458 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1459 authenticated or an ident call has been made. Suppress the default
1460 values for $authenticated_id and $authenticated_sender (but permit -oMai
1461 and -oMas) when testing with -bh.
1463 PH/24 Re-jigged the order of the tests in the default configuration so that the
1464 tests for valid domains and recipients precede the DNS black list and CSA
1465 tests, on the grounds that those ones are more expensive.
1467 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1468 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1469 This bug exists in every version of Exim that I still have, right back to
1472 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1473 However, an attempt to turn on case-sensitivity in a regex key by
1474 including (?-i) didn't work because the subject string was already
1475 lowercased, and the effects were non-intuitive. It turns out that a
1476 one-line patch can be used to allow (?-i) to work as expected.
1482 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1483 other effects) broke the use of negated acl sub-conditions.
1485 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1488 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1489 "Deny" causes Exim to reject the incoming connection with a 554 error.
1490 Unfortunately, if there is a major crisis, such as a disk failure,
1491 tcp-wrappers gives "deny", whereas what one would like would be some
1492 kind of temporary error. A kludge has been added to help with this.
1493 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1494 554 error is used if errno is still zero or contains ENOENT (which occurs
1495 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1498 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1500 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1501 errors. Otherwise a message that provokes a temporary error (when other
1502 messages do not) can cause a whole host to time out.
1504 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1505 addresses were routed directly to files or pipes from a redirect router.
1506 File deliveries just didn't batch; pipe deliveries might have suffered
1509 PH/06 A failure to get a lock for a hints database would erroneously always say
1510 "Failed to get write lock", even when it was really a read lock.
1512 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1513 of 0600. This has been changed to use the value of the lockfile_mode
1514 option (which defaults to 0600).
1516 PH/08 Applied small patch from the Sieve maintainer.
1518 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1519 folder from quota calculations, a direct delivery into this folder messed
1520 up the contents of the maildirsize file. This was because the regex was
1521 used only to exclude .Trash (or whatever) when the size of the mailbox
1522 was calculated. There was no check that a delivery was happening into an
1523 excluded directory. This bug has been fixed by ignoring all quota
1524 processing for deliveries into excluded directories.
1526 PH/10 Added the maildirfolder_create_regex option to appendfile.
1532 PH/01 The code for finding all the local interface addresses on a FreeBSD
1533 system running IPv6 was broken. This may well have applied to all BSD
1534 systems, as well as to others that have similar system calls. The broken
1535 code found IPv4 interfaces correctly, but gave incorrect values for the
1536 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1537 that it would not match correctly against @[] and not recognize the IPv6
1540 PH/02 The ipliteral router was not recognizing addresses of the form user@
1541 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1543 PH/03 Added disable_ipv6.
1545 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1546 decoded form, because it is most often used to construct To: headers
1547 lines in autoreplies, and the decoded form may well be syntactically
1548 invalid. However, $reply_address has leading white space removed, and all
1549 newlines turned into spaces so that the autoreply transport does not
1552 PH/05 If group was specified without a user on a router, and no group or user
1553 was specified on a transport, the group from the router was ignored.
1555 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1556 for visible compile-time settings that can be used to change these
1557 numbers, for those that want even more. Backwards compatibility with old
1558 spool files has been maintained. However, going back to a previous Exim
1559 release will lost any variables that are in spool files.
1561 PH/07 Two small changes when running in the test harness: increase delay when
1562 passing a TCP/IP connection to a new process, in case the original
1563 process has to generate a bounce, and remove special handling of
1564 127.0.0.2 (sic), which is no longer necessary.
1566 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1567 be the same on different OS.
1569 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1572 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1573 whether --show-vars was specified or not
1575 JJ/02 exipick: Added support for new ACL variable spool format introduced
1578 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1579 syntactically invalid From: or Reply-to: line, and a filter used this to
1580 generate an autoreply, and therefore failed to obtain an address for the
1581 autoreply, Exim could try to deliver to a non-existent relative file
1582 name, causing unrelated and misleading errors. What now happens is that
1583 it logs this as a hard delivery error, but does not attempt to create a
1586 PH/11 The exinext utility has a -C option for testing purposes, but although
1587 the given file was scanned by exinext itself; it wasn't being passed on
1588 when Exim was called.
1590 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1591 an end-of-file indication when reading a command response.
1593 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1594 compiled. In many other places in Exim, IPv6 addresses are always
1595 recognized, so I have changed this. It also means that IPv4 domain
1596 literals of the form [IPV4:n.n.n.n] are now always recognized.
1598 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1599 used if the router is not running as root, for example, when verifying at
1600 ACL time, or when using -bh. The debugging output from this situation was
1601 non-existent - all you got was a failure to exec. I have made two
1604 (a) Failures to set uid/gid, the current directory, or a process leader
1605 in a subprocess such as that created by queryprogram now generate
1606 suitable debugging ouput when -d is set.
1608 (b) The queryprogram router detects when it is not running as root,
1609 outputs suitable debugging information if -d is set, and then runs
1610 the subprocess without attempting to change uid/gid.
1612 PH/15 Minor change to Makefile for building test_host (undocumented testing
1615 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1616 additional section of a DNS packet that returns MX or SRV records.
1617 Instead, it always explicitly searches for A/AAAA records. This avoids
1618 major problems that occur when a DNS server includes only records of one
1619 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1620 fixed another bug: if SRV records were looked up and the corresponding
1621 address records were *not* found in the additional section, the port
1622 values from the SRV records were lost.
1624 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1625 using the correct key (the original address) when searching the retry
1626 rules in order to find which one to use for generating the retry hint.
1628 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1629 adding the default one. Similarly, if it contains a Reply-To: header, the
1630 errors_reply_to option, if set, is not used.
1632 PH/19 When calculating a retry time, Exim used to measure the "time since
1633 failure" by looking at the "first failed" field in the retry record. Now
1634 it does not use this if it is later than than the arrival time of the
1635 message. Instead it uses the arrival time. This makes for better
1636 behaviour in cases where some deliveries succeed, thus re-setting the
1637 "first failed" field. An example is a quota failure for a huge message
1638 when small messages continue to be delivered. Without this change, the
1639 "time since failure" will always be short, possible causing more frequent
1640 delivery attempts for the huge message than are intended.
1641 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1643 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1644 $1, $2, $3) because the numerical variables can be reset during some
1645 expansion items (e.g. "match"), thereby losing the authentication data.
1647 PH/21 Make -bV show the size of off_t variables so that the test suite can
1648 decide whether to run tests for quotas > 2G.
1650 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1651 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1652 filecount value is greater than 2G or if a quota value is greater than 2G
1653 on a system where the size of off_t is not greater than 4, a panic error
1656 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1657 never match. The debug and -bh output now contains an explicit error
1658 message indicating a malformed IPv4 address or mask.
1660 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1661 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1662 PH/23 above applies.
1664 PH/25 Do not write to syslog when running in the test harness. The only
1665 occasion when this arises is a failure to open the main or panic logs
1666 (for which there is an explicit test).
1668 PH/26 Added the /no_tell option to "control=freeze".
1670 PH/27 If a host name lookup failed very early in a connection, for example, if
1671 the IP address matched host_lookup and the reverse lookup yielded a name
1672 that did not have a forward lookup, an error message of the form "no IP
1673 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1674 could be logged. Now it outputs the IP address instead of "NULL".
1676 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1677 allows the sender and the authenticated sender to be set when
1678 submitting a message from within Exim. Since child_open_exim() is
1679 documented for local_scan(), the new function should be too.
1681 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1682 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1683 results in an empty string is now treated as unset.
1685 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1687 PH/31 Added sender_verify_fail logging option.
1689 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1690 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1691 tidied the source and removed it altogether.
1693 PH/33 When a queue run was abandoned because the load average was too high, a
1694 log line was always written; now it is written only if the queue_run log
1695 selector is set. In addition, the log line for abandonment now contains
1696 information about the queue run such as the pid. This is always present
1697 in "start" and "stop" lines but was omitted from the "abandon" line.
1699 PH/34 Omit spaces between a header name and the colon in the error message that
1700 is given when verify = headers_syntax fails (if there are lots of them,
1701 the message gets confusing).
1703 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1704 names, as there are now some PTR records that contain slashes. This check
1705 is only to protect against broken name servers that fall over on strange
1706 characters, so the fact that it applies to all lookups doesn't matter.
1708 PH/36 Now that the new test suite is complete, we can remove some of the
1709 special code in Exim that was needed for the old test suite. For example,
1710 sorting DNS records because real resolvers return them in an arbitrary
1711 order. The new test suite's fake resolver always returns records in the
1714 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1715 bounces) except when queue_only is set, to avoid logging races between
1716 the different processes.
1718 PH/38 Panic-die if .include specifies a non-absolute path.
1720 PH/39 A tweak to the "H" retry rule from its user.
1722 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1723 a label. They prevented compilation on older perls.
1725 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1726 a warning to be raised on newish perls.
1728 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1729 on queue. Changes to match documented behaviour of showing count of
1730 messages matching specified criteria.
1732 PH/40 Changed the default ident timeout from 30s to 5s.
1734 PH/41 Added support for the use of login_cap features, on those BSD systems
1735 that have them, for controlling the resources used by pipe deliveries.
1737 PH/42 The content-scanning code uses fopen() to create files in which to put
1738 message data. Previously it was not paying any attention to the mode of
1739 the files. Exim runs with umask(0) because the rest of the code creates
1740 files with open(), and sets the required mode explicitly. Thus, these
1741 files were ending up world-writeable. This was not a big issue, because,
1742 being within the spool directory, they were not world-accessible. I have
1743 created a function called modefopen, which takes an additional mode
1744 argument. It sets umask(777), creates the file, chmods it to the required
1745 mode, then resets the umask. All the relevant calls to fopen() in the
1746 content scanning code have been changed to use this function.
1748 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1749 to 24 hours. This avoids potential overflow problems when processing G
1750 and H retry rules. I suspect nobody ever tinkers with this value.
1752 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1754 PH/45 When the plaintext authenticator is running as a client, the server's
1755 challenges are checked to ensure they are valid base64 strings. By
1756 default, the authentication attempt is cancelled if an invalid string is
1757 received. Setting client_ignore_invalid_base64 true ignores these errors.
1758 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1759 they are received. Thus, the responses can be made to depend on the
1760 challenges. If an invalid string is ignored, an empty string is placed in
1763 PH/46 Messages that are created by the autoreply transport now contains a
1764 References: header, in accordance with RFCs 2822 and 3834.
1766 PH/47 Added authenticated_sender_force to the smtp transport.
1768 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1770 PH/49 Installed latest patch from the Sieve maintainer.
1772 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1773 was also set, the appendfile transport was unnecessarily scanning a
1774 directory of message files (e.g. for maildir delivery) to find the count
1775 of files (along with the size), even though it did not need this
1776 information. It now does the scan only if it needs to find either the
1777 size of the count of files.
1779 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1781 PH/52 Two bugs concerned with error handling when the smtp transport is
1784 (i) Exim was not creating retry information for temporary errors given
1785 for individual recipients after the DATA command when the smtp transport
1786 was used in LMTP mode. This meant that they could be retried too
1787 frequently, and not timed out correctly.
1789 (ii) Exim was setting the flag that allows error details to be returned
1790 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1791 recipients that were returned after the DATA command.
1793 PH/53 This is related to PH/52, but is more general: for any failing address,
1794 when detailed error information was permitted to be returned to the
1795 sender, but the error was temporary, then after the final timeout, only
1796 "retry timeout exceeded" was returned. Now it returns the full error as
1797 well as "retry timeout exceeded".
1799 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1800 do this, and (what is worse) MTAs that accept it.
1802 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1803 will now be deprecated.
1805 PH/56 New os.c-cygwin from the Cygwin maintainer.
1807 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1808 formats (previously only available in exim formats via -bpr, -bpru,
1809 and -bpra. Now also available in native and exiqgrep formats)
1811 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1812 with very large, slow to parse queues
1814 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1816 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1818 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1819 responses to authentication challenges, though it was showing the
1820 challenges; (ii) I've removed the CR characters from the debug output for
1823 PH/58 Allow for the insertion of a newline as well as a space when a string
1824 is turned into more than one encoded-word during RFC 2047 encoding. The
1825 Sieve code now uses this.
1827 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1828 data_4xx, lost_connection, tls_required.
1830 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1831 message was being sent as an SMTP response.
1833 PH/61 Add -l and -k options to exicyclog.
1835 PH/62 When verifying, if an address was redirected to one new address, so that
1836 verification continued, and the new address failed or deferred after
1837 having set something in $address_data, the value of $address_data was not
1838 passed back to the ACL. This was different to the case when no
1839 redirection occurred. The value is now passed back in both cases.
1841 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1842 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1843 use login_cap.h, so on its own it isn't the distinguishing feature. The
1844 new name refers directly to the setclassresources() function.
1846 PH/65 Added configuration files for NetBSD3.
1848 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1850 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1851 is preferred over IPv4.
1853 PH/68 The bounce_return_message and bounce_return_body options were not being
1854 honoured for bounces generated during the reception of non-SMTP messages.
1855 In particular, this applied to messages rejected by the ACL. This bug has
1856 been fixed. However, if bounce_return_message is true and bounce_return_
1857 body is false, the headers that are returned for a non-SMTP message
1858 include only those that have been read before the error was detected.
1859 (In the case of an ACL rejection, they have all been read.)
1861 PH/69 The HTML version of the specification is now built in a directory called
1862 spec_html instead of spec.html, because the latter looks like a path with
1863 a MIME-type, and this confuses some software.
1865 PH/70 Catch two compiler warnings in sieve.c.
1867 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1868 function verify_get_ident() calls ip_connect() to connect a socket, but
1869 if the "connect()" function timed out, ip_connect() used to close the
1870 socket. However, verify_get_ident() also closes the socket later, and in
1871 between Exim writes to the log, which may get opened at this point. When
1872 the socket was closed in ip_connect(), the log could get the same file
1873 descriptor number as the socket. This naturally causes chaos. The fix is
1874 not to close the socket in ip_connect(); the socket should be closed by
1875 the function that creates it. There was only one place in the code where
1876 this was missing, in the iplookup router, which I don't think anybody now
1877 uses, but I've fixed it anyway.
1879 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1880 well as to direct DNS lookups. Otherwise the handling of names in host
1881 lists is inconsistent and therefore confusing.
1887 PH/01 Two changes to the default runtime configuration:
1889 (1) Move the checks for relay_from_hosts and authenticated clients from
1890 after to before the (commented out) DNS black list checks.
1892 (2) Add control=submission to the relay_from_hosts and authenticated
1893 clients checks, on the grounds that messages accepted by these
1894 statements are most likely to be submissions.
1896 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1898 (1) Generate an error if the third argument for the ${prvs expansion is
1901 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1904 (3) Reset the variables that are obtained from the first argument of
1905 ${prvscheck and used in the second argument before leaving the code,
1906 because their memory is reclaimed, so using them afterwards may do
1909 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1910 one (it's much easier than Tom thought :-).
1912 (5) Because of (4), we can now allow for the use of $prvscheck_result
1913 inside the third argument.
1915 PH/03 For some reason, the default setting of PATH when running a command from
1916 a pipe transport was just "/usr/bin". I have changed it to
1919 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1920 anything to be listed in the output from -bV.
1922 PH/05 When a filter generated an autoreply, the entire To: header line was
1923 quoted in the delivery log line, like this:
1925 => >A.N.Other <ano@some.domain> <original@ddress> ...
1927 This has been changed so that it extracts the operative address. There
1928 may be more than one such address. If so, they are comma-separated, like
1931 => >ano@some.domain,ona@other.domain <original@ddress> ...
1933 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1934 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1935 not being looked up in the rDNS to get a host name, Exim was showing the
1936 IP address twice in Received: lines, even though the IP addresses were
1937 identical. For example:
1939 Received: from [1.2.3.4] (helo=[1.2.3.4])
1941 However, if the real host name was known, it was omitting the HELO data
1942 if it matched the actual IP address. This has been tidied up so that it
1943 doesn't show the same IP address twice.
1945 PH/07 When both +timestamp and +memory debugging was on, the value given by
1946 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1947 called by the debug printing, thereby overwriting the timestamp buffer.
1948 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1950 PH/08 When the original message was included in an autoreply transport, it
1951 always said "this is a copy of the message, including all the headers",
1952 even if body_only or headers_only was set. It now gives an appropriate
1955 PH/09 Applied a patch from the Sieve maintainer which:
1957 o fixes some comments
1958 o adds the (disabled) notify extension core
1959 o adds some debug output for the result of if/elsif tests
1960 o points to the current vacation draft in the documentation
1961 and documents the missing references header update
1965 o fixes a bug in processing the envelope test (when testing
1966 multiple envelope elements, the last element determinted the
1969 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1970 Electronic Mail") by including:
1972 Auto-submitted: auto-generated
1974 in the messages that it generates (bounce messages and others, such as
1975 warnings). In the case of bounce messages for non-SMTP mesages, there was
1976 also a typo: it was using "Auto_submitted" (underscore instead of
1977 hyphen). Since every message generated by Exim is necessarily in response
1978 to another message, thes have all been changed to:
1980 Auto-Submitted: auto-replied
1982 in accordance with these statements in the RFC:
1984 The auto-replied keyword:
1986 - SHOULD be used on messages sent in direct response to another
1987 message by an automatic process,
1989 - MUST NOT be used on manually-generated messages,
1991 - MAY be used on Delivery Status Notifications (DSNs) and Message
1992 Disposition Notifications (MDNs),
1994 - MUST NOT be used on messages generated by automatic or periodic
1995 processes, except for messages which are automatic responses to
1998 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1999 to the default Received: header definition.
2001 PH/12 Added log selector acl_warn_skipped (default on).
2003 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2004 variables because (a) they are in the wrong storage pool and (b) even if
2005 they were copied, it wouldn't work properly because of the caching.
2007 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2008 checking when decoding. Apparently there are clients that generate
2009 overlong encoded strings. Why am I not surprised?
2011 PH/15 If the first argument of "${if match_address" was not empty, but did not
2012 contain an "@" character, Exim crashed. Now it writes a panic log message
2013 and treats the condition as false.
2015 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2017 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2018 extension "envelope-auth". The code is finished and in agreement with
2019 other implementations, but there is no documentation so far and in fact,
2020 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2021 not changing the active code.
2023 Print executed "if" and "elsif" statements when debugging is used. This
2024 helps a great deal to understand what a filter does.
2026 Document more things not specified clearly in RFC3028. I had all this
2027 sorted out, when out of a sudden new issues came to my mind. Oops."
2029 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2032 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2033 canonical form (as documented). However, after a host name lookup from
2034 the IP address, check_host() was doing a simple string comparison with
2035 addresses acquired from the DNS when checking that the found name did
2036 have the original IP as one of its addresses. Since any found IPv6
2037 addresses are likely to be in abbreviated form, the comparison could
2038 fail. Luckily, there already exists a function for doing the comparison
2039 by converting both addresses to binary, so now that is used instead of
2040 the text comparison.
2042 PH/20 There was another similar case to PH/19, when a complete host name was
2043 given in a host list; looking up its IP address could give an abbreviated
2044 form, whereas the current host's name might or might not be abbreviated.
2045 The same fix has been applied.
2051 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2052 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2055 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2056 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2058 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2060 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2061 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2062 Also, for "Auto-Submitted", treat anything other than "no" as
2063 non-personal, in accordance with RFC 3834. (Previously it treated
2064 anything starting "auto-" as non-personal.)
2066 TF/01 The control=submission/name=... option had a problem with syntax
2067 errors if the name included a slash character. The /name= option
2068 now slurps the rest of the string, so it can include any characters
2069 but it must come last in the list of options (after /sender_retain
2072 PH/05 Some modifications to the interface to the fake nameserver for the new
2080 TK/01 Added the "success_on_redirect" address verification option. See
2081 NewStuff for rationale and an example.
2083 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2085 PH/02 Patch to exigrep to allow it to work on syslog lines.
2087 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2088 fread() to skip over the body file's header line, because in Cygwin the
2089 header line is locked and is inaccessible.
2091 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2092 co-exist for some time) to make it clear that it is the Exim ID that is
2093 referenced, not the Message-ID: header line.
2095 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2096 string_format() function, because snprintf() does not exist on all
2099 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2100 ${stat: expansion item.
2102 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2103 protocol synchronization error", to keep the pedants happy.
2105 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2106 well as for IRIX systems, when gcc is being used. See the host.c source
2109 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2111 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2114 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2115 transport and to the smtp transport in LMTP mode.
2117 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2119 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2120 run a filter in a subprocess. This could lead to confusion in subsequent
2121 lookups in the parent process. There should also be a search_tidyup() at
2122 the end of the subprocess.
2124 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2125 only if the host matched helo_try_verify_hosts, which caused the
2126 verification to occur when the EHLO/HELO command was issued. The ACL just
2127 tested the remembered result. Now, if a previous verification attempt has
2128 not happened, "verify = helo" does it there and then.
2130 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2132 TK/03 Fix log output including CR from clamd.
2134 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2135 exist provoked a memory error which could cause a segfault.
2137 PH/15 Installed PCRE 6.2
2139 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2141 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2142 of the problem. Specifically, suggested +O2 rather than +O1 for the
2145 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2147 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2148 record, the port was not correctly logged when the outgoing_port log
2149 selector was set (it logged the transort's default port).
2151 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2152 fallback_hosts, and "hosts" in the smtp transport.
2154 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2155 host errors such as "Connection refused".
2157 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2158 authentication with radiusclient 0.4.9:
2160 - Error returned from rc_read_config was caught wrongly
2161 - Username/password not passed on to radius server due to wrong length.
2163 The presumption is that some radiusclient API changes for 4.51/PH/17
2164 were not taken care of correctly. The code is still untested by me (my
2165 Linux distribution still has 0.3.2 of radiusclient), but it was
2166 contributed by a Radius user.
2168 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2169 expanding the "port" option of the smtp transport.
2171 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2172 while reading a MIME header. Thanks to Tom Hughes for a patch.
2174 PH/24 Include config.h inside local_scan.h so that configuration settings are
2177 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2178 that in an ACL for RCPT (for example), you can examine exactly what was
2181 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2182 commands, but it was not correctly comparing the address with the actual
2183 client host address. Thus, it would show the EHLO address in Received:
2184 header lines when this was not necessary.
2186 PH/27 Added the % operator to ${eval:}.
2188 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2189 it should be ignoring failures (because with -C, for example, it has lost
2190 privilege). It wasn't ignoring creation failures other than "already
2193 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2196 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2197 because an input line was too long, either on its own, or by virtue of
2198 too many continuations, the temporary file was not being removed, and the
2199 return code was incorrect.
2201 PH/31 Missing "BOOL" in function definition in filtertest.c.
2203 PH/32 Applied Sieve patches from the maintainer.
2205 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2207 PH/33 Added "verify = not_blind".
2209 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2210 Local/Makefile (with some defaults set). These are used in built scripts
2211 such as exicyclog, but they have never been used in the exim_install
2212 script (though there are many overriding facilities there). I have
2213 arranged that the exim_install script now takes note of these two
2216 PH/35 Installed configuration files for Dragonfly.
2218 PH/36 When a locally submitted message by a trusted user did not contain a
2219 From: header, and the sender address was obtained from -f or from an SMTP
2220 MAIL command, and the trusted user did not use -F to supply a sender
2221 name, $originator_name was incorrectly used when constructing a From:
2222 header. Furthermore, $originator_name was used for submission mode
2223 messages from external hosts without From: headers in a similar way,
2224 which is clearly wrong.
2226 PH/37 Added control=suppress_local_fixups.
2228 PH/38 When log_selector = +received_sender was set, and the addition of the
2229 sender made the log line's construction buffer exactly full, or one byte
2230 less than full, an overflow happened when the terminating "\n" was
2233 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2234 when the result of a list match is failure because a DNS lookup failed.
2236 PH/40 RM_COMMAND is now used in the building process.
2238 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2239 the "build-* directories that it finds.
2241 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2242 address in a domain literal was a prefix of an interface address.
2244 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2245 when verifying a sender address, unless rewrite_headers is false.
2247 PH/44 Wrote a long comment about why errors_to addresses are verified as
2248 recipients, not senders.
2250 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2251 the ratelimit ACL was added.
2253 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2255 PH/46 Added extra information about PostgreSQL errors to the error string.
2257 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2258 avoiding the need to install special zones in a real server. This is
2259 backwards compatible; if it can't find the fake resolver, it drops back.
2260 Thus, both old and new test suites can be run.
2262 TF/02 Added util/ratelimit.pl
2264 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2265 clock is set back in time.
2267 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2268 Candler <B.Candler@pobox.com>.
2270 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2271 OK for addresses that are the result of redirections.
2273 PH/48 A number of further additions for the benefit of the new test suite,
2274 including a fake gethostbyname() that interfaces to the fake DNS resolver
2277 TF/06 The fix for widen_domains has also been applied to qualify_single and
2278 search_parents which are the other dnslookup options that can cause
2281 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2284 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2286 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2287 service" if it encountered an MX record with an empty target hostname.
2288 The message is now "an MX or SRV record indicated no SMTP service".
2290 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2291 if the DNS of the sending site is misconfigured. This is quite a
2292 common situation. This change restores the behaviour of treating a
2293 helo verification defer as a failure.
2295 PH/51 If self=fail was set on a router, the bounce message did not include the
2296 actual error message.
2302 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2304 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2305 command itself ended in error, the underlying message about the transport
2306 filter timeout was being overwritten with the pipe command error. Now the
2307 underlying error message should be appended to the second error message.
2309 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2311 PH/02 Reduce the amount of output that "make" produces by default. Full output
2312 can still be requested.
2314 PH/03 The warning log line about a condition test deferring for a "warn" verb
2315 was being output only once per connection, rather than after each
2316 occurrence (because it was using the same function as for successful
2317 "warn" verbs). This seems wrong, so I have changed it.
2319 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2320 it should not have, which might have caused a crash in the right
2321 circumstances, but probably never did.
2323 PH/04 Installed a modified version of Tony Finch's patch to make submission
2324 mode fix the return path as well as the Sender: header line, and to
2325 add a /name= option so that you can make the user's friendly name appear
2328 TF/03 Added the control = fakedefer ACL modifier.
2330 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2331 Mark Lowes for thorough testing.
2333 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2335 TK/03 Merged latest SRS patch from Miles Wilton.
2337 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2338 with the definition in sysexits.h (which is #included earlier).
2339 Fortunately, Exim does not actually use EX_OK. The code used to try to
2340 preserve the sysexits.h value, by assumimg that macro definitions were
2341 scanned for macro replacements. I have been disabused of this notion,
2342 so now the code just undefines EX_OK before #including unistd.h.
2344 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2345 in the smtp transport. When a block could not be written in a single
2346 write() function, the timeout was being re-applied to each part-write.
2347 This seems wrong - if the receiver was accepting one byte at a time it
2348 would take for ever. The timeout is now adjusted when this happens. It
2349 doesn't have to be particularly precise.
2351 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2352 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2354 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2355 without a domain if the check to <postmaster@domain> fails.
2357 SC/01 Eximstats: added -xls and the ability to specify output files
2358 (patch written by Frank Heydlauf).
2360 SC/02 Eximstats: use FileHandles for outputing results.
2362 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2364 SC/04 Eximstats: fixed display of large numbers with -nvr option
2366 SC/05 Eximstats: fixed merging of reports with empty tables.
2368 SC/06 Eximstats: added the -include_original_destination flag
2370 SC/07 Eximstats: removed tabs and trailing whitespace.
2372 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2374 TK/06 MBOX spool code: Add real "From " MBOX separator line
2375 so the .eml file is really in mbox format (even though
2376 most programs do not really care). Patch from Alex Miller.
2378 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2379 The latter is generated from $received_to and is only set if the
2380 message has one envelope recipient. SA can use these headers,
2381 obviously out-of-the-box. Patch from Alex Miller.
2383 PH/08 The ${def test on a variable was returning false if the variable's
2384 value was "0", contrary to what the specification has always said!
2385 The result should be true unless the variable is empty.
2387 PH/09 The syntax error of a character other than { following "${if
2388 def:variable_name" (after optional whitespace) was not being diagnosed.
2389 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2390 accidental colon was present, for example, could give incorrect results.
2392 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2393 result is used (not including appendfile, where other changes are about
2396 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2397 This involved changing a lot of size variables from int to off_t. It
2398 should work with maildirs and everything.
2400 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2401 spamd dying while we are connected to it.
2403 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2406 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2407 function for BDB 4.3.
2409 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2411 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2412 the sources for PCRE have been re-arranged and more files are now
2415 PH/15 The code I had for printing potentially long long variables in PH/11
2416 above was not the best (it lost precision). The length of off_t variables
2417 is now inspected at build time, and an appropriate printing format (%ld
2418 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2419 to be "long long int" or "long int". This is needed for the internal
2420 formatting function string_vformat().
2422 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2423 the configuration file to be ":syslog", then the script "guesses" where
2424 the logs files are, rather than using the compiled in default. In our
2425 case the guess is not the same as the compiled default, so the script
2426 suddenly stopped working when I started to use syslog. The patch checks
2427 to see if log_file_path is "". If so, it attempts to read it from exim
2428 with no configuration file to get the compiled in version, before it
2429 falls back to the previous guessing code."
2431 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2432 implementing BATV in an Exim configuration. See NewStuff for the gory
2435 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2436 Makefile that are specific to HP-UX.
2438 PH/18 If the "use_postmaster" option was set for a recipient callout together
2439 with the "random" option, the postmaster address was used as the MAIL
2440 FROM address for the random test, but not for the subsequent recipient
2441 test. It is now used for both.
2443 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2444 patch removes a few documentation additions to RFC 3028, because the
2445 latest draft now contains them. It adds the new en;ascii-case comparator
2446 and a new error check for 8bit text in MIME parts. Comparator and
2447 require names are now matched exactly. I enabled the subaddress
2448 extension, but it is not well tested yet (read: it works for me)."
2450 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2451 rework some of the code of TK/09 above to avoid the hardwired use of
2452 "%lld" and "long long". Replaced the call to snprintf() with a call to
2455 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2456 records point to non-existent hosts", "retry timeout exceeded", and
2457 "retry time not reached for any host after a long failure period".
2459 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2460 experimental DomainKeys support:
2462 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2463 (2) On an error such as an illegally used "control", the wrong name for
2464 the control was given.
2466 These problems did NOT occur unless DomainKeys support was compiled.
2468 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2470 PH/24 Added ${if match_ip condition.
2472 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2473 databases so that it will be absolutely obvious if a crash occurs in the
2474 DB library. This is a regular occurrence (often caused by mis-matched
2477 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2478 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2479 picked up on a user's system that detects such things. There doesn't seem
2480 to be a gcc warning option for this - only an attribute that has to be
2481 put on the function's prototype. It seems that in Fedora Core 4 they have
2482 set this on a number of new functions. No doubt there will be more in due
2485 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2486 specify a transport. However, if an address that was verified by such a
2487 router was the subject of a callout, Exim crashed because it tried to
2488 read the rcpt_include_affixes from the non-existent transport. Now it
2489 just assumes that the setting of that option is false. This bug was
2490 introduced by 4.51/PH/31.
2492 PH/28 Changed -d+all to exclude +memory, because that information is very
2493 rarely of interest, but it makes the output a lot bigger. People tend to
2494 do -d+all out of habit.
2496 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2497 code in os-type was giving problems when libc.so lives in lib64, like on
2500 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2501 aren't the modern standard, and it seems that some systems' include files
2502 don't always have them. Exim was already checking for some of the newer
2503 ones like T_AAAA, and defining it itself. I've added checks for all the
2504 record types that Exim uses.
2506 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2507 not automatically generating a new one, as it is supposed to. This
2508 prevented TLS from working. If the file did exist, but contained invalid
2509 data, a new version was generated, as expected. It was only the case of a
2510 non-existent file that was broken.
2512 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2513 with a change in libdomainkeys > 0.64.
2515 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2516 from DNS. If the selector record carries the flag, it now has
2517 precedence over the domain-wide flag.
2519 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2521 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2522 the use of a port name that isn't defined in /etc/services) occurred, the
2523 message was deferred as in a normal delivery, and thus remained on the
2524 spool, instead of being failed because of the mua_wrapper setting. This
2525 is now fixed, and I tidied up some of the mua_wrapper messages at the
2528 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2529 the messages in a hash of arrays rather than using individual hashes.
2530 This is a bit cleaner and results in dramatic memory savings, albeit
2531 at a slight CPU cost.
2533 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2534 as requested by Marc Sherman.
2536 SC/10 Eximstats: added histograms for user specified patterns as requested
2539 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2541 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2542 fopen() in the content-scanning modules that did not already have it.
2548 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2549 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2551 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2553 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2555 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2556 to test Sieve filters that use "vacation".
2558 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2559 that changes the way the GnuTLS parameters are stored in the cache file.
2560 The new format can be generated externally. For backward compatibility,
2561 if the data in the cache doesn't make sense, Exim assumes it has read an
2562 old-format file, and it generates new data and writes a new file. This
2563 means that you can't go back to an older release without removing the
2566 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2567 work if there are any delivery delays because "one_time" forces the
2568 parent to be marked "delivered", so its unseen clone is never tried
2569 again. For this reason, Exim now forbids the simultaneous setting of
2572 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2573 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2574 entry if you want to know the details. The fix, however, neglected to
2575 consider the case when local delivery batching is involved. The test for
2576 "previously delivered" was not happening when checking to see if an
2577 address could be batched with a previous (undelivered) one; under
2578 certain circumstances this could lead to multiple deliveries to the same
2581 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2582 in its include files, and this causes problems building Exim.
2584 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2585 header_syntax) but Exim was just ignoring anything given after a slash.
2586 In particular, this caused confusion with an attempt to use "verify =
2587 reverse_host_lookup/defer_ok". An error is now given when options are
2588 supplied for verify items that do not have them. (Maybe reverse_host_
2589 lookup should have a defer_ok option, but that's a different point.)
2591 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2592 defined by RFC 821) to 2048, because there were problems with some AUTH
2593 commands, and RFC 1869 says the size should be increased for extended
2594 SMTP commands that take arguments.
2596 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2599 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2600 "unknown" error; now it says that the functionality isn't in the binary.
2602 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2603 an address' error message when a string expansion fails (syntax or
2604 whatever). Otherwise the password may appear in the log. Following change
2605 PH/42 below, there is no longer a chance of it appearing in a bounce
2608 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2610 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2611 of its addresses was ever tried. (Bugzilla bug #2.)
2613 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2614 the result incorrectly in the debug output. (It correctly added a newline
2615 to what was transported.)
2617 TF/01 Added $received_time.
2619 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2620 commented out examples of how to interface to a virus scanner and to
2621 SpamAssassin. Also added commented examples of av_scanner and
2622 spamd_address settings.
2624 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2625 and controls are allowed in which ACLs. There were a couple of minor
2626 errors. Some of the entries in the conditions table (which is a table of
2627 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2628 negation of where the condition IS allowed.
2630 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2632 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2633 header file does not have a version number, so I've had to invent a new
2634 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2635 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2636 radiusclient), but it was contributed by a Radius user.
2638 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2639 files or directories, for OpenSSL.
2641 PH/19 When an Exim process that is running as root has to create an Exim log
2642 file, it does so in a subprocess that runs as exim:exim so as to get the
2643 ownership right at creation (otherwise, other Exim processes might see
2644 the file with the wrong ownership). There was no test for failure of this
2645 fork() call, which would lead to the process getting stuck as it waited
2646 for a non-existent subprocess. Forks do occasionally fail when resources
2647 run out. I reviewed all the other calls to fork(); they all seem to check
2650 PH/20 When checking for unexpected SMTP input at connect time (before writing
2651 the banner), Exim was not dealing correctly with a non-positive return
2652 from the read() function. If the client had disconnected by this time,
2653 the result was a log entry for a synchronization error with an empty
2654 string after "input=" when read() returned zero. If read() returned -1
2655 (an event I could not check), uninitialized data bytes were printed.
2656 There were reports of junk text (parts of files, etc) appearing after
2659 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2661 PH/22 Added support for macro redefinition, and (re)definition in between
2662 driver and ACL definitions.
2664 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2665 forgetting to use the resulting value; it was using the unexpanded value.
2667 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2668 hadn't been configured. The fix is from Juergen Kreileder, who
2669 understands it better than I do:
2671 "Here's what I see happening with three configured cyrus_sasl
2672 authenticators configured (plain, login, cram-md5):
2674 On startup auth_cyrus_sasl_init() gets called for each of these.
2675 This means three calls to sasl_listmech() without a specified mech_list.
2676 => SASL tests which mechs of all available mechs actually work
2677 => three warnings about OTP not working
2678 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2680 With the patch, sasl_listmech() also gets called three times. But now
2681 SASL's mech_list option is set to the server_mech specified in the the
2682 authenticator. Or in other words, the answer from sasl_listmech()
2683 gets limited to just the mech you're testing for (which is different
2685 => the return list contains just 'plain' or 'login', 'cram-md5' or
2686 nothing depending on the value of ob->server_mech.
2688 I've just tested the patch: Authentication still works fine,
2689 unavailable mechs specified in the exim configuration are still
2690 caught, and the auth.log warnings about OTP are gone."
2692 PH/25 When debugging is enabled, the contents of the command line are added
2693 to the debugging output, even when log_selector=+arguments is not
2696 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2697 answer is "GNU", and only if the return is "GNU/something" is the answer
2700 PH/27 $acl_verify_message is now set immediately after the failure of a
2701 verification in an ACL, and so is available in subsequent modifiers. In
2702 particular, the message can be preserved by coding like this:
2704 warn !verify = sender
2705 set acl_m0 = $acl_verify_message
2707 Previously, $acl_verify_message was set only while expanding "message"
2708 and "log_message" when a very denied access.
2710 PH/28 Modified OS/os.c-Linux with
2712 -#ifndef OS_LOAD_AVERAGE
2713 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2715 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2716 nomenclature these days.)
2718 PH/29 Installed patch from the Sieve maintainer that adds the options
2719 sieve_useraddress and sieve_subaddress to the redirect router.
2721 PH/30 In these circumstances:
2722 . Two addresses routed to the same list of hosts;
2723 . First host does not offer TLS;
2724 . First host accepts first address;
2725 . First host gives temporary error to second address;
2726 . Second host offers TLS and a TLS session is established;
2727 . Second host accepts second address.
2728 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2729 and peerdn, if requested) that were in fact used only for the second
2732 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2733 attention to any local part prefix or suffix that was matched by the
2734 router that accepted the address. It now behaves in the same way as it
2735 does for delivery: the affixes are removed from the local part unless
2736 rcpt_include_affixes is set on the transport.
2738 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2739 timeout during the DATA phase of an incoming message.
2741 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2742 applied a patch sanctioned by the Sieve maintainer.
2744 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2745 the uid or gid is negative. A case of a negative gid caused this to be
2746 noticed. The fix allows for either to be negative.
2748 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2749 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2752 PH/36 The change PH/12 above was broken. Fixed it.
2754 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2755 the grounds that routing the same address twice would always produce the
2756 same answer. This might have been true once, but it is certainly no
2757 longer true now. Routing a child address may depend on the previous
2758 routing that produced that child. Some complicated redirection strategies
2759 went wrong when messages had multiple recipients, and made Exim's
2760 behaviour dependent on the order in which the addresses were given.
2762 I have moved the duplicate checking until after the routing is complete.
2763 Exim scans the addresses that are assigned to local and remote
2764 transports, and removes any duplicates. This means that more work will be
2765 done, as duplicates will always all be routed, but duplicates are
2766 presumably rare, so I don't expect this is of any significance.
2768 For deliveries to pipes, files, and autoreplies, the duplicate checking
2769 still happens during the routing process, since they are not going to be
2772 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2773 It corrects a timeout issue with spamd. This is Ian's comment: "The
2774 background is that sometimes spamd either never reads data from a
2775 connection it has accepted, or it never writes response data. The exiscan
2776 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2777 blindly assumes that writes won't block so it may never time out."
2779 PH/39 Allow G after quota size as well as K and M.
2781 PH/40 The value set for $authenticated_id in an authenticator may not contain
2782 binary zeroes or newlines because the value is written to log lines and
2783 to spool files. There was no check on this. Now the value is run through
2784 the string_printing() function so that such characters are converted to
2785 printable escape sequences.
2787 PH/41 $message_linecount is a new variable that contains the total number of
2788 lines in the message. Compare $body_linecount, which is the count for the
2791 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2792 bounce and delay warning messages, except in certain special cases, which
2795 (a) An SMTP error message from a remote host;
2796 (b) A message specified in a :fail: redirection;
2797 (c) A message specified in a "fail" command in a system filter;
2798 (d) A message specified in a FAIL return from the queryprogram router;
2799 (e) A message specified by the cannot_route_message router option.
2801 In these cases only, Exim does include the error details in bounce and
2802 warning messages. There are also a few cases where bland messages such
2803 as "unrouteable address" or "local delivery error" are given.
2805 PH/43 $value is now also set for the "else" part of a ${run expansion.
2807 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2808 being worked on, but at least Exim now implements the latest version to
2811 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2812 process to complete was treated as a delivery failure, a timeout while
2813 writing the message to the pipe was logged, but erroneously treated as a
2814 successful delivery. Such timeouts include transport filter timeouts. For
2815 consistency with the overall process timeout, these timeouts are now
2816 treated as errors, giving rise to delivery failures by default. However,
2817 there is now a new Boolean option for the pipe transport called
2818 timeout_defer, which, if set TRUE, converts the failures into defers for
2819 both kinds of timeout. A transport filter timeout is now identified in
2822 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2823 systems where "make" and "gmake" are different, calling "gmake" at top
2824 level broke things. I've arranged for the value of $(MAKE) to be passed
2825 from the Makefile to this script so that it can call the same version of
2829 A note about Exim versions 4.44 and 4.50
2830 ----------------------------------------
2832 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2833 changes of various kinds. As a consequence, a big documentation update was
2834 needed. This delayed the release for rather longer than seemed good, especially
2835 in the light of a couple of (minor) security issues. Therefore, the changes
2836 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2837 release. So 4.44 and 4.50 are in effect two different branches that both start
2840 I have left the 4.50 change log unchanged; it contains all the changes since
2841 4.43. The change log for 4.44 is below; many of its items are identical to
2842 those for 4.50. This seems to be the most sensible way to preserve the
2843 historical information.
2849 1. Minor wording change to the doc/README.SIEVE file.
2851 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2852 computation of the current number of files was incorrect.
2854 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2855 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2858 4. Give more explanation in the error message when the command for a transport
2859 filter fails to execute.
2861 5. There are several places where Exim runs a non-Exim command in a
2862 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2863 was being done only for the command run by the queryprogram router. It is
2864 now done for all such subprocesses. The other cases are: ${run, transport
2865 filters, and the commands run by the lmtp and pipe transports.
2867 6. Added CONFIGURE_GROUP build-time option.
2869 7. Some older OS have a limit of 256 on the maximum number of file
2870 descriptors. Exim was using setrlimit() to set 1000 as a large value
2871 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2872 systems. I've change it so that if it can't get 1000, it tries for 256.
2874 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2875 was an oversight, and furthermore, ever since the addition of extra
2876 controls (e.g. 4.43/32), the checks on when to allow different forms of
2877 "control" were broken. There should now be diagnostics for all cases when a
2878 control that does not make sense is encountered.
2880 9. Added the /retain_sender option to "control=submission".
2882 10. $recipients is now available in the predata ACL (oversight).
2884 11. Tidy the search cache before the fork to do a delivery from a message
2885 received from the command line. Otherwise the child will trigger a lookup
2886 failure and thereby defer the delivery if it tries to use (for example) a
2887 cached ldap connection that the parent has called unbind on.
2889 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2890 of $address_data from the recipient verification was clobbered by the
2891 sender verification.
2893 13. The value of address_data from a sender verification is now available in
2894 $sender_address_data in subsequent conditions in the ACL statement.
2896 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2898 15. Added a new option "connect=<time>" to callout options, to set a different
2901 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2902 was its contents. (It was OK if the option was not defined at all.)
2904 17. A "Completed" log line is now written for messages that are removed from
2905 the spool by the -Mrm option.
2907 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2908 information about exactly what failed.
2910 19. Added -dd to debug only the daemon process.
2912 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2913 handles timeouts, both on the server side and network timeouts. Renamed the
2914 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2916 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2917 It is now set to "smtps".
2919 22. $host_address is now set to the target address during the checking of
2920 ignore_target_hosts.
2922 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2923 being passed; this would have caused $sender_host_name to have been used if
2924 matching the list had actually called for a host name (not very likely,
2925 since this list is usually IP addresses). A host name is now passed as
2928 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2929 code that specifies a non-restarting handler (typically sigaction() in
2930 modern systems) in an attempt to fix a rare and obscure crash bug.
2932 25. Narrowed the window for a race in the daemon that could cause it to ignore
2933 SIGCHLD signals. This is not a major problem, because they are used only to
2934 wake it up if nothing else does.
2936 26. A malformed maildirsize file could cause Exim to calculate negative values
2937 for the mailbox size or file count. Odd effects could occur as a result.
2938 The maildirsize information is now recalculated if the size or filecount
2941 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2942 support for a long time. Removed HAVE_SYS_VFS_H.
2944 28. Installed the latest version of exipick from John Jetmore.
2946 29. In an address list, if the pattern was not a regular expression, an empty
2947 subject address (from a bounce message) matched only if the pattern was an
2948 empty string. Non-empty patterns were not even tested. This was the wrong
2949 because it is perfectly reasonable to use an empty address as part of a
2950 database query. An empty address is now tested by patterns that are
2951 lookups. However, all the other forms of pattern expect the subject to
2952 contain a local part and a domain, and therefore, for them, an empty
2953 address still always fails if the pattern is not itself empty.
2955 30. Exim went into a mad DNS loop when attempting to do a callout where the
2956 host was specified on an smtp transport, and looking it up yielded more
2957 than one IP address.
2959 31. Re-factored the code for checking spool and log partition space into a
2960 function that finds that data and another that does the check. The former
2961 is then used to implement four new variables: $spool_space, $log_space,
2962 $spool_inodes, and $log_inodes.
2964 32. The RFC2047 encoding function was originally intended for short strings
2965 such as real names; it was not keeping to the 75-character limit for
2966 encoded words that the RFC imposes. It now respects the limit, and
2967 generates multiple encoded words if necessary. To be on the safe side, I
2968 have increased the buffer size for the ${rfc2047: expansion operator from
2971 33. It is now permitted to omit both strings after an "if" condition; if the
2972 condition is true, the result is "true". As before, when the second string
2973 is omitted, a false condition yields an empty string. This makes it less
2974 cumbersome to write custom ACL and router conditions.
2976 34. Failure to deliver a bounce message always caused it to be frozen, even if
2977 there was an errors_to setting on the router. The errors_to setting is now
2980 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2981 canonical form (fully expanded) before being placed in
2982 $sender_host_address.
2984 36. The table in the code that translates DNS record types into text (T_A to
2985 "A" for instance) was missing entries for NS and CNAME. It is just possible
2986 that this could have caused confusion if both these types were looked up
2987 for the same domain, because the text type is used as part of Exim's
2988 per-process caching. But the chance of anyone hitting this buglet seems
2991 37. The dnsdb lookup has been extended in a number of ways.
2993 (1) There is a new type, "zns", which walks up the domain tree until it
2994 finds some nameserver records. It should be used with care.
2996 (2) There is a new type, "mxh", which is like "mx" except that it returns
2997 just the host names, not the priorities.
2999 (3) It is now possible to give a list of domains (or IP addresses) to be
3000 looked up. The behaviour when one of the lookups defers can be
3001 controlled by a keyword.
3003 (4) It is now possible to specify the separator character for use when
3004 multiple records are returned.
3006 38. The dnslists ACL condition has been extended: it is now possible to supply
3007 a list of IP addresses and/or domains to be looked up in a particular DNS
3010 39. Added log_selector=+queue_time_overall.
3012 40. When running the queue in the test harness, wait just a tad after forking a
3013 delivery process, to get repeatability of debugging output.
3015 41. Include certificate and key file names in error message when GnuTLS fails
3016 to set them up, because the GnuTLS error message doesn't include the name
3017 of the failing file when there is a problem reading it.
3019 42. Allow both -bf and -bF in the same test run.
3021 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3023 44. The "Exiscan patch" is now merged into the mainline Exim source.
3025 45. Sometimes the final signoff response after QUIT could fail to get
3026 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3027 < 0 before doing a fflush(). This bug looks as though it goes back to the
3028 introduction of TLS in release 3.20, but "sometimes" must have been rare
3029 because the tests only now provoked it.
3031 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3032 (this can affect the format of dates).
3034 47. exim_tidydb, when checking for the continued existence of a message for
3035 which it has found a message-specific retry record, was not finding
3036 messages that were in split spool directories. Consequently, it was
3037 deleting retry records that should have stayed in existence.
3039 48. Steve fixed some bugs in eximstats.
3041 49. The SPA authentication driver was not abandoning authentication and moving
3042 on to the next authenticator when an expansion was forced to fail,
3043 contradicting the general specification for all authenticators. Instead it
3044 was generating a temporary error. It now behaves as specified.
3046 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3047 (the order specifies the preference for clients). The order is now AES256,
3048 AES128, 3DES, ARCFOUR128.
3050 51. Small patch to Sieve code - explicitly set From: when generating an
3053 52. Exim crashed if a remote delivery caused a very long error message to be
3054 recorded - for instance if somebody sent an entire SpamAssassin report back
3055 as a large number of 550 error lines. This bug was coincidentally fixed by
3056 increasing the size of one of Exim's internal buffers (big_buffer) that
3057 happened as part of the Exiscan merge. However, to be on the safe side, I
3058 have made the code more robust (and fixed the comments that describe what
3061 53. Now that there can be additional text after "Completed" in log lines (if
3062 the queue_time_overall log selector is set), a one-byte patch to exigrep
3063 was needed to allow it to recognize "Completed" as not the last thing in
3066 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3067 patch that reportedly fixes this has been added. I am not expert enough to
3068 create a test for it. This is what the patch creator wrote:
3070 "I found a little strange behaviour of ldap code when working with
3071 Windows 2003 AD Domain, where users was placed in more than one
3072 Organization Units. When I tried to give exim partial DN, the exit code
3073 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3074 But simultaneously result of request was absolutely normal ldap result,
3075 so I produce this patch..."
3077 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3078 REFERENCE, so I have modified the code to exclude the patch when that macro
3081 55. Some experimental protocols are using DNS PTR records for new purposes. The
3082 keys for these records are domain names, not reversed IP addresses. The
3083 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3084 leaves it alone. Component reversal etc. now happens only for IP addresses.
3087 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3089 57. Double the size of the debug message buffer (to 2048) so that more of very
3090 long debug lines gets shown.
3092 58. The exicyclog utility now does better if the number of log files to keep
3093 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3095 59. Two changes related to the smtp_active_hostname option:
3097 (1) $smtp_active_hostname is now available as a variable.
3098 (2) The default for smtp_banner uses $smtp_active_hostname instead
3099 of $primary_hostname.
3101 60. The host_aton() function is supposed to be passed a string that is known
3102 to be a valid IP address. However, in the case of IPv6 addresses, it was
3103 not checking this. This is a hostage to fortune. Exim now panics and dies
3104 if the condition is not met. A case was found where this could be provoked
3105 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3106 components; fortuitously, this particular loophole had already been fixed
3107 by change 4.50/55 above.
3109 If there are any other similar loopholes, the new check in host_aton()
3110 itself should stop them being exploited. The report I received stated that
3111 data on the command line could provoke the exploit when Exim was running as
3112 exim, but did not say which command line option was involved. All I could
3113 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3114 running as the user.
3117 61. There was a buffer overflow vulnerability in the SPA authentication code
3118 (which came originally from the Samba project). I have added a test to the
3119 spa_base64_to_bits() function which I hope fixes it.
3122 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3123 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3125 63. The daemon start-up calls getloadavg() while still root for those OS that
3126 need the first call to be done as root, but it missed one case: when
3127 deliver_queue_load_max is set with deliver_drop_privilege. This is
3128 necessary for the benefit of the queue runner, because there is no re-exec
3129 when deliver_drop_privilege is set.
3131 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3132 This has been fixed.
3134 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3135 was in use, was not putting the data itself into the right store pool;
3136 consequently, it could be overwritten for a subsequent message in the same
3137 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3140 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3142 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3143 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3144 the function were treating the return as a boolean value, which happened to
3145 work because 0=false and not-0=true, but is not correct code.
3147 68. The host_aton() function was not handling scoped IPv6 addresses (those
3148 with, for example, "%eth0" on the end) correctly.
3150 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3151 negated items (that is, ~something) in unsigned ints. Some compilers
3152 apparently mutter when there is no cast.
3154 70. If an address verification called from an ACL failed, and did not produce a
3155 user-specific message (i.e. there was only a "system" message), nothing was
3156 put in $acl_verify_message. In this situation, it now puts the system
3159 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3160 session; change 4.31/43 added the unwanted input to the log line - except
3161 that it did not do this in the start of session case. It now does.
3163 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3164 This is wrong and can cause the other end to generate a synchronization
3165 error if it is another Exim or anything else that does the synchronization
3166 check. A QUIT command is no longer sent after a timeout.
3168 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3169 during host lookups.
3171 74. The defer_ok option of callout verification was not working if it was used
3172 when verifying addresses in header lines, that is, for this case:
3174 verify = header_sender/callout=defer_ok
3176 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3177 those file descriptors could be used for SMTP connections. If anything
3178 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3179 could be sent to the SMTP client, causing chaos. The daemon now opens
3180 stdin, stdout, and stderr to /dev/null when it puts itself into the
3183 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3184 log by default. The user can override this with suitable Perl magic.
3186 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3187 the log message when discard triggers, was not working for the DATA ACL or
3188 for the non-SMTP ACL.
3190 78. Error message wording change in sieve.c.
3192 79. If smtp_accept_max_per_host was set, the number of connections could be
3193 restricted to fewer than expected, because the daemon was trying to set up
3194 a new connection before checking whether the processes handling previous
3195 connections had finished. The check for completed processes is now done
3196 earlier. On busy systems, this bug wouldn't be noticed because something
3197 else would have woken the daemon, and it would have reaped the completed
3200 80. If a message was submitted locally by a user whose login name contained one
3201 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3202 It caused a spool format error. I have fixed the spool reading code. A
3203 related problem was that the "from" clause in the Received: line became
3204 illegal because of the space(s). It is now covered by ${quote_local_part.
3206 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3209 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3211 83. Patch from Sieve maintainer to fix unterminated string problem in
3212 "vacation" handling.
3214 84. Some minor changes to the Linux configuration files to help with other
3215 OS variants using glibc.
3217 85. One more patch for Sieve to update vacation handling to latest spec.
3220 ----------------------------------------------------
3221 See the note above about the 4.44 and 4.50 releases.
3222 ----------------------------------------------------
3228 1. Change 4.43/35 introduced a bug that caused file counts to be
3229 incorrectly computed when quota_filecount was set in an appendfile
3232 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3233 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3236 3. Give more explanation in the error message when the command for a transport
3237 filter fails to execute.
3239 4. There are several places where Exim runs a non-Exim command in a
3240 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3241 was being done only for the command run by the queryprogram router. It is
3242 now done for all such subprocesses. The other cases are: ${run, transport
3243 filters, and the commands run by the lmtp and pipe transports.
3245 5. Some older OS have a limit of 256 on the maximum number of file
3246 descriptors. Exim was using setrlimit() to set 1000 as a large value
3247 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3248 systems. I've change it so that if it can't get 1000, it tries for 256.
3250 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3251 was an oversight, and furthermore, ever since the addition of extra
3252 controls (e.g. 4.43/32), the checks on when to allow different forms of
3253 "control" were broken. There should now be diagnostics for all cases when a
3254 control that does not make sense is encountered.
3256 7. $recipients is now available in the predata ACL (oversight).
3258 8. Tidy the search cache before the fork to do a delivery from a message
3259 received from the command line. Otherwise the child will trigger a lookup
3260 failure and thereby defer the delivery if it tries to use (for example) a
3261 cached ldap connection that the parent has called unbind on.
3263 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3264 of $address_data from the recipient verification was clobbered by the
3265 sender verification.
3267 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3268 was its contents. (It was OK if the option was not defined at all.)
3270 11. A "Completed" log line is now written for messages that are removed from
3271 the spool by the -Mrm option.
3273 12. $host_address is now set to the target address during the checking of
3274 ignore_target_hosts.
3276 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3277 being passed; this would have caused $sender_host_name to have been used if
3278 matching the list had actually called for a host name (not very likely,
3279 since this list is usually IP addresses). A host name is now passed as
3282 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3283 code that specifies a non-restarting handler (typically sigaction() in
3284 modern systems) in an attempt to fix a rare and obscure crash bug.
3286 15. Narrowed the window for a race in the daemon that could cause it to ignore
3287 SIGCHLD signals. This is not a major problem, because they are used only to
3288 wake it up if nothing else does.
3290 16. A malformed maildirsize file could cause Exim to calculate negative values
3291 for the mailbox size or file count. Odd effects could occur as a result.
3292 The maildirsize information is now recalculated if the size or filecount
3295 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3296 support for a long time. Removed HAVE_SYS_VFS_H.
3298 18. Updated exipick to current release from John Jetmore.
3300 19. Allow an empty sender to be matched against a lookup in an address list.
3301 Previously the only cases considered were a regular expression, or an
3304 20. Exim went into a mad DNS lookup loop when doing a callout where the
3305 host was specified on the transport, if the DNS lookup yielded more than
3308 21. The RFC2047 encoding function was originally intended for short strings
3309 such as real names; it was not keeping to the 75-character limit for
3310 encoded words that the RFC imposes. It now respects the limit, and
3311 generates multiple encoded words if necessary. To be on the safe side, I
3312 have increased the buffer size for the ${rfc2047: expansion operator from
3315 22. Failure to deliver a bounce message always caused it to be frozen, even if
3316 there was an errors_to setting on the router. The errors_to setting is now
3319 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3320 canonical form (fully expanded) before being placed in
3321 $sender_host_address.
3323 24. Updated eximstats to version 1.33
3325 25. Include certificate and key file names in error message when GnuTLS fails
3326 to set them up, because the GnuTLS error message doesn't include the name
3327 of the failing file when there is a problem reading it.
3329 26. Expand error message when OpenSSL has problems setting up cert/key files.
3332 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3333 (this can affect the format of dates).
3335 28. exim_tidydb, when checking for the continued existence of a message for
3336 which it has found a message-specific retry record, was not finding
3337 messages that were in split spool directories. Consequently, it was
3338 deleting retry records that should have stayed in existence.
3340 29. eximstats updated to version 1.35
3341 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3342 1.35 - bugfix such that pie charts by volume are generated correctly
3344 30. The SPA authentication driver was not abandoning authentication and moving
3345 on to the next authenticator when an expansion was forced to fail,
3346 contradicting the general specification for all authenticators. Instead it
3347 was generating a temporary error. It now behaves as specified.
3349 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3350 (the order specifies the preference for clients). The order is now AES256,
3351 AES128, 3DES, ARCFOUR128.
3353 31. Small patch to Sieve code - explicitly set From: when generating an
3356 32. Exim crashed if a remote delivery caused a very long error message to be
3357 recorded - for instance if somebody sent an entire SpamAssassin report back
3358 as a large number of 550 error lines. This bug was coincidentally fixed by
3359 increasing the size of one of Exim's internal buffers (big_buffer) that
3360 happened as part of the Exiscan merge. However, to be on the safe side, I
3361 have made the code more robust (and fixed the comments that describe what
3364 33. Some experimental protocols are using DNS PTR records for new purposes. The
3365 keys for these records are domain names, not reversed IP addresses. The
3366 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3367 leaves it alone. Component reversal etc. now happens only for IP addresses.
3370 34. The host_aton() function is supposed to be passed a string that is known
3371 to be a valid IP address. However, in the case of IPv6 addresses, it was
3372 not checking this. This is a hostage to fortune. Exim now panics and dies
3373 if the condition is not met. A case was found where this could be provoked
3374 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3375 components; fortuitously, this particular loophole had already been fixed
3376 by change 4.50/55 or 4.44/33 above.
3378 If there are any other similar loopholes, the new check in host_aton()
3379 itself should stop them being exploited. The report I received stated that
3380 data on the command line could provoke the exploit when Exim was running as
3381 exim, but did not say which command line option was involved. All I could
3382 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3383 running as the user.
3386 35. There was a buffer overflow vulnerability in the SPA authentication code
3387 (which came originally from the Samba project). I have added a test to the
3388 spa_base64_to_bits() function which I hope fixes it.
3391 36. The daemon start-up calls getloadavg() while still root for those OS that
3392 need the first call to be done as root, but it missed one case: when
3393 deliver_queue_load_max is set with deliver_drop_privilege. This is
3394 necessary for the benefit of the queue runner, because there is no re-exec
3395 when deliver_drop_privilege is set.
3397 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3398 was in use, was not putting the data itself into the right store pool;
3399 consequently, it could be overwritten for a subsequent message in the same
3400 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3403 38. Sometimes the final signoff response after QUIT could fail to get
3404 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3405 < 0 before doing a fflush(). This bug looks as though it goes back to the
3406 introduction of TLS in release 3.20, but "sometimes" must have been rare
3407 because the tests only now provoked it.
3413 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3414 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3415 Now it returns an integer. A number of calls were still expecting a T/F
3416 return. Fortuitously, in all cases, the tests worked in OK situations,
3417 which is the norm. However, things would have gone wrong on any write
3418 failures on the smtp file descriptor. This function is used when sending
3419 messages over SMTP and also when doing verify callouts.
3421 2. When Exim is called to do synchronous delivery of a locally submitted
3422 message (the -odf or -odi options), it no longer closes stderr before doing
3425 3. Implemented the mua_wrapper option.
3427 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3429 5. Implemented the functions header_remove(), header_testname(),
3430 header_add_at_position(), and receive_remove_recipient(), and exported them
3433 6. If an ACL "warn" statement specified the addition of headers, Exim already
3434 inserted X-ACL-Warn: at the start if there was no header name. However, it
3435 was not making this test for the second and subsequent header lines if
3436 there were newlines in the string. This meant that an invalid header could
3437 be inserted if Exim was badly configured.
3439 7. Allow an ACL "warn" statement to add header lines at the start or after all
3440 the Received: headers, as well as at the end.
3442 8. Added the rcpt_4xx retry error code.
3444 9. Added postmaster_mailfrom=xxx to callout verification option.
3446 10. Added mailfrom=xxxx to the callout verification option, for verify=
3449 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3450 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3452 12. Inserted some casts to stop certain compilers warning when using pointer
3453 differences as field lengths or precisions in printf-type calls (mostly
3454 affecting debugging statements).
3456 13. Added optional readline() support for -be (dynamically loaded).
3458 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3459 same clock tick as a message's arrival, so that its received time was the
3460 same as the "first fail" time on the retry record, and that message
3461 remained on the queue past the ultimate address timeout, every queue runner
3462 would try a delivery (because it was past the ultimate address timeout) but
3463 after another failure, the ultimate address timeout, which should have then
3464 bounced the address, did not kick in. This was a "< instead of <=" error;
3465 in most cases the first failure would have been in the next clock tick
3466 after the received time, and all would be well.
3468 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3469 being recognized when the domain list was tested by the match_domain
3470 condition in an expansion string.
3472 16. Added the ${str2b64: operator.
3474 17. Exim was always calling setrlimit() to set a large limit for the number of
3475 processes, without checking whether the existing limit was already
3476 adequate. (It did check for the limit on file descriptors.) Furthermore,
3477 errors from getrlimit() and setrlimit() were being ignored. Now they are
3478 logged to the main and panic logs, but Exim does carry on, to try to do its
3479 job under whatever limits there are.
3481 18. Imported PCRE 5.0.
3483 19. Trivial typo in log message " temporarily refused connection" (the leading
3486 20. If the log selector return_path_on_delivery was set and an address was
3487 redirected to /dev/null, the delivery process crashed because it assumed
3488 that a return path would always be set for a "successful" delivery. In this
3489 case, the whole delivery is bypassed as an optimization, and therefore no
3492 21. Internal re-arrangement: the function for sending a challenge and reading
3493 a response while authentication was assuming a zero-terminated challenge
3494 string. It's now changed to take a pointer and a length, to allow for
3495 binary data in such strings.
3497 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3499 23. Exim was not respecting finduser_retries when seeking the login of the
3500 uid under which it was called; it was always trying 10 times. (The default
3501 setting of finduser_retries is zero.) Also, it was sleeping after the final
3502 failure, which is pointless.
3504 24. Implemented tls_on_connect_ports.
3506 25. Implemented acl_smtp_predata.
3508 26. If the domain in control=submission is set empty, Exim assumes that the
3509 authenticated id is a complete email address when it generates From: or
3510 Sender: header lines.
3512 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3513 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3514 chown and chgrp in /bin and hostname in /usr/bin.
3516 28. Exim was keeping the "process log" file open after each use, just as it
3517 does for the main log. This opens the possibility of it remaining open for
3518 long periods when the USR1 signal hits a daemon. Occasional processlog
3519 errors were reported, that could have been caused by this. Anyway, it seems
3520 much more sensible not to leave this file open at all, so that is what now
3523 29. The long-running daemon process does not normally write to the log once it
3524 has entered its main loop, and it closes the log before doing so. This is
3525 so that log files can straightforwardly be renamed and moved. However,
3526 there are a couple of unusual error situations where the daemon does write
3527 log entries, and I had neglected to close the log afterwards.
3529 30. The text of an SMTP error response that was received during a remote
3530 delivery was being truncated at 512 bytes. This is too short for some of
3531 the long messages that one sometimes sees. I've increased the limit to
3534 31. It is now possible to make retry rules that apply only when a message has a
3535 specific sender, in particular, an empty sender.
3537 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3538 it possible to be selective about when SMTP synchronization is enforced.
3540 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3542 32. Implemented hosts_connection_nolog.
3544 33. Added an ACL for QUIT.
3546 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3549 35. Added mailbox_size and mailbox_filecount to appendfile.
3551 36. Added control = no_multiline_responses to ACLs.
3553 37. There was a bug in the logic of the code that waits for the clock to tick
3554 in the case where the clock went backwards by a substantial amount such
3555 that the microsecond fraction of "now" was more than the microsecond
3556 fraction of "then" (but the whole seconds number was less).
3558 38. Added support for the libradius Radius client library this is found on
3559 FreeBSD (previously only the radiusclient library was supported).
3565 1. When certain lookups returned multiple values in the form name=value, the
3566 quoting of the values was not always being done properly. Specifically:
3567 (a) If the value started with a double quote, but contained no whitespace,
3569 (b) If the value contained whitespace other than a space character (i.e.
3570 tabs or newlines or carriage returns) it was not quoted.
3571 This fix has been applied to the mysql and pgsql lookups by writing a
3572 separate quoting function and calling it from the lookup code. The fix
3573 should probably also be applied to nisplus, ibase and oracle lookups, but
3574 since I cannot test any of those, I have not disturbed their existing code.
3576 2. A hit in the callout cache for a specific address caused a log line with no
3577 reason for rejecting RCPT. Now it says "Previous (cached) callout
3578 verification failure".
3580 3. There was an off-by-one bug in the queryprogram router. An over-long
3581 return line was truncated at 256 instead of 255 characters, thereby
3582 overflowing its buffer with the terminating zero. As well as fixing this, I
3583 have increased the buffer size to 1024 (and made a note to document this).
3585 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3586 when Exim is waiting for an SMTP response from a remote server, Exim
3587 restarts its select() call on the socket, thereby resetting its timeout.
3588 This is not a problem when such interrupts are rare. Somebody set up a cron
3589 job to run exiwhat every 2 minutes, which is less than the normal select()
3590 timeout (5 or 10 minutes). This meant that the select() timeout never
3591 kicked in because it was always reset. I have fixed this by comparing the
3592 time when an interrupt arrives with the time at the start of the first call
3593 to select(). If more time than the timeout has elapsed, the interrupt is
3594 treated as a timeout.
3596 5. Some internal re-factoring in preparation for the addition of Sieve
3597 extensions (by MH). In particular, the "personal" test is moved to a
3598 separate function, and given an option for scanning Cc: and Bcc: (which is
3599 not set for Exim filters).
3601 6. When Exim created an email address using the login of the caller as the
3602 local part (e.g. when creating a From: or Sender: header line), it was not
3603 quoting the local part when it contained special characters such as @.
3605 7. Installed new OpenBSD configuration files.
3607 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3608 try to make them clearer.
3610 9. Callout options, other than the timeout value, were being ignored when
3611 verifying sender addresses in header lines. For example, when using
3613 verify = header_sender/callout=no_cache
3615 the cache was (incorrectly) being used.
3617 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3618 only the Cygwin environment.
3620 11. When return_path_on_delivery was set as a log selector, if different remote
3621 addresses in the same message used different return paths and parallel
3622 remote delivery occurred, the wrong values would sometimes be logged.
3623 (Whenever a remote delivery process finished, the return path value from
3624 the most recently started remote delivery process was logged.)
3626 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3627 lines when AUTH and/or TLS are in use. This is the "received protocol"
3628 field. Exim used to use "asmtp" for authenticated SMTP, without any
3629 indication (in the protocol name) for TLS use. Now it follows the RFC and
3630 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3631 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3632 names appear in log lines as well as in Received: header lines.
3634 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3635 extensions, and comparison tests, and to fix some bugs.
3637 14. Changes to the "personal" filter test:
3639 (1) The test was buggy in that it was just doing the equivalent of
3640 "contains" tests on header lines. For example, if a user's address was
3641 anne@some.where, the "personal" test would incorrectly be true for
3643 To: susanne@some.where
3645 This test is now done by extracting each address from the header in turn,
3646 and checking the entire address. Other tests that are part of "personal"
3647 are now done using regular expressions (for example, to check local parts
3648 of addresses in From: header lines).
3650 (2) The list of non-personal local parts in From: addresses has been
3651 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3652 taken from the Sieve specification recommendations.
3654 (3) If the message contains any header line starting with "List-" it is
3655 treated as non-personal.
3657 (4) The test for "circular" in the Subject: header line has been removed
3658 because it now seems ill-conceived.
3660 15. Minor typos in src/EDITME comments corrected.
3662 16. Installed latest exipick from John Jetmore.
3664 17. If headers_add on a router specified a text string that was too long for
3665 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3666 of string_sprintf() is now avoided.
3668 18. $message_body_size was not set (it was always zero) when running the DATA
3669 ACL and the local_scan() function.
3671 19. For the "mail" command in an Exim filter, no default was being set for
3672 the once_repeat time, causing a random time value to be used if "once" was
3673 specified. (If the value happened to be <= 0, no repeat happened.) The
3674 default is now 0s, meaning "never repeat". The "vacation" command was OK
3675 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3676 (I found it when inspecting the code).
3678 20. There is now an overall timeout for performing a callout verification. It
3679 defaults to 4 times the callout timeout, which applies to individual SMTP
3680 commands during the callout. The overall timeout applies when there is more
3681 than one host that can be tried. The timeout is checked before trying the
3682 next host. This prevents very long delays if there are a large number of
3683 hosts and all are timing out (e.g. when the network connections are timing
3684 out). The value of the overall timeout can be changed by specifying an
3685 additional sub-option for "callout", called "maxwait". For example:
3687 verify = sender/callout=5s,maxwait=20s
3689 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3690 to the end before writing, but this should make it even safer).
3692 22. Exim was forgetting that it had advertised PIPELINING for the second and
3693 subsequent messages on an SMTP connection. It was also not resetting its
3694 memory on STARTTLS and an internal HELO.
3696 23. When Exim logs an SMTP synchronization error within a session, it now
3697 records whether PIPELINING has been advertised or not.
3699 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3700 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3701 rather than long int.
3703 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3705 26. Added the never_mail option to autoreply.
3711 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3712 crash if the getsockname() call failed; this can happen if a connection is
3713 closed very soon after it is established. The problem was simply in the
3714 order in which certain operations were done, causing Exim to try to write
3715 to the SMTP stream before it had set up the file descriptor. The bug has
3716 been fixed by making things happen in the correct order.
3722 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3723 before the connection was closed, thus losing the rejection response.
3725 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3726 some early Solaris releases, but causes trouble in current releases where
3727 socklen_t is defined.
3729 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3732 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3735 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3736 that releasing the top of it at the end releases what was used for sub-
3737 expansions (unless the block got too big). However, discard this block if
3738 the first thing is a variable or header, so that we can use its block when
3739 it is dynamic (useful for very large $message_headers, for example).
3741 6. Lookups now cache *every* query, not just the most recent. A new, separate
3742 store pool is used for this. It can be recovered when all lookup caches are
3743 flushed. Lookups now release memory at the end of their result strings.
3744 This has involved some general refactoring of the lookup sources.
3746 7. Some code has been added to the store_xxx() functions to reduce the amount
3747 of flapping under certain conditions.
3749 8. log_incoming_interface used to affect only the <= reception log lines. Now
3750 it causes the local interface and port to be added to several more SMTP log
3751 lines, for example "SMTP connection from", and rejection lines.
3753 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3755 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3757 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3758 could be overwritten at the end of the current message (or the start of a
3759 new message if it was set in a HELO ACL). The value is now preserved for
3760 the duration of the SMTP connection.
3762 12. If a transport had a headers_rewrite setting, and a matching header line
3763 contained an unqualified address, that address was qualified, even if it
3764 did not match any rewriting rules. The underlying bug was that the values
3765 of the flags that permit the existence of unqualified sender and recipient
3766 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3767 non-local messages, and by -bnq for local messages) were not being
3768 preserved with the message after it was received.
3770 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3771 "next input=" as part of the text comprising the host identity instead of
3772 the correct text. The code was using the same buffer for two different
3773 strings. However, depending on which order the printing function evaluated
3774 its arguments, the bug did not always show up. Under Linux, for example, my
3775 test suite worked just fine.
3777 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3778 doesn't work with some older versions of Perl. It has been changed to "my",
3779 which in any case is probably the better facility to use.
3781 15. A really picky compiler found some instances of statements for creating
3782 error messages that either had too many or two few arguments for the format
3785 16. The size of the buffer for calls to the DNS resolver has been increased
3786 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3787 for addresses that have a lot of PTR records. This alleviates a problem; it
3788 does not fully solve it.
3790 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3791 buffer now truncates the list and logs the incident, which is the same
3792 action as happens when Exim is looking up a host name and its aliases.
3793 Previously in this situation something unpredictable would happen;
3794 sometimes it was "internal error: store_reset failed".
3796 18. If a server dropped the connection unexpectedly when an Exim client was
3797 using GnuTLS and trying to read a response, the client delivery process
3798 crashed while trying to generate an error log message.
3800 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3801 string, for example:
3803 warn message = H1: something\nH2: something
3805 the text was added as a single header line from Exim's point of view
3806 though it ended up OK in the delivered message. However, searching for the
3807 second and subsequent header lines using $h_h2: did not work. This has been
3808 fixed. Similarly, if a system filter added multiple headers in this way,
3809 the routers could not see them.
3811 20. Expanded the error message when iplsearch is called with an invalid key to
3812 suggest using net-iplsearch in a host list.
3814 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3815 ACLs are no longer actually imposed (and a message to that effect is
3818 22. If a "gecos" field in a passwd entry contained escaped characters, in
3819 particular, if it contained a \" sequence, Exim got it wrong when building
3820 a From: or a Sender: header from that name. A second bug also caused
3821 incorrect handling when an unquoted " was present following a character
3822 that needed quoting.
3824 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3825 was not being matched caselessly.
3827 24. Arranged for all hyphens in the exim.8 source to be escaped with
3830 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3831 properly. Recipient callout cache records were still being keyed to include
3832 the sender, even when use_sender was set false. This led to far more
3833 callouts that were necessary. The sender is no longer included in the key
3834 when use_sender is false.
3836 26. Added "control = submission" modifier to ACLs.
3838 27. Added the ${base62d: operator to decode base 62 numbers.
3840 28. dnsdb lookups can now access SRV records.
3842 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3843 the configuration file.
3845 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3846 (-v) mode. This makes the output for a verbose queue run more intelligible.
3848 31. Added a use_postmaster feature to recipient callouts.
3850 32. Added the $body_zerocount variable, containing the number of binary zero
3851 bytes in the message body.
3853 33. The time of last modification of the "new" subdirectory is now used as the
3854 "mailbox time last read" when there is a quota error for a maildir
3857 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3859 35. Added +ignore_unknown as a special item in host lists.
3861 36. Code for decoding IPv6 addresses in host lists is now included, even if
3862 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3863 address was recognized as an IP address, but was then not correctly decoded
3864 into binary, causing unexpected and incorrect effects when compared with
3871 1. Very minor rewording of debugging text in manualroute to say "list of
3872 hosts" instead of "hostlist".
3874 2. If verify=header_syntax was set, and a header line with an unqualified
3875 address (no domain) and a large number of spaces between the end of the
3876 name and the colon was received, the reception process suffered a buffer
3877 overflow, and (when I tested it) crashed. This was caused by some obsolete
3878 code that should have been removed. The fix is to remove it!
3880 3. When running in the test harness, delay a bit after writing a bounce
3881 message to get a bit more predictability in the log output.
3883 4. Added a call to search_tidyup() just before forking a reception process. In
3884 theory, someone could use a lookup in the expansion of smtp_accept_max_
3885 per_host which, without the tidyup, could leave open a database connection.
3887 5. Added the variables $recipient_data and $sender_data which get set from a
3888 lookup success in an ACL "recipients" or "senders" condition, or a router
3889 "senders" option, similar to $domain_data and $local_part_data.
3891 6. Moved the writing of debug_print from before to after the "senders" test
3894 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3895 problems for message scanning, either using a data ACL, or using
3896 local_scan() because the Received: header was not generated till after they
3897 were called (in order to set the time as the time of reception completion).
3898 I have revised the way this works. The header is now generated after the
3899 body is received, but before the ACL or local_scan() are called. After they
3900 are run, the timestamp in the header is updated.
3906 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3907 before starting a queue runner without re-exec. This happened only when
3908 deliver_drop_privilege was set or when the Exim user was set to root. The
3909 effect of the bug was that timeouts during subsequent deliveries caused
3910 crashes instead of being properly handled. The handler is now left at its
3911 default (and expected) setting.
3913 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3914 message, again when deliver_drop_privilege is set or Exim is run as root.
3915 The bug described in (1) was not present in this case, but the tidying up
3916 of the other signals was missing. I have made the two cases consistent.
3918 3. The ignore_target_hosts setting on a manualroute router was being ignored
3919 for hosts that were looked up using the /MX notation.
3921 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3924 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3925 operated on the sender address. After changing the $sender_address to <>
3926 for the sender address verify, Exim was re-instated it as the original
3927 (before rewriting) address, but remembering that it had rewritten it, so it
3928 wasn't rewriting it again. This bug also had the effect of breaking the
3929 sender address verification caching when the sender address was rewritten.
3931 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3932 This has been changed so that if the ip literal address matches
3933 ignore_target_hosts, the router declines.
3935 7. Added expansion conditions match_domain, match_address, and match_local_
3936 part (NOT match_host).
3938 8. The placeholder for the Received: header didn't have a length field set.
3940 9. Added code to Exim itself and to exim_lock to test for a specific race
3941 condition that could lead to file corruption when using MBX delivery. The
3942 issue is with the lockfile that is created in /tmp. If this file is removed
3943 after a process has opened it but before that process has acquired a lock,
3944 there is the potential for a second process to recreate the file and also
3945 acquire a lock. This could lead to two Exim processes writing to the file
3946 at the same time. The added code performs the same test as UW imapd; it
3947 checks after acquiring the lock that its file descriptor still refers to
3948 the same named file.
3950 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3951 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3954 11. Added the smtp_active_hostname option. If used, this will typically be made
3955 to depend on the incoming interface address. Because $interface_address is
3956 not set up until the daemon has forked a reception process, error responses
3957 that can happen earlier (such as "too many connections") no longer contain
3960 12. If an expansion in a condition on a "warn" statement fails because a lookup
3961 defers, the "warn" statement is abandoned, and the next ACL statement is
3962 processed. Previously this caused the whole ACL to be aborted.
3964 13. Added the iplsearch lookup type.
3966 14. Added ident_timeout as a log selector.
3968 15. Added tls_certificate_verified as a log selector.
3970 16. Added a global option tls_require_ciphers (compare the smtp transport
3971 option of the same name). This controls incoming TLS connections.
3973 17. I finally figured out how to make tls_require_ciphers do a similar thing
3974 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3975 before starting the TLS session.
3977 18. Tabs are now shown as \t in -bP output.
3979 19. If the log selector return_path_on_delivery was set, Exim crashed when
3980 bouncing a message because it had too many Received: header lines.
3982 20. If two routers both had headers_remove settings, and the first one included
3983 a superfluous trailing colon, the final name in the first list and the
3984 first name in the second list were incorrectly joined into one item (with a
3985 colon in the middle).
3991 1. Added -C and -D options to the exinext utility, mainly to make it easier
3992 to include in the automated testing, but these could be helpful when
3993 multiple configurations are in use.
3995 2. The exinext utility was not formatting the output nicely when there was
3996 an alternate port involved in the retry record key, nor when there was a
3997 message id as well (for retries that were specific to a specific message
3998 and a specific host). It was also confused by IPv6 addresses, because of
3999 the additional colons they contain. I have fixed the IPv4 problem, and
4000 patched it up to do a reasonable job for IPv6.
4002 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4003 delivery, the log line now contains "pipelined" if PIPELINING was used.
4005 4. An SMTP transport process used to panic and die if the bind() call to set
4006 an explicit outgoing interface failed. This has been changed; it is now
4007 treated in the same way as a connect() failure.
4009 5. A reference to $sender_host_name in the part of a conditional expansion
4010 that was being skipped was still causing a DNS lookup. This no longer
4013 6. The def: expansion condition was not recognizing references to header lines
4014 that used bh_ and bheader_.
4016 7. Added the _cache feature to named lists.
4018 8. The code for checking quota_filecount in the appendfile transport was
4019 allowing one more file than it should have been.
4021 9. For compatibility with Sendmail, the command line option
4029 and sets the incoming protocol and host name (for trusted callers). The
4030 host name and its colon can be omitted when only the protocol is to be set.
4031 Note the Exim already has two private options, -pd and -ps, that refer to
4032 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4033 "s", but I don't think that's a major issue.
4035 10. A number of refactoring changes to the code, none of which should affect
4038 (a) The number of logging options was getting close to filling up the
4039 32-bit word that was used as a bit map. I have split them into two classes:
4040 those that are passed in the argument to log_write(), and those that are
4041 only ever tested independently outside of that function. These are now in
4042 separate 32-bit words, so there is plenty of room for expansion again.
4043 There is no change in the user interface or the logging behaviour.
4045 (b) When building, for example, log lines, the code previously used a
4046 macro that called string_cat() twice, in order to add two strings. This is
4047 not really sufficiently general. Furthermore, there was one instance where
4048 it was actually wrong because one of the argument was used twice, and in
4049 one call a function was used. (As it happened, calling the function twice
4050 did not affect the overall behaviour.) The macro has been replaced by a
4051 function that can join an arbitrary number of extra strings onto a growing
4054 (c) The code for expansion conditions now uses a table and a binary chop
4055 instead of a serial search (which was left over from when there were very
4056 few conditions). Also, it now recognizes conditions like "pam" even when
4057 the relevant support is not compiled in: a suitably worded error message is
4058 given if an attempt is made to use such a condition.
4060 11. Added ${time_interval:xxxxx}.
4062 12. A bug was causing one of the ddress fields not to be passed back correctly
4063 from remote delivery subprocesses. The field in question was not being
4064 subsequently used, so this caused to problems in practice.
4066 13. Added new log selectors queue_time and deliver_time.
4068 14. Might have fixed a bug in maildirsizefile handling that threw up
4069 "unexpected character" debug warnings, and recalculated the data
4070 unnecessarily. In any case, I expanded the warning message to give more
4073 15. Added the message "Restricted characters in address" to the statements in
4074 the default ACL that block characters like @ and % in local parts.
4076 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4077 Three changes have been made:
4079 (a) There was a serious bug; a negative response to MAIL caused the whole
4080 recipient domain to be cached as invalid, thereby blocking all messages
4081 to all local parts at the same domain, from all senders. This bug has
4082 been fixed. The domain is no longer cached after a negative response to
4083 MAIL if the sender used is not empty.
4085 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4088 (c) A new callout option, "use_sender" has been added for people who want
4089 the modified behaviour.
4095 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4098 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4099 indeed breaks things for older releases.
4101 3. Added additional logging to the case where there is a problem reading data
4102 from a filter that is running in a subprocess using a pipe, in order to
4103 try to track down a specific problem.
4105 4. Testing facility fudge: when running in the test harness and attempting
4106 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4107 getting "No route to host". Convert this to a timeout.
4109 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4112 6. Some OS don't have socklen_t but use size_t instead. This affects the
4113 fifth argument of getsockopt() amongst other things. This is now
4114 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4115 can be set for individual OS. I have set it for SunOS5, OSF1, and
4116 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4117 some earlier ones do not.
4119 7. Change 4.30/15 was not doing the test caselessly.
4121 8. The standard form for an IPv6 address literal was being rejected by address
4122 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4123 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4124 this, as well as the form without the "IPv6" on the front (but only when
4125 address literals are enabled, of course).
4127 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4129 10. Exim crashed if a message with an empty sender address specified by -f
4130 encountered a router with an errors_to setting. This could be provoked only
4131 by a command such as
4135 where an empty string was supplied; "<>" did not hit this bug.
4137 11. Installed PCRE release 4.5.
4139 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4140 remained set. It is now erased.
4142 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4143 times from message ids (which are base 36 rather than the normal 62).
4145 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4146 were being counted as actual protocol errors, and logged if the log
4147 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4148 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4149 and DATA following a set of rejected RCPTs do not count as protocol errors.
4150 In other words, Exim assumes they were pipelined, though this may not
4151 actually be the case. Of course, in all cases the client gets an
4152 appropriate error code.
4154 15. If a lookup fails in an ACL condition, a message about the failure may
4155 be available; it is used if testing the ACL cannot continue, because most
4156 such messages specify what the cause of the deferral is. However, some
4157 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4158 that caused an old message to be retained and used if a later statement
4159 caused a defer, replacing the real cause of the deferral.
4161 16. If an IP address had so many PTR records that the DNS lookup buffer
4162 was not large enough to hold them, Exim could crash while trying to process
4163 the truncated data. It now detects and logs this case.
4165 17. Further to 4.21/58, another change has been made: if (and only if) the
4166 first line of a message (the first header line) ends with CRLF, a bare LF
4167 in a subsequent header line has a space inserted after it, so as not to
4168 terminate the header.
4170 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4171 unnecessarily, used atoi() instead of strtol(), and didn't check the
4172 termination when getting file sizes from file names by regex.
4174 19. Completely re-implemented the support for maildirsize files, in the light
4175 of a number of problems with the previous contributed implementation
4176 (4.30/29). In particular:
4178 . If the quota is zero, the maildirsize file is maintained, but no quota is
4181 . If the maildir directory does not exist, it is created before any attempt
4182 to write a maildirsize file.
4184 . The quota value in the file is just a cache; if the quota is changed in
4185 the transport, the new value overrides.
4187 . A regular expression is available for excluding directories from the
4190 20. The autoreply transport checks the characters in options that define the
4191 message's headers; it allows continued headers, but it was checking with
4192 isspace() after an embedded newline instead of explicitly looking for a
4195 21. If all the "regular" hosts to which an address was routed had passed their
4196 expiry times, and had not reached their retry times, the address was
4197 bounced, even if fallback hosts were defined. Now Exim should go on to try
4200 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4201 equivalent code in the SMTP transport. Some hosts send humungous responses
4202 to HELO/EHLO, more than 1024 it seems.
4204 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4205 gives compiler warnings in some environments. I've now done it "properly",
4208 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4209 (because of problems with the built-in one) was declared to return uschar *
4210 instead of char *, causing compiler failure.
4212 25. Fixed a file descriptor leak when processing alias/forward files.
4214 26. Fixed a minor format string issue in dbfn.c.
4216 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4218 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4219 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4222 29. When a sender address is verified, it is cached, to save repeating the test
4223 when there is more than one recipient in a message. However, when the
4224 verification involves a callout, it is possible for different callout
4225 options to be set for different recipients. It is too complicated to keep
4226 track of this in the cache, so now Exim always runs a verification when a
4227 callout is required, relying on the callout cache for the optimization.
4228 The overhead is duplication of the address routing, but this should not be
4231 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4232 to be verified with callout=postmaster, and the main callout worked but the
4233 postmaster check failed, the verification correctly failed. However, if a
4234 subsequent RCPT command asked for sender verification *without* the
4235 postmaster check, incorrect caching caused this verification also to fail,
4238 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4239 it was not caching the DNS options (qualify_single, search_parents) that
4240 were used when the lookup failed. A subsequent lookup with different
4241 options therefore always gave the same answer, though there were cases
4242 where it should not have. (Example: a "domains = !$mx_any" option on a
4243 dnslookup router: the "domains" option is always processed without any
4244 widening, but the router might have qualify_single set.) Now Exim uses the
4245 cached value only when the same options are set.
4247 32. Added John Jetmore's "exipick" utility to the distribution.
4249 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4250 than a timeout (e.g. a certificate is required, and is not provided), an
4251 Exim server now closes the connection immediately. Previously it waited for
4252 the client to close - but if the client is SSL, it seems that they each
4253 wait for each other, leading to a delay before one of them times out.
4255 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4256 maintained 0.8.x compatibility because I don't think many are using it, and
4257 it is clearly obsolete.
4259 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4262 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4263 client certificate was expired. A simple patch fixes this, though I don't
4264 understand the full logic of why the verify callback is called multiple
4267 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4268 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4269 which causes problems with some clients (such as the Certicom SSL Plus
4270 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4271 disables the coutermeasure allowing Eudora to connect."
4273 38. Exim was not checking that a write() to a log file succeeded. This could
4274 lead to Bad Things if a log got too big, in particular if it hit a file
4275 size limit. Exim now panics and dies if it cannot write to a log file, just
4276 as it does if it cannot open a log file.
4278 39. Modified OS/Makefile-Linux so that it now contains
4280 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4282 The two -D definitions ensure that Exim is compiled with large file
4283 support, which makes it possible to handle log files that are bigger than
4286 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4287 instance) a domain was checked against a named list that involved a lookup,
4288 causing $domain_data to be set, then another domain was checked against the
4289 same list, then the first domain was re-checked, the value of $domain_data
4290 after the final check could be wrong. In particular, if the second check
4291 failed, it could be set empty. This bug probably also applied to
4294 41. The strip_trailing_dot option was not being applied to the address given
4295 with the -f command-line option.
4297 42. The code for reading a message's header from the spool was incrementing
4298 $received_count, but never initializing it. This meant that the value was
4299 incorrect (doubled) while delivering a message in the same process in which
4300 it was received. In the most common configuration of Exim, this never
4301 happens - a fresh exec is done - but it can happen when
4302 deliver_drop_privilege is set.
4304 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4305 it now includes up to 150 characters of the unexpected data in the log
4308 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4309 and building data strings. The size of both of these buffers was 10 000
4310 bytes - far larger than anybody would *ever* want, thought I. Needless to
4311 say, somebody hit the limit. I have increased the maximum line length to
4312 20 000 and the maximum data length of concatenated lines to 100 000. I have
4313 also fixed two bugs, because there was no checking on these buffers. Tsk,
4314 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4315 buffer is too small.
4317 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4318 lsearch lookups. Now it does.
4320 46. When parsing a route_list item in a manualroute router, a fixed-length
4321 buffer was used for the list of hosts. I made this 1024 bytes long,
4322 thinking that nobody would ever have a list of hosts that long. Wrong.
4323 Somebody had a whole pile of complicated expansion conditions, and the
4324 string was silently truncated, leading to an expansion error. It turns out
4325 that it is easier to change to an unlimited length (owing to other changes
4326 that have happened since this code was originally written) than to build
4327 structure for giving a limitation error. The length of the item that
4328 expands into the list of hosts is now unlimited.
4330 47. The lsearch lookup could not handle data where the length of text line was
4331 more than 4095 characters. Such lines were truncated, leading to shortened
4332 data being returned. It should now handle lines of any length.
4334 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4335 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4338 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4339 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4340 no longer happens in comment lines. A list of replacements is now placed
4341 at the head of all of the source files, except those whose only change is
4342 to replace PERL_COMMAND in the very first #! line.
4344 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4345 messages on the queue, with a bottom-up merge sort, using code contributed
4346 by Michael Haardt. This should make operations like -bp somewhat faster on
4347 large queues. It won't affect queue runners, except when queue_run_in_order
4350 51. Installed eximstats 1.31 in the distribution.
4352 52. Added support for SRV lookups to the dnslookup router.
4354 53. If an ACL referred to $message_body or $message_body_end, the value was not
4355 reset for any messages that followed in the same SMTP session.
4357 54. The store-handling optimization for building very long strings was not
4358 differentiating between the different store pools. I don't think this
4359 actually made any difference in practice, but I've tidied it.
4361 55. While running the routers to verify a sender address, $sender_address
4362 was still set to the sender address. This is wrong, because when routing to
4363 send a bounce to the sender, it would be empty. Therefore, I have changed
4364 it so that, while verifying a sender address, $sender_address is set to <>.
4365 (There is no change to what happens when verifying a recipient address.)
4367 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4368 target A or AAAA records (if not already returned) without resetting the
4369 qualify_single or search_parents options of the DNS resolver. These are
4370 inappropriate in this case because the targets of MX and SRV records must
4371 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4372 target that, when qualified, matched something in the local domain. These
4373 two options are now turned off when doing these lookups.
4375 57. It seems that at least some releases of Reiserfs (which does not have the
4376 concept of a fixed number of inodes) returns zero and not -1 for the
4377 number of available inodes. This interacted badly with check_spool_inodes,
4378 which assumed that -1 was the "no such thing" setting. What I have done is
4379 to check that the total number of inodes is greater than zero before doing
4380 the test of how many are available.
4382 58. When a "warn" ACL statement has a log_message modifier, the message is
4383 remembered, and not repeated. This is to avoid a lot of repetition when a
4384 message has many recipients that cause the same warning to be written.
4385 Howewer, Exim was preserving the list of already written lines for an
4386 entire SMTP session, which doesn't seem right. The memory is now reset if a
4387 new message is started.
4389 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4390 debugging output unless log_rewrite was also set.
4392 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4393 of (int)(handle) when we know that handle contains (void *)(-1).
4395 61. The Exim daemon panic-logs an error return when it closes the incoming
4396 connection. However "connection reset by peer" seems to be common, and
4397 isn't really an error worthy of noting specially, so that particular error
4400 62. When Exim is trying to find all the local interfaces, it used to panic and
4401 die if the ioctl to get the interface flags failed. However, it seems that
4402 on at least one OS (Solaris 9) it is possible to have an interface that is
4403 included in the list of interfaces, but for which you get a failure error
4404 for this call. This happens when the interface is not "plumbed" into a
4405 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4406 failure of the "get flags" call assumes that the interface is down.
4408 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4409 makes life easier for people who are doing arithmetic on fields extracted
4410 from dates, where you often get leading zeros that should not be
4411 interpreted as octal.
4413 64. Added qualify_domain to the redirect router, to override the global
4416 65. If a pathologically long header line contained very many addresses (the
4417 report of this problem mentioned 10 000) and each of them was rewritten,
4418 Exim could use up a very large amount of memory. (It kept on making new
4419 copies of the header line as it rewrote, and never released the old ones.)
4420 At the expense of a bit more processing, the header rewriting function has
4421 been changed so that it no longer eats memory in this way.
4423 66. The generation of the Received: header has been moved from the time that a
4424 message starts to be received, to the time that it finishes. The timestamp
4425 in the Received: header should now be very close to that of the <= log
4426 line. There are two side-effects of this change:
4428 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4429 logged header lines no longer include the local Received: line, because
4430 it has not yet been created. The same applies to a copy of the message
4431 that is returned to a non-SMTP sender when a message is rejected.
4433 (b) When a filter file is tested using -bf, no additional Received: header
4434 is added to the test message. After some thought, I decided that this
4437 This change does not affect the value of $received_for. It is still set
4438 after address rewriting, but before local_scan() is called.
4440 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4442 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4443 gave an unhelpful panic error message, and a defer error. I have managed to
4444 change this behaviour so that it now rejects any supplied certificate,
4445 which seems right, as the list of acceptable certificates is empty.
4447 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4448 gave an unhelpful defer error. I have not managed to make this reject any
4449 supplied certificates, but the error message it gives is "no certificate
4450 supplied", which is not helpful.
4452 70. exigrep's output now also includes lines that are not associated with any
4453 message, but which match the given pattern. Implemented by a patch from
4454 Martin Sluka, which also tidied up the Perl a bit.
4456 71. Recipient callout verification, like sender verification, was using <> in
4457 the MAIL FROM command. This isn't really the right thing, since the actual
4458 sender may affect whether the remote host accepts the recipient or not. I
4459 have changed it to use the actual sender in the callout; this means that
4460 the cache record is now keyed on a recipient/sender pair, not just the
4461 recipient address. There doesn't seem to be a real danger of callout loops,
4462 since a callout by the remote host to check the sender would use <>.
4463 [SEE ABOVE: changed after hitting problems.]
4465 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4466 temporary errors. However, in the case of such a code being given after
4467 the end of a data transmission (i.e. after ".") Exim was failing to write
4468 a retry record for the message. (Yes, there was some broken host that was
4469 actually sending 8xx at this point.)
4471 73. An unknown lookup type in a host list could cause Exim to panic-die when
4472 the list was checked. (An example that provoked this was putting <; in the
4473 middle of a list instead of at the start.) If this happened during a DATA
4474 ACL check, a -D file could be left lying around. This kind of configuration
4475 error no longer causes Exim to die; instead it causes a defer errror. The
4476 incident is still logged to the main and panic logs.
4478 74. Buglet left over from Exim 3 conversion. The message "too many messages
4479 in one connection" was written to the rejectlog but not the mainlog, except
4480 when address rewriting (yes!) was being logged.
4482 75. Added write_rejectlog option.
4484 76. When a system filter was run not as root (that is, when system_filter_user
4485 was set), the values of the $n variables were not being returned to the
4486 main process; thus, they were not subsequently available in the $sn
4489 77. Added +return_path_on_delivery log selector.
4491 78. A connection timeout was being treated differently from recipients deferred
4492 when testing hosts_max_try with a message that was older than the host's
4493 retry timeout. (The host should not be counted, thus allowing all hosts to
4494 be tried at least once before bouncing.) This may have been the cause of an
4495 occasionally reported bug whereby a message would remain on the queue
4496 longer than the retry timeout, but would be bounced if a delivery was
4497 forced. I say "may" because I never totally pinned down the problem;
4498 setting up timeout/retry tests is difficult. See also the next item.
4500 79. The ultimate address timeout was not being applied to errors that involved
4501 a combination of host plus message (for example, a timeout on a MAIL
4502 command). When an address resolved to a number of possible hosts, and they
4503 were not all tried for each delivery (e.g. because of hosts_max_try), a
4504 message could remain on the queue longer than the retry timeout.
4506 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4509 81. Fixed an obscure SMTP outgoing bug which required at least the following
4510 conditions: (a) there was another message waiting for the same server;
4511 (b) the server returned 5xx to all RCPT commands in the first message so
4512 that the message was not completed; (c) the server dropped the connection
4513 or gave a negative response to the RSET that Exim sends to abort the
4514 transaction. The observed case was a dropped connection after DATA that had
4515 been sent in pipelining mode. That is, the server had advertised PIPELINING
4516 but was not implementing it correctly. The effect of the bug was incorrect
4517 behaviour, such as trying another host, and this could lead to a crash.
4523 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4524 and daemon.c were passed as pointers to ints; they should have been
4525 pointers to socklen_t variables (which are typically unsigned ints).
4527 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4530 3. Fixed a really odd bug that affected only the testing scheme; patching a
4531 certain fixed string in the binary changed the value of another string that
4532 happened to be identical to the end of the original first string.
4534 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4535 name", it returns that address as the IP address. On some operating
4536 systems (e.g. Solaris), it also passes back the IP address string as the
4537 "host name". However, on others (e.g. Linux), it passes back an empty
4538 string. Exim wasn't checking for this, and was changing the host name to an
4539 empty string, assuming it had been canonicized.
4541 5. Although rare, it is permitted to have more than one PTR record for a given
4542 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4543 all the names associated with an address, because they do in Solaris.
4544 However, it seems that they do not in Linux for data that comes from the
4545 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4546 I found this out when I moved to a new Linux workstation and tried to run
4547 the Exim test suite.
4549 To get round this problem I have changed the code so that it now does its
4550 own call to the DNS to look up PTR records when searching for a host name.
4551 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4552 addresses that are only in /etc/hosts are still found.
4554 This behaviour is, however, controlled by an option called host_lookup_
4555 order, which defaults to "bydns:byaddr". If people want to use the other
4556 order, or indeed, just use one or the other means of lookup, they can
4557 specify it in this variable.
4559 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4560 some operating systems, this comes back from gethostbyaddr() as an empty
4561 string, and this is what Exim used to test for. However, it seems that in
4562 other systems, "." is yielded. Exim now tests for this case too.
4564 7. The values of check_spool_space and check_log_space are now held internally
4565 as a number of kilobytes instead of an absolute number of bytes. If a
4566 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4567 kilobyte. This means that much larger values can be stored.
4569 8. Exim monitor: an attempt to get the action menu when not actually pointing
4570 at a message produces an empty menu entitled "No message selected". This
4571 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4572 no entries in it ("Shell widget menu has zero width and/or height"). So I
4573 have added a single, blank menu entry in this case.
4575 9. Added ${quote_local_part.
4577 10. MIME decoding is now applied to the contents of Subject: header lines when
4580 11. Now that a reference to $sender_host_address automatically causes a reverse
4581 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4582 host lookup before query-style lookups in lists that might use this
4583 variable. This has therefore been abolished, and the "net-" prefix is no
4584 longer necessary for query-style lookups.
4586 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4587 have been a typo for LFLAGS, so it has been changed.
4589 13. The install script calls Exim with "-C /dev/null" in order to find the
4590 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4591 to be output. Howeve, since Exim outputs its version number before the
4592 error, it didn't break the script. It just looked ugly. I fixed this by
4593 always allowing "-C /dev/null" if the caller is root.
4595 14. Ignore overlarge ACL variable number when reading spool file - insurance
4596 against a later release with more variables having written the file.
4598 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4599 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4600 this, as well as the form without the "IPv6" on the front.
4602 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4603 OS/Makefile-Darwin file.
4605 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4606 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4608 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4609 variable. However, some OS use ints for inodes. I've added cast to long int
4610 to get rid of the compiler warning.
4612 19. I had forgotten to lock out "/../" in configuration file names when
4613 ALT_CONFIG_PREFIX was set.
4615 20. Routers used for verification do not need to specify transports. However,
4616 if such a router generated a host list, and callout was configured, Exim
4617 crashed, because it could not find a port number from the (non-existent)
4618 transport. It now assumes port 25 in this circumstance.
4620 21. Added the -t option to exigrep.
4622 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4623 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4624 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4625 them is set, LOOKUP_LSEARCH is forced.
4627 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4629 24. Added sender and host information to the "rejected by local_scan()" log
4630 line; previously there was no indication of these.
4632 25. Added .include_if_exists.
4634 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4635 Linux. It turns out that not all file systems support this. Apparently some
4636 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4637 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4638 not supported on the file descriptor, is now ignored when Exim is trying to
4639 sync a directory. This applies only to Linux.
4641 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4643 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4644 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4647 29. Added support for maildirsize files from supplied patch (modified a bit).
4649 30. The use of :fail: followed by an empty string could lead Exim to respond to
4650 sender verification failures with (e.g.):
4652 550 Verification failed for <xxx>
4653 550 Sender verify failed
4655 where the first response line was missing the '-' that indicates it is not
4656 the final line of the response.
4658 31. The loop for finding the name of the user that called Exim had a hardwired
4659 limit of 10; it now uses the value of finduser_retries, which is used for
4660 all other user lookups.
4662 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4665 33. Exim was neglecting to zero errno before one call of strtol() when
4666 expanding a string and expecting an integer value. On some systems this
4667 resulted in spurious "integer overflow" errors. Also, it was casting the
4668 result into an int without checking.
4670 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4671 did not work. The code looks as if it has *never* worked, though it appears
4672 to have been documented since at least releast 1.62. I have made it work.
4674 35. The "timeout_DNS" error in retry rules, also documented since at least
4675 1.62, also never worked. As it isn't clear exactly what this means, and
4676 clearly it isn't a major issue, I have abolished the feature by treating it
4677 as "timeout", and writing a warning to the main and panic logs.
4679 36. The display of retry rules for -brt wasn't always showing the error code
4682 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4683 timeout_connect_A, timeout_connect_MX.
4685 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4686 to the empty sender.
4688 39. The daemon was not analysing the content of -oX till after it had closed
4689 stderr and disconnected from the controlling terminal. This meant that any
4690 syntax errors were only noted on the panic log, and the return code from
4691 the command was 0. By re-arranging the code a little, I've made the
4692 decoding happen first, so such errors now appear on stderr, and the return
4693 code is 1. However, the actual setting up of the sockets still happens in
4694 the disconnected process, so errors there are still only recorded on the
4697 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4698 connections (as happens on some IP stacks) was logged at start up time as
4699 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4700 it from "IPv6 and IPv4", which means that two separate sockets are being
4703 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4704 says whether AF_INET or AF_INET6 was passed as an argument.
4706 42. Exiwhat output was messed up when time zones were included in log
4709 43. Exiwhat now gives more information about the daemon's listening ports,
4710 and whether -tls-on-connect was used.
4712 44. The "port" option of the smtp transport is now expanded.
4714 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4715 silently ignored. Now an error message is written to the main and panic
4718 46. There's a new ACL modifier called "logwrite" which writes to a log file
4719 as soon as it is encountered.
4721 47. Added $local_user_uid and $local_user_gid at routing time.
4723 48. Exim crashed when trying to verify a sender address that was being
4726 49. Exim was recognizing only a space character after ".include". It now also
4727 recognizes a tab character.
4729 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4730 extracting the relevant information from the specification. The man page no
4731 longer contains scrambled data for the -d option, and I've added a section
4732 at the front about calling Exim under different names.
4734 51. Added "extra_headers" argument to the "mail" command in filter files.
4736 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4739 53. Installed eximstats 1.29.
4741 54. Added transport_filter_timeout as a generic transport option.
4743 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4744 Cc: header lines. This was required by RFC 822, but it not required by RFC
4747 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4748 incoming messages that did not have them. Now it does so only if the
4749 message originates locally, that is, if there is no associated remote host
4750 address. When Resent- header lines are present, this applies to the Resent-
4751 lines rather than the non-Resent- lines.
4753 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4754 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4756 58. Messages for configuration errors now include the name of the main
4757 configuration file - useful now that there may be more than one file in a
4758 list (.included file names were always shown).
4760 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4761 for those rare installations that do not start the daemon as root or run it
4762 setuid root. I've cut out the call to initgroups() if the daemon is not
4765 60. The Exim user and group can now be bound into the binary as text strings
4766 that are looked up at the start of Exim's processing.
4768 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4770 62. Added $mailstore_basename variable.
4772 63. Installed patch to sieve.c from Michael Haardt.
4774 64. When Exim failed to open the panic log after failing to open the main log,
4775 the original message it was trying to log was written to stderr and debug
4776 output, but if they were not available (the usual case in production), it
4777 was lost. Now it is written to syslog before the two lines that record the
4778 failures to open the logs.
4780 65. Users' Exim filters run in subprocesses under the user's uid. It is
4781 possible for a "deliver" command or an alias in a "personal" command to
4782 provoke an address rewrite. If logging of address rewriting is configured,
4783 this fails because the process is not running as root or exim. There may be
4784 a better way of dealing with this, but for the moment (because 4.30 needs
4785 to be released), I have disabled address rewrite logging when running a
4786 filter in a non-root, non-exim process.
4792 1. The buildconfig auxiliary program wasn't quoting the value set for
4793 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4794 not defined. This bug was masked in 4.22 by the effect that was fixed in
4797 2. Some messages that were rejected after a message id was allocated were
4798 shown as "incomplete" by exigrep. It no longer does this for messages that
4799 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4801 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4802 have allow_domain_literals set, the ID did not get logged in the <= line.
4803 Domain literals are now always recognized in Message-ID: header lines.
4805 4. The first argument for a ${extract expansion item is the key name or field
4806 number. Leading and trailing spaces in this item were not being ignored,
4807 causing some misleading effects.
4809 5. When deliver_drop_privilege was set, single queue runner processes started
4810 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4811 same command in the process it spins off) were not dropping privilege.
4813 6. When the daemon running as "exim" started a queue runner, it always
4814 re-executed Exim in the spun-off process. This is a waste of effort when
4815 deliver_drop_privilege is set. The new process now just calls the
4816 queue-runner function directly.
4822 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4825 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4826 ignored. Though the use of -oP was forcing the writing of a pid file, it
4827 was always written to the default place.
4829 3. If the message "no IP address found for host xxxx" is generated during
4830 incoming verification, it is now followed by identification of the incoming
4831 connection (so you can more easily find what provoked it).
4833 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4835 5. Added some features to "harden" Exim a bit more against certain attacks:
4837 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4838 be put in Local/Makefile. This is like the never_users runtime option,
4839 but it cannot be overridden. The default setting is "root".
4841 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4842 prefix string with which any file named in a -C command line option
4845 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4846 is retained for -C and -D only if the caller of Exim is root. Without
4847 it, the exim user may also use -C and -D and retain privilege.
4849 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4850 command line option is disabled.
4852 6. Macro names set by the -D option must start with an upper case letter, just
4853 like macro names defined in the configuration file.
4855 7. Added "dereference=" facility to LDAP.
4857 8. Two instances of the typo "uknown" in the source files are fixed.
4859 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4860 the Configure-Makefile script screwed up while processing it.
4862 10. Incorporated PCRE 4.4.
4864 11. The SMTP synchronization check was not operating right at the start of an
4865 SMTP session. For example, it could not catch a HELO sent before the client
4866 waited for the greeting. There is now a check for outstanding input at the
4867 point when the greeting is written. Because of the duplex, asynchronous
4868 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4869 way, but not yet received, when the check is performed.
4871 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4872 on TCP/IP sockets, because this apparently causes some broken clients to
4875 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4876 unchanged) from the Cygwin maintainer.
4878 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4879 was supported instead of testing for mbx. Effectively a typo.
4881 15. The spa authenticator server code was not checking that the input it
4882 received was valid base64.
4884 16. The debug output line for the "set" modifier in ACLs was not showing the
4885 name of the variable that was being set.
4887 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4889 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4890 lookup on its own. Something else was needed to trigger the lookup. For
4891 example, a match in host_lookup or the need for a host name in a host list.
4892 Now, if $sender_host_name is referenced and the host name has not yet been
4893 looked up, a lookup is performed. If the lookup fails, the variable remains
4894 empty, and $host_lookup_failed is set to "1".
4896 19. Added "eqi" as a case-independent comparison operator.
4898 20. The saslauthd authentication condition could segfault if neither service
4899 nor realm was specified.
4901 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4902 error message that was logged was misleading, and incoming SMTP
4903 connections were dropped. The message is now more accurate, and temporary
4904 errors are given to SMTP connections.
4906 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4907 (except RSET) with a 421 error, until QUIT is received. However, it was
4908 failing to send a response to QUIT.
4910 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4911 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4912 ACL is now run after the helo_try_verify_hosts code.
4914 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4915 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4916 case-independent, so other case variants are also recognized). Apparently
4917 some systems use these upper case variants.
4919 25. If more than two messages were waiting for the same host, and a transport
4920 filter was specified for the transport, Exim sent two messages over the
4921 same TCP/IP connection, and then failed with "socket operation on non-
4922 socket" when it tried to send the third.
4924 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4926 27. The extern definition of crypt16() in expand.c was not being excluded when
4927 the OS had its own crypt16() function.
4929 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4930 as a preferred synonym for return_size_limit, both as an option and as an
4933 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4935 30. Changed the default configuration ACL to relax the local part checking rule
4936 for addresses that are not in any local domains. For these addresses,
4937 slashes and pipe symbols are allowed within local parts, but the sequence
4938 /../ is explicitly forbidden.
4940 31. SPA server authentication was not clearing the challenge buffer before
4943 32. log_message in a "warn" ACL statement was writing to the reject log as
4944 well as to the main log, which contradicts the documentation and doesn't
4945 seem right (because no rejection is happening). So I have stopped it.
4947 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4948 However, I am unable to do any testing of this.
4950 34. Fixed an infelicity in the appendfile transport. When checking directories
4951 for a mailbox, to see if any needed to be created, it was accidentally
4952 using path names with one or more superfluous leading slashes; tracing
4953 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4955 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4956 contents are added to the log line that is written for every discarded
4957 recipient. (Previously a log_message setting was ignored.)
4959 36. The ${quote: operator now quotes the string if it is empty.
4961 37. The install script runs exim in order to find its version number. If for
4962 some reason other than non-existence or emptiness, which it checks, it
4963 could not run './exim', it was installing it with an empty version number,
4964 i.e. as "exim-". This error state is now caught, and the installation is
4967 38. An argument was missing from the function that creates an error message
4968 when Exim fails to connect to the socket for saslauthd authentication.
4969 This could cause Exim to crash, or give a corrupted message.
4971 39. Added isip, isip4, and isip6 to ${if conditions.
4973 40. The ACL variables $acl_xx are now saved with the message, and can be
4974 accessed later in routers, transports, and filters.
4976 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4977 strings in the file are not string-expanded.
4979 42. If a MAIL command specified a SIZE value that was too large to fit into an
4980 int variable, the check against message_size_limit failed. Such values are
4981 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4982 day this will have to be increased, but I don't think I want to be around
4983 when emails are that large.
4990 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4991 iconv() is not standard in FreeBSD.
4993 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4994 IPv6 enabled. The observed symptom was a segmentation fault on return from
4995 the function os_common_find_running_interfaces() in src/os.c.
4997 3. In the check_special_case() function in daemon.c I had used "errno" as an
4998 argument name, which causes warnings on some systems. This was basically a
4999 typo, since it was named "eno" in the comments!
5001 4. The code that waits for the clock to tick (at a resolution of some fraction
5002 of a second) so as to ensure message-id uniqueness was always waiting for
5003 at least one whole tick, when it could have waited for less. [This is
5004 almost certainly not relevant at current processor speeds, where it is
5005 unlikely to ever wait at all. But we try to future-proof.]
5007 5. The function that sleeps for a time interval that includes fractions of a
5008 second contained a race. It did not block SIGALRM between setting the
5009 timer, and suspending (a couple of lines later). If the interval was short
5010 and the sigsuspend() was delayed until after it had expired, the suspension
5011 never ended. On busy systems this could lead to processes getting stuck for
5014 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5015 process, before it forks any delivery processes. The open lookup caching
5016 mechanism meant that the open file or database connection was passed into
5017 the delivery process. The problem was that delivery processes always tidy
5018 up cached lookup data. This could cause a problem for the next delivery
5019 process started by the queue runner, because the external queue runner
5020 process does not know about the closure. So the next delivery process
5021 still has data in the lookup cache. In the case of a file lookup, there was
5022 no problem because closing a file descriptor in a subprocess doesn't affect
5023 the parent. However, if the lookup was caching a connection to a database,
5024 the connection was closed, and the second delivery process was likely to
5025 see errors such as "PGSQL: query failed: server closed the connection
5026 unexpectedly". The problem has been fixed by closing all cached lookups
5027 in a queue runner before running a delivery process.
5029 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5030 seem to have the "const" qualifier which it has on other OS. I've
5033 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5034 *of the same type* (client or server) with the same public name that an
5035 error should be diagnosed.
5037 9. When Exim looked up a host name for an IP address, but failed to find the
5038 original IP address when looking up the host name (a safety check), it
5039 output the message "<ip address> does not match any IP for NULL", which was
5040 confusing, to say the least. The bug was that the host name should have
5041 appeared instead of "NULL".
5043 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5044 user that is built into the binary, and the -C or -D options is used, root
5045 privilege is dropped before the configuration file is read. In addition,
5046 logging is switched to stderr instead of the normal log files. If the
5047 configuration then re-defines the Exim user, the unprivileged environment
5048 is probably not what is expected, so Exim logs a panic warning message (but
5051 However, if deliver_drop_privilege is set, the unprivileged state may well
5052 be exactly what is intended, so the warning has been cut out in that case,
5053 and Exim is allowed to try to write to its normal log files.
5059 1. smtp_return_error_details was not giving details for temporary sender
5060 or receiver verification errors.
5062 2. Diagnose a configuration error if two authenticators have the same public
5065 3. Exim used not to create the message log file for a message until the first
5066 delivery attempt. This could be confusing when incoming messages were held
5067 for policy or load reasons. The message log file is now created at the time
5068 the message is received, and an initial "Received" line is written to it.
5070 4. The automatically generated man page for command line options had a minor
5071 bug that caused no ill effects; however, a more serious problem was that
5072 the procedure for building the man page automatically didn't always
5073 operate. Consequently, release 4.20 contains an out-of-date version. This
5074 shouldn't happen again.
5076 5. When building Exim with embedded Perl support, the script that builds the
5077 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5078 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5080 6. The freeze_tell option was not being used for messages that were frozen on
5081 arrival, either by an ACL or by local_scan().
5083 7. Added the smtp_incomplete_transaction log selector.
5085 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5086 was accepting AUTH without a new EHLO.
5088 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5089 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5092 10. Logging of TCP/IP connections (when configured) now happens in the main
5093 daemon process instead of the child process, so that the TCP/IP connection
5094 count is more accurate (but it can never be perfect).
5096 11. The use of "drop" in a nested ACL was not being handled correctly in the
5097 outer ACL. Now, if condition failure induced by the nested "drop" causes
5098 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5099 or "require"), the connection is dropped.
5101 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5102 that yield "discard" can now be used with an "accept" or a "discard" verb,
5103 but an error is generated for any others (because I can't see a useful way
5104 to define what should happen).
5106 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5107 are now processed in the same way as lines in the Exim configuration file.
5108 In particular, continuation lines are supported.
5110 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5112 15. Added -ti meaning -t -i.
5114 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5115 domains, and warn by logging if others are found.
5117 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5118 returned from GIFCONF when Exim is trying to find the list of interfaces on
5119 a host. The code in os.c has been modified to copy each ifreq to an aligned
5120 structure in all cases.
5122 Also, in some cases, the returned ifreq's were being copied to a 'struct
5123 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5124 means the last couple of bytes of an IPv6 address could be chopped if the
5125 ifreq contained only a normal sockaddr (14 bytes storage).
5127 18. Named domain lists were not supported in the hosts_treat_as_local option.
5128 An entry such as +xxxx was not recognized, and was treated as a literal
5131 19. Ensure that header lines added by a DATA ACL are included in the reject log
5132 if the ACL subsequently rejects the message.
5134 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5135 MD5 (which is deprecated).
5137 21. When testing a filter file using -bf, Exim was writing a message when it
5138 took the sender from a "From " line in the message, but it was not doing so
5139 when it took $return_path from a Return-Path: header line. It now does.
5141 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5142 with a valid header line field name (a series of printing characters
5143 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5145 23. Changed "disc" in the source to "disk" to conform to the documentation and
5146 the book and for uniformity.
5148 24. Ignore Sendmail's -Ooption=value command line item.
5150 25. When execve() failed while trying to run a command in a pipe transport,
5151 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5152 could be confused with a return value of 69 from the command itself. This
5153 has been changed to 127, the value the shell returns if it is asked to run
5154 a non-existent command. The wording for the related log line suggests a
5155 non-existent command as the problem.
5157 26. If received_header_text expands to an empty string, do not add a Received:
5158 header line to the message. (Well, it adds a token one on the spool, but
5159 marks it "old" so that it doesn't get used or transmitted.)
5161 27. Installed eximstats 1.28 (addition of -nt option).
5163 28. There was no check for failure on the call to getsockname() in the daemon
5164 code. This can fail if there is a shortage of resources on the system, with
5165 ENOMEM, for example. A temporary error is now given on failure.
5167 29. Contrary to the C standard, it seems that in some environments, the
5168 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5169 program. Exim now does this explicitly; it affects the formatting of
5170 timestamps using strftime().
5172 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5173 complaints. I've now initialized all the variables, to avoid this.
5175 32. Header lines added by a system filter were not being "seen" during
5176 transport-time rewrites.
5178 33. The info_callback() function passed to OpenSSL is set up with type void
5179 (*)(SSL *, int, int), as described somewhere. However, when calling the
5180 function (actually a macro) that sets it up, the type void(*)() is
5181 expected. I've put in a cast to prevent warnings from picky compilers.
5183 34. If a DNS black list lookup found a CNAME record, but there were no A
5184 records associated with the domain it pointed at, Exim crashed.
5186 35. If a DNS black list lookup returned more than one A record, Exim ignored
5187 all but the first. It now scans all returned addresses if a particular IP
5188 value is being sought. In this situation, the contents of the
5189 $dnslist_value variable are a list of all the addresses, separated by a
5192 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5193 to accept a host name and all its aliases if the forward lookup for any of
5194 them yielded the IP address of the incoming connection. Now it accepts only
5195 those names whose forward lookup yields the correct IP address. Any other
5196 names are discarded. This closes a loophole whereby a rogue DNS
5197 administrator could create reverse DNS records to break through a
5198 wildcarded host restriction in an ACL.
5200 37. If a user filter or a system filter that ran in a subprocess used any of
5201 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5202 the wrong values were passed to the pipe command ($thisaddress had the
5203 value of $0, $0 had the value of $1, etc). This bug was introduced by
5204 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5206 38. Improved the line breaking for long SMTP error messages from ACLs.
5207 Previously, if there was no break point between 40 and 75 characters, Exim
5208 left the rest of the message alone. Two changes have been made: (a) I've
5209 reduced the minimum length to 35 characters; (b) if it can't find a break
5210 point between 35 and 75 characters, it looks ahead and uses the first one
5211 that it finds. This may give the occasional overlong line, but at least the
5212 remaining text gets split now.
5214 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5215 file descriptors might be low, and that setting 1000 would always raise it.
5216 It turns out that in some environments, the limit is already over 1000 and
5217 that lowering it causes trouble. So now Exim takes care not to decrease it.
5219 40. When delivering a message, the value of $return_path is set to $sender_
5220 address at the start of routing (routers may change the value). By an
5221 oversight, this default was not being set up when an address was tested by
5222 -bt or -bv, which affected the outcome if any router or filter referred to
5225 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5226 log only when "message" or "log_message" are set. However, if one of the
5227 conditions was an address verification, or a call to a nested ACL, the
5228 messages generated by the underlying test were being passed through. This
5229 no longer happens. The underlying message is available in $acl_verify_
5230 message for both "message" and "log_message" expansions, so it can be
5231 passed through if needed.
5233 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5234 new expansion $bh_ to give the encoded byte string without charset
5235 translation. Translation happens only if iconv() is available; HAVE_ICONV
5236 indicates this at build time. HEADERS_CHARSET gives the charset to
5237 translate to; headers_charset can change it in the configuration, and
5238 "headers charset" can change it in an individual filter file.
5240 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5241 that creates RFC 2047 encoded "words" labels them as that charset instead
5242 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5243 expansion operator; (ii) when Exim creates a From: line for a local
5244 message; (iii) when a header line is rewritten to include a "phrase" part.
5246 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5247 buggy, causing it to skip the first lines of messages whose message ID
5248 ended in 'D'. This would not have bitten before Exim release 4.14, because
5249 message IDs were unlikely to end in 'D' before then. The effect was to have
5250 incorrect size information for certain domains.
5252 45. #include "config.h" was missing at the start of the crypt16.c module. This
5253 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5256 46. If there was a timeout during a "random" callout check, Exim treated it as
5257 a failure of the random address, and carried on sending RSET and the real
5258 address. If the delay was just some slowness somewhere, the response to the
5259 original RCPT would be taken as a response to RSET and so on, causing
5260 mayhem of various kinds.
5262 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5263 when I implemented it. It didn't allow for the fact that some option values
5264 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5265 the right test for positive values.
5267 48. Domain names in DNS records are case-independent. Exim always looks them up
5268 in lower case. Some resolvers return domain names in exactly the case they
5269 appear in the zone file, that is, they may contain uppercase letters. Not
5270 all resolvers do this - some return always lower case. Exim was treating a
5271 change of case by a resolver as a change of domain, similar to a widening
5272 of a domain abbreviation. This triggered its re-routing code and so it was
5273 trying to route what was effectively the same domain again. This normally
5274 caused routing to fail (because the router wouldn't handle the domain
5275 twice). Now Exim checks for this case specially, and just changes the
5276 casing of the domain that it ultimately uses when it transmits the message
5279 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5282 50. If a filter generated a file delivery with a non-absolute name (possible if
5283 no home directory exists for the router), the forbid_file option was not
5286 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5287 the existing equality matching.
5289 52. Exim was using ints instead of ino_t variables in some places where it was
5290 dealing with inode numbers.
5292 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5293 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5294 called TMPDIR, and if it finds it is different, it changes its value.
5296 54. The smtp_printf() function is now made available to local_scan() so
5297 additional output lines can be written before returning. There is also an
5298 smtp_fflush() function to enable the detection of a dropped connection.
5299 The variables smtp_input and smtp_batched_input are exported to
5302 55. Changed the default runtime configuration: the message "Unknown user"
5303 has been removed from the ACL, and instead placed on the localuser router,
5304 using the cannot_route_message feature. This means that any verification
5305 failures that generate their own messages won't get overridden. Similarly,
5306 the "Unrouteable address" message that was in the ACL for unverifiable
5307 relay addresses has also been removed.
5309 56. Added hosts_avoid_esmtp to the smtp transport.
5311 57. The exicyclog script was not checking for the esoteric option
5312 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5313 will work only if exicyclog is run under the appropriate euid.
5315 58. Following a discussion on the list, the rules by which Exim recognises line
5316 endings on incoming messages have been changed. The -dropcr and drop_cr
5317 options are now no-ops, retained only for backwards compatibility. The
5318 following line terminators are recognized: LF CRLF CR. However, special
5319 processing applies to CR:
5321 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5322 nor a local message in the state where . is a terminator.
5324 (ii) If a bare CR is encountered in a header line, an extra space is added
5325 after the line terminator so as not to end the header. The reasoning
5326 behind this is that bare CRs in header lines are most likely either
5327 to be mistakes, or people trying to play silly games.
5329 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5330 was being incorrectly given as 18 bytes larger than it should have been.
5331 This is a VOB (very old bug).
5333 60. This may never have affected anything current, but just in case it has:
5334 When the local host is found other than at the start of a list of hosts,
5335 the local host, those with the same MX, and any that follow, are discarded.
5336 When the list in question was part of a longer list of hosts, the following
5337 hosts (not currently being processed) were also being discarded. This no
5338 longer happens. I'm not sure if this situation could ever has previously
5341 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5344 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5345 header. This is something that is recommended in a new Internet Draft, and
5346 is something that is documented as being done by Sendmail. There are two
5347 possible values. For messages generated by the autoreply transport, Exim
5350 Auto-Submitted: auto-replied
5352 whereas for all other generated messages (e.g. bounces) it adds
5354 Auto-Submitted: auto-generated
5356 63. The "personal" condition in filters now includes a test for the
5357 Auto-Submitted: header. If it contains the string "auto-" the message it
5358 not considered personal.
5360 64. Added rcpt_include_affixes as a generic transport option.
5362 65. Added queue_only_override (default true).
5364 66. Added the syslog_duplication option.
5366 67. If what should have been the first header line of a message consisted of
5367 a space followed by a colon, Exim was mis-interpreting it as a header line.
5368 It isn't of course - it is syntactically invalid and should therefore be
5369 treated as the start of the message body. The misbehaviour could have
5370 caused a number of strange effects, including loss of data in subsequent
5371 header lines, and spool format errors.
5373 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5374 client host had authenticated. This control can now be exercised by an ACL
5375 for more flexibility.
5377 69. By default, callouts do not happen when testing with -bh. There is now a
5378 variant, -bhc, which does actually run the callout code, including
5379 consulting and updating the callout cache.
5381 70. Added support for saslauthd authentication, courtesy of Alexander
5384 71. If statvfs() failed on the spool or log directories while checking their
5385 size for availability, Exim confusingly gave the error "space shortage".
5386 Furthermore, in debugging mode it crashed with a floating point exception.
5387 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5388 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5389 serious problem, Exim now writes to the main and panic logs when this
5390 happens, with details of the failure. It then refuses to accept the
5391 incoming message, giving the message "spool directory problem" or "log
5392 directory problem" with a 421 code for SMTP messages.
5394 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5395 0, 1, and 2 exist, because some OS complain for execs without them (see
5396 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5397 descriptors. However, the code omitted to check that the open succeeded,
5398 causing mysterious errors if for some reason the permissions on /dev/null
5399 got screwed. Now Exim writes a message to the main and panic logs, and
5400 bombs out if it can't open /dev/null.
5402 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5403 interact so that it is all more flexible. It is supposed to remain
5404 backwards compatible. Also added extra_local_interfaces.
5406 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5407 to bomb out with an assertion failure - to the client this appears as a
5408 connection drop. This problem occurs in the part of the code that was taken
5409 from the Samba project. Fortunately, the assertion is in a very simple
5410 function, so I have fixed this by reproducing the function inline in the
5411 one place where it is called, and arranging for authentication to fail
5412 instead of killing the process with assert().
5414 75. The SPA client code was not working when the server requested OEM rather
5415 than Unicode encoding.
5417 76. Added code to make require_files with a specific uid setting more usable in
5418 the case where statting the file as root fails - usually a non-root-mounted
5419 NFS file system. When this happens and the failure is EACCES, Exim now
5420 forks a subprocess and does the per-uid checking as the relevant uid.
5422 77. Added process_log_path.
5424 78. If log_file_path was not explicitly set, a setting of check_log_space or
5425 check_log_inodes was ignored.
5427 79. If a space check for the spool or log partitions fails, the incident is now
5428 logged. Of course, in the latter case the data may get lost...
5430 80. Added the %p formatting code to string_format() so that it can be used to
5431 print addresses in debug_print(). Adjusted all the address printing in the
5432 debugging in store.c to use %p rather than %d.
5434 81. There was a concern that a line of code in smtp_in.c could overflow a
5435 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5436 initially expressed, the concern was not well-founded, because trailing
5437 spaces are removed early. However, if the trailing spaces were followed by
5438 a NULL, they did not get removed, so the overflow was possible. Two fixes
5441 (a) I re-wrote the offending code in a cleaner fashion.
5442 (b) If an incoming SMTP command contains a NULL character, it is rejected
5445 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5446 runs initgroups(), so that if the Exim user is in any additional groups,
5447 they will be used during message reception.
5453 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob