1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
23 TF/01 Failure to get a lock on a hints database can have serious
24 consequences so log it to the panic log.
26 TF/02 Log LMTP confirmation messages in the same way as SMTP,
27 controlled using the smtp_confirmation log selector.
29 TF/03 Include the error message when we fail to unlink a spool file.
31 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
32 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
33 for maintaining out-of-tree patches for some time.
35 PP/01 Bugzilla 139: Documentation and portability issues.
36 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
37 Handle per-OS dynamic-module compilation flags.
39 PP/02 Let /dev/null have normal permissions.
40 The 4.73 fixes were a little too stringent and complained about the
41 permissions on /dev/null. Exempt it from some checks.
42 Reported by Andreas M. Kirchwitz.
44 PP/03 Report version information for many libraries, including
45 Exim version information for dynamically loaded libraries. Created
46 version.h, now support a version extension string for distributors
47 who patch heavily. Dynamic module ABI change.
49 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
50 privilege escalation vulnerability whereby the Exim run-time user
51 can cause root to append content of the attacker's choosing to
54 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
57 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
58 If dropping privileges for untrusted macros, we disabled normal logging
59 on the basis that it would fail; for the Exim run-time user, this is not
60 the case, and it resulted in successful deliveries going unlogged.
61 Fixed. Reported by Andreas Metzler.
67 PP/01 Date: & Message-Id: revert to normally being appended to a message,
68 only prepend for the Resent-* case. Fixes regression introduced in
69 Exim 4.70 by NM/22 for Bugzilla 607.
71 PP/02 Include check_rfc2047_length in configure.default because we're seeing
72 increasing numbers of administrators be bitten by this.
74 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
76 PP/03 Bugzilla 994: added openssl_options main configuration option.
78 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
80 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
82 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
84 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
85 without a peer certificate, leading to a segfault because of an
86 assumption that peers always have certificates. Be a little more
87 paranoid. Problem reported by Martin Tscholak.
89 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
90 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
91 NB: ClamAV planning to remove STREAM in "middle of 2010".
92 CL also introduces -bmalware, various -d+acl logging additions and
93 more caution in buffer sizes.
95 PP/09 Implemented reverse_ip expansion operator.
97 PP/10 Bugzilla 937: provide a "debug" ACL control.
99 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
101 PP/12 Bugzilla 973: Implement --version.
103 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
105 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
107 PP/15 Bugzilla 816: support multiple condition rules on Routers.
109 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
110 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
111 ignore trailing whitespace.
113 JJ/02 prevent non-panic DKIM error from being sent to paniclog
115 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
118 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
119 Notification from Dr Andrew Aitchison.
121 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
122 ExtendedDetectionInfo response format.
123 Notification from John Horne.
125 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
128 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
129 XSL and documented dependency on system catalogs, with examples of how
132 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
135 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
136 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
137 configuration file which is writeable by the Exim user or group.
139 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
140 of configuration files to cover files specified with the -C option if
141 they are going to be used with root privileges, not just the default
144 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
145 option (effectively making it always true).
147 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
148 files to be used while preserving root privileges.
150 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
151 that rogue child processes cannot use them.
153 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
154 run-time user, instead of root.
156 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
157 Exim run-time user without dropping privileges.
159 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
160 result string, instead of calling string_vformat() twice with the same
163 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
164 for other users. Others should always drop root privileges if they use
165 -C on the command line, even for a whitelisted configure file.
167 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
169 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
175 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
176 $data_path, and $header_path variables; fixed documentation bugs and
179 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
180 exipick to access non-standard spools, including the "frozen" queue
183 NM/01 Bugzilla 965: Support mysql stored procedures.
184 Patch from Alain Williams
186 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
188 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
189 Patch from Andreas Metzler
191 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
192 Patch from Kirill Miazine
194 NM/05 Bugzilla 671: Added umask to procmail example.
196 JJ/03 installed exipick 20100323.0, fixing doc bug
198 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
199 directory. Notification and patch from Dan Rosenberg.
201 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
203 TK/02 Improve log output when DKIM signing operation fails.
205 MH/01 Treat the transport option dkim_domain as a colon separated
206 list, not as a single string, and sign the message with each element,
207 omitting multiple occurences of the same signer.
209 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
210 Bugzilla 985, 986. Patch by Simon Arlott
212 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
213 Patch by Simon Arlott
215 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
216 MBX locking. Notification from Dan Rosenberg.
222 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
224 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
226 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
228 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
230 NM/04 Bugzilla 915: Flag broken perl installation during build.
236 TK/01 Added patch by Johannes Berg that expands the main option
237 "spamd_address" if it starts with a dollar sign.
239 TK/02 Write list of recipients to X-Envelope-Sender header when building
240 the mbox-format spool file for content scanning (suggested by Jakob
243 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
244 (http://www.dcc-servers.net/) support via dccifd. Activated by
245 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
247 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
248 by Mark Daniel Reidel <mr@df.eu>.
250 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
251 When building exim an external PCRE library is now needed -
252 PCRE is a system library on the majority of modern systems.
253 See entry on PCRE_LIBS in EDITME file.
255 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
256 conversation. Added nologin parameter to request.
257 Patch contributed by Kirill Miazine.
259 TF/01 Do not log submission mode rewrites if they do not change the address.
261 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
263 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
264 log files in place. Contributed by Roberto Lima.
266 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
268 TF/03 Bugzilla 615: When checking the local_parts router precondition
269 after a local_part_suffix or local_part_prefix option, Exim now
270 does not use the address's named list lookup cache, since this
271 contains cached lookups for the whole local part.
273 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
274 Robert Millan. Documentation is in experimental-spec.txt.
276 TF/04 Bugzilla 668: Fix parallel build (make -j).
278 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
280 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
281 Patch provided by Jan Srzednicki.
283 TF/05 Leading white space used to be stripped from $spam_report which
284 wrecked the formatting. Now it is preserved.
286 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
287 that they are available at delivery time.
289 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
291 TF/08 TLS error reporting now respects the incoming_interface and
292 incoming_port log selectors.
294 TF/09 Produce a more useful error message if an SMTP transport's hosts
295 setting expands to an empty string.
297 NM/06 Bugzilla 744: EXPN did not work under TLS.
298 Patch provided by Phil Pennock.
300 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
301 Patch provided by Richard Godbee.
303 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
304 acl_smtp_notquit, added index entry.
306 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
307 Patch provided by Eugene Bujak.
309 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
310 accept(). Patch provided by Maxim Dounin.
312 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
313 Patch provided by Phil Pennock.
315 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
317 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
318 Patch provided by Brad "anomie" Jorsch.
320 NM/14 Bugzilla 622: Added timeout setting to transport filter.
321 Patch provided by Dean Brooks.
323 TK/05 Add native DKIM support (does not depend on external libraries).
325 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
326 Patch provided by Graeme Fowler.
328 NM/16 Bugzilla 851: Documentation example syntax fix.
330 NM/17 Changed NOTICE file to remove references to embedded PCRE.
332 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
335 NM/19 Bugzilla 745: TLS version reporting.
336 Patch provided by Phil Pennock.
338 NM/20 Bugzilla 167: bool: condition support.
339 Patch provided by Phil Pennock.
341 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
342 clients. Patch provided by Phil Pennock.
344 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
345 Patch provided by Brad "anomie" Jorsch.
347 NM/23 Bugzilla 687: Fix misparses in eximstats.
348 Patch provided by Heiko Schlittermann.
350 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
351 Patch provided by Heiko Schlittermann.
353 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
354 plus update to original patch.
356 NM/26 Bugzilla 799: Documentation correction for ratelimit.
358 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
359 Patch provided by David Brownlee.
361 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
363 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
365 NM/30 Bugzilla 888: TLS documentation bugfixes.
367 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
369 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
370 Unlike the original bugzilla I have changed all shell scripts in src tree.
372 NM/33 Bugzilla 898: Transport filter timeout fix.
373 Patch by Todd Rinaldo.
375 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
376 Patch by Serge Demonchaux.
378 NM/35 Bugzilla 39: Base64 decode bug fixes.
379 Patch by Jakob Hirsch.
381 NM/36 Bugzilla 909: Correct connect() call in dcc code.
383 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
385 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
387 NM/39 Bugzilla 911: Fixed MakeLinks build script.
393 TK/01 Add preliminary DKIM support. Currently requires a forked version of
394 ALT-N's libdkim that I have put here:
395 http://duncanthrax.net/exim-experimental/
397 Note to Michael Haardt: I had to rename some vars in sieve.c. They
398 were called 'true' and it seems that C99 defines that as a reserved
399 keyword to be used with 'bool' variable types. That means you could
400 not include C99-style headers which use bools without triggering
401 build errors in sieve.c.
403 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
404 as mailq or other aliases. Changed the --help handling significantly
405 to do whats expected. exim_usage() emits usage/help information.
407 SC/01 Added the -bylocaldomain option to eximstats.
409 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
411 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
413 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
419 PH/01 Another patch from the Sieve maintainer.
421 PH/02 When an IPv6 address is converted to a string for single-key lookup
422 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
423 dots are used instead of colons so that keys in lsearch files need not
424 contain colons. This was done some time before quoting was made available
425 in lsearch files. However, iplsearch files do require colons in IPv6 keys
426 (notated using the quote facility) so as to distinguish them from IPv4
427 keys. This meant that lookups for IP addresses in host lists did not work
428 for iplsearch lookups.
430 This has been fixed by arranging for IPv6 addresses to be expressed with
431 colons if the lookup type is iplsearch. This is not incompatible, because
432 previously such lookups could never work.
434 The situation is now rather anomolous, since one *can* have colons in
435 ordinary lsearch keys. However, making the change in all cases is
436 incompatible and would probably break a number of configurations.
438 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
441 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
442 conversion specification without a maximum field width, thereby enabling
443 a rogue spamd server to cause a buffer overflow. While nobody in their
444 right mind would setup Exim to query an untrusted spamd server, an
445 attacker that gains access to a server running spamd could potentially
446 exploit this vulnerability to run arbitrary code as the Exim user.
448 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
449 $primary_hostname instead of what libspf2 thinks the hosts name is.
451 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
452 a directory entry by the name of the lookup key. Previously, if a
453 symlink pointed to a non-existing file or a file in a directory that
454 Exim lacked permissions to read, a lookup for a key matching that
455 symlink would fail. Now it is enough that a matching directory entry
456 exists, symlink or not. (Bugzilla 503.)
458 PH/03 The body_linecount and body_zerocount variables are now exported in the
461 PH/04 Added the $dnslist_matched variable.
463 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
464 This means they are set thereafter only if the connection becomes
467 PH/06 Added the client_condition to authenticators so that some can be skipped
468 by clients under certain conditions.
470 PH/07 The error message for a badly-placed control=no_multiline_responses left
471 "_responses" off the end of the name.
473 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
475 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
476 (without spaces) instead of just copying the configuration text.
478 PH/10 Added the /noupdate option to the ratelimit ACL condition.
480 PH/11 Added $max_received_linelength.
482 PH/12 Added +ignore_defer and +include_defer to host lists.
484 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
485 way in which PCRE > 7.0 is built.
487 PH/14 Implemented queue_only_load_latch.
489 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
490 MAIL command. The effect was to mangle the value on 64-bit systems.
492 PH/16 Another patch from the Sieve maintainer.
494 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
496 PH/18 If a system quota error occurred while trying to create the file for
497 a maildir delivery, the message "Mailbox is full" was not appended to the
498 bounce if the delivery eventually timed out. Change 4.67/27 below applied
499 only to a quota excession during the actual writing of the file.
501 PH/19 It seems that peer DN values may contain newlines (and other non-printing
502 characters?) which causes problems in log lines. The DN values are now
503 passed through string_printing() before being added to log lines.
505 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
506 and InterBase are left for another time.)
508 PH/21 Added message_body_newlines option.
510 PH/22 Guard against possible overflow in moan_check_errorcopy().
512 PH/23 POSIX allows open() to be a macro; guard against that.
514 PH/24 If the recipient of an error message contained an @ in the local part
515 (suitably quoted, of course), incorrect values were put in $domain and
516 $local_part during the evaluation of errors_copy.
522 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
523 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
526 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
527 issue a MAIL command.
529 PH/02 In an ACL statement such as
531 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
533 if a client was not listed at all, or was listed with a value other than
534 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
535 the condition was not true (as it should be), so access was not denied.
536 The bug was that the ! inversion was incorrectly passed on to the second
537 item. This has been fixed.
539 PH/03 Added additional dnslists conditions == and =& which are different from
540 = and & when the dns lookup returns more than one IP address.
542 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
543 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
545 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
546 FSYNC, which compiles an option called disable_fsync that allows for
547 bypassing fsync(). The documentation is heavily laced with warnings.
549 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
551 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
552 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
553 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
554 including adding "make clean"; (3) Added -fPIC when compiling the test
555 dynamically loaded module, to get rid of a warning.
557 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
558 message fails, move_frozen_messages = true and ignore_bounce_errors_after
559 = 0s. The bug is otherwise harmless.
561 PH/07 There was a bug in the dovecot authenticator such that the value of
562 $auth1 could be overwritten, and so not correctly preserved, after a
563 successful authentication. This usually meant that the value preserved by
564 the server_setid option was incorrect.
566 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
568 PH/09 Installed PCRE release 7.0.
570 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
571 run for batched SMTP input. It is now run at the start of every message
572 in the batch. While fixing this I discovered that the process information
573 (output by running exiwhat) was not always getting set for -bs and -bS
574 input. This is fixed, and it now also says "batched" for BSMTP.
576 PH/11 Added control=no_pipelining.
578 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
579 patch, slightly modified), and move the expansion of helo_data till after
580 the connection is made in the smtp transport (so it can use these
583 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
585 PH/14 Added log_selector = +pid.
587 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
589 PH/16 Add ${if forany and ${if forall.
591 PH/17 Added dsn_from option to vary the From: line in DSNs.
593 PH/18 Flush SMTP output before performing a callout, unless control =
594 no_callout_flush is set.
596 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
597 was true (the default) a successful delivery failed to delete the retry
598 item, thus causing premature timeout of the address. The bug is now
601 PH/20 Added hosts_avoid_pipelining to the smtp transport.
603 PH/21 Long custom messages for fakedefer and fakereject are now split up
604 into multiline reponses in the same way that messages for "deny" and
605 other ACL rejections are.
607 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
608 with slight modification.
610 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
611 draft, changing the syntax and factoring some duplicate code".
613 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
614 for deliveries of the second and subsequent messages over the same SMTP
617 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
618 ${reduce, with only minor "tidies".
620 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
622 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
623 expansion side effects.
625 PH/27 When a message times out after an over-quota error from an Exim-imposed
626 quota, the bounce message says "mailbox is full". This message was not
627 being given when it was a system quota that was exceeded. It now should
630 MH/03 Made $recipients available in local_scan(). local_scan() already has
631 better access to the recipient list through recipients_list[], but
632 $recipients can be useful in postmaster-provided expansion strings.
634 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
635 in the case of a MAIL command with additional options following the
636 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
637 were accidentally chopped off.
639 PH/29 SMTP synchronization checks are implemented when a command is read -
640 there is a check that no more input is waiting when there shouldn't be
641 any. However, for some commands, a delay in an ACL can mean that it is
642 some time before the response is written. In this time, more input might
643 arrive, invalidly. So now there are extra checks after an ACL has run for
644 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
645 pipelining has not been advertised.
647 PH/30 MH's patch to allow iscntrl() characters to be list separators.
649 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
650 returned in the SMTP response when smtp_return_error_details was false.
653 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
654 instead of the C I/O that was originally supplied, because problems were
657 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
658 Exim which did not show up earlier: it was assuming that a call to
659 SSL_CTX_set_info_callback() might give an error value. In fact, there is
660 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
661 was a macro that became an assignment, so it seemed to work. This has
662 changed to a proper function call with a void return, hence the compile
663 error. Exim's code has been fixed.
665 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
668 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
670 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
672 PH/37 If a message is not accepted after it has had an id assigned (e.g.
673 because it turns out to be too big or there is a timeout) there is no
674 "Completed" line in the log. When some messages of this type were
675 selected by exigrep, they were listed as "not completed". Others were
676 picked up by some special patterns. I have improved the selection
677 criteria to be more general.
679 PH/38 The host_find_failed option in the manualroute router can now be set
680 to "ignore", to completely ignore a host whose IP address cannot be
681 found. If all hosts are ignored, the behaviour is controlled by the new
682 host_all_ignored option.
684 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
685 homing or because of multiple MX records with /mx) generated more than
686 one IP address, and the following item turned out to be the local host,
687 all the secondary addresses of the first item were incorrectly removed
688 from the list, along with the local host and any following hosts (which
689 is what is supposed to happen).
691 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
692 whoever called Exim into the -H file. In the case of the daemon it was
693 behaving confusingly. When first started, it used values for whoever
694 started the daemon, but after a SIGHUP it used the Exim user (because it
695 calls itself on a restart). I have changed the code so that it now always
698 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
699 message are rejected with the same error (e.g. no authentication or bad
700 sender address), and a DATA command is nevertheless sent (as can happen
701 with PIPELINING or a stupid MUA), the error message that was given to the
702 RCPT commands is included in the rejection of the DATA command. This is
703 intended to be helpful for MUAs that show only the final error to their
706 PH/42 Another patch from the Sieve maintainer.
708 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
709 Eximstats - Fixed some broken HTML links and added missing column headers
711 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
712 columns for Rejects, Temp Rejects, Ham, and Spam rows.
714 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
716 PH/43 Yet another patch from the Sieve maintainer.
718 PH/44 I found a way to check for a TCP/IP connection going away before sending
719 the response to the final '.' that terminates a message, but only in the
720 case where the client has not sent further data following the '.'
721 (unfortunately, this is allowed). However, in many cases there won't be
722 any further data because there won't be any more messages to send. A call
723 to select() can be used: if it shows that the input is "ready", there is
724 either input waiting, or the socket has been closed. An attempt to read
725 the next input character can distinguish the two cases. Previously, Exim
726 would have sent an OK response which the client would never have see.
727 This could lead to message repetition. This fix should cure that, at
728 least in a lot of common cases.
730 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
731 advertised in response to EHLO.
737 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
738 fixed by 4.65/MH/01 (is this a record?) are fixed:
740 (i) An empty string was always treated as zero by the numeric comparison
741 operators. This behaviour has been restored.
743 (ii) It is documented that the numeric comparison operators always treat
744 their arguments as decimal numbers. This was broken in that numbers
745 starting with 0 were being interpreted as octal.
747 While fixing these problems I realized that there was another issue that
748 hadn't been noticed. Values of message_size_limit (both the global option
749 and the transport option) were treated as octal if they started with 0.
750 The documentation was vague. These values are now always treated as
751 decimal, and I will make that clear in the documentation.
757 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
758 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
761 MH/01 Don't check that the operands of numeric comparison operators are
762 integers when their expansion is in "skipping" mode (fixes bug
763 introduced by 4.64-PH/07).
765 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
766 child addresses, Exim now panics and dies. Previously, because the count
767 is held in a short int, deliveries were likely to be lost. As such a
768 large number of recipients for a single message is ridiculous
769 (performance will be very, very poor), I have chosen to impose a limit
770 rather than extend the field.
776 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
777 leftover -K file (the existence of which was triggered by #402).
778 While we were at it, introduced process PID as part of the -K
779 filename. This should rule out race conditions when creating
782 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
783 processing considerably. Previous code took too long for large mails,
784 triggering a timeout which in turn triggers #401.
786 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
787 in the DK code in transports.c. sendfile() is not really portable,
788 hence the _LINUX specificness.
790 TF/01 In the add_headers option to the mail command in an Exim filter,
791 there was a bug that Exim would claim a syntax error in any
792 header after the first one which had an odd number of characters
795 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
796 callout verification, Exim cached a "reject" for the entire domain. This
797 is correct for most verifications, but it is not correct for a recipient
798 verification with use_sender or use_postmaster set, because in that case
799 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
800 case of MAIL FROM:<> rejection from other early rejections (e.g.
801 rejection of HELO). When verifying a recipient using a non-null MAIL
802 address, the cache is ignored if it shows MAIL FROM:<> rejection.
803 Whatever the result of the callout, the value of the domain cache is
804 left unchanged (for any other kind of callout, getting as far as trying
805 RCPT means that the domain itself is ok).
807 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
810 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
811 manifest itself as EPIPE rather than ECONNECT. When tidying away a
812 session, the daemon ignores ECONNECT errors and logs others; it now
813 ignores EPIPE as well.
815 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
816 (quoted-printable decoding).
818 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
819 later the small subsequent patch to fix an introduced bug.
821 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
823 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
825 PH/08 An error is now given if message_size_limit is specified negative.
827 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
828 to be given (somewhat) arbitrary names.
830 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
833 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
834 miscellaneous code fixes
836 PH/10 Added the log_reject_target ACL modifier to specify where to log
839 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
840 hostname. This is wrong, because it relates to the incoming message (and
841 probably the interface on which it is arriving) and not to the outgoing
842 callout (which could be using a different interface). This has been
843 changed to use the value of the helo_data option from the smtp transport
844 instead - this is what is used when a message is actually being sent. If
845 there is no remote transport (possible with a router that sets up host
846 addresses), $smtp_active_hostname is used.
848 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
849 tweaks were necessary in order to get it to work (see also 21 below):
850 (a) The code assumed that strncpy() returns a negative number on buffer
851 overflow, which isn't the case. Replaced with Exim's string_format()
853 (b) There were several signed/unsigned issues. I just did the minimum
854 hacking in of casts. There is scope for a larger refactoring.
855 (c) The code used strcasecmp() which is not a standard C function.
856 Replaced with Exim's strcmpic() function.
857 (d) The code set only $1; it now sets $auth1 as well.
858 (e) A simple test gave the error "authentication client didn't specify
859 service in request". It would seem that Dovecot has changed its
860 interface. Fortunately there's a specification; I followed it and
861 changed what the client sends and it appears to be working now.
863 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
866 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
867 address A is aliased to B and C, where B exists and C does not. Without
868 -v the output is "A verified" because verification stops after a
869 successful redirection if more than one address is generated. However,
870 with -v the child addresses are also verified. Exim was outputting "A
871 failed to verify" and then showing the successful verification for C,
872 with its parentage. It now outputs "B failed to verify", showing B's
873 parentage before showing the successful verification of C.
875 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
876 look up a TXT record in a specific list after matching in a combined
879 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
880 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
881 they consult the DNS. I had assumed they would set it the way they
882 wanted; and indeed my experiments on Linux seem to show that in some
883 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
884 To be on the safe side, however, I have now made the interface to
885 host_find_byname() similar to host_find_bydns(), with an argument
886 containing the DNS resolver options. The host_find_byname() function now
887 sets these options at its start, just as host_find_bydns() does. The smtp
888 transport options dns_qualify_single and dns_search_parents are passed to
889 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
890 of host_find_byname() use the default settings of RES_DEFNAMES
891 (qualify_single) but not RES_DNSRCH (search_parents).
893 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
894 spool_read_header() do less string testing, by means of a preliminary
895 switch on the second character of optional "-foo" lines. (This is
896 overdue, caused by the large number of possibilities that now exist.
897 Originally there were few.) While I was there, I also converted the
898 str(n)cmp tests so they don't re-test the leading "-" and the first
899 character, in the hope this might squeeze out yet more improvement.
901 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
902 flag allowing group syntax was set by the header_syntax check but not
903 turned off, possible causing trouble later; (2) The flag was not being
904 set at all for the header_verify test, causing "group"-style headers to
905 be rejected. I have now set it in this case, and also caused header_
906 verify to ignore an empty address taken from a group. While doing this, I
907 came across some other cases where the code for allowing group syntax
908 while scanning a header line wasn't quite right (mostly, not resetting
909 the flag correctly in the right place). These bugs could have caused
910 trouble for malformed header lines. I hope it is now all correct.
912 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
913 with the "reply" argument non-NULL. The code, however (which originally
914 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
915 but it didn't always do it. This confused somebody who was copying the
916 code for some other use. I have removed all the tests.
918 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
919 feature that was used to support insecure browsers during the U.S. crypto
920 embargo. It requires special client support, and Exim is probably the
921 only MTA that supported it -- and would never use it because real RSA is
922 always available. This code has been removed, because it had the bad
923 effect of slowing Exim down by computing (never used) parameters for the
924 RSA_EXPORT functionality.
926 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
927 authenticator to fail if there's a tab character in the incoming data
928 (there should never be unless someone is messing about, as it's supposed
929 to be base64-encoded). Also added, on Timo's advice, the "secured" option
930 if the connection is using TLS or if the remote IP is the same as the
931 local IP, and the "valid-client-cert option" if a client certificate has
934 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
935 authenticators. This can be used for authorization after authentication
936 succeeds. (In the case of plaintext, it servers for both authentication
939 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
940 if any retry times were supplied.
942 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
943 connection, where there is no client IP address to check. In this
944 situation, the verify now always succeeds.
946 PH/25 Applied John Jetmore's -Mset patch.
948 PH/26 Added -bem to be like -Mset, but loading a message from a file.
950 PH/27 In a string expansion for a processed (not raw) header when multiple
951 headers of the same name were present, leading whitespace was being
952 removed from all of them, but trailing whitespace was being removed only
953 from the last one. Now trailing whitespace is removed from each header
954 before concatenation. Completely empty headers in a concatenation (as
957 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
958 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
960 PH/29 [Removed. This was a change that I later backed out, and forgot to
961 correct the ChangeLog entry (that I had efficiently created) before
962 committing the later change.]
964 PH/30 Exim was sometimes attempting to deliver messages that had suffered
965 address errors (4xx response to RCPT) over the same connection as other
966 messages routed to the same hosts. Such deliveries are always "forced",
967 so retry times are not inspected. This resulted in far too many retries
968 for the affected addresses. The effect occurred only when there were more
969 hosts than the hosts_max_try setting in the smtp transport when it had
970 the 4xx errors. Those hosts that it had tried were not added to the list
971 of hosts for which the message was waiting, so if all were tried, there
972 was no problem. Two fixes have been applied:
974 (i) If there are any address or message errors in an SMTP delivery, none
975 of the hosts (tried or untried) are now added to the list of hosts
976 for which the message is waiting, so the message should not be a
977 candidate for sending over the same connection that was used for a
978 successful delivery of some other message. This seems entirely
979 reasonable: after all the message is NOT "waiting for some host".
980 This is so "obvious" that I'm not sure why it wasn't done
981 previously. Hope I haven't missed anything, but it can't do any
982 harm, as the worst effect is to miss an optimization.
984 (ii) If, despite (i), such a delivery is accidentally attempted, the
985 routing retry time is respected, so at least it doesn't keep
986 hammering the server.
988 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
989 in ${readsocket because some servers need this prod.
991 PH/32 Added some extra debug output when updating a wait-xxx database.
993 PH/33 The hint "could be header name not terminated by colon", which has been
994 given for certain expansion errors for a long time, was not being given
995 for the ${if def:h_colon_omitted{... case.
997 PH/34 The spec says: "With one important exception, whenever a domain list is
998 being scanned, $domain contains the subject domain." There was at least
999 one case where this was not true.
1001 PH/35 The error "getsockname() failed: connection reset by peer" was being
1002 written to the panic log as well as the main log, but it isn't really
1003 panic-worthy as it just means the connection died rather early on. I have
1004 removed the panic log writing for the ECONNRESET error when getsockname()
1007 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1008 runs only) independently of the message's sender address. This meant
1009 that, if the 4xx error was in fact related to the sender, a different
1010 message to the same recipient with a different sender could confuse
1011 things. In particualar, this can happen when sending to a greylisting
1012 server, but other circumstances could also provoke similar problems.
1013 I have changed the default so that the retry time for these errors is now
1014 based a combination of the sender and recipient addresses. This change
1015 can be overridden by setting address_retry_include_sender=false in the
1018 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1019 remote server are returned as part of bounce messages. This was not
1020 happening for LMTP over a pipe (the lmtp transport), but now it is the
1021 same for both kinds of LMTP.
1023 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1024 in header lines that were in fact CNAMEs. This is no longer the case.
1026 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1027 and queue runs started by the daemon processed all messages. This has
1028 been fixed so that -R and -S can now usefully be given with -q<time>.
1030 PH/40 Import PCRE release 6.7 (fixes some bugs).
1032 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1034 PH/42 Give an error if -q is specified more than once.
1036 PH/43 Renamed the variables $interface_address and $interface_port as
1037 $received_ip_address and $received_port, to make it clear that these
1038 values apply to message reception, and not to the outgoing interface when
1039 a message is delivered. (The old names remain recognized, of course.)
1041 PH/44 There was no timeout on the connect() call when using a Unix domain
1042 socket in the ${readsocket expansion. There now is.
1044 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1045 be meaningful with "accept".
1047 SC/01 Eximstats V1.43
1048 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1050 SC/02 Eximstats V1.44
1051 Use a glob alias rather than an array ref in the generated
1052 parser. This improves both readability and performance.
1054 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1055 Collect SpamAssassin and rejection statistics.
1056 Don't display local sender or destination tables unless
1057 there is data to show.
1058 Added average volumes into the top table text output.
1060 SC/04 Eximstats V1.46
1061 Collect data on the number of addresses (recipients)
1062 as well as the number of messages.
1064 SC/05 Eximstats V1.47
1065 Added 'Message too big' to the list of mail rejection
1066 reasons (thanks to Marco Gaiarin).
1068 SC/06 Eximstats V1.48
1069 Mainlog lines which have GMT offsets and are too short to
1070 have a flag are now skipped.
1072 SC/07 Eximstats V1.49 (Alain Williams)
1073 Added the -emptyok flag.
1075 SC/08 Eximstats V1.50
1076 Fixes for obtaining the IP address from reject messages.
1078 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1079 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1080 whitesspace changes from 4.64-PH/27
1082 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1085 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1086 are found, allow negative numbers in numeric criteria)
1088 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1090 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1091 to match changes made in 4.64-PH/43
1093 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1095 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1096 "domain" is actually an IP address, and force a failure. This locks out
1097 those revolvers/nameservers that support "A-for-A" lookups, in
1098 contravention of the specifications.
1100 PH/48 When a host name was looked up from an IP address, and the subsequent
1101 forward lookup of the name timed out, the host name was left in
1102 $sender_host_name, contrary to the specification.
1104 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1105 restricted to single-key lookups, Exim was not diagnosing an error if
1106 * or *@ was used with a query-style lookup.
1108 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1110 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1111 long ago, but noone interested enough thought of it. Let's just say that
1112 the "1.1" means that there are some new functions that weren't there at
1113 some point in the past.
1115 PH/51 Error processing for expansion failure of helo_data from an smtp
1116 transport during callout processing was broken.
1118 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1119 tested/used via the -bh/-bhc/-bs options.
1121 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1122 bug, fixed in subsequent PCRE releases).
1124 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1125 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1127 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1133 SC/01 Use a glob alias rather than an array ref in eximstats generated
1134 parser. This improves both readability and performance.
1136 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1137 Don't display local sender or destination tables in eximstats unless
1138 there is data to show.
1139 Added average volumes into the eximstats top table text output.
1141 SC/03 Collect data on the number of addresses (recipients) as well
1142 as the number of messages in eximstats.
1144 TF/01 Correct an error in the documentation for the redirect router. Exim
1145 does (usually) call initgroups() when daemonizing.
1147 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1148 with consistent privilege compared to when running as a daemon.
1150 TF/03 Note in the spec that $authenticated_id is not set for local
1151 submissions from trusted users.
1153 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1154 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1156 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1157 by adding some example configuration directives to the default
1158 configuration file. A little bit of work is required to uncomment the
1159 directives and define how usernames and passwords are checked, but
1160 there is now a framework to start from.
1162 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1163 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1164 without this. I don't know how relevant this is to other LDAP libraries.
1166 PH/02 Add the verb name to the "unknown ACL verb" error.
1168 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1170 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1172 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1173 directory not expanded when it should be if an expanded home directory
1174 was set for the address (which is overridden by the transport).
1176 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1179 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1180 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1181 because it is too late at that time, and has no effect.
1183 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1184 security issue with \' (bugzilla #107). I could not use the
1185 PQescapeStringConn() function, because it needs a PGconn value as one of
1188 PH/08 When testing addresses using -bt, indicate those final addresses that
1189 are duplicates that would not cause an additional delivery. At least one
1190 person was confused, thinking that -bt output corresponded to deliveries.
1191 (Suppressing duplicates isn't a good idea as you lose the information
1192 about possibly different redirections that led to the duplicates.)
1194 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1195 systems where poll() doesn't work, in particular OS X.
1197 PH/10 Added more information to debugging output for retry time not reached.
1199 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1200 operations in malware.c.
1202 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1205 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1206 syslog_duplication set false, log lines that would normally be written
1207 both the the main log and to the reject log were not written to syslog at
1210 PH/14 In the default configuration, change the use of "message" in ACL warn
1211 statements to "add_header".
1213 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1214 not followed by a command (e.g. "seen endif").
1216 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1217 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1220 PH/17 Added extra conditions to the default value of delay_warning_condition
1224 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1225 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1226 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1229 The Auto-Submitted: and various List- headers are standardised, whereas I
1230 don't think Precedence: ever was.
1232 PH/18 Refactored debugging code in route_finduser() to show more information,
1233 in particular, the error code if getpwnam() issues one.
1235 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1236 This is apparently needed in addition to the PH/07 change above to avoid
1237 any possible encoding problems.
1239 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1240 but not after initializing Perl.
1242 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1243 output them only if debugging. By default they are written stderr,
1244 apparently, which is not desirable.
1246 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1249 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1252 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1254 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1255 authenticated or an ident call has been made. Suppress the default
1256 values for $authenticated_id and $authenticated_sender (but permit -oMai
1257 and -oMas) when testing with -bh.
1259 PH/24 Re-jigged the order of the tests in the default configuration so that the
1260 tests for valid domains and recipients precede the DNS black list and CSA
1261 tests, on the grounds that those ones are more expensive.
1263 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1264 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1265 This bug exists in every version of Exim that I still have, right back to
1268 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1269 However, an attempt to turn on case-sensitivity in a regex key by
1270 including (?-i) didn't work because the subject string was already
1271 lowercased, and the effects were non-intuitive. It turns out that a
1272 one-line patch can be used to allow (?-i) to work as expected.
1278 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1279 other effects) broke the use of negated acl sub-conditions.
1281 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1284 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1285 "Deny" causes Exim to reject the incoming connection with a 554 error.
1286 Unfortunately, if there is a major crisis, such as a disk failure,
1287 tcp-wrappers gives "deny", whereas what one would like would be some
1288 kind of temporary error. A kludge has been added to help with this.
1289 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1290 554 error is used if errno is still zero or contains ENOENT (which occurs
1291 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1294 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1296 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1297 errors. Otherwise a message that provokes a temporary error (when other
1298 messages do not) can cause a whole host to time out.
1300 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1301 addresses were routed directly to files or pipes from a redirect router.
1302 File deliveries just didn't batch; pipe deliveries might have suffered
1305 PH/06 A failure to get a lock for a hints database would erroneously always say
1306 "Failed to get write lock", even when it was really a read lock.
1308 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1309 of 0600. This has been changed to use the value of the lockfile_mode
1310 option (which defaults to 0600).
1312 PH/08 Applied small patch from the Sieve maintainer.
1314 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1315 folder from quota calculations, a direct delivery into this folder messed
1316 up the contents of the maildirsize file. This was because the regex was
1317 used only to exclude .Trash (or whatever) when the size of the mailbox
1318 was calculated. There was no check that a delivery was happening into an
1319 excluded directory. This bug has been fixed by ignoring all quota
1320 processing for deliveries into excluded directories.
1322 PH/10 Added the maildirfolder_create_regex option to appendfile.
1328 PH/01 The code for finding all the local interface addresses on a FreeBSD
1329 system running IPv6 was broken. This may well have applied to all BSD
1330 systems, as well as to others that have similar system calls. The broken
1331 code found IPv4 interfaces correctly, but gave incorrect values for the
1332 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1333 that it would not match correctly against @[] and not recognize the IPv6
1336 PH/02 The ipliteral router was not recognizing addresses of the form user@
1337 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1339 PH/03 Added disable_ipv6.
1341 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1342 decoded form, because it is most often used to construct To: headers
1343 lines in autoreplies, and the decoded form may well be syntactically
1344 invalid. However, $reply_address has leading white space removed, and all
1345 newlines turned into spaces so that the autoreply transport does not
1348 PH/05 If group was specified without a user on a router, and no group or user
1349 was specified on a transport, the group from the router was ignored.
1351 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1352 for visible compile-time settings that can be used to change these
1353 numbers, for those that want even more. Backwards compatibility with old
1354 spool files has been maintained. However, going back to a previous Exim
1355 release will lost any variables that are in spool files.
1357 PH/07 Two small changes when running in the test harness: increase delay when
1358 passing a TCP/IP connection to a new process, in case the original
1359 process has to generate a bounce, and remove special handling of
1360 127.0.0.2 (sic), which is no longer necessary.
1362 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1363 be the same on different OS.
1365 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1368 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1369 whether --show-vars was specified or not
1371 JJ/02 exipick: Added support for new ACL variable spool format introduced
1374 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1375 syntactically invalid From: or Reply-to: line, and a filter used this to
1376 generate an autoreply, and therefore failed to obtain an address for the
1377 autoreply, Exim could try to deliver to a non-existent relative file
1378 name, causing unrelated and misleading errors. What now happens is that
1379 it logs this as a hard delivery error, but does not attempt to create a
1382 PH/11 The exinext utility has a -C option for testing purposes, but although
1383 the given file was scanned by exinext itself; it wasn't being passed on
1384 when Exim was called.
1386 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1387 an end-of-file indication when reading a command response.
1389 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1390 compiled. In many other places in Exim, IPv6 addresses are always
1391 recognized, so I have changed this. It also means that IPv4 domain
1392 literals of the form [IPV4:n.n.n.n] are now always recognized.
1394 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1395 used if the router is not running as root, for example, when verifying at
1396 ACL time, or when using -bh. The debugging output from this situation was
1397 non-existent - all you got was a failure to exec. I have made two
1400 (a) Failures to set uid/gid, the current directory, or a process leader
1401 in a subprocess such as that created by queryprogram now generate
1402 suitable debugging ouput when -d is set.
1404 (b) The queryprogram router detects when it is not running as root,
1405 outputs suitable debugging information if -d is set, and then runs
1406 the subprocess without attempting to change uid/gid.
1408 PH/15 Minor change to Makefile for building test_host (undocumented testing
1411 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1412 additional section of a DNS packet that returns MX or SRV records.
1413 Instead, it always explicitly searches for A/AAAA records. This avoids
1414 major problems that occur when a DNS server includes only records of one
1415 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1416 fixed another bug: if SRV records were looked up and the corresponding
1417 address records were *not* found in the additional section, the port
1418 values from the SRV records were lost.
1420 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1421 using the correct key (the original address) when searching the retry
1422 rules in order to find which one to use for generating the retry hint.
1424 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1425 adding the default one. Similarly, if it contains a Reply-To: header, the
1426 errors_reply_to option, if set, is not used.
1428 PH/19 When calculating a retry time, Exim used to measure the "time since
1429 failure" by looking at the "first failed" field in the retry record. Now
1430 it does not use this if it is later than than the arrival time of the
1431 message. Instead it uses the arrival time. This makes for better
1432 behaviour in cases where some deliveries succeed, thus re-setting the
1433 "first failed" field. An example is a quota failure for a huge message
1434 when small messages continue to be delivered. Without this change, the
1435 "time since failure" will always be short, possible causing more frequent
1436 delivery attempts for the huge message than are intended.
1437 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1439 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1440 $1, $2, $3) because the numerical variables can be reset during some
1441 expansion items (e.g. "match"), thereby losing the authentication data.
1443 PH/21 Make -bV show the size of off_t variables so that the test suite can
1444 decide whether to run tests for quotas > 2G.
1446 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1447 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1448 filecount value is greater than 2G or if a quota value is greater than 2G
1449 on a system where the size of off_t is not greater than 4, a panic error
1452 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1453 never match. The debug and -bh output now contains an explicit error
1454 message indicating a malformed IPv4 address or mask.
1456 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1457 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1458 PH/23 above applies.
1460 PH/25 Do not write to syslog when running in the test harness. The only
1461 occasion when this arises is a failure to open the main or panic logs
1462 (for which there is an explicit test).
1464 PH/26 Added the /no_tell option to "control=freeze".
1466 PH/27 If a host name lookup failed very early in a connection, for example, if
1467 the IP address matched host_lookup and the reverse lookup yielded a name
1468 that did not have a forward lookup, an error message of the form "no IP
1469 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1470 could be logged. Now it outputs the IP address instead of "NULL".
1472 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1473 allows the sender and the authenticated sender to be set when
1474 submitting a message from within Exim. Since child_open_exim() is
1475 documented for local_scan(), the new function should be too.
1477 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1478 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1479 results in an empty string is now treated as unset.
1481 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1483 PH/31 Added sender_verify_fail logging option.
1485 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1486 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1487 tidied the source and removed it altogether.
1489 PH/33 When a queue run was abandoned because the load average was too high, a
1490 log line was always written; now it is written only if the queue_run log
1491 selector is set. In addition, the log line for abandonment now contains
1492 information about the queue run such as the pid. This is always present
1493 in "start" and "stop" lines but was omitted from the "abandon" line.
1495 PH/34 Omit spaces between a header name and the colon in the error message that
1496 is given when verify = headers_syntax fails (if there are lots of them,
1497 the message gets confusing).
1499 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1500 names, as there are now some PTR records that contain slashes. This check
1501 is only to protect against broken name servers that fall over on strange
1502 characters, so the fact that it applies to all lookups doesn't matter.
1504 PH/36 Now that the new test suite is complete, we can remove some of the
1505 special code in Exim that was needed for the old test suite. For example,
1506 sorting DNS records because real resolvers return them in an arbitrary
1507 order. The new test suite's fake resolver always returns records in the
1510 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1511 bounces) except when queue_only is set, to avoid logging races between
1512 the different processes.
1514 PH/38 Panic-die if .include specifies a non-absolute path.
1516 PH/39 A tweak to the "H" retry rule from its user.
1518 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1519 a label. They prevented compilation on older perls.
1521 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1522 a warning to be raised on newish perls.
1524 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1525 on queue. Changes to match documented behaviour of showing count of
1526 messages matching specified criteria.
1528 PH/40 Changed the default ident timeout from 30s to 5s.
1530 PH/41 Added support for the use of login_cap features, on those BSD systems
1531 that have them, for controlling the resources used by pipe deliveries.
1533 PH/42 The content-scanning code uses fopen() to create files in which to put
1534 message data. Previously it was not paying any attention to the mode of
1535 the files. Exim runs with umask(0) because the rest of the code creates
1536 files with open(), and sets the required mode explicitly. Thus, these
1537 files were ending up world-writeable. This was not a big issue, because,
1538 being within the spool directory, they were not world-accessible. I have
1539 created a function called modefopen, which takes an additional mode
1540 argument. It sets umask(777), creates the file, chmods it to the required
1541 mode, then resets the umask. All the relevant calls to fopen() in the
1542 content scanning code have been changed to use this function.
1544 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1545 to 24 hours. This avoids potential overflow problems when processing G
1546 and H retry rules. I suspect nobody ever tinkers with this value.
1548 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1550 PH/45 When the plaintext authenticator is running as a client, the server's
1551 challenges are checked to ensure they are valid base64 strings. By
1552 default, the authentication attempt is cancelled if an invalid string is
1553 received. Setting client_ignore_invalid_base64 true ignores these errors.
1554 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1555 they are received. Thus, the responses can be made to depend on the
1556 challenges. If an invalid string is ignored, an empty string is placed in
1559 PH/46 Messages that are created by the autoreply transport now contains a
1560 References: header, in accordance with RFCs 2822 and 3834.
1562 PH/47 Added authenticated_sender_force to the smtp transport.
1564 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1566 PH/49 Installed latest patch from the Sieve maintainer.
1568 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1569 was also set, the appendfile transport was unnecessarily scanning a
1570 directory of message files (e.g. for maildir delivery) to find the count
1571 of files (along with the size), even though it did not need this
1572 information. It now does the scan only if it needs to find either the
1573 size of the count of files.
1575 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1577 PH/52 Two bugs concerned with error handling when the smtp transport is
1580 (i) Exim was not creating retry information for temporary errors given
1581 for individual recipients after the DATA command when the smtp transport
1582 was used in LMTP mode. This meant that they could be retried too
1583 frequently, and not timed out correctly.
1585 (ii) Exim was setting the flag that allows error details to be returned
1586 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1587 recipients that were returned after the DATA command.
1589 PH/53 This is related to PH/52, but is more general: for any failing address,
1590 when detailed error information was permitted to be returned to the
1591 sender, but the error was temporary, then after the final timeout, only
1592 "retry timeout exceeded" was returned. Now it returns the full error as
1593 well as "retry timeout exceeded".
1595 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1596 do this, and (what is worse) MTAs that accept it.
1598 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1599 will now be deprecated.
1601 PH/56 New os.c-cygwin from the Cygwin maintainer.
1603 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1604 formats (previously only available in exim formats via -bpr, -bpru,
1605 and -bpra. Now also available in native and exiqgrep formats)
1607 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1608 with very large, slow to parse queues
1610 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1612 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1614 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1615 responses to authentication challenges, though it was showing the
1616 challenges; (ii) I've removed the CR characters from the debug output for
1619 PH/58 Allow for the insertion of a newline as well as a space when a string
1620 is turned into more than one encoded-word during RFC 2047 encoding. The
1621 Sieve code now uses this.
1623 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1624 data_4xx, lost_connection, tls_required.
1626 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1627 message was being sent as an SMTP response.
1629 PH/61 Add -l and -k options to exicyclog.
1631 PH/62 When verifying, if an address was redirected to one new address, so that
1632 verification continued, and the new address failed or deferred after
1633 having set something in $address_data, the value of $address_data was not
1634 passed back to the ACL. This was different to the case when no
1635 redirection occurred. The value is now passed back in both cases.
1637 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1638 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1639 use login_cap.h, so on its own it isn't the distinguishing feature. The
1640 new name refers directly to the setclassresources() function.
1642 PH/65 Added configuration files for NetBSD3.
1644 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1646 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1647 is preferred over IPv4.
1649 PH/68 The bounce_return_message and bounce_return_body options were not being
1650 honoured for bounces generated during the reception of non-SMTP messages.
1651 In particular, this applied to messages rejected by the ACL. This bug has
1652 been fixed. However, if bounce_return_message is true and bounce_return_
1653 body is false, the headers that are returned for a non-SMTP message
1654 include only those that have been read before the error was detected.
1655 (In the case of an ACL rejection, they have all been read.)
1657 PH/69 The HTML version of the specification is now built in a directory called
1658 spec_html instead of spec.html, because the latter looks like a path with
1659 a MIME-type, and this confuses some software.
1661 PH/70 Catch two compiler warnings in sieve.c.
1663 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1664 function verify_get_ident() calls ip_connect() to connect a socket, but
1665 if the "connect()" function timed out, ip_connect() used to close the
1666 socket. However, verify_get_ident() also closes the socket later, and in
1667 between Exim writes to the log, which may get opened at this point. When
1668 the socket was closed in ip_connect(), the log could get the same file
1669 descriptor number as the socket. This naturally causes chaos. The fix is
1670 not to close the socket in ip_connect(); the socket should be closed by
1671 the function that creates it. There was only one place in the code where
1672 this was missing, in the iplookup router, which I don't think anybody now
1673 uses, but I've fixed it anyway.
1675 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1676 well as to direct DNS lookups. Otherwise the handling of names in host
1677 lists is inconsistent and therefore confusing.
1683 PH/01 Two changes to the default runtime configuration:
1685 (1) Move the checks for relay_from_hosts and authenticated clients from
1686 after to before the (commented out) DNS black list checks.
1688 (2) Add control=submission to the relay_from_hosts and authenticated
1689 clients checks, on the grounds that messages accepted by these
1690 statements are most likely to be submissions.
1692 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1694 (1) Generate an error if the third argument for the ${prvs expansion is
1697 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1700 (3) Reset the variables that are obtained from the first argument of
1701 ${prvscheck and used in the second argument before leaving the code,
1702 because their memory is reclaimed, so using them afterwards may do
1705 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1706 one (it's much easier than Tom thought :-).
1708 (5) Because of (4), we can now allow for the use of $prvscheck_result
1709 inside the third argument.
1711 PH/03 For some reason, the default setting of PATH when running a command from
1712 a pipe transport was just "/usr/bin". I have changed it to
1715 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1716 anything to be listed in the output from -bV.
1718 PH/05 When a filter generated an autoreply, the entire To: header line was
1719 quoted in the delivery log line, like this:
1721 => >A.N.Other <ano@some.domain> <original@ddress> ...
1723 This has been changed so that it extracts the operative address. There
1724 may be more than one such address. If so, they are comma-separated, like
1727 => >ano@some.domain,ona@other.domain <original@ddress> ...
1729 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1730 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1731 not being looked up in the rDNS to get a host name, Exim was showing the
1732 IP address twice in Received: lines, even though the IP addresses were
1733 identical. For example:
1735 Received: from [1.2.3.4] (helo=[1.2.3.4])
1737 However, if the real host name was known, it was omitting the HELO data
1738 if it matched the actual IP address. This has been tidied up so that it
1739 doesn't show the same IP address twice.
1741 PH/07 When both +timestamp and +memory debugging was on, the value given by
1742 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1743 called by the debug printing, thereby overwriting the timestamp buffer.
1744 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1746 PH/08 When the original message was included in an autoreply transport, it
1747 always said "this is a copy of the message, including all the headers",
1748 even if body_only or headers_only was set. It now gives an appropriate
1751 PH/09 Applied a patch from the Sieve maintainer which:
1753 o fixes some comments
1754 o adds the (disabled) notify extension core
1755 o adds some debug output for the result of if/elsif tests
1756 o points to the current vacation draft in the documentation
1757 and documents the missing references header update
1761 o fixes a bug in processing the envelope test (when testing
1762 multiple envelope elements, the last element determinted the
1765 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1766 Electronic Mail") by including:
1768 Auto-submitted: auto-generated
1770 in the messages that it generates (bounce messages and others, such as
1771 warnings). In the case of bounce messages for non-SMTP mesages, there was
1772 also a typo: it was using "Auto_submitted" (underscore instead of
1773 hyphen). Since every message generated by Exim is necessarily in response
1774 to another message, thes have all been changed to:
1776 Auto-Submitted: auto-replied
1778 in accordance with these statements in the RFC:
1780 The auto-replied keyword:
1782 - SHOULD be used on messages sent in direct response to another
1783 message by an automatic process,
1785 - MUST NOT be used on manually-generated messages,
1787 - MAY be used on Delivery Status Notifications (DSNs) and Message
1788 Disposition Notifications (MDNs),
1790 - MUST NOT be used on messages generated by automatic or periodic
1791 processes, except for messages which are automatic responses to
1794 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1795 to the default Received: header definition.
1797 PH/12 Added log selector acl_warn_skipped (default on).
1799 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1800 variables because (a) they are in the wrong storage pool and (b) even if
1801 they were copied, it wouldn't work properly because of the caching.
1803 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1804 checking when decoding. Apparently there are clients that generate
1805 overlong encoded strings. Why am I not surprised?
1807 PH/15 If the first argument of "${if match_address" was not empty, but did not
1808 contain an "@" character, Exim crashed. Now it writes a panic log message
1809 and treats the condition as false.
1811 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1813 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1814 extension "envelope-auth". The code is finished and in agreement with
1815 other implementations, but there is no documentation so far and in fact,
1816 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1817 not changing the active code.
1819 Print executed "if" and "elsif" statements when debugging is used. This
1820 helps a great deal to understand what a filter does.
1822 Document more things not specified clearly in RFC3028. I had all this
1823 sorted out, when out of a sudden new issues came to my mind. Oops."
1825 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1828 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1829 canonical form (as documented). However, after a host name lookup from
1830 the IP address, check_host() was doing a simple string comparison with
1831 addresses acquired from the DNS when checking that the found name did
1832 have the original IP as one of its addresses. Since any found IPv6
1833 addresses are likely to be in abbreviated form, the comparison could
1834 fail. Luckily, there already exists a function for doing the comparison
1835 by converting both addresses to binary, so now that is used instead of
1836 the text comparison.
1838 PH/20 There was another similar case to PH/19, when a complete host name was
1839 given in a host list; looking up its IP address could give an abbreviated
1840 form, whereas the current host's name might or might not be abbreviated.
1841 The same fix has been applied.
1847 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1848 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1851 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1852 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1854 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1856 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1857 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1858 Also, for "Auto-Submitted", treat anything other than "no" as
1859 non-personal, in accordance with RFC 3834. (Previously it treated
1860 anything starting "auto-" as non-personal.)
1862 TF/01 The control=submission/name=... option had a problem with syntax
1863 errors if the name included a slash character. The /name= option
1864 now slurps the rest of the string, so it can include any characters
1865 but it must come last in the list of options (after /sender_retain
1868 PH/05 Some modifications to the interface to the fake nameserver for the new
1876 TK/01 Added the "success_on_redirect" address verification option. See
1877 NewStuff for rationale and an example.
1879 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1881 PH/02 Patch to exigrep to allow it to work on syslog lines.
1883 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1884 fread() to skip over the body file's header line, because in Cygwin the
1885 header line is locked and is inaccessible.
1887 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1888 co-exist for some time) to make it clear that it is the Exim ID that is
1889 referenced, not the Message-ID: header line.
1891 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1892 string_format() function, because snprintf() does not exist on all
1895 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1896 ${stat: expansion item.
1898 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1899 protocol synchronization error", to keep the pedants happy.
1901 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1902 well as for IRIX systems, when gcc is being used. See the host.c source
1905 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1907 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1910 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1911 transport and to the smtp transport in LMTP mode.
1913 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1915 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1916 run a filter in a subprocess. This could lead to confusion in subsequent
1917 lookups in the parent process. There should also be a search_tidyup() at
1918 the end of the subprocess.
1920 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1921 only if the host matched helo_try_verify_hosts, which caused the
1922 verification to occur when the EHLO/HELO command was issued. The ACL just
1923 tested the remembered result. Now, if a previous verification attempt has
1924 not happened, "verify = helo" does it there and then.
1926 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1928 TK/03 Fix log output including CR from clamd.
1930 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1931 exist provoked a memory error which could cause a segfault.
1933 PH/15 Installed PCRE 6.2
1935 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1937 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1938 of the problem. Specifically, suggested +O2 rather than +O1 for the
1941 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1943 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1944 record, the port was not correctly logged when the outgoing_port log
1945 selector was set (it logged the transort's default port).
1947 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1948 fallback_hosts, and "hosts" in the smtp transport.
1950 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1951 host errors such as "Connection refused".
1953 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1954 authentication with radiusclient 0.4.9:
1956 - Error returned from rc_read_config was caught wrongly
1957 - Username/password not passed on to radius server due to wrong length.
1959 The presumption is that some radiusclient API changes for 4.51/PH/17
1960 were not taken care of correctly. The code is still untested by me (my
1961 Linux distribution still has 0.3.2 of radiusclient), but it was
1962 contributed by a Radius user.
1964 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1965 expanding the "port" option of the smtp transport.
1967 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1968 while reading a MIME header. Thanks to Tom Hughes for a patch.
1970 PH/24 Include config.h inside local_scan.h so that configuration settings are
1973 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1974 that in an ACL for RCPT (for example), you can examine exactly what was
1977 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1978 commands, but it was not correctly comparing the address with the actual
1979 client host address. Thus, it would show the EHLO address in Received:
1980 header lines when this was not necessary.
1982 PH/27 Added the % operator to ${eval:}.
1984 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1985 it should be ignoring failures (because with -C, for example, it has lost
1986 privilege). It wasn't ignoring creation failures other than "already
1989 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1992 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1993 because an input line was too long, either on its own, or by virtue of
1994 too many continuations, the temporary file was not being removed, and the
1995 return code was incorrect.
1997 PH/31 Missing "BOOL" in function definition in filtertest.c.
1999 PH/32 Applied Sieve patches from the maintainer.
2001 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2003 PH/33 Added "verify = not_blind".
2005 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2006 Local/Makefile (with some defaults set). These are used in built scripts
2007 such as exicyclog, but they have never been used in the exim_install
2008 script (though there are many overriding facilities there). I have
2009 arranged that the exim_install script now takes note of these two
2012 PH/35 Installed configuration files for Dragonfly.
2014 PH/36 When a locally submitted message by a trusted user did not contain a
2015 From: header, and the sender address was obtained from -f or from an SMTP
2016 MAIL command, and the trusted user did not use -F to supply a sender
2017 name, $originator_name was incorrectly used when constructing a From:
2018 header. Furthermore, $originator_name was used for submission mode
2019 messages from external hosts without From: headers in a similar way,
2020 which is clearly wrong.
2022 PH/37 Added control=suppress_local_fixups.
2024 PH/38 When log_selector = +received_sender was set, and the addition of the
2025 sender made the log line's construction buffer exactly full, or one byte
2026 less than full, an overflow happened when the terminating "\n" was
2029 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2030 when the result of a list match is failure because a DNS lookup failed.
2032 PH/40 RM_COMMAND is now used in the building process.
2034 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2035 the "build-* directories that it finds.
2037 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2038 address in a domain literal was a prefix of an interface address.
2040 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2041 when verifying a sender address, unless rewrite_headers is false.
2043 PH/44 Wrote a long comment about why errors_to addresses are verified as
2044 recipients, not senders.
2046 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2047 the ratelimit ACL was added.
2049 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2051 PH/46 Added extra information about PostgreSQL errors to the error string.
2053 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2054 avoiding the need to install special zones in a real server. This is
2055 backwards compatible; if it can't find the fake resolver, it drops back.
2056 Thus, both old and new test suites can be run.
2058 TF/02 Added util/ratelimit.pl
2060 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2061 clock is set back in time.
2063 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2064 Candler <B.Candler@pobox.com>.
2066 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2067 OK for addresses that are the result of redirections.
2069 PH/48 A number of further additions for the benefit of the new test suite,
2070 including a fake gethostbyname() that interfaces to the fake DNS resolver
2073 TF/06 The fix for widen_domains has also been applied to qualify_single and
2074 search_parents which are the other dnslookup options that can cause
2077 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2080 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2082 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2083 service" if it encountered an MX record with an empty target hostname.
2084 The message is now "an MX or SRV record indicated no SMTP service".
2086 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2087 if the DNS of the sending site is misconfigured. This is quite a
2088 common situation. This change restores the behaviour of treating a
2089 helo verification defer as a failure.
2091 PH/51 If self=fail was set on a router, the bounce message did not include the
2092 actual error message.
2098 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2100 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2101 command itself ended in error, the underlying message about the transport
2102 filter timeout was being overwritten with the pipe command error. Now the
2103 underlying error message should be appended to the second error message.
2105 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2107 PH/02 Reduce the amount of output that "make" produces by default. Full output
2108 can still be requested.
2110 PH/03 The warning log line about a condition test deferring for a "warn" verb
2111 was being output only once per connection, rather than after each
2112 occurrence (because it was using the same function as for successful
2113 "warn" verbs). This seems wrong, so I have changed it.
2115 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2116 it should not have, which might have caused a crash in the right
2117 circumstances, but probably never did.
2119 PH/04 Installed a modified version of Tony Finch's patch to make submission
2120 mode fix the return path as well as the Sender: header line, and to
2121 add a /name= option so that you can make the user's friendly name appear
2124 TF/03 Added the control = fakedefer ACL modifier.
2126 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2127 Mark Lowes for thorough testing.
2129 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2131 TK/03 Merged latest SRS patch from Miles Wilton.
2133 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2134 with the definition in sysexits.h (which is #included earlier).
2135 Fortunately, Exim does not actually use EX_OK. The code used to try to
2136 preserve the sysexits.h value, by assumimg that macro definitions were
2137 scanned for macro replacements. I have been disabused of this notion,
2138 so now the code just undefines EX_OK before #including unistd.h.
2140 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2141 in the smtp transport. When a block could not be written in a single
2142 write() function, the timeout was being re-applied to each part-write.
2143 This seems wrong - if the receiver was accepting one byte at a time it
2144 would take for ever. The timeout is now adjusted when this happens. It
2145 doesn't have to be particularly precise.
2147 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2148 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2150 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2151 without a domain if the check to <postmaster@domain> fails.
2153 SC/01 Eximstats: added -xls and the ability to specify output files
2154 (patch written by Frank Heydlauf).
2156 SC/02 Eximstats: use FileHandles for outputing results.
2158 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2160 SC/04 Eximstats: fixed display of large numbers with -nvr option
2162 SC/05 Eximstats: fixed merging of reports with empty tables.
2164 SC/06 Eximstats: added the -include_original_destination flag
2166 SC/07 Eximstats: removed tabs and trailing whitespace.
2168 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2170 TK/06 MBOX spool code: Add real "From " MBOX separator line
2171 so the .eml file is really in mbox format (even though
2172 most programs do not really care). Patch from Alex Miller.
2174 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2175 The latter is generated from $received_to and is only set if the
2176 message has one envelope recipient. SA can use these headers,
2177 obviously out-of-the-box. Patch from Alex Miller.
2179 PH/08 The ${def test on a variable was returning false if the variable's
2180 value was "0", contrary to what the specification has always said!
2181 The result should be true unless the variable is empty.
2183 PH/09 The syntax error of a character other than { following "${if
2184 def:variable_name" (after optional whitespace) was not being diagnosed.
2185 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2186 accidental colon was present, for example, could give incorrect results.
2188 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2189 result is used (not including appendfile, where other changes are about
2192 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2193 This involved changing a lot of size variables from int to off_t. It
2194 should work with maildirs and everything.
2196 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2197 spamd dying while we are connected to it.
2199 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2202 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2203 function for BDB 4.3.
2205 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2207 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2208 the sources for PCRE have been re-arranged and more files are now
2211 PH/15 The code I had for printing potentially long long variables in PH/11
2212 above was not the best (it lost precision). The length of off_t variables
2213 is now inspected at build time, and an appropriate printing format (%ld
2214 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2215 to be "long long int" or "long int". This is needed for the internal
2216 formatting function string_vformat().
2218 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2219 the configuration file to be ":syslog", then the script "guesses" where
2220 the logs files are, rather than using the compiled in default. In our
2221 case the guess is not the same as the compiled default, so the script
2222 suddenly stopped working when I started to use syslog. The patch checks
2223 to see if log_file_path is "". If so, it attempts to read it from exim
2224 with no configuration file to get the compiled in version, before it
2225 falls back to the previous guessing code."
2227 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2228 implementing BATV in an Exim configuration. See NewStuff for the gory
2231 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2232 Makefile that are specific to HP-UX.
2234 PH/18 If the "use_postmaster" option was set for a recipient callout together
2235 with the "random" option, the postmaster address was used as the MAIL
2236 FROM address for the random test, but not for the subsequent recipient
2237 test. It is now used for both.
2239 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2240 patch removes a few documentation additions to RFC 3028, because the
2241 latest draft now contains them. It adds the new en;ascii-case comparator
2242 and a new error check for 8bit text in MIME parts. Comparator and
2243 require names are now matched exactly. I enabled the subaddress
2244 extension, but it is not well tested yet (read: it works for me)."
2246 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2247 rework some of the code of TK/09 above to avoid the hardwired use of
2248 "%lld" and "long long". Replaced the call to snprintf() with a call to
2251 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2252 records point to non-existent hosts", "retry timeout exceeded", and
2253 "retry time not reached for any host after a long failure period".
2255 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2256 experimental DomainKeys support:
2258 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2259 (2) On an error such as an illegally used "control", the wrong name for
2260 the control was given.
2262 These problems did NOT occur unless DomainKeys support was compiled.
2264 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2266 PH/24 Added ${if match_ip condition.
2268 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2269 databases so that it will be absolutely obvious if a crash occurs in the
2270 DB library. This is a regular occurrence (often caused by mis-matched
2273 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2274 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2275 picked up on a user's system that detects such things. There doesn't seem
2276 to be a gcc warning option for this - only an attribute that has to be
2277 put on the function's prototype. It seems that in Fedora Core 4 they have
2278 set this on a number of new functions. No doubt there will be more in due
2281 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2282 specify a transport. However, if an address that was verified by such a
2283 router was the subject of a callout, Exim crashed because it tried to
2284 read the rcpt_include_affixes from the non-existent transport. Now it
2285 just assumes that the setting of that option is false. This bug was
2286 introduced by 4.51/PH/31.
2288 PH/28 Changed -d+all to exclude +memory, because that information is very
2289 rarely of interest, but it makes the output a lot bigger. People tend to
2290 do -d+all out of habit.
2292 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2293 code in os-type was giving problems when libc.so lives in lib64, like on
2296 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2297 aren't the modern standard, and it seems that some systems' include files
2298 don't always have them. Exim was already checking for some of the newer
2299 ones like T_AAAA, and defining it itself. I've added checks for all the
2300 record types that Exim uses.
2302 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2303 not automatically generating a new one, as it is supposed to. This
2304 prevented TLS from working. If the file did exist, but contained invalid
2305 data, a new version was generated, as expected. It was only the case of a
2306 non-existent file that was broken.
2308 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2309 with a change in libdomainkeys > 0.64.
2311 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2312 from DNS. If the selector record carries the flag, it now has
2313 precedence over the domain-wide flag.
2315 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2317 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2318 the use of a port name that isn't defined in /etc/services) occurred, the
2319 message was deferred as in a normal delivery, and thus remained on the
2320 spool, instead of being failed because of the mua_wrapper setting. This
2321 is now fixed, and I tidied up some of the mua_wrapper messages at the
2324 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2325 the messages in a hash of arrays rather than using individual hashes.
2326 This is a bit cleaner and results in dramatic memory savings, albeit
2327 at a slight CPU cost.
2329 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2330 as requested by Marc Sherman.
2332 SC/10 Eximstats: added histograms for user specified patterns as requested
2335 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2337 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2338 fopen() in the content-scanning modules that did not already have it.
2344 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2345 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2347 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2349 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2351 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2352 to test Sieve filters that use "vacation".
2354 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2355 that changes the way the GnuTLS parameters are stored in the cache file.
2356 The new format can be generated externally. For backward compatibility,
2357 if the data in the cache doesn't make sense, Exim assumes it has read an
2358 old-format file, and it generates new data and writes a new file. This
2359 means that you can't go back to an older release without removing the
2362 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2363 work if there are any delivery delays because "one_time" forces the
2364 parent to be marked "delivered", so its unseen clone is never tried
2365 again. For this reason, Exim now forbids the simultaneous setting of
2368 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2369 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2370 entry if you want to know the details. The fix, however, neglected to
2371 consider the case when local delivery batching is involved. The test for
2372 "previously delivered" was not happening when checking to see if an
2373 address could be batched with a previous (undelivered) one; under
2374 certain circumstances this could lead to multiple deliveries to the same
2377 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2378 in its include files, and this causes problems building Exim.
2380 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2381 header_syntax) but Exim was just ignoring anything given after a slash.
2382 In particular, this caused confusion with an attempt to use "verify =
2383 reverse_host_lookup/defer_ok". An error is now given when options are
2384 supplied for verify items that do not have them. (Maybe reverse_host_
2385 lookup should have a defer_ok option, but that's a different point.)
2387 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2388 defined by RFC 821) to 2048, because there were problems with some AUTH
2389 commands, and RFC 1869 says the size should be increased for extended
2390 SMTP commands that take arguments.
2392 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2395 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2396 "unknown" error; now it says that the functionality isn't in the binary.
2398 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2399 an address' error message when a string expansion fails (syntax or
2400 whatever). Otherwise the password may appear in the log. Following change
2401 PH/42 below, there is no longer a chance of it appearing in a bounce
2404 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2406 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2407 of its addresses was ever tried. (Bugzilla bug #2.)
2409 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2410 the result incorrectly in the debug output. (It correctly added a newline
2411 to what was transported.)
2413 TF/01 Added $received_time.
2415 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2416 commented out examples of how to interface to a virus scanner and to
2417 SpamAssassin. Also added commented examples of av_scanner and
2418 spamd_address settings.
2420 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2421 and controls are allowed in which ACLs. There were a couple of minor
2422 errors. Some of the entries in the conditions table (which is a table of
2423 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2424 negation of where the condition IS allowed.
2426 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2428 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2429 header file does not have a version number, so I've had to invent a new
2430 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2431 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2432 radiusclient), but it was contributed by a Radius user.
2434 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2435 files or directories, for OpenSSL.
2437 PH/19 When an Exim process that is running as root has to create an Exim log
2438 file, it does so in a subprocess that runs as exim:exim so as to get the
2439 ownership right at creation (otherwise, other Exim processes might see
2440 the file with the wrong ownership). There was no test for failure of this
2441 fork() call, which would lead to the process getting stuck as it waited
2442 for a non-existent subprocess. Forks do occasionally fail when resources
2443 run out. I reviewed all the other calls to fork(); they all seem to check
2446 PH/20 When checking for unexpected SMTP input at connect time (before writing
2447 the banner), Exim was not dealing correctly with a non-positive return
2448 from the read() function. If the client had disconnected by this time,
2449 the result was a log entry for a synchronization error with an empty
2450 string after "input=" when read() returned zero. If read() returned -1
2451 (an event I could not check), uninitialized data bytes were printed.
2452 There were reports of junk text (parts of files, etc) appearing after
2455 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2457 PH/22 Added support for macro redefinition, and (re)definition in between
2458 driver and ACL definitions.
2460 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2461 forgetting to use the resulting value; it was using the unexpanded value.
2463 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2464 hadn't been configured. The fix is from Juergen Kreileder, who
2465 understands it better than I do:
2467 "Here's what I see happening with three configured cyrus_sasl
2468 authenticators configured (plain, login, cram-md5):
2470 On startup auth_cyrus_sasl_init() gets called for each of these.
2471 This means three calls to sasl_listmech() without a specified mech_list.
2472 => SASL tests which mechs of all available mechs actually work
2473 => three warnings about OTP not working
2474 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2476 With the patch, sasl_listmech() also gets called three times. But now
2477 SASL's mech_list option is set to the server_mech specified in the the
2478 authenticator. Or in other words, the answer from sasl_listmech()
2479 gets limited to just the mech you're testing for (which is different
2481 => the return list contains just 'plain' or 'login', 'cram-md5' or
2482 nothing depending on the value of ob->server_mech.
2484 I've just tested the patch: Authentication still works fine,
2485 unavailable mechs specified in the exim configuration are still
2486 caught, and the auth.log warnings about OTP are gone."
2488 PH/25 When debugging is enabled, the contents of the command line are added
2489 to the debugging output, even when log_selector=+arguments is not
2492 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2493 answer is "GNU", and only if the return is "GNU/something" is the answer
2496 PH/27 $acl_verify_message is now set immediately after the failure of a
2497 verification in an ACL, and so is available in subsequent modifiers. In
2498 particular, the message can be preserved by coding like this:
2500 warn !verify = sender
2501 set acl_m0 = $acl_verify_message
2503 Previously, $acl_verify_message was set only while expanding "message"
2504 and "log_message" when a very denied access.
2506 PH/28 Modified OS/os.c-Linux with
2508 -#ifndef OS_LOAD_AVERAGE
2509 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2511 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2512 nomenclature these days.)
2514 PH/29 Installed patch from the Sieve maintainer that adds the options
2515 sieve_useraddress and sieve_subaddress to the redirect router.
2517 PH/30 In these circumstances:
2518 . Two addresses routed to the same list of hosts;
2519 . First host does not offer TLS;
2520 . First host accepts first address;
2521 . First host gives temporary error to second address;
2522 . Second host offers TLS and a TLS session is established;
2523 . Second host accepts second address.
2524 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2525 and peerdn, if requested) that were in fact used only for the second
2528 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2529 attention to any local part prefix or suffix that was matched by the
2530 router that accepted the address. It now behaves in the same way as it
2531 does for delivery: the affixes are removed from the local part unless
2532 rcpt_include_affixes is set on the transport.
2534 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2535 timeout during the DATA phase of an incoming message.
2537 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2538 applied a patch sanctioned by the Sieve maintainer.
2540 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2541 the uid or gid is negative. A case of a negative gid caused this to be
2542 noticed. The fix allows for either to be negative.
2544 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2545 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2548 PH/36 The change PH/12 above was broken. Fixed it.
2550 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2551 the grounds that routing the same address twice would always produce the
2552 same answer. This might have been true once, but it is certainly no
2553 longer true now. Routing a child address may depend on the previous
2554 routing that produced that child. Some complicated redirection strategies
2555 went wrong when messages had multiple recipients, and made Exim's
2556 behaviour dependent on the order in which the addresses were given.
2558 I have moved the duplicate checking until after the routing is complete.
2559 Exim scans the addresses that are assigned to local and remote
2560 transports, and removes any duplicates. This means that more work will be
2561 done, as duplicates will always all be routed, but duplicates are
2562 presumably rare, so I don't expect this is of any significance.
2564 For deliveries to pipes, files, and autoreplies, the duplicate checking
2565 still happens during the routing process, since they are not going to be
2568 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2569 It corrects a timeout issue with spamd. This is Ian's comment: "The
2570 background is that sometimes spamd either never reads data from a
2571 connection it has accepted, or it never writes response data. The exiscan
2572 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2573 blindly assumes that writes won't block so it may never time out."
2575 PH/39 Allow G after quota size as well as K and M.
2577 PH/40 The value set for $authenticated_id in an authenticator may not contain
2578 binary zeroes or newlines because the value is written to log lines and
2579 to spool files. There was no check on this. Now the value is run through
2580 the string_printing() function so that such characters are converted to
2581 printable escape sequences.
2583 PH/41 $message_linecount is a new variable that contains the total number of
2584 lines in the message. Compare $body_linecount, which is the count for the
2587 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2588 bounce and delay warning messages, except in certain special cases, which
2591 (a) An SMTP error message from a remote host;
2592 (b) A message specified in a :fail: redirection;
2593 (c) A message specified in a "fail" command in a system filter;
2594 (d) A message specified in a FAIL return from the queryprogram router;
2595 (e) A message specified by the cannot_route_message router option.
2597 In these cases only, Exim does include the error details in bounce and
2598 warning messages. There are also a few cases where bland messages such
2599 as "unrouteable address" or "local delivery error" are given.
2601 PH/43 $value is now also set for the "else" part of a ${run expansion.
2603 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2604 being worked on, but at least Exim now implements the latest version to
2607 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2608 process to complete was treated as a delivery failure, a timeout while
2609 writing the message to the pipe was logged, but erroneously treated as a
2610 successful delivery. Such timeouts include transport filter timeouts. For
2611 consistency with the overall process timeout, these timeouts are now
2612 treated as errors, giving rise to delivery failures by default. However,
2613 there is now a new Boolean option for the pipe transport called
2614 timeout_defer, which, if set TRUE, converts the failures into defers for
2615 both kinds of timeout. A transport filter timeout is now identified in
2618 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2619 systems where "make" and "gmake" are different, calling "gmake" at top
2620 level broke things. I've arranged for the value of $(MAKE) to be passed
2621 from the Makefile to this script so that it can call the same version of
2625 A note about Exim versions 4.44 and 4.50
2626 ----------------------------------------
2628 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2629 changes of various kinds. As a consequence, a big documentation update was
2630 needed. This delayed the release for rather longer than seemed good, especially
2631 in the light of a couple of (minor) security issues. Therefore, the changes
2632 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2633 release. So 4.44 and 4.50 are in effect two different branches that both start
2636 I have left the 4.50 change log unchanged; it contains all the changes since
2637 4.43. The change log for 4.44 is below; many of its items are identical to
2638 those for 4.50. This seems to be the most sensible way to preserve the
2639 historical information.
2645 1. Minor wording change to the doc/README.SIEVE file.
2647 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2648 computation of the current number of files was incorrect.
2650 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2651 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2654 4. Give more explanation in the error message when the command for a transport
2655 filter fails to execute.
2657 5. There are several places where Exim runs a non-Exim command in a
2658 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2659 was being done only for the command run by the queryprogram router. It is
2660 now done for all such subprocesses. The other cases are: ${run, transport
2661 filters, and the commands run by the lmtp and pipe transports.
2663 6. Added CONFIGURE_GROUP build-time option.
2665 7. Some older OS have a limit of 256 on the maximum number of file
2666 descriptors. Exim was using setrlimit() to set 1000 as a large value
2667 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2668 systems. I've change it so that if it can't get 1000, it tries for 256.
2670 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2671 was an oversight, and furthermore, ever since the addition of extra
2672 controls (e.g. 4.43/32), the checks on when to allow different forms of
2673 "control" were broken. There should now be diagnostics for all cases when a
2674 control that does not make sense is encountered.
2676 9. Added the /retain_sender option to "control=submission".
2678 10. $recipients is now available in the predata ACL (oversight).
2680 11. Tidy the search cache before the fork to do a delivery from a message
2681 received from the command line. Otherwise the child will trigger a lookup
2682 failure and thereby defer the delivery if it tries to use (for example) a
2683 cached ldap connection that the parent has called unbind on.
2685 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2686 of $address_data from the recipient verification was clobbered by the
2687 sender verification.
2689 13. The value of address_data from a sender verification is now available in
2690 $sender_address_data in subsequent conditions in the ACL statement.
2692 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2694 15. Added a new option "connect=<time>" to callout options, to set a different
2697 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2698 was its contents. (It was OK if the option was not defined at all.)
2700 17. A "Completed" log line is now written for messages that are removed from
2701 the spool by the -Mrm option.
2703 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2704 information about exactly what failed.
2706 19. Added -dd to debug only the daemon process.
2708 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2709 handles timeouts, both on the server side and network timeouts. Renamed the
2710 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2712 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2713 It is now set to "smtps".
2715 22. $host_address is now set to the target address during the checking of
2716 ignore_target_hosts.
2718 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2719 being passed; this would have caused $sender_host_name to have been used if
2720 matching the list had actually called for a host name (not very likely,
2721 since this list is usually IP addresses). A host name is now passed as
2724 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2725 code that specifies a non-restarting handler (typically sigaction() in
2726 modern systems) in an attempt to fix a rare and obscure crash bug.
2728 25. Narrowed the window for a race in the daemon that could cause it to ignore
2729 SIGCHLD signals. This is not a major problem, because they are used only to
2730 wake it up if nothing else does.
2732 26. A malformed maildirsize file could cause Exim to calculate negative values
2733 for the mailbox size or file count. Odd effects could occur as a result.
2734 The maildirsize information is now recalculated if the size or filecount
2737 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2738 support for a long time. Removed HAVE_SYS_VFS_H.
2740 28. Installed the latest version of exipick from John Jetmore.
2742 29. In an address list, if the pattern was not a regular expression, an empty
2743 subject address (from a bounce message) matched only if the pattern was an
2744 empty string. Non-empty patterns were not even tested. This was the wrong
2745 because it is perfectly reasonable to use an empty address as part of a
2746 database query. An empty address is now tested by patterns that are
2747 lookups. However, all the other forms of pattern expect the subject to
2748 contain a local part and a domain, and therefore, for them, an empty
2749 address still always fails if the pattern is not itself empty.
2751 30. Exim went into a mad DNS loop when attempting to do a callout where the
2752 host was specified on an smtp transport, and looking it up yielded more
2753 than one IP address.
2755 31. Re-factored the code for checking spool and log partition space into a
2756 function that finds that data and another that does the check. The former
2757 is then used to implement four new variables: $spool_space, $log_space,
2758 $spool_inodes, and $log_inodes.
2760 32. The RFC2047 encoding function was originally intended for short strings
2761 such as real names; it was not keeping to the 75-character limit for
2762 encoded words that the RFC imposes. It now respects the limit, and
2763 generates multiple encoded words if necessary. To be on the safe side, I
2764 have increased the buffer size for the ${rfc2047: expansion operator from
2767 33. It is now permitted to omit both strings after an "if" condition; if the
2768 condition is true, the result is "true". As before, when the second string
2769 is omitted, a false condition yields an empty string. This makes it less
2770 cumbersome to write custom ACL and router conditions.
2772 34. Failure to deliver a bounce message always caused it to be frozen, even if
2773 there was an errors_to setting on the router. The errors_to setting is now
2776 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2777 canonical form (fully expanded) before being placed in
2778 $sender_host_address.
2780 36. The table in the code that translates DNS record types into text (T_A to
2781 "A" for instance) was missing entries for NS and CNAME. It is just possible
2782 that this could have caused confusion if both these types were looked up
2783 for the same domain, because the text type is used as part of Exim's
2784 per-process caching. But the chance of anyone hitting this buglet seems
2787 37. The dnsdb lookup has been extended in a number of ways.
2789 (1) There is a new type, "zns", which walks up the domain tree until it
2790 finds some nameserver records. It should be used with care.
2792 (2) There is a new type, "mxh", which is like "mx" except that it returns
2793 just the host names, not the priorities.
2795 (3) It is now possible to give a list of domains (or IP addresses) to be
2796 looked up. The behaviour when one of the lookups defers can be
2797 controlled by a keyword.
2799 (4) It is now possible to specify the separator character for use when
2800 multiple records are returned.
2802 38. The dnslists ACL condition has been extended: it is now possible to supply
2803 a list of IP addresses and/or domains to be looked up in a particular DNS
2806 39. Added log_selector=+queue_time_overall.
2808 40. When running the queue in the test harness, wait just a tad after forking a
2809 delivery process, to get repeatability of debugging output.
2811 41. Include certificate and key file names in error message when GnuTLS fails
2812 to set them up, because the GnuTLS error message doesn't include the name
2813 of the failing file when there is a problem reading it.
2815 42. Allow both -bf and -bF in the same test run.
2817 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2819 44. The "Exiscan patch" is now merged into the mainline Exim source.
2821 45. Sometimes the final signoff response after QUIT could fail to get
2822 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2823 < 0 before doing a fflush(). This bug looks as though it goes back to the
2824 introduction of TLS in release 3.20, but "sometimes" must have been rare
2825 because the tests only now provoked it.
2827 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2828 (this can affect the format of dates).
2830 47. exim_tidydb, when checking for the continued existence of a message for
2831 which it has found a message-specific retry record, was not finding
2832 messages that were in split spool directories. Consequently, it was
2833 deleting retry records that should have stayed in existence.
2835 48. Steve fixed some bugs in eximstats.
2837 49. The SPA authentication driver was not abandoning authentication and moving
2838 on to the next authenticator when an expansion was forced to fail,
2839 contradicting the general specification for all authenticators. Instead it
2840 was generating a temporary error. It now behaves as specified.
2842 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2843 (the order specifies the preference for clients). The order is now AES256,
2844 AES128, 3DES, ARCFOUR128.
2846 51. Small patch to Sieve code - explicitly set From: when generating an
2849 52. Exim crashed if a remote delivery caused a very long error message to be
2850 recorded - for instance if somebody sent an entire SpamAssassin report back
2851 as a large number of 550 error lines. This bug was coincidentally fixed by
2852 increasing the size of one of Exim's internal buffers (big_buffer) that
2853 happened as part of the Exiscan merge. However, to be on the safe side, I
2854 have made the code more robust (and fixed the comments that describe what
2857 53. Now that there can be additional text after "Completed" in log lines (if
2858 the queue_time_overall log selector is set), a one-byte patch to exigrep
2859 was needed to allow it to recognize "Completed" as not the last thing in
2862 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2863 patch that reportedly fixes this has been added. I am not expert enough to
2864 create a test for it. This is what the patch creator wrote:
2866 "I found a little strange behaviour of ldap code when working with
2867 Windows 2003 AD Domain, where users was placed in more than one
2868 Organization Units. When I tried to give exim partial DN, the exit code
2869 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2870 But simultaneously result of request was absolutely normal ldap result,
2871 so I produce this patch..."
2873 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2874 REFERENCE, so I have modified the code to exclude the patch when that macro
2877 55. Some experimental protocols are using DNS PTR records for new purposes. The
2878 keys for these records are domain names, not reversed IP addresses. The
2879 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2880 leaves it alone. Component reversal etc. now happens only for IP addresses.
2883 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2885 57. Double the size of the debug message buffer (to 2048) so that more of very
2886 long debug lines gets shown.
2888 58. The exicyclog utility now does better if the number of log files to keep
2889 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2891 59. Two changes related to the smtp_active_hostname option:
2893 (1) $smtp_active_hostname is now available as a variable.
2894 (2) The default for smtp_banner uses $smtp_active_hostname instead
2895 of $primary_hostname.
2897 60. The host_aton() function is supposed to be passed a string that is known
2898 to be a valid IP address. However, in the case of IPv6 addresses, it was
2899 not checking this. This is a hostage to fortune. Exim now panics and dies
2900 if the condition is not met. A case was found where this could be provoked
2901 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2902 components; fortuitously, this particular loophole had already been fixed
2903 by change 4.50/55 above.
2905 If there are any other similar loopholes, the new check in host_aton()
2906 itself should stop them being exploited. The report I received stated that
2907 data on the command line could provoke the exploit when Exim was running as
2908 exim, but did not say which command line option was involved. All I could
2909 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2910 running as the user.
2913 61. There was a buffer overflow vulnerability in the SPA authentication code
2914 (which came originally from the Samba project). I have added a test to the
2915 spa_base64_to_bits() function which I hope fixes it.
2918 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2919 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2921 63. The daemon start-up calls getloadavg() while still root for those OS that
2922 need the first call to be done as root, but it missed one case: when
2923 deliver_queue_load_max is set with deliver_drop_privilege. This is
2924 necessary for the benefit of the queue runner, because there is no re-exec
2925 when deliver_drop_privilege is set.
2927 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2928 This has been fixed.
2930 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2931 was in use, was not putting the data itself into the right store pool;
2932 consequently, it could be overwritten for a subsequent message in the same
2933 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2936 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2938 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2939 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2940 the function were treating the return as a boolean value, which happened to
2941 work because 0=false and not-0=true, but is not correct code.
2943 68. The host_aton() function was not handling scoped IPv6 addresses (those
2944 with, for example, "%eth0" on the end) correctly.
2946 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2947 negated items (that is, ~something) in unsigned ints. Some compilers
2948 apparently mutter when there is no cast.
2950 70. If an address verification called from an ACL failed, and did not produce a
2951 user-specific message (i.e. there was only a "system" message), nothing was
2952 put in $acl_verify_message. In this situation, it now puts the system
2955 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2956 session; change 4.31/43 added the unwanted input to the log line - except
2957 that it did not do this in the start of session case. It now does.
2959 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2960 This is wrong and can cause the other end to generate a synchronization
2961 error if it is another Exim or anything else that does the synchronization
2962 check. A QUIT command is no longer sent after a timeout.
2964 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2965 during host lookups.
2967 74. The defer_ok option of callout verification was not working if it was used
2968 when verifying addresses in header lines, that is, for this case:
2970 verify = header_sender/callout=defer_ok
2972 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2973 those file descriptors could be used for SMTP connections. If anything
2974 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2975 could be sent to the SMTP client, causing chaos. The daemon now opens
2976 stdin, stdout, and stderr to /dev/null when it puts itself into the
2979 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2980 log by default. The user can override this with suitable Perl magic.
2982 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2983 the log message when discard triggers, was not working for the DATA ACL or
2984 for the non-SMTP ACL.
2986 78. Error message wording change in sieve.c.
2988 79. If smtp_accept_max_per_host was set, the number of connections could be
2989 restricted to fewer than expected, because the daemon was trying to set up
2990 a new connection before checking whether the processes handling previous
2991 connections had finished. The check for completed processes is now done
2992 earlier. On busy systems, this bug wouldn't be noticed because something
2993 else would have woken the daemon, and it would have reaped the completed
2996 80. If a message was submitted locally by a user whose login name contained one
2997 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2998 It caused a spool format error. I have fixed the spool reading code. A
2999 related problem was that the "from" clause in the Received: line became
3000 illegal because of the space(s). It is now covered by ${quote_local_part.
3002 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3005 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3007 83. Patch from Sieve maintainer to fix unterminated string problem in
3008 "vacation" handling.
3010 84. Some minor changes to the Linux configuration files to help with other
3011 OS variants using glibc.
3013 85. One more patch for Sieve to update vacation handling to latest spec.
3016 ----------------------------------------------------
3017 See the note above about the 4.44 and 4.50 releases.
3018 ----------------------------------------------------
3024 1. Change 4.43/35 introduced a bug that caused file counts to be
3025 incorrectly computed when quota_filecount was set in an appendfile
3028 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3029 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3032 3. Give more explanation in the error message when the command for a transport
3033 filter fails to execute.
3035 4. There are several places where Exim runs a non-Exim command in a
3036 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3037 was being done only for the command run by the queryprogram router. It is
3038 now done for all such subprocesses. The other cases are: ${run, transport
3039 filters, and the commands run by the lmtp and pipe transports.
3041 5. Some older OS have a limit of 256 on the maximum number of file
3042 descriptors. Exim was using setrlimit() to set 1000 as a large value
3043 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3044 systems. I've change it so that if it can't get 1000, it tries for 256.
3046 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3047 was an oversight, and furthermore, ever since the addition of extra
3048 controls (e.g. 4.43/32), the checks on when to allow different forms of
3049 "control" were broken. There should now be diagnostics for all cases when a
3050 control that does not make sense is encountered.
3052 7. $recipients is now available in the predata ACL (oversight).
3054 8. Tidy the search cache before the fork to do a delivery from a message
3055 received from the command line. Otherwise the child will trigger a lookup
3056 failure and thereby defer the delivery if it tries to use (for example) a
3057 cached ldap connection that the parent has called unbind on.
3059 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3060 of $address_data from the recipient verification was clobbered by the
3061 sender verification.
3063 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3064 was its contents. (It was OK if the option was not defined at all.)
3066 11. A "Completed" log line is now written for messages that are removed from
3067 the spool by the -Mrm option.
3069 12. $host_address is now set to the target address during the checking of
3070 ignore_target_hosts.
3072 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3073 being passed; this would have caused $sender_host_name to have been used if
3074 matching the list had actually called for a host name (not very likely,
3075 since this list is usually IP addresses). A host name is now passed as
3078 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3079 code that specifies a non-restarting handler (typically sigaction() in
3080 modern systems) in an attempt to fix a rare and obscure crash bug.
3082 15. Narrowed the window for a race in the daemon that could cause it to ignore
3083 SIGCHLD signals. This is not a major problem, because they are used only to
3084 wake it up if nothing else does.
3086 16. A malformed maildirsize file could cause Exim to calculate negative values
3087 for the mailbox size or file count. Odd effects could occur as a result.
3088 The maildirsize information is now recalculated if the size or filecount
3091 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3092 support for a long time. Removed HAVE_SYS_VFS_H.
3094 18. Updated exipick to current release from John Jetmore.
3096 19. Allow an empty sender to be matched against a lookup in an address list.
3097 Previously the only cases considered were a regular expression, or an
3100 20. Exim went into a mad DNS lookup loop when doing a callout where the
3101 host was specified on the transport, if the DNS lookup yielded more than
3104 21. The RFC2047 encoding function was originally intended for short strings
3105 such as real names; it was not keeping to the 75-character limit for
3106 encoded words that the RFC imposes. It now respects the limit, and
3107 generates multiple encoded words if necessary. To be on the safe side, I
3108 have increased the buffer size for the ${rfc2047: expansion operator from
3111 22. Failure to deliver a bounce message always caused it to be frozen, even if
3112 there was an errors_to setting on the router. The errors_to setting is now
3115 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3116 canonical form (fully expanded) before being placed in
3117 $sender_host_address.
3119 24. Updated eximstats to version 1.33
3121 25. Include certificate and key file names in error message when GnuTLS fails
3122 to set them up, because the GnuTLS error message doesn't include the name
3123 of the failing file when there is a problem reading it.
3125 26. Expand error message when OpenSSL has problems setting up cert/key files.
3128 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3129 (this can affect the format of dates).
3131 28. exim_tidydb, when checking for the continued existence of a message for
3132 which it has found a message-specific retry record, was not finding
3133 messages that were in split spool directories. Consequently, it was
3134 deleting retry records that should have stayed in existence.
3136 29. eximstats updated to version 1.35
3137 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3138 1.35 - bugfix such that pie charts by volume are generated correctly
3140 30. The SPA authentication driver was not abandoning authentication and moving
3141 on to the next authenticator when an expansion was forced to fail,
3142 contradicting the general specification for all authenticators. Instead it
3143 was generating a temporary error. It now behaves as specified.
3145 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3146 (the order specifies the preference for clients). The order is now AES256,
3147 AES128, 3DES, ARCFOUR128.
3149 31. Small patch to Sieve code - explicitly set From: when generating an
3152 32. Exim crashed if a remote delivery caused a very long error message to be
3153 recorded - for instance if somebody sent an entire SpamAssassin report back
3154 as a large number of 550 error lines. This bug was coincidentally fixed by
3155 increasing the size of one of Exim's internal buffers (big_buffer) that
3156 happened as part of the Exiscan merge. However, to be on the safe side, I
3157 have made the code more robust (and fixed the comments that describe what
3160 33. Some experimental protocols are using DNS PTR records for new purposes. The
3161 keys for these records are domain names, not reversed IP addresses. The
3162 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3163 leaves it alone. Component reversal etc. now happens only for IP addresses.
3166 34. The host_aton() function is supposed to be passed a string that is known
3167 to be a valid IP address. However, in the case of IPv6 addresses, it was
3168 not checking this. This is a hostage to fortune. Exim now panics and dies
3169 if the condition is not met. A case was found where this could be provoked
3170 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3171 components; fortuitously, this particular loophole had already been fixed
3172 by change 4.50/55 or 4.44/33 above.
3174 If there are any other similar loopholes, the new check in host_aton()
3175 itself should stop them being exploited. The report I received stated that
3176 data on the command line could provoke the exploit when Exim was running as
3177 exim, but did not say which command line option was involved. All I could
3178 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3179 running as the user.
3182 35. There was a buffer overflow vulnerability in the SPA authentication code
3183 (which came originally from the Samba project). I have added a test to the
3184 spa_base64_to_bits() function which I hope fixes it.
3187 36. The daemon start-up calls getloadavg() while still root for those OS that
3188 need the first call to be done as root, but it missed one case: when
3189 deliver_queue_load_max is set with deliver_drop_privilege. This is
3190 necessary for the benefit of the queue runner, because there is no re-exec
3191 when deliver_drop_privilege is set.
3193 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3194 was in use, was not putting the data itself into the right store pool;
3195 consequently, it could be overwritten for a subsequent message in the same
3196 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3199 38. Sometimes the final signoff response after QUIT could fail to get
3200 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3201 < 0 before doing a fflush(). This bug looks as though it goes back to the
3202 introduction of TLS in release 3.20, but "sometimes" must have been rare
3203 because the tests only now provoked it.
3209 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3210 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3211 Now it returns an integer. A number of calls were still expecting a T/F
3212 return. Fortuitously, in all cases, the tests worked in OK situations,
3213 which is the norm. However, things would have gone wrong on any write
3214 failures on the smtp file descriptor. This function is used when sending
3215 messages over SMTP and also when doing verify callouts.
3217 2. When Exim is called to do synchronous delivery of a locally submitted
3218 message (the -odf or -odi options), it no longer closes stderr before doing
3221 3. Implemented the mua_wrapper option.
3223 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3225 5. Implemented the functions header_remove(), header_testname(),
3226 header_add_at_position(), and receive_remove_recipient(), and exported them
3229 6. If an ACL "warn" statement specified the addition of headers, Exim already
3230 inserted X-ACL-Warn: at the start if there was no header name. However, it
3231 was not making this test for the second and subsequent header lines if
3232 there were newlines in the string. This meant that an invalid header could
3233 be inserted if Exim was badly configured.
3235 7. Allow an ACL "warn" statement to add header lines at the start or after all
3236 the Received: headers, as well as at the end.
3238 8. Added the rcpt_4xx retry error code.
3240 9. Added postmaster_mailfrom=xxx to callout verification option.
3242 10. Added mailfrom=xxxx to the callout verification option, for verify=
3245 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3246 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3248 12. Inserted some casts to stop certain compilers warning when using pointer
3249 differences as field lengths or precisions in printf-type calls (mostly
3250 affecting debugging statements).
3252 13. Added optional readline() support for -be (dynamically loaded).
3254 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3255 same clock tick as a message's arrival, so that its received time was the
3256 same as the "first fail" time on the retry record, and that message
3257 remained on the queue past the ultimate address timeout, every queue runner
3258 would try a delivery (because it was past the ultimate address timeout) but
3259 after another failure, the ultimate address timeout, which should have then
3260 bounced the address, did not kick in. This was a "< instead of <=" error;
3261 in most cases the first failure would have been in the next clock tick
3262 after the received time, and all would be well.
3264 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3265 being recognized when the domain list was tested by the match_domain
3266 condition in an expansion string.
3268 16. Added the ${str2b64: operator.
3270 17. Exim was always calling setrlimit() to set a large limit for the number of
3271 processes, without checking whether the existing limit was already
3272 adequate. (It did check for the limit on file descriptors.) Furthermore,
3273 errors from getrlimit() and setrlimit() were being ignored. Now they are
3274 logged to the main and panic logs, but Exim does carry on, to try to do its
3275 job under whatever limits there are.
3277 18. Imported PCRE 5.0.
3279 19. Trivial typo in log message " temporarily refused connection" (the leading
3282 20. If the log selector return_path_on_delivery was set and an address was
3283 redirected to /dev/null, the delivery process crashed because it assumed
3284 that a return path would always be set for a "successful" delivery. In this
3285 case, the whole delivery is bypassed as an optimization, and therefore no
3288 21. Internal re-arrangement: the function for sending a challenge and reading
3289 a response while authentication was assuming a zero-terminated challenge
3290 string. It's now changed to take a pointer and a length, to allow for
3291 binary data in such strings.
3293 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3295 23. Exim was not respecting finduser_retries when seeking the login of the
3296 uid under which it was called; it was always trying 10 times. (The default
3297 setting of finduser_retries is zero.) Also, it was sleeping after the final
3298 failure, which is pointless.
3300 24. Implemented tls_on_connect_ports.
3302 25. Implemented acl_smtp_predata.
3304 26. If the domain in control=submission is set empty, Exim assumes that the
3305 authenticated id is a complete email address when it generates From: or
3306 Sender: header lines.
3308 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3309 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3310 chown and chgrp in /bin and hostname in /usr/bin.
3312 28. Exim was keeping the "process log" file open after each use, just as it
3313 does for the main log. This opens the possibility of it remaining open for
3314 long periods when the USR1 signal hits a daemon. Occasional processlog
3315 errors were reported, that could have been caused by this. Anyway, it seems
3316 much more sensible not to leave this file open at all, so that is what now
3319 29. The long-running daemon process does not normally write to the log once it
3320 has entered its main loop, and it closes the log before doing so. This is
3321 so that log files can straightforwardly be renamed and moved. However,
3322 there are a couple of unusual error situations where the daemon does write
3323 log entries, and I had neglected to close the log afterwards.
3325 30. The text of an SMTP error response that was received during a remote
3326 delivery was being truncated at 512 bytes. This is too short for some of
3327 the long messages that one sometimes sees. I've increased the limit to
3330 31. It is now possible to make retry rules that apply only when a message has a
3331 specific sender, in particular, an empty sender.
3333 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3334 it possible to be selective about when SMTP synchronization is enforced.
3336 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3338 32. Implemented hosts_connection_nolog.
3340 33. Added an ACL for QUIT.
3342 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3345 35. Added mailbox_size and mailbox_filecount to appendfile.
3347 36. Added control = no_multiline_responses to ACLs.
3349 37. There was a bug in the logic of the code that waits for the clock to tick
3350 in the case where the clock went backwards by a substantial amount such
3351 that the microsecond fraction of "now" was more than the microsecond
3352 fraction of "then" (but the whole seconds number was less).
3354 38. Added support for the libradius Radius client library this is found on
3355 FreeBSD (previously only the radiusclient library was supported).
3361 1. When certain lookups returned multiple values in the form name=value, the
3362 quoting of the values was not always being done properly. Specifically:
3363 (a) If the value started with a double quote, but contained no whitespace,
3365 (b) If the value contained whitespace other than a space character (i.e.
3366 tabs or newlines or carriage returns) it was not quoted.
3367 This fix has been applied to the mysql and pgsql lookups by writing a
3368 separate quoting function and calling it from the lookup code. The fix
3369 should probably also be applied to nisplus, ibase and oracle lookups, but
3370 since I cannot test any of those, I have not disturbed their existing code.
3372 2. A hit in the callout cache for a specific address caused a log line with no
3373 reason for rejecting RCPT. Now it says "Previous (cached) callout
3374 verification failure".
3376 3. There was an off-by-one bug in the queryprogram router. An over-long
3377 return line was truncated at 256 instead of 255 characters, thereby
3378 overflowing its buffer with the terminating zero. As well as fixing this, I
3379 have increased the buffer size to 1024 (and made a note to document this).
3381 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3382 when Exim is waiting for an SMTP response from a remote server, Exim
3383 restarts its select() call on the socket, thereby resetting its timeout.
3384 This is not a problem when such interrupts are rare. Somebody set up a cron
3385 job to run exiwhat every 2 minutes, which is less than the normal select()
3386 timeout (5 or 10 minutes). This meant that the select() timeout never
3387 kicked in because it was always reset. I have fixed this by comparing the
3388 time when an interrupt arrives with the time at the start of the first call
3389 to select(). If more time than the timeout has elapsed, the interrupt is
3390 treated as a timeout.
3392 5. Some internal re-factoring in preparation for the addition of Sieve
3393 extensions (by MH). In particular, the "personal" test is moved to a
3394 separate function, and given an option for scanning Cc: and Bcc: (which is
3395 not set for Exim filters).
3397 6. When Exim created an email address using the login of the caller as the
3398 local part (e.g. when creating a From: or Sender: header line), it was not
3399 quoting the local part when it contained special characters such as @.
3401 7. Installed new OpenBSD configuration files.
3403 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3404 try to make them clearer.
3406 9. Callout options, other than the timeout value, were being ignored when
3407 verifying sender addresses in header lines. For example, when using
3409 verify = header_sender/callout=no_cache
3411 the cache was (incorrectly) being used.
3413 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3414 only the Cygwin environment.
3416 11. When return_path_on_delivery was set as a log selector, if different remote
3417 addresses in the same message used different return paths and parallel
3418 remote delivery occurred, the wrong values would sometimes be logged.
3419 (Whenever a remote delivery process finished, the return path value from
3420 the most recently started remote delivery process was logged.)
3422 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3423 lines when AUTH and/or TLS are in use. This is the "received protocol"
3424 field. Exim used to use "asmtp" for authenticated SMTP, without any
3425 indication (in the protocol name) for TLS use. Now it follows the RFC and
3426 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3427 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3428 names appear in log lines as well as in Received: header lines.
3430 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3431 extensions, and comparison tests, and to fix some bugs.
3433 14. Changes to the "personal" filter test:
3435 (1) The test was buggy in that it was just doing the equivalent of
3436 "contains" tests on header lines. For example, if a user's address was
3437 anne@some.where, the "personal" test would incorrectly be true for
3439 To: susanne@some.where
3441 This test is now done by extracting each address from the header in turn,
3442 and checking the entire address. Other tests that are part of "personal"
3443 are now done using regular expressions (for example, to check local parts
3444 of addresses in From: header lines).
3446 (2) The list of non-personal local parts in From: addresses has been
3447 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3448 taken from the Sieve specification recommendations.
3450 (3) If the message contains any header line starting with "List-" it is
3451 treated as non-personal.
3453 (4) The test for "circular" in the Subject: header line has been removed
3454 because it now seems ill-conceived.
3456 15. Minor typos in src/EDITME comments corrected.
3458 16. Installed latest exipick from John Jetmore.
3460 17. If headers_add on a router specified a text string that was too long for
3461 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3462 of string_sprintf() is now avoided.
3464 18. $message_body_size was not set (it was always zero) when running the DATA
3465 ACL and the local_scan() function.
3467 19. For the "mail" command in an Exim filter, no default was being set for
3468 the once_repeat time, causing a random time value to be used if "once" was
3469 specified. (If the value happened to be <= 0, no repeat happened.) The
3470 default is now 0s, meaning "never repeat". The "vacation" command was OK
3471 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3472 (I found it when inspecting the code).
3474 20. There is now an overall timeout for performing a callout verification. It
3475 defaults to 4 times the callout timeout, which applies to individual SMTP
3476 commands during the callout. The overall timeout applies when there is more
3477 than one host that can be tried. The timeout is checked before trying the
3478 next host. This prevents very long delays if there are a large number of
3479 hosts and all are timing out (e.g. when the network connections are timing
3480 out). The value of the overall timeout can be changed by specifying an
3481 additional sub-option for "callout", called "maxwait". For example:
3483 verify = sender/callout=5s,maxwait=20s
3485 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3486 to the end before writing, but this should make it even safer).
3488 22. Exim was forgetting that it had advertised PIPELINING for the second and
3489 subsequent messages on an SMTP connection. It was also not resetting its
3490 memory on STARTTLS and an internal HELO.
3492 23. When Exim logs an SMTP synchronization error within a session, it now
3493 records whether PIPELINING has been advertised or not.
3495 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3496 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3497 rather than long int.
3499 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3501 26. Added the never_mail option to autoreply.
3507 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3508 crash if the getsockname() call failed; this can happen if a connection is
3509 closed very soon after it is established. The problem was simply in the
3510 order in which certain operations were done, causing Exim to try to write
3511 to the SMTP stream before it had set up the file descriptor. The bug has
3512 been fixed by making things happen in the correct order.
3518 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3519 before the connection was closed, thus losing the rejection response.
3521 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3522 some early Solaris releases, but causes trouble in current releases where
3523 socklen_t is defined.
3525 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3528 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3531 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3532 that releasing the top of it at the end releases what was used for sub-
3533 expansions (unless the block got too big). However, discard this block if
3534 the first thing is a variable or header, so that we can use its block when
3535 it is dynamic (useful for very large $message_headers, for example).
3537 6. Lookups now cache *every* query, not just the most recent. A new, separate
3538 store pool is used for this. It can be recovered when all lookup caches are
3539 flushed. Lookups now release memory at the end of their result strings.
3540 This has involved some general refactoring of the lookup sources.
3542 7. Some code has been added to the store_xxx() functions to reduce the amount
3543 of flapping under certain conditions.
3545 8. log_incoming_interface used to affect only the <= reception log lines. Now
3546 it causes the local interface and port to be added to several more SMTP log
3547 lines, for example "SMTP connection from", and rejection lines.
3549 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3551 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3553 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3554 could be overwritten at the end of the current message (or the start of a
3555 new message if it was set in a HELO ACL). The value is now preserved for
3556 the duration of the SMTP connection.
3558 12. If a transport had a headers_rewrite setting, and a matching header line
3559 contained an unqualified address, that address was qualified, even if it
3560 did not match any rewriting rules. The underlying bug was that the values
3561 of the flags that permit the existence of unqualified sender and recipient
3562 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3563 non-local messages, and by -bnq for local messages) were not being
3564 preserved with the message after it was received.
3566 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3567 "next input=" as part of the text comprising the host identity instead of
3568 the correct text. The code was using the same buffer for two different
3569 strings. However, depending on which order the printing function evaluated
3570 its arguments, the bug did not always show up. Under Linux, for example, my
3571 test suite worked just fine.
3573 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3574 doesn't work with some older versions of Perl. It has been changed to "my",
3575 which in any case is probably the better facility to use.
3577 15. A really picky compiler found some instances of statements for creating
3578 error messages that either had too many or two few arguments for the format
3581 16. The size of the buffer for calls to the DNS resolver has been increased
3582 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3583 for addresses that have a lot of PTR records. This alleviates a problem; it
3584 does not fully solve it.
3586 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3587 buffer now truncates the list and logs the incident, which is the same
3588 action as happens when Exim is looking up a host name and its aliases.
3589 Previously in this situation something unpredictable would happen;
3590 sometimes it was "internal error: store_reset failed".
3592 18. If a server dropped the connection unexpectedly when an Exim client was
3593 using GnuTLS and trying to read a response, the client delivery process
3594 crashed while trying to generate an error log message.
3596 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3597 string, for example:
3599 warn message = H1: something\nH2: something
3601 the text was added as a single header line from Exim's point of view
3602 though it ended up OK in the delivered message. However, searching for the
3603 second and subsequent header lines using $h_h2: did not work. This has been
3604 fixed. Similarly, if a system filter added multiple headers in this way,
3605 the routers could not see them.
3607 20. Expanded the error message when iplsearch is called with an invalid key to
3608 suggest using net-iplsearch in a host list.
3610 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3611 ACLs are no longer actually imposed (and a message to that effect is
3614 22. If a "gecos" field in a passwd entry contained escaped characters, in
3615 particular, if it contained a \" sequence, Exim got it wrong when building
3616 a From: or a Sender: header from that name. A second bug also caused
3617 incorrect handling when an unquoted " was present following a character
3618 that needed quoting.
3620 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3621 was not being matched caselessly.
3623 24. Arranged for all hyphens in the exim.8 source to be escaped with
3626 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3627 properly. Recipient callout cache records were still being keyed to include
3628 the sender, even when use_sender was set false. This led to far more
3629 callouts that were necessary. The sender is no longer included in the key
3630 when use_sender is false.
3632 26. Added "control = submission" modifier to ACLs.
3634 27. Added the ${base62d: operator to decode base 62 numbers.
3636 28. dnsdb lookups can now access SRV records.
3638 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3639 the configuration file.
3641 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3642 (-v) mode. This makes the output for a verbose queue run more intelligible.
3644 31. Added a use_postmaster feature to recipient callouts.
3646 32. Added the $body_zerocount variable, containing the number of binary zero
3647 bytes in the message body.
3649 33. The time of last modification of the "new" subdirectory is now used as the
3650 "mailbox time last read" when there is a quota error for a maildir
3653 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3655 35. Added +ignore_unknown as a special item in host lists.
3657 36. Code for decoding IPv6 addresses in host lists is now included, even if
3658 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3659 address was recognized as an IP address, but was then not correctly decoded
3660 into binary, causing unexpected and incorrect effects when compared with
3667 1. Very minor rewording of debugging text in manualroute to say "list of
3668 hosts" instead of "hostlist".
3670 2. If verify=header_syntax was set, and a header line with an unqualified
3671 address (no domain) and a large number of spaces between the end of the
3672 name and the colon was received, the reception process suffered a buffer
3673 overflow, and (when I tested it) crashed. This was caused by some obsolete
3674 code that should have been removed. The fix is to remove it!
3676 3. When running in the test harness, delay a bit after writing a bounce
3677 message to get a bit more predictability in the log output.
3679 4. Added a call to search_tidyup() just before forking a reception process. In
3680 theory, someone could use a lookup in the expansion of smtp_accept_max_
3681 per_host which, without the tidyup, could leave open a database connection.
3683 5. Added the variables $recipient_data and $sender_data which get set from a
3684 lookup success in an ACL "recipients" or "senders" condition, or a router
3685 "senders" option, similar to $domain_data and $local_part_data.
3687 6. Moved the writing of debug_print from before to after the "senders" test
3690 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3691 problems for message scanning, either using a data ACL, or using
3692 local_scan() because the Received: header was not generated till after they
3693 were called (in order to set the time as the time of reception completion).
3694 I have revised the way this works. The header is now generated after the
3695 body is received, but before the ACL or local_scan() are called. After they
3696 are run, the timestamp in the header is updated.
3702 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3703 before starting a queue runner without re-exec. This happened only when
3704 deliver_drop_privilege was set or when the Exim user was set to root. The
3705 effect of the bug was that timeouts during subsequent deliveries caused
3706 crashes instead of being properly handled. The handler is now left at its
3707 default (and expected) setting.
3709 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3710 message, again when deliver_drop_privilege is set or Exim is run as root.
3711 The bug described in (1) was not present in this case, but the tidying up
3712 of the other signals was missing. I have made the two cases consistent.
3714 3. The ignore_target_hosts setting on a manualroute router was being ignored
3715 for hosts that were looked up using the /MX notation.
3717 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3720 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3721 operated on the sender address. After changing the $sender_address to <>
3722 for the sender address verify, Exim was re-instated it as the original
3723 (before rewriting) address, but remembering that it had rewritten it, so it
3724 wasn't rewriting it again. This bug also had the effect of breaking the
3725 sender address verification caching when the sender address was rewritten.
3727 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3728 This has been changed so that if the ip literal address matches
3729 ignore_target_hosts, the router declines.
3731 7. Added expansion conditions match_domain, match_address, and match_local_
3732 part (NOT match_host).
3734 8. The placeholder for the Received: header didn't have a length field set.
3736 9. Added code to Exim itself and to exim_lock to test for a specific race
3737 condition that could lead to file corruption when using MBX delivery. The
3738 issue is with the lockfile that is created in /tmp. If this file is removed
3739 after a process has opened it but before that process has acquired a lock,
3740 there is the potential for a second process to recreate the file and also
3741 acquire a lock. This could lead to two Exim processes writing to the file
3742 at the same time. The added code performs the same test as UW imapd; it
3743 checks after acquiring the lock that its file descriptor still refers to
3744 the same named file.
3746 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3747 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3750 11. Added the smtp_active_hostname option. If used, this will typically be made
3751 to depend on the incoming interface address. Because $interface_address is
3752 not set up until the daemon has forked a reception process, error responses
3753 that can happen earlier (such as "too many connections") no longer contain
3756 12. If an expansion in a condition on a "warn" statement fails because a lookup
3757 defers, the "warn" statement is abandoned, and the next ACL statement is
3758 processed. Previously this caused the whole ACL to be aborted.
3760 13. Added the iplsearch lookup type.
3762 14. Added ident_timeout as a log selector.
3764 15. Added tls_certificate_verified as a log selector.
3766 16. Added a global option tls_require_ciphers (compare the smtp transport
3767 option of the same name). This controls incoming TLS connections.
3769 17. I finally figured out how to make tls_require_ciphers do a similar thing
3770 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3771 before starting the TLS session.
3773 18. Tabs are now shown as \t in -bP output.
3775 19. If the log selector return_path_on_delivery was set, Exim crashed when
3776 bouncing a message because it had too many Received: header lines.
3778 20. If two routers both had headers_remove settings, and the first one included
3779 a superfluous trailing colon, the final name in the first list and the
3780 first name in the second list were incorrectly joined into one item (with a
3781 colon in the middle).
3787 1. Added -C and -D options to the exinext utility, mainly to make it easier
3788 to include in the automated testing, but these could be helpful when
3789 multiple configurations are in use.
3791 2. The exinext utility was not formatting the output nicely when there was
3792 an alternate port involved in the retry record key, nor when there was a
3793 message id as well (for retries that were specific to a specific message
3794 and a specific host). It was also confused by IPv6 addresses, because of
3795 the additional colons they contain. I have fixed the IPv4 problem, and
3796 patched it up to do a reasonable job for IPv6.
3798 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3799 delivery, the log line now contains "pipelined" if PIPELINING was used.
3801 4. An SMTP transport process used to panic and die if the bind() call to set
3802 an explicit outgoing interface failed. This has been changed; it is now
3803 treated in the same way as a connect() failure.
3805 5. A reference to $sender_host_name in the part of a conditional expansion
3806 that was being skipped was still causing a DNS lookup. This no longer
3809 6. The def: expansion condition was not recognizing references to header lines
3810 that used bh_ and bheader_.
3812 7. Added the _cache feature to named lists.
3814 8. The code for checking quota_filecount in the appendfile transport was
3815 allowing one more file than it should have been.
3817 9. For compatibility with Sendmail, the command line option
3825 and sets the incoming protocol and host name (for trusted callers). The
3826 host name and its colon can be omitted when only the protocol is to be set.
3827 Note the Exim already has two private options, -pd and -ps, that refer to
3828 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3829 "s", but I don't think that's a major issue.
3831 10. A number of refactoring changes to the code, none of which should affect
3834 (a) The number of logging options was getting close to filling up the
3835 32-bit word that was used as a bit map. I have split them into two classes:
3836 those that are passed in the argument to log_write(), and those that are
3837 only ever tested independently outside of that function. These are now in
3838 separate 32-bit words, so there is plenty of room for expansion again.
3839 There is no change in the user interface or the logging behaviour.
3841 (b) When building, for example, log lines, the code previously used a
3842 macro that called string_cat() twice, in order to add two strings. This is
3843 not really sufficiently general. Furthermore, there was one instance where
3844 it was actually wrong because one of the argument was used twice, and in
3845 one call a function was used. (As it happened, calling the function twice
3846 did not affect the overall behaviour.) The macro has been replaced by a
3847 function that can join an arbitrary number of extra strings onto a growing
3850 (c) The code for expansion conditions now uses a table and a binary chop
3851 instead of a serial search (which was left over from when there were very
3852 few conditions). Also, it now recognizes conditions like "pam" even when
3853 the relevant support is not compiled in: a suitably worded error message is
3854 given if an attempt is made to use such a condition.
3856 11. Added ${time_interval:xxxxx}.
3858 12. A bug was causing one of the ddress fields not to be passed back correctly
3859 from remote delivery subprocesses. The field in question was not being
3860 subsequently used, so this caused to problems in practice.
3862 13. Added new log selectors queue_time and deliver_time.
3864 14. Might have fixed a bug in maildirsizefile handling that threw up
3865 "unexpected character" debug warnings, and recalculated the data
3866 unnecessarily. In any case, I expanded the warning message to give more
3869 15. Added the message "Restricted characters in address" to the statements in
3870 the default ACL that block characters like @ and % in local parts.
3872 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3873 Three changes have been made:
3875 (a) There was a serious bug; a negative response to MAIL caused the whole
3876 recipient domain to be cached as invalid, thereby blocking all messages
3877 to all local parts at the same domain, from all senders. This bug has
3878 been fixed. The domain is no longer cached after a negative response to
3879 MAIL if the sender used is not empty.
3881 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3884 (c) A new callout option, "use_sender" has been added for people who want
3885 the modified behaviour.
3891 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3894 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3895 indeed breaks things for older releases.
3897 3. Added additional logging to the case where there is a problem reading data
3898 from a filter that is running in a subprocess using a pipe, in order to
3899 try to track down a specific problem.
3901 4. Testing facility fudge: when running in the test harness and attempting
3902 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3903 getting "No route to host". Convert this to a timeout.
3905 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3908 6. Some OS don't have socklen_t but use size_t instead. This affects the
3909 fifth argument of getsockopt() amongst other things. This is now
3910 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3911 can be set for individual OS. I have set it for SunOS5, OSF1, and
3912 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3913 some earlier ones do not.
3915 7. Change 4.30/15 was not doing the test caselessly.
3917 8. The standard form for an IPv6 address literal was being rejected by address
3918 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3919 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3920 this, as well as the form without the "IPv6" on the front (but only when
3921 address literals are enabled, of course).
3923 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3925 10. Exim crashed if a message with an empty sender address specified by -f
3926 encountered a router with an errors_to setting. This could be provoked only
3927 by a command such as
3931 where an empty string was supplied; "<>" did not hit this bug.
3933 11. Installed PCRE release 4.5.
3935 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3936 remained set. It is now erased.
3938 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3939 times from message ids (which are base 36 rather than the normal 62).
3941 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3942 were being counted as actual protocol errors, and logged if the log
3943 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3944 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3945 and DATA following a set of rejected RCPTs do not count as protocol errors.
3946 In other words, Exim assumes they were pipelined, though this may not
3947 actually be the case. Of course, in all cases the client gets an
3948 appropriate error code.
3950 15. If a lookup fails in an ACL condition, a message about the failure may
3951 be available; it is used if testing the ACL cannot continue, because most
3952 such messages specify what the cause of the deferral is. However, some
3953 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3954 that caused an old message to be retained and used if a later statement
3955 caused a defer, replacing the real cause of the deferral.
3957 16. If an IP address had so many PTR records that the DNS lookup buffer
3958 was not large enough to hold them, Exim could crash while trying to process
3959 the truncated data. It now detects and logs this case.
3961 17. Further to 4.21/58, another change has been made: if (and only if) the
3962 first line of a message (the first header line) ends with CRLF, a bare LF
3963 in a subsequent header line has a space inserted after it, so as not to
3964 terminate the header.
3966 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3967 unnecessarily, used atoi() instead of strtol(), and didn't check the
3968 termination when getting file sizes from file names by regex.
3970 19. Completely re-implemented the support for maildirsize files, in the light
3971 of a number of problems with the previous contributed implementation
3972 (4.30/29). In particular:
3974 . If the quota is zero, the maildirsize file is maintained, but no quota is
3977 . If the maildir directory does not exist, it is created before any attempt
3978 to write a maildirsize file.
3980 . The quota value in the file is just a cache; if the quota is changed in
3981 the transport, the new value overrides.
3983 . A regular expression is available for excluding directories from the
3986 20. The autoreply transport checks the characters in options that define the
3987 message's headers; it allows continued headers, but it was checking with
3988 isspace() after an embedded newline instead of explicitly looking for a
3991 21. If all the "regular" hosts to which an address was routed had passed their
3992 expiry times, and had not reached their retry times, the address was
3993 bounced, even if fallback hosts were defined. Now Exim should go on to try
3996 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3997 equivalent code in the SMTP transport. Some hosts send humungous responses
3998 to HELO/EHLO, more than 1024 it seems.
4000 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4001 gives compiler warnings in some environments. I've now done it "properly",
4004 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4005 (because of problems with the built-in one) was declared to return uschar *
4006 instead of char *, causing compiler failure.
4008 25. Fixed a file descriptor leak when processing alias/forward files.
4010 26. Fixed a minor format string issue in dbfn.c.
4012 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4014 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4015 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4018 29. When a sender address is verified, it is cached, to save repeating the test
4019 when there is more than one recipient in a message. However, when the
4020 verification involves a callout, it is possible for different callout
4021 options to be set for different recipients. It is too complicated to keep
4022 track of this in the cache, so now Exim always runs a verification when a
4023 callout is required, relying on the callout cache for the optimization.
4024 The overhead is duplication of the address routing, but this should not be
4027 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4028 to be verified with callout=postmaster, and the main callout worked but the
4029 postmaster check failed, the verification correctly failed. However, if a
4030 subsequent RCPT command asked for sender verification *without* the
4031 postmaster check, incorrect caching caused this verification also to fail,
4034 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4035 it was not caching the DNS options (qualify_single, search_parents) that
4036 were used when the lookup failed. A subsequent lookup with different
4037 options therefore always gave the same answer, though there were cases
4038 where it should not have. (Example: a "domains = !$mx_any" option on a
4039 dnslookup router: the "domains" option is always processed without any
4040 widening, but the router might have qualify_single set.) Now Exim uses the
4041 cached value only when the same options are set.
4043 32. Added John Jetmore's "exipick" utility to the distribution.
4045 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4046 than a timeout (e.g. a certificate is required, and is not provided), an
4047 Exim server now closes the connection immediately. Previously it waited for
4048 the client to close - but if the client is SSL, it seems that they each
4049 wait for each other, leading to a delay before one of them times out.
4051 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4052 maintained 0.8.x compatibility because I don't think many are using it, and
4053 it is clearly obsolete.
4055 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4058 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4059 client certificate was expired. A simple patch fixes this, though I don't
4060 understand the full logic of why the verify callback is called multiple
4063 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4064 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4065 which causes problems with some clients (such as the Certicom SSL Plus
4066 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4067 disables the coutermeasure allowing Eudora to connect."
4069 38. Exim was not checking that a write() to a log file succeeded. This could
4070 lead to Bad Things if a log got too big, in particular if it hit a file
4071 size limit. Exim now panics and dies if it cannot write to a log file, just
4072 as it does if it cannot open a log file.
4074 39. Modified OS/Makefile-Linux so that it now contains
4076 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4078 The two -D definitions ensure that Exim is compiled with large file
4079 support, which makes it possible to handle log files that are bigger than
4082 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4083 instance) a domain was checked against a named list that involved a lookup,
4084 causing $domain_data to be set, then another domain was checked against the
4085 same list, then the first domain was re-checked, the value of $domain_data
4086 after the final check could be wrong. In particular, if the second check
4087 failed, it could be set empty. This bug probably also applied to
4090 41. The strip_trailing_dot option was not being applied to the address given
4091 with the -f command-line option.
4093 42. The code for reading a message's header from the spool was incrementing
4094 $received_count, but never initializing it. This meant that the value was
4095 incorrect (doubled) while delivering a message in the same process in which
4096 it was received. In the most common configuration of Exim, this never
4097 happens - a fresh exec is done - but it can happen when
4098 deliver_drop_privilege is set.
4100 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4101 it now includes up to 150 characters of the unexpected data in the log
4104 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4105 and building data strings. The size of both of these buffers was 10 000
4106 bytes - far larger than anybody would *ever* want, thought I. Needless to
4107 say, somebody hit the limit. I have increased the maximum line length to
4108 20 000 and the maximum data length of concatenated lines to 100 000. I have
4109 also fixed two bugs, because there was no checking on these buffers. Tsk,
4110 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4111 buffer is too small.
4113 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4114 lsearch lookups. Now it does.
4116 46. When parsing a route_list item in a manualroute router, a fixed-length
4117 buffer was used for the list of hosts. I made this 1024 bytes long,
4118 thinking that nobody would ever have a list of hosts that long. Wrong.
4119 Somebody had a whole pile of complicated expansion conditions, and the
4120 string was silently truncated, leading to an expansion error. It turns out
4121 that it is easier to change to an unlimited length (owing to other changes
4122 that have happened since this code was originally written) than to build
4123 structure for giving a limitation error. The length of the item that
4124 expands into the list of hosts is now unlimited.
4126 47. The lsearch lookup could not handle data where the length of text line was
4127 more than 4095 characters. Such lines were truncated, leading to shortened
4128 data being returned. It should now handle lines of any length.
4130 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4131 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4134 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4135 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4136 no longer happens in comment lines. A list of replacements is now placed
4137 at the head of all of the source files, except those whose only change is
4138 to replace PERL_COMMAND in the very first #! line.
4140 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4141 messages on the queue, with a bottom-up merge sort, using code contributed
4142 by Michael Haardt. This should make operations like -bp somewhat faster on
4143 large queues. It won't affect queue runners, except when queue_run_in_order
4146 51. Installed eximstats 1.31 in the distribution.
4148 52. Added support for SRV lookups to the dnslookup router.
4150 53. If an ACL referred to $message_body or $message_body_end, the value was not
4151 reset for any messages that followed in the same SMTP session.
4153 54. The store-handling optimization for building very long strings was not
4154 differentiating between the different store pools. I don't think this
4155 actually made any difference in practice, but I've tidied it.
4157 55. While running the routers to verify a sender address, $sender_address
4158 was still set to the sender address. This is wrong, because when routing to
4159 send a bounce to the sender, it would be empty. Therefore, I have changed
4160 it so that, while verifying a sender address, $sender_address is set to <>.
4161 (There is no change to what happens when verifying a recipient address.)
4163 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4164 target A or AAAA records (if not already returned) without resetting the
4165 qualify_single or search_parents options of the DNS resolver. These are
4166 inappropriate in this case because the targets of MX and SRV records must
4167 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4168 target that, when qualified, matched something in the local domain. These
4169 two options are now turned off when doing these lookups.
4171 57. It seems that at least some releases of Reiserfs (which does not have the
4172 concept of a fixed number of inodes) returns zero and not -1 for the
4173 number of available inodes. This interacted badly with check_spool_inodes,
4174 which assumed that -1 was the "no such thing" setting. What I have done is
4175 to check that the total number of inodes is greater than zero before doing
4176 the test of how many are available.
4178 58. When a "warn" ACL statement has a log_message modifier, the message is
4179 remembered, and not repeated. This is to avoid a lot of repetition when a
4180 message has many recipients that cause the same warning to be written.
4181 Howewer, Exim was preserving the list of already written lines for an
4182 entire SMTP session, which doesn't seem right. The memory is now reset if a
4183 new message is started.
4185 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4186 debugging output unless log_rewrite was also set.
4188 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4189 of (int)(handle) when we know that handle contains (void *)(-1).
4191 61. The Exim daemon panic-logs an error return when it closes the incoming
4192 connection. However "connection reset by peer" seems to be common, and
4193 isn't really an error worthy of noting specially, so that particular error
4196 62. When Exim is trying to find all the local interfaces, it used to panic and
4197 die if the ioctl to get the interface flags failed. However, it seems that
4198 on at least one OS (Solaris 9) it is possible to have an interface that is
4199 included in the list of interfaces, but for which you get a failure error
4200 for this call. This happens when the interface is not "plumbed" into a
4201 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4202 failure of the "get flags" call assumes that the interface is down.
4204 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4205 makes life easier for people who are doing arithmetic on fields extracted
4206 from dates, where you often get leading zeros that should not be
4207 interpreted as octal.
4209 64. Added qualify_domain to the redirect router, to override the global
4212 65. If a pathologically long header line contained very many addresses (the
4213 report of this problem mentioned 10 000) and each of them was rewritten,
4214 Exim could use up a very large amount of memory. (It kept on making new
4215 copies of the header line as it rewrote, and never released the old ones.)
4216 At the expense of a bit more processing, the header rewriting function has
4217 been changed so that it no longer eats memory in this way.
4219 66. The generation of the Received: header has been moved from the time that a
4220 message starts to be received, to the time that it finishes. The timestamp
4221 in the Received: header should now be very close to that of the <= log
4222 line. There are two side-effects of this change:
4224 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4225 logged header lines no longer include the local Received: line, because
4226 it has not yet been created. The same applies to a copy of the message
4227 that is returned to a non-SMTP sender when a message is rejected.
4229 (b) When a filter file is tested using -bf, no additional Received: header
4230 is added to the test message. After some thought, I decided that this
4233 This change does not affect the value of $received_for. It is still set
4234 after address rewriting, but before local_scan() is called.
4236 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4238 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4239 gave an unhelpful panic error message, and a defer error. I have managed to
4240 change this behaviour so that it now rejects any supplied certificate,
4241 which seems right, as the list of acceptable certificates is empty.
4243 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4244 gave an unhelpful defer error. I have not managed to make this reject any
4245 supplied certificates, but the error message it gives is "no certificate
4246 supplied", which is not helpful.
4248 70. exigrep's output now also includes lines that are not associated with any
4249 message, but which match the given pattern. Implemented by a patch from
4250 Martin Sluka, which also tidied up the Perl a bit.
4252 71. Recipient callout verification, like sender verification, was using <> in
4253 the MAIL FROM command. This isn't really the right thing, since the actual
4254 sender may affect whether the remote host accepts the recipient or not. I
4255 have changed it to use the actual sender in the callout; this means that
4256 the cache record is now keyed on a recipient/sender pair, not just the
4257 recipient address. There doesn't seem to be a real danger of callout loops,
4258 since a callout by the remote host to check the sender would use <>.
4259 [SEE ABOVE: changed after hitting problems.]
4261 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4262 temporary errors. However, in the case of such a code being given after
4263 the end of a data transmission (i.e. after ".") Exim was failing to write
4264 a retry record for the message. (Yes, there was some broken host that was
4265 actually sending 8xx at this point.)
4267 73. An unknown lookup type in a host list could cause Exim to panic-die when
4268 the list was checked. (An example that provoked this was putting <; in the
4269 middle of a list instead of at the start.) If this happened during a DATA
4270 ACL check, a -D file could be left lying around. This kind of configuration
4271 error no longer causes Exim to die; instead it causes a defer errror. The
4272 incident is still logged to the main and panic logs.
4274 74. Buglet left over from Exim 3 conversion. The message "too many messages
4275 in one connection" was written to the rejectlog but not the mainlog, except
4276 when address rewriting (yes!) was being logged.
4278 75. Added write_rejectlog option.
4280 76. When a system filter was run not as root (that is, when system_filter_user
4281 was set), the values of the $n variables were not being returned to the
4282 main process; thus, they were not subsequently available in the $sn
4285 77. Added +return_path_on_delivery log selector.
4287 78. A connection timeout was being treated differently from recipients deferred
4288 when testing hosts_max_try with a message that was older than the host's
4289 retry timeout. (The host should not be counted, thus allowing all hosts to
4290 be tried at least once before bouncing.) This may have been the cause of an
4291 occasionally reported bug whereby a message would remain on the queue
4292 longer than the retry timeout, but would be bounced if a delivery was
4293 forced. I say "may" because I never totally pinned down the problem;
4294 setting up timeout/retry tests is difficult. See also the next item.
4296 79. The ultimate address timeout was not being applied to errors that involved
4297 a combination of host plus message (for example, a timeout on a MAIL
4298 command). When an address resolved to a number of possible hosts, and they
4299 were not all tried for each delivery (e.g. because of hosts_max_try), a
4300 message could remain on the queue longer than the retry timeout.
4302 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4305 81. Fixed an obscure SMTP outgoing bug which required at least the following
4306 conditions: (a) there was another message waiting for the same server;
4307 (b) the server returned 5xx to all RCPT commands in the first message so
4308 that the message was not completed; (c) the server dropped the connection
4309 or gave a negative response to the RSET that Exim sends to abort the
4310 transaction. The observed case was a dropped connection after DATA that had
4311 been sent in pipelining mode. That is, the server had advertised PIPELINING
4312 but was not implementing it correctly. The effect of the bug was incorrect
4313 behaviour, such as trying another host, and this could lead to a crash.
4319 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4320 and daemon.c were passed as pointers to ints; they should have been
4321 pointers to socklen_t variables (which are typically unsigned ints).
4323 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4326 3. Fixed a really odd bug that affected only the testing scheme; patching a
4327 certain fixed string in the binary changed the value of another string that
4328 happened to be identical to the end of the original first string.
4330 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4331 name", it returns that address as the IP address. On some operating
4332 systems (e.g. Solaris), it also passes back the IP address string as the
4333 "host name". However, on others (e.g. Linux), it passes back an empty
4334 string. Exim wasn't checking for this, and was changing the host name to an
4335 empty string, assuming it had been canonicized.
4337 5. Although rare, it is permitted to have more than one PTR record for a given
4338 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4339 all the names associated with an address, because they do in Solaris.
4340 However, it seems that they do not in Linux for data that comes from the
4341 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4342 I found this out when I moved to a new Linux workstation and tried to run
4343 the Exim test suite.
4345 To get round this problem I have changed the code so that it now does its
4346 own call to the DNS to look up PTR records when searching for a host name.
4347 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4348 addresses that are only in /etc/hosts are still found.
4350 This behaviour is, however, controlled by an option called host_lookup_
4351 order, which defaults to "bydns:byaddr". If people want to use the other
4352 order, or indeed, just use one or the other means of lookup, they can
4353 specify it in this variable.
4355 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4356 some operating systems, this comes back from gethostbyaddr() as an empty
4357 string, and this is what Exim used to test for. However, it seems that in
4358 other systems, "." is yielded. Exim now tests for this case too.
4360 7. The values of check_spool_space and check_log_space are now held internally
4361 as a number of kilobytes instead of an absolute number of bytes. If a
4362 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4363 kilobyte. This means that much larger values can be stored.
4365 8. Exim monitor: an attempt to get the action menu when not actually pointing
4366 at a message produces an empty menu entitled "No message selected". This
4367 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4368 no entries in it ("Shell widget menu has zero width and/or height"). So I
4369 have added a single, blank menu entry in this case.
4371 9. Added ${quote_local_part.
4373 10. MIME decoding is now applied to the contents of Subject: header lines when
4376 11. Now that a reference to $sender_host_address automatically causes a reverse
4377 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4378 host lookup before query-style lookups in lists that might use this
4379 variable. This has therefore been abolished, and the "net-" prefix is no
4380 longer necessary for query-style lookups.
4382 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4383 have been a typo for LFLAGS, so it has been changed.
4385 13. The install script calls Exim with "-C /dev/null" in order to find the
4386 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4387 to be output. Howeve, since Exim outputs its version number before the
4388 error, it didn't break the script. It just looked ugly. I fixed this by
4389 always allowing "-C /dev/null" if the caller is root.
4391 14. Ignore overlarge ACL variable number when reading spool file - insurance
4392 against a later release with more variables having written the file.
4394 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4395 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4396 this, as well as the form without the "IPv6" on the front.
4398 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4399 OS/Makefile-Darwin file.
4401 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4402 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4404 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4405 variable. However, some OS use ints for inodes. I've added cast to long int
4406 to get rid of the compiler warning.
4408 19. I had forgotten to lock out "/../" in configuration file names when
4409 ALT_CONFIG_PREFIX was set.
4411 20. Routers used for verification do not need to specify transports. However,
4412 if such a router generated a host list, and callout was configured, Exim
4413 crashed, because it could not find a port number from the (non-existent)
4414 transport. It now assumes port 25 in this circumstance.
4416 21. Added the -t option to exigrep.
4418 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4419 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4420 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4421 them is set, LOOKUP_LSEARCH is forced.
4423 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4425 24. Added sender and host information to the "rejected by local_scan()" log
4426 line; previously there was no indication of these.
4428 25. Added .include_if_exists.
4430 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4431 Linux. It turns out that not all file systems support this. Apparently some
4432 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4433 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4434 not supported on the file descriptor, is now ignored when Exim is trying to
4435 sync a directory. This applies only to Linux.
4437 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4439 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4440 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4443 29. Added support for maildirsize files from supplied patch (modified a bit).
4445 30. The use of :fail: followed by an empty string could lead Exim to respond to
4446 sender verification failures with (e.g.):
4448 550 Verification failed for <xxx>
4449 550 Sender verify failed
4451 where the first response line was missing the '-' that indicates it is not
4452 the final line of the response.
4454 31. The loop for finding the name of the user that called Exim had a hardwired
4455 limit of 10; it now uses the value of finduser_retries, which is used for
4456 all other user lookups.
4458 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4461 33. Exim was neglecting to zero errno before one call of strtol() when
4462 expanding a string and expecting an integer value. On some systems this
4463 resulted in spurious "integer overflow" errors. Also, it was casting the
4464 result into an int without checking.
4466 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4467 did not work. The code looks as if it has *never* worked, though it appears
4468 to have been documented since at least releast 1.62. I have made it work.
4470 35. The "timeout_DNS" error in retry rules, also documented since at least
4471 1.62, also never worked. As it isn't clear exactly what this means, and
4472 clearly it isn't a major issue, I have abolished the feature by treating it
4473 as "timeout", and writing a warning to the main and panic logs.
4475 36. The display of retry rules for -brt wasn't always showing the error code
4478 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4479 timeout_connect_A, timeout_connect_MX.
4481 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4482 to the empty sender.
4484 39. The daemon was not analysing the content of -oX till after it had closed
4485 stderr and disconnected from the controlling terminal. This meant that any
4486 syntax errors were only noted on the panic log, and the return code from
4487 the command was 0. By re-arranging the code a little, I've made the
4488 decoding happen first, so such errors now appear on stderr, and the return
4489 code is 1. However, the actual setting up of the sockets still happens in
4490 the disconnected process, so errors there are still only recorded on the
4493 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4494 connections (as happens on some IP stacks) was logged at start up time as
4495 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4496 it from "IPv6 and IPv4", which means that two separate sockets are being
4499 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4500 says whether AF_INET or AF_INET6 was passed as an argument.
4502 42. Exiwhat output was messed up when time zones were included in log
4505 43. Exiwhat now gives more information about the daemon's listening ports,
4506 and whether -tls-on-connect was used.
4508 44. The "port" option of the smtp transport is now expanded.
4510 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4511 silently ignored. Now an error message is written to the main and panic
4514 46. There's a new ACL modifier called "logwrite" which writes to a log file
4515 as soon as it is encountered.
4517 47. Added $local_user_uid and $local_user_gid at routing time.
4519 48. Exim crashed when trying to verify a sender address that was being
4522 49. Exim was recognizing only a space character after ".include". It now also
4523 recognizes a tab character.
4525 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4526 extracting the relevant information from the specification. The man page no
4527 longer contains scrambled data for the -d option, and I've added a section
4528 at the front about calling Exim under different names.
4530 51. Added "extra_headers" argument to the "mail" command in filter files.
4532 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4535 53. Installed eximstats 1.29.
4537 54. Added transport_filter_timeout as a generic transport option.
4539 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4540 Cc: header lines. This was required by RFC 822, but it not required by RFC
4543 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4544 incoming messages that did not have them. Now it does so only if the
4545 message originates locally, that is, if there is no associated remote host
4546 address. When Resent- header lines are present, this applies to the Resent-
4547 lines rather than the non-Resent- lines.
4549 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4550 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4552 58. Messages for configuration errors now include the name of the main
4553 configuration file - useful now that there may be more than one file in a
4554 list (.included file names were always shown).
4556 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4557 for those rare installations that do not start the daemon as root or run it
4558 setuid root. I've cut out the call to initgroups() if the daemon is not
4561 60. The Exim user and group can now be bound into the binary as text strings
4562 that are looked up at the start of Exim's processing.
4564 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4566 62. Added $mailstore_basename variable.
4568 63. Installed patch to sieve.c from Michael Haardt.
4570 64. When Exim failed to open the panic log after failing to open the main log,
4571 the original message it was trying to log was written to stderr and debug
4572 output, but if they were not available (the usual case in production), it
4573 was lost. Now it is written to syslog before the two lines that record the
4574 failures to open the logs.
4576 65. Users' Exim filters run in subprocesses under the user's uid. It is
4577 possible for a "deliver" command or an alias in a "personal" command to
4578 provoke an address rewrite. If logging of address rewriting is configured,
4579 this fails because the process is not running as root or exim. There may be
4580 a better way of dealing with this, but for the moment (because 4.30 needs
4581 to be released), I have disabled address rewrite logging when running a
4582 filter in a non-root, non-exim process.
4588 1. The buildconfig auxiliary program wasn't quoting the value set for
4589 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4590 not defined. This bug was masked in 4.22 by the effect that was fixed in
4593 2. Some messages that were rejected after a message id was allocated were
4594 shown as "incomplete" by exigrep. It no longer does this for messages that
4595 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4597 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4598 have allow_domain_literals set, the ID did not get logged in the <= line.
4599 Domain literals are now always recognized in Message-ID: header lines.
4601 4. The first argument for a ${extract expansion item is the key name or field
4602 number. Leading and trailing spaces in this item were not being ignored,
4603 causing some misleading effects.
4605 5. When deliver_drop_privilege was set, single queue runner processes started
4606 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4607 same command in the process it spins off) were not dropping privilege.
4609 6. When the daemon running as "exim" started a queue runner, it always
4610 re-executed Exim in the spun-off process. This is a waste of effort when
4611 deliver_drop_privilege is set. The new process now just calls the
4612 queue-runner function directly.
4618 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4621 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4622 ignored. Though the use of -oP was forcing the writing of a pid file, it
4623 was always written to the default place.
4625 3. If the message "no IP address found for host xxxx" is generated during
4626 incoming verification, it is now followed by identification of the incoming
4627 connection (so you can more easily find what provoked it).
4629 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4631 5. Added some features to "harden" Exim a bit more against certain attacks:
4633 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4634 be put in Local/Makefile. This is like the never_users runtime option,
4635 but it cannot be overridden. The default setting is "root".
4637 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4638 prefix string with which any file named in a -C command line option
4641 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4642 is retained for -C and -D only if the caller of Exim is root. Without
4643 it, the exim user may also use -C and -D and retain privilege.
4645 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4646 command line option is disabled.
4648 6. Macro names set by the -D option must start with an upper case letter, just
4649 like macro names defined in the configuration file.
4651 7. Added "dereference=" facility to LDAP.
4653 8. Two instances of the typo "uknown" in the source files are fixed.
4655 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4656 the Configure-Makefile script screwed up while processing it.
4658 10. Incorporated PCRE 4.4.
4660 11. The SMTP synchronization check was not operating right at the start of an
4661 SMTP session. For example, it could not catch a HELO sent before the client
4662 waited for the greeting. There is now a check for outstanding input at the
4663 point when the greeting is written. Because of the duplex, asynchronous
4664 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4665 way, but not yet received, when the check is performed.
4667 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4668 on TCP/IP sockets, because this apparently causes some broken clients to
4671 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4672 unchanged) from the Cygwin maintainer.
4674 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4675 was supported instead of testing for mbx. Effectively a typo.
4677 15. The spa authenticator server code was not checking that the input it
4678 received was valid base64.
4680 16. The debug output line for the "set" modifier in ACLs was not showing the
4681 name of the variable that was being set.
4683 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4685 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4686 lookup on its own. Something else was needed to trigger the lookup. For
4687 example, a match in host_lookup or the need for a host name in a host list.
4688 Now, if $sender_host_name is referenced and the host name has not yet been
4689 looked up, a lookup is performed. If the lookup fails, the variable remains
4690 empty, and $host_lookup_failed is set to "1".
4692 19. Added "eqi" as a case-independent comparison operator.
4694 20. The saslauthd authentication condition could segfault if neither service
4695 nor realm was specified.
4697 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4698 error message that was logged was misleading, and incoming SMTP
4699 connections were dropped. The message is now more accurate, and temporary
4700 errors are given to SMTP connections.
4702 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4703 (except RSET) with a 421 error, until QUIT is received. However, it was
4704 failing to send a response to QUIT.
4706 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4707 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4708 ACL is now run after the helo_try_verify_hosts code.
4710 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4711 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4712 case-independent, so other case variants are also recognized). Apparently
4713 some systems use these upper case variants.
4715 25. If more than two messages were waiting for the same host, and a transport
4716 filter was specified for the transport, Exim sent two messages over the
4717 same TCP/IP connection, and then failed with "socket operation on non-
4718 socket" when it tried to send the third.
4720 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4722 27. The extern definition of crypt16() in expand.c was not being excluded when
4723 the OS had its own crypt16() function.
4725 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4726 as a preferred synonym for return_size_limit, both as an option and as an
4729 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4731 30. Changed the default configuration ACL to relax the local part checking rule
4732 for addresses that are not in any local domains. For these addresses,
4733 slashes and pipe symbols are allowed within local parts, but the sequence
4734 /../ is explicitly forbidden.
4736 31. SPA server authentication was not clearing the challenge buffer before
4739 32. log_message in a "warn" ACL statement was writing to the reject log as
4740 well as to the main log, which contradicts the documentation and doesn't
4741 seem right (because no rejection is happening). So I have stopped it.
4743 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4744 However, I am unable to do any testing of this.
4746 34. Fixed an infelicity in the appendfile transport. When checking directories
4747 for a mailbox, to see if any needed to be created, it was accidentally
4748 using path names with one or more superfluous leading slashes; tracing
4749 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4751 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4752 contents are added to the log line that is written for every discarded
4753 recipient. (Previously a log_message setting was ignored.)
4755 36. The ${quote: operator now quotes the string if it is empty.
4757 37. The install script runs exim in order to find its version number. If for
4758 some reason other than non-existence or emptiness, which it checks, it
4759 could not run './exim', it was installing it with an empty version number,
4760 i.e. as "exim-". This error state is now caught, and the installation is
4763 38. An argument was missing from the function that creates an error message
4764 when Exim fails to connect to the socket for saslauthd authentication.
4765 This could cause Exim to crash, or give a corrupted message.
4767 39. Added isip, isip4, and isip6 to ${if conditions.
4769 40. The ACL variables $acl_xx are now saved with the message, and can be
4770 accessed later in routers, transports, and filters.
4772 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4773 strings in the file are not string-expanded.
4775 42. If a MAIL command specified a SIZE value that was too large to fit into an
4776 int variable, the check against message_size_limit failed. Such values are
4777 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4778 day this will have to be increased, but I don't think I want to be around
4779 when emails are that large.
4786 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4787 iconv() is not standard in FreeBSD.
4789 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4790 IPv6 enabled. The observed symptom was a segmentation fault on return from
4791 the function os_common_find_running_interfaces() in src/os.c.
4793 3. In the check_special_case() function in daemon.c I had used "errno" as an
4794 argument name, which causes warnings on some systems. This was basically a
4795 typo, since it was named "eno" in the comments!
4797 4. The code that waits for the clock to tick (at a resolution of some fraction
4798 of a second) so as to ensure message-id uniqueness was always waiting for
4799 at least one whole tick, when it could have waited for less. [This is
4800 almost certainly not relevant at current processor speeds, where it is
4801 unlikely to ever wait at all. But we try to future-proof.]
4803 5. The function that sleeps for a time interval that includes fractions of a
4804 second contained a race. It did not block SIGALRM between setting the
4805 timer, and suspending (a couple of lines later). If the interval was short
4806 and the sigsuspend() was delayed until after it had expired, the suspension
4807 never ended. On busy systems this could lead to processes getting stuck for
4810 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4811 process, before it forks any delivery processes. The open lookup caching
4812 mechanism meant that the open file or database connection was passed into
4813 the delivery process. The problem was that delivery processes always tidy
4814 up cached lookup data. This could cause a problem for the next delivery
4815 process started by the queue runner, because the external queue runner
4816 process does not know about the closure. So the next delivery process
4817 still has data in the lookup cache. In the case of a file lookup, there was
4818 no problem because closing a file descriptor in a subprocess doesn't affect
4819 the parent. However, if the lookup was caching a connection to a database,
4820 the connection was closed, and the second delivery process was likely to
4821 see errors such as "PGSQL: query failed: server closed the connection
4822 unexpectedly". The problem has been fixed by closing all cached lookups
4823 in a queue runner before running a delivery process.
4825 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4826 seem to have the "const" qualifier which it has on other OS. I've
4829 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4830 *of the same type* (client or server) with the same public name that an
4831 error should be diagnosed.
4833 9. When Exim looked up a host name for an IP address, but failed to find the
4834 original IP address when looking up the host name (a safety check), it
4835 output the message "<ip address> does not match any IP for NULL", which was
4836 confusing, to say the least. The bug was that the host name should have
4837 appeared instead of "NULL".
4839 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4840 user that is built into the binary, and the -C or -D options is used, root
4841 privilege is dropped before the configuration file is read. In addition,
4842 logging is switched to stderr instead of the normal log files. If the
4843 configuration then re-defines the Exim user, the unprivileged environment
4844 is probably not what is expected, so Exim logs a panic warning message (but
4847 However, if deliver_drop_privilege is set, the unprivileged state may well
4848 be exactly what is intended, so the warning has been cut out in that case,
4849 and Exim is allowed to try to write to its normal log files.
4855 1. smtp_return_error_details was not giving details for temporary sender
4856 or receiver verification errors.
4858 2. Diagnose a configuration error if two authenticators have the same public
4861 3. Exim used not to create the message log file for a message until the first
4862 delivery attempt. This could be confusing when incoming messages were held
4863 for policy or load reasons. The message log file is now created at the time
4864 the message is received, and an initial "Received" line is written to it.
4866 4. The automatically generated man page for command line options had a minor
4867 bug that caused no ill effects; however, a more serious problem was that
4868 the procedure for building the man page automatically didn't always
4869 operate. Consequently, release 4.20 contains an out-of-date version. This
4870 shouldn't happen again.
4872 5. When building Exim with embedded Perl support, the script that builds the
4873 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4874 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4876 6. The freeze_tell option was not being used for messages that were frozen on
4877 arrival, either by an ACL or by local_scan().
4879 7. Added the smtp_incomplete_transaction log selector.
4881 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4882 was accepting AUTH without a new EHLO.
4884 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4885 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4888 10. Logging of TCP/IP connections (when configured) now happens in the main
4889 daemon process instead of the child process, so that the TCP/IP connection
4890 count is more accurate (but it can never be perfect).
4892 11. The use of "drop" in a nested ACL was not being handled correctly in the
4893 outer ACL. Now, if condition failure induced by the nested "drop" causes
4894 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4895 or "require"), the connection is dropped.
4897 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4898 that yield "discard" can now be used with an "accept" or a "discard" verb,
4899 but an error is generated for any others (because I can't see a useful way
4900 to define what should happen).
4902 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4903 are now processed in the same way as lines in the Exim configuration file.
4904 In particular, continuation lines are supported.
4906 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4908 15. Added -ti meaning -t -i.
4910 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4911 domains, and warn by logging if others are found.
4913 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4914 returned from GIFCONF when Exim is trying to find the list of interfaces on
4915 a host. The code in os.c has been modified to copy each ifreq to an aligned
4916 structure in all cases.
4918 Also, in some cases, the returned ifreq's were being copied to a 'struct
4919 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4920 means the last couple of bytes of an IPv6 address could be chopped if the
4921 ifreq contained only a normal sockaddr (14 bytes storage).
4923 18. Named domain lists were not supported in the hosts_treat_as_local option.
4924 An entry such as +xxxx was not recognized, and was treated as a literal
4927 19. Ensure that header lines added by a DATA ACL are included in the reject log
4928 if the ACL subsequently rejects the message.
4930 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4931 MD5 (which is deprecated).
4933 21. When testing a filter file using -bf, Exim was writing a message when it
4934 took the sender from a "From " line in the message, but it was not doing so
4935 when it took $return_path from a Return-Path: header line. It now does.
4937 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4938 with a valid header line field name (a series of printing characters
4939 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4941 23. Changed "disc" in the source to "disk" to conform to the documentation and
4942 the book and for uniformity.
4944 24. Ignore Sendmail's -Ooption=value command line item.
4946 25. When execve() failed while trying to run a command in a pipe transport,
4947 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4948 could be confused with a return value of 69 from the command itself. This
4949 has been changed to 127, the value the shell returns if it is asked to run
4950 a non-existent command. The wording for the related log line suggests a
4951 non-existent command as the problem.
4953 26. If received_header_text expands to an empty string, do not add a Received:
4954 header line to the message. (Well, it adds a token one on the spool, but
4955 marks it "old" so that it doesn't get used or transmitted.)
4957 27. Installed eximstats 1.28 (addition of -nt option).
4959 28. There was no check for failure on the call to getsockname() in the daemon
4960 code. This can fail if there is a shortage of resources on the system, with
4961 ENOMEM, for example. A temporary error is now given on failure.
4963 29. Contrary to the C standard, it seems that in some environments, the
4964 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4965 program. Exim now does this explicitly; it affects the formatting of
4966 timestamps using strftime().
4968 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4969 complaints. I've now initialized all the variables, to avoid this.
4971 32. Header lines added by a system filter were not being "seen" during
4972 transport-time rewrites.
4974 33. The info_callback() function passed to OpenSSL is set up with type void
4975 (*)(SSL *, int, int), as described somewhere. However, when calling the
4976 function (actually a macro) that sets it up, the type void(*)() is
4977 expected. I've put in a cast to prevent warnings from picky compilers.
4979 34. If a DNS black list lookup found a CNAME record, but there were no A
4980 records associated with the domain it pointed at, Exim crashed.
4982 35. If a DNS black list lookup returned more than one A record, Exim ignored
4983 all but the first. It now scans all returned addresses if a particular IP
4984 value is being sought. In this situation, the contents of the
4985 $dnslist_value variable are a list of all the addresses, separated by a
4988 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4989 to accept a host name and all its aliases if the forward lookup for any of
4990 them yielded the IP address of the incoming connection. Now it accepts only
4991 those names whose forward lookup yields the correct IP address. Any other
4992 names are discarded. This closes a loophole whereby a rogue DNS
4993 administrator could create reverse DNS records to break through a
4994 wildcarded host restriction in an ACL.
4996 37. If a user filter or a system filter that ran in a subprocess used any of
4997 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4998 the wrong values were passed to the pipe command ($thisaddress had the
4999 value of $0, $0 had the value of $1, etc). This bug was introduced by
5000 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5002 38. Improved the line breaking for long SMTP error messages from ACLs.
5003 Previously, if there was no break point between 40 and 75 characters, Exim
5004 left the rest of the message alone. Two changes have been made: (a) I've
5005 reduced the minimum length to 35 characters; (b) if it can't find a break
5006 point between 35 and 75 characters, it looks ahead and uses the first one
5007 that it finds. This may give the occasional overlong line, but at least the
5008 remaining text gets split now.
5010 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5011 file descriptors might be low, and that setting 1000 would always raise it.
5012 It turns out that in some environments, the limit is already over 1000 and
5013 that lowering it causes trouble. So now Exim takes care not to decrease it.
5015 40. When delivering a message, the value of $return_path is set to $sender_
5016 address at the start of routing (routers may change the value). By an
5017 oversight, this default was not being set up when an address was tested by
5018 -bt or -bv, which affected the outcome if any router or filter referred to
5021 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5022 log only when "message" or "log_message" are set. However, if one of the
5023 conditions was an address verification, or a call to a nested ACL, the
5024 messages generated by the underlying test were being passed through. This
5025 no longer happens. The underlying message is available in $acl_verify_
5026 message for both "message" and "log_message" expansions, so it can be
5027 passed through if needed.
5029 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5030 new expansion $bh_ to give the encoded byte string without charset
5031 translation. Translation happens only if iconv() is available; HAVE_ICONV
5032 indicates this at build time. HEADERS_CHARSET gives the charset to
5033 translate to; headers_charset can change it in the configuration, and
5034 "headers charset" can change it in an individual filter file.
5036 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5037 that creates RFC 2047 encoded "words" labels them as that charset instead
5038 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5039 expansion operator; (ii) when Exim creates a From: line for a local
5040 message; (iii) when a header line is rewritten to include a "phrase" part.
5042 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5043 buggy, causing it to skip the first lines of messages whose message ID
5044 ended in 'D'. This would not have bitten before Exim release 4.14, because
5045 message IDs were unlikely to end in 'D' before then. The effect was to have
5046 incorrect size information for certain domains.
5048 45. #include "config.h" was missing at the start of the crypt16.c module. This
5049 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5052 46. If there was a timeout during a "random" callout check, Exim treated it as
5053 a failure of the random address, and carried on sending RSET and the real
5054 address. If the delay was just some slowness somewhere, the response to the
5055 original RCPT would be taken as a response to RSET and so on, causing
5056 mayhem of various kinds.
5058 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5059 when I implemented it. It didn't allow for the fact that some option values
5060 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5061 the right test for positive values.
5063 48. Domain names in DNS records are case-independent. Exim always looks them up
5064 in lower case. Some resolvers return domain names in exactly the case they
5065 appear in the zone file, that is, they may contain uppercase letters. Not
5066 all resolvers do this - some return always lower case. Exim was treating a
5067 change of case by a resolver as a change of domain, similar to a widening
5068 of a domain abbreviation. This triggered its re-routing code and so it was
5069 trying to route what was effectively the same domain again. This normally
5070 caused routing to fail (because the router wouldn't handle the domain
5071 twice). Now Exim checks for this case specially, and just changes the
5072 casing of the domain that it ultimately uses when it transmits the message
5075 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5078 50. If a filter generated a file delivery with a non-absolute name (possible if
5079 no home directory exists for the router), the forbid_file option was not
5082 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5083 the existing equality matching.
5085 52. Exim was using ints instead of ino_t variables in some places where it was
5086 dealing with inode numbers.
5088 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5089 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5090 called TMPDIR, and if it finds it is different, it changes its value.
5092 54. The smtp_printf() function is now made available to local_scan() so
5093 additional output lines can be written before returning. There is also an
5094 smtp_fflush() function to enable the detection of a dropped connection.
5095 The variables smtp_input and smtp_batched_input are exported to
5098 55. Changed the default runtime configuration: the message "Unknown user"
5099 has been removed from the ACL, and instead placed on the localuser router,
5100 using the cannot_route_message feature. This means that any verification
5101 failures that generate their own messages won't get overridden. Similarly,
5102 the "Unrouteable address" message that was in the ACL for unverifiable
5103 relay addresses has also been removed.
5105 56. Added hosts_avoid_esmtp to the smtp transport.
5107 57. The exicyclog script was not checking for the esoteric option
5108 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5109 will work only if exicyclog is run under the appropriate euid.
5111 58. Following a discussion on the list, the rules by which Exim recognises line
5112 endings on incoming messages have been changed. The -dropcr and drop_cr
5113 options are now no-ops, retained only for backwards compatibility. The
5114 following line terminators are recognized: LF CRLF CR. However, special
5115 processing applies to CR:
5117 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5118 nor a local message in the state where . is a terminator.
5120 (ii) If a bare CR is encountered in a header line, an extra space is added
5121 after the line terminator so as not to end the header. The reasoning
5122 behind this is that bare CRs in header lines are most likely either
5123 to be mistakes, or people trying to play silly games.
5125 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5126 was being incorrectly given as 18 bytes larger than it should have been.
5127 This is a VOB (very old bug).
5129 60. This may never have affected anything current, but just in case it has:
5130 When the local host is found other than at the start of a list of hosts,
5131 the local host, those with the same MX, and any that follow, are discarded.
5132 When the list in question was part of a longer list of hosts, the following
5133 hosts (not currently being processed) were also being discarded. This no
5134 longer happens. I'm not sure if this situation could ever has previously
5137 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5140 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5141 header. This is something that is recommended in a new Internet Draft, and
5142 is something that is documented as being done by Sendmail. There are two
5143 possible values. For messages generated by the autoreply transport, Exim
5146 Auto-Submitted: auto-replied
5148 whereas for all other generated messages (e.g. bounces) it adds
5150 Auto-Submitted: auto-generated
5152 63. The "personal" condition in filters now includes a test for the
5153 Auto-Submitted: header. If it contains the string "auto-" the message it
5154 not considered personal.
5156 64. Added rcpt_include_affixes as a generic transport option.
5158 65. Added queue_only_override (default true).
5160 66. Added the syslog_duplication option.
5162 67. If what should have been the first header line of a message consisted of
5163 a space followed by a colon, Exim was mis-interpreting it as a header line.
5164 It isn't of course - it is syntactically invalid and should therefore be
5165 treated as the start of the message body. The misbehaviour could have
5166 caused a number of strange effects, including loss of data in subsequent
5167 header lines, and spool format errors.
5169 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5170 client host had authenticated. This control can now be exercised by an ACL
5171 for more flexibility.
5173 69. By default, callouts do not happen when testing with -bh. There is now a
5174 variant, -bhc, which does actually run the callout code, including
5175 consulting and updating the callout cache.
5177 70. Added support for saslauthd authentication, courtesy of Alexander
5180 71. If statvfs() failed on the spool or log directories while checking their
5181 size for availability, Exim confusingly gave the error "space shortage".
5182 Furthermore, in debugging mode it crashed with a floating point exception.
5183 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5184 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5185 serious problem, Exim now writes to the main and panic logs when this
5186 happens, with details of the failure. It then refuses to accept the
5187 incoming message, giving the message "spool directory problem" or "log
5188 directory problem" with a 421 code for SMTP messages.
5190 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5191 0, 1, and 2 exist, because some OS complain for execs without them (see
5192 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5193 descriptors. However, the code omitted to check that the open succeeded,
5194 causing mysterious errors if for some reason the permissions on /dev/null
5195 got screwed. Now Exim writes a message to the main and panic logs, and
5196 bombs out if it can't open /dev/null.
5198 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5199 interact so that it is all more flexible. It is supposed to remain
5200 backwards compatible. Also added extra_local_interfaces.
5202 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5203 to bomb out with an assertion failure - to the client this appears as a
5204 connection drop. This problem occurs in the part of the code that was taken
5205 from the Samba project. Fortunately, the assertion is in a very simple
5206 function, so I have fixed this by reproducing the function inline in the
5207 one place where it is called, and arranging for authentication to fail
5208 instead of killing the process with assert().
5210 75. The SPA client code was not working when the server requested OEM rather
5211 than Unicode encoding.
5213 76. Added code to make require_files with a specific uid setting more usable in
5214 the case where statting the file as root fails - usually a non-root-mounted
5215 NFS file system. When this happens and the failure is EACCES, Exim now
5216 forks a subprocess and does the per-uid checking as the relevant uid.
5218 77. Added process_log_path.
5220 78. If log_file_path was not explicitly set, a setting of check_log_space or
5221 check_log_inodes was ignored.
5223 79. If a space check for the spool or log partitions fails, the incident is now
5224 logged. Of course, in the latter case the data may get lost...
5226 80. Added the %p formatting code to string_format() so that it can be used to
5227 print addresses in debug_print(). Adjusted all the address printing in the
5228 debugging in store.c to use %p rather than %d.
5230 81. There was a concern that a line of code in smtp_in.c could overflow a
5231 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5232 initially expressed, the concern was not well-founded, because trailing
5233 spaces are removed early. However, if the trailing spaces were followed by
5234 a NULL, they did not get removed, so the overflow was possible. Two fixes
5237 (a) I re-wrote the offending code in a cleaner fashion.
5238 (b) If an incoming SMTP command contains a NULL character, it is rejected
5241 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5242 runs initgroups(), so that if the Exim user is in any additional groups,
5243 they will be used during message reception.
5249 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob