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.
20 TF/01 Failure to get a lock on a hints database can have serious
21 consequences so log it to the panic log.
23 TF/02 Log LMTP confirmation messages in the same way as SMTP,
24 controlled using the smtp_confirmation log selector.
26 TF/03 Include the error message when we fail to unlink a spool file.
28 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
29 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
30 for maintaining out-of-tree patches for some time.
32 PP/01 Bugzilla 139: Documentation and portability issues.
33 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
34 Handle per-OS dynamic-module compilation flags.
36 PP/02 Let /dev/null have normal permissions.
37 The 4.73 fixes were a little too stringent and complained about the
38 permissions on /dev/null. Exempt it from some checks.
39 Reported by Andreas M. Kirchwitz.
41 PP/03 Report version information for many libraries, including
42 Exim version information for dynamically loaded libraries. Created
43 version.h, now support a version extension string for distributors
44 who patch heavily. Dynamic module ABI change.
46 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
47 privilege escalation vulnerability whereby the Exim run-time user
48 can cause root to append content of the attacker's choosing to
51 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
54 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
55 If dropping privileges for untrusted macros, we disabled normal logging
56 on the basis that it would fail; for the Exim run-time user, this is not
57 the case, and it resulted in successful deliveries going unlogged.
58 Fixed. Reported by Andreas Metzler.
64 PP/01 Date: & Message-Id: revert to normally being appended to a message,
65 only prepend for the Resent-* case. Fixes regression introduced in
66 Exim 4.70 by NM/22 for Bugzilla 607.
68 PP/02 Include check_rfc2047_length in configure.default because we're seeing
69 increasing numbers of administrators be bitten by this.
71 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
73 PP/03 Bugzilla 994: added openssl_options main configuration option.
75 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
77 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
79 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
81 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
82 without a peer certificate, leading to a segfault because of an
83 assumption that peers always have certificates. Be a little more
84 paranoid. Problem reported by Martin Tscholak.
86 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
87 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
88 NB: ClamAV planning to remove STREAM in "middle of 2010".
89 CL also introduces -bmalware, various -d+acl logging additions and
90 more caution in buffer sizes.
92 PP/09 Implemented reverse_ip expansion operator.
94 PP/10 Bugzilla 937: provide a "debug" ACL control.
96 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
98 PP/12 Bugzilla 973: Implement --version.
100 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
102 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
104 PP/15 Bugzilla 816: support multiple condition rules on Routers.
106 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
107 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
108 ignore trailing whitespace.
110 JJ/02 prevent non-panic DKIM error from being sent to paniclog
112 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
115 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
116 Notification from Dr Andrew Aitchison.
118 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
119 ExtendedDetectionInfo response format.
120 Notification from John Horne.
122 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
125 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
126 XSL and documented dependency on system catalogs, with examples of how
129 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
132 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
133 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
134 configuration file which is writeable by the Exim user or group.
136 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
137 of configuration files to cover files specified with the -C option if
138 they are going to be used with root privileges, not just the default
141 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
142 option (effectively making it always true).
144 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
145 files to be used while preserving root privileges.
147 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
148 that rogue child processes cannot use them.
150 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
151 run-time user, instead of root.
153 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
154 Exim run-time user without dropping privileges.
156 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
157 result string, instead of calling string_vformat() twice with the same
160 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
161 for other users. Others should always drop root privileges if they use
162 -C on the command line, even for a whitelisted configure file.
164 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
166 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
172 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
173 $data_path, and $header_path variables; fixed documentation bugs and
176 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
177 exipick to access non-standard spools, including the "frozen" queue
180 NM/01 Bugzilla 965: Support mysql stored procedures.
181 Patch from Alain Williams
183 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
185 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
186 Patch from Andreas Metzler
188 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
189 Patch from Kirill Miazine
191 NM/05 Bugzilla 671: Added umask to procmail example.
193 JJ/03 installed exipick 20100323.0, fixing doc bug
195 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
196 directory. Notification and patch from Dan Rosenberg.
198 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
200 TK/02 Improve log output when DKIM signing operation fails.
202 MH/01 Treat the transport option dkim_domain as a colon separated
203 list, not as a single string, and sign the message with each element,
204 omitting multiple occurences of the same signer.
206 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
207 Bugzilla 985, 986. Patch by Simon Arlott
209 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
210 Patch by Simon Arlott
212 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
213 MBX locking. Notification from Dan Rosenberg.
219 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
221 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
223 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
225 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
227 NM/04 Bugzilla 915: Flag broken perl installation during build.
233 TK/01 Added patch by Johannes Berg that expands the main option
234 "spamd_address" if it starts with a dollar sign.
236 TK/02 Write list of recipients to X-Envelope-Sender header when building
237 the mbox-format spool file for content scanning (suggested by Jakob
240 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
241 (http://www.dcc-servers.net/) support via dccifd. Activated by
242 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
244 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
245 by Mark Daniel Reidel <mr@df.eu>.
247 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
248 When building exim an external PCRE library is now needed -
249 PCRE is a system library on the majority of modern systems.
250 See entry on PCRE_LIBS in EDITME file.
252 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
253 conversation. Added nologin parameter to request.
254 Patch contributed by Kirill Miazine.
256 TF/01 Do not log submission mode rewrites if they do not change the address.
258 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
260 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
261 log files in place. Contributed by Roberto Lima.
263 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
265 TF/03 Bugzilla 615: When checking the local_parts router precondition
266 after a local_part_suffix or local_part_prefix option, Exim now
267 does not use the address's named list lookup cache, since this
268 contains cached lookups for the whole local part.
270 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
271 Robert Millan. Documentation is in experimental-spec.txt.
273 TF/04 Bugzilla 668: Fix parallel build (make -j).
275 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
277 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
278 Patch provided by Jan Srzednicki.
280 TF/05 Leading white space used to be stripped from $spam_report which
281 wrecked the formatting. Now it is preserved.
283 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
284 that they are available at delivery time.
286 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
288 TF/08 TLS error reporting now respects the incoming_interface and
289 incoming_port log selectors.
291 TF/09 Produce a more useful error message if an SMTP transport's hosts
292 setting expands to an empty string.
294 NM/06 Bugzilla 744: EXPN did not work under TLS.
295 Patch provided by Phil Pennock.
297 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
298 Patch provided by Richard Godbee.
300 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
301 acl_smtp_notquit, added index entry.
303 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
304 Patch provided by Eugene Bujak.
306 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
307 accept(). Patch provided by Maxim Dounin.
309 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
310 Patch provided by Phil Pennock.
312 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
314 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
315 Patch provided by Brad "anomie" Jorsch.
317 NM/14 Bugzilla 622: Added timeout setting to transport filter.
318 Patch provided by Dean Brooks.
320 TK/05 Add native DKIM support (does not depend on external libraries).
322 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
323 Patch provided by Graeme Fowler.
325 NM/16 Bugzilla 851: Documentation example syntax fix.
327 NM/17 Changed NOTICE file to remove references to embedded PCRE.
329 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
332 NM/19 Bugzilla 745: TLS version reporting.
333 Patch provided by Phil Pennock.
335 NM/20 Bugzilla 167: bool: condition support.
336 Patch provided by Phil Pennock.
338 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
339 clients. Patch provided by Phil Pennock.
341 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
342 Patch provided by Brad "anomie" Jorsch.
344 NM/23 Bugzilla 687: Fix misparses in eximstats.
345 Patch provided by Heiko Schlittermann.
347 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
348 Patch provided by Heiko Schlittermann.
350 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
351 plus update to original patch.
353 NM/26 Bugzilla 799: Documentation correction for ratelimit.
355 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
356 Patch provided by David Brownlee.
358 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
360 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
362 NM/30 Bugzilla 888: TLS documentation bugfixes.
364 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
366 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
367 Unlike the original bugzilla I have changed all shell scripts in src tree.
369 NM/33 Bugzilla 898: Transport filter timeout fix.
370 Patch by Todd Rinaldo.
372 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
373 Patch by Serge Demonchaux.
375 NM/35 Bugzilla 39: Base64 decode bug fixes.
376 Patch by Jakob Hirsch.
378 NM/36 Bugzilla 909: Correct connect() call in dcc code.
380 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
382 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
384 NM/39 Bugzilla 911: Fixed MakeLinks build script.
390 TK/01 Add preliminary DKIM support. Currently requires a forked version of
391 ALT-N's libdkim that I have put here:
392 http://duncanthrax.net/exim-experimental/
394 Note to Michael Haardt: I had to rename some vars in sieve.c. They
395 were called 'true' and it seems that C99 defines that as a reserved
396 keyword to be used with 'bool' variable types. That means you could
397 not include C99-style headers which use bools without triggering
398 build errors in sieve.c.
400 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
401 as mailq or other aliases. Changed the --help handling significantly
402 to do whats expected. exim_usage() emits usage/help information.
404 SC/01 Added the -bylocaldomain option to eximstats.
406 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
408 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
410 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
416 PH/01 Another patch from the Sieve maintainer.
418 PH/02 When an IPv6 address is converted to a string for single-key lookup
419 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
420 dots are used instead of colons so that keys in lsearch files need not
421 contain colons. This was done some time before quoting was made available
422 in lsearch files. However, iplsearch files do require colons in IPv6 keys
423 (notated using the quote facility) so as to distinguish them from IPv4
424 keys. This meant that lookups for IP addresses in host lists did not work
425 for iplsearch lookups.
427 This has been fixed by arranging for IPv6 addresses to be expressed with
428 colons if the lookup type is iplsearch. This is not incompatible, because
429 previously such lookups could never work.
431 The situation is now rather anomolous, since one *can* have colons in
432 ordinary lsearch keys. However, making the change in all cases is
433 incompatible and would probably break a number of configurations.
435 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
438 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
439 conversion specification without a maximum field width, thereby enabling
440 a rogue spamd server to cause a buffer overflow. While nobody in their
441 right mind would setup Exim to query an untrusted spamd server, an
442 attacker that gains access to a server running spamd could potentially
443 exploit this vulnerability to run arbitrary code as the Exim user.
445 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
446 $primary_hostname instead of what libspf2 thinks the hosts name is.
448 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
449 a directory entry by the name of the lookup key. Previously, if a
450 symlink pointed to a non-existing file or a file in a directory that
451 Exim lacked permissions to read, a lookup for a key matching that
452 symlink would fail. Now it is enough that a matching directory entry
453 exists, symlink or not. (Bugzilla 503.)
455 PH/03 The body_linecount and body_zerocount variables are now exported in the
458 PH/04 Added the $dnslist_matched variable.
460 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
461 This means they are set thereafter only if the connection becomes
464 PH/06 Added the client_condition to authenticators so that some can be skipped
465 by clients under certain conditions.
467 PH/07 The error message for a badly-placed control=no_multiline_responses left
468 "_responses" off the end of the name.
470 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
472 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
473 (without spaces) instead of just copying the configuration text.
475 PH/10 Added the /noupdate option to the ratelimit ACL condition.
477 PH/11 Added $max_received_linelength.
479 PH/12 Added +ignore_defer and +include_defer to host lists.
481 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
482 way in which PCRE > 7.0 is built.
484 PH/14 Implemented queue_only_load_latch.
486 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
487 MAIL command. The effect was to mangle the value on 64-bit systems.
489 PH/16 Another patch from the Sieve maintainer.
491 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
493 PH/18 If a system quota error occurred while trying to create the file for
494 a maildir delivery, the message "Mailbox is full" was not appended to the
495 bounce if the delivery eventually timed out. Change 4.67/27 below applied
496 only to a quota excession during the actual writing of the file.
498 PH/19 It seems that peer DN values may contain newlines (and other non-printing
499 characters?) which causes problems in log lines. The DN values are now
500 passed through string_printing() before being added to log lines.
502 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
503 and InterBase are left for another time.)
505 PH/21 Added message_body_newlines option.
507 PH/22 Guard against possible overflow in moan_check_errorcopy().
509 PH/23 POSIX allows open() to be a macro; guard against that.
511 PH/24 If the recipient of an error message contained an @ in the local part
512 (suitably quoted, of course), incorrect values were put in $domain and
513 $local_part during the evaluation of errors_copy.
519 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
520 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
523 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
524 issue a MAIL command.
526 PH/02 In an ACL statement such as
528 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
530 if a client was not listed at all, or was listed with a value other than
531 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
532 the condition was not true (as it should be), so access was not denied.
533 The bug was that the ! inversion was incorrectly passed on to the second
534 item. This has been fixed.
536 PH/03 Added additional dnslists conditions == and =& which are different from
537 = and & when the dns lookup returns more than one IP address.
539 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
540 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
542 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
543 FSYNC, which compiles an option called disable_fsync that allows for
544 bypassing fsync(). The documentation is heavily laced with warnings.
546 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
548 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
549 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
550 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
551 including adding "make clean"; (3) Added -fPIC when compiling the test
552 dynamically loaded module, to get rid of a warning.
554 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
555 message fails, move_frozen_messages = true and ignore_bounce_errors_after
556 = 0s. The bug is otherwise harmless.
558 PH/07 There was a bug in the dovecot authenticator such that the value of
559 $auth1 could be overwritten, and so not correctly preserved, after a
560 successful authentication. This usually meant that the value preserved by
561 the server_setid option was incorrect.
563 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
565 PH/09 Installed PCRE release 7.0.
567 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
568 run for batched SMTP input. It is now run at the start of every message
569 in the batch. While fixing this I discovered that the process information
570 (output by running exiwhat) was not always getting set for -bs and -bS
571 input. This is fixed, and it now also says "batched" for BSMTP.
573 PH/11 Added control=no_pipelining.
575 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
576 patch, slightly modified), and move the expansion of helo_data till after
577 the connection is made in the smtp transport (so it can use these
580 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
582 PH/14 Added log_selector = +pid.
584 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
586 PH/16 Add ${if forany and ${if forall.
588 PH/17 Added dsn_from option to vary the From: line in DSNs.
590 PH/18 Flush SMTP output before performing a callout, unless control =
591 no_callout_flush is set.
593 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
594 was true (the default) a successful delivery failed to delete the retry
595 item, thus causing premature timeout of the address. The bug is now
598 PH/20 Added hosts_avoid_pipelining to the smtp transport.
600 PH/21 Long custom messages for fakedefer and fakereject are now split up
601 into multiline reponses in the same way that messages for "deny" and
602 other ACL rejections are.
604 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
605 with slight modification.
607 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
608 draft, changing the syntax and factoring some duplicate code".
610 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
611 for deliveries of the second and subsequent messages over the same SMTP
614 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
615 ${reduce, with only minor "tidies".
617 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
619 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
620 expansion side effects.
622 PH/27 When a message times out after an over-quota error from an Exim-imposed
623 quota, the bounce message says "mailbox is full". This message was not
624 being given when it was a system quota that was exceeded. It now should
627 MH/03 Made $recipients available in local_scan(). local_scan() already has
628 better access to the recipient list through recipients_list[], but
629 $recipients can be useful in postmaster-provided expansion strings.
631 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
632 in the case of a MAIL command with additional options following the
633 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
634 were accidentally chopped off.
636 PH/29 SMTP synchronization checks are implemented when a command is read -
637 there is a check that no more input is waiting when there shouldn't be
638 any. However, for some commands, a delay in an ACL can mean that it is
639 some time before the response is written. In this time, more input might
640 arrive, invalidly. So now there are extra checks after an ACL has run for
641 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
642 pipelining has not been advertised.
644 PH/30 MH's patch to allow iscntrl() characters to be list separators.
646 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
647 returned in the SMTP response when smtp_return_error_details was false.
650 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
651 instead of the C I/O that was originally supplied, because problems were
654 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
655 Exim which did not show up earlier: it was assuming that a call to
656 SSL_CTX_set_info_callback() might give an error value. In fact, there is
657 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
658 was a macro that became an assignment, so it seemed to work. This has
659 changed to a proper function call with a void return, hence the compile
660 error. Exim's code has been fixed.
662 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
665 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
667 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
669 PH/37 If a message is not accepted after it has had an id assigned (e.g.
670 because it turns out to be too big or there is a timeout) there is no
671 "Completed" line in the log. When some messages of this type were
672 selected by exigrep, they were listed as "not completed". Others were
673 picked up by some special patterns. I have improved the selection
674 criteria to be more general.
676 PH/38 The host_find_failed option in the manualroute router can now be set
677 to "ignore", to completely ignore a host whose IP address cannot be
678 found. If all hosts are ignored, the behaviour is controlled by the new
679 host_all_ignored option.
681 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
682 homing or because of multiple MX records with /mx) generated more than
683 one IP address, and the following item turned out to be the local host,
684 all the secondary addresses of the first item were incorrectly removed
685 from the list, along with the local host and any following hosts (which
686 is what is supposed to happen).
688 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
689 whoever called Exim into the -H file. In the case of the daemon it was
690 behaving confusingly. When first started, it used values for whoever
691 started the daemon, but after a SIGHUP it used the Exim user (because it
692 calls itself on a restart). I have changed the code so that it now always
695 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
696 message are rejected with the same error (e.g. no authentication or bad
697 sender address), and a DATA command is nevertheless sent (as can happen
698 with PIPELINING or a stupid MUA), the error message that was given to the
699 RCPT commands is included in the rejection of the DATA command. This is
700 intended to be helpful for MUAs that show only the final error to their
703 PH/42 Another patch from the Sieve maintainer.
705 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
706 Eximstats - Fixed some broken HTML links and added missing column headers
708 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
709 columns for Rejects, Temp Rejects, Ham, and Spam rows.
711 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
713 PH/43 Yet another patch from the Sieve maintainer.
715 PH/44 I found a way to check for a TCP/IP connection going away before sending
716 the response to the final '.' that terminates a message, but only in the
717 case where the client has not sent further data following the '.'
718 (unfortunately, this is allowed). However, in many cases there won't be
719 any further data because there won't be any more messages to send. A call
720 to select() can be used: if it shows that the input is "ready", there is
721 either input waiting, or the socket has been closed. An attempt to read
722 the next input character can distinguish the two cases. Previously, Exim
723 would have sent an OK response which the client would never have see.
724 This could lead to message repetition. This fix should cure that, at
725 least in a lot of common cases.
727 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
728 advertised in response to EHLO.
734 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
735 fixed by 4.65/MH/01 (is this a record?) are fixed:
737 (i) An empty string was always treated as zero by the numeric comparison
738 operators. This behaviour has been restored.
740 (ii) It is documented that the numeric comparison operators always treat
741 their arguments as decimal numbers. This was broken in that numbers
742 starting with 0 were being interpreted as octal.
744 While fixing these problems I realized that there was another issue that
745 hadn't been noticed. Values of message_size_limit (both the global option
746 and the transport option) were treated as octal if they started with 0.
747 The documentation was vague. These values are now always treated as
748 decimal, and I will make that clear in the documentation.
754 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
755 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
758 MH/01 Don't check that the operands of numeric comparison operators are
759 integers when their expansion is in "skipping" mode (fixes bug
760 introduced by 4.64-PH/07).
762 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
763 child addresses, Exim now panics and dies. Previously, because the count
764 is held in a short int, deliveries were likely to be lost. As such a
765 large number of recipients for a single message is ridiculous
766 (performance will be very, very poor), I have chosen to impose a limit
767 rather than extend the field.
773 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
774 leftover -K file (the existence of which was triggered by #402).
775 While we were at it, introduced process PID as part of the -K
776 filename. This should rule out race conditions when creating
779 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
780 processing considerably. Previous code took too long for large mails,
781 triggering a timeout which in turn triggers #401.
783 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
784 in the DK code in transports.c. sendfile() is not really portable,
785 hence the _LINUX specificness.
787 TF/01 In the add_headers option to the mail command in an Exim filter,
788 there was a bug that Exim would claim a syntax error in any
789 header after the first one which had an odd number of characters
792 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
793 callout verification, Exim cached a "reject" for the entire domain. This
794 is correct for most verifications, but it is not correct for a recipient
795 verification with use_sender or use_postmaster set, because in that case
796 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
797 case of MAIL FROM:<> rejection from other early rejections (e.g.
798 rejection of HELO). When verifying a recipient using a non-null MAIL
799 address, the cache is ignored if it shows MAIL FROM:<> rejection.
800 Whatever the result of the callout, the value of the domain cache is
801 left unchanged (for any other kind of callout, getting as far as trying
802 RCPT means that the domain itself is ok).
804 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
807 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
808 manifest itself as EPIPE rather than ECONNECT. When tidying away a
809 session, the daemon ignores ECONNECT errors and logs others; it now
810 ignores EPIPE as well.
812 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
813 (quoted-printable decoding).
815 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
816 later the small subsequent patch to fix an introduced bug.
818 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
820 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
822 PH/08 An error is now given if message_size_limit is specified negative.
824 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
825 to be given (somewhat) arbitrary names.
827 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
830 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
831 miscellaneous code fixes
833 PH/10 Added the log_reject_target ACL modifier to specify where to log
836 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
837 hostname. This is wrong, because it relates to the incoming message (and
838 probably the interface on which it is arriving) and not to the outgoing
839 callout (which could be using a different interface). This has been
840 changed to use the value of the helo_data option from the smtp transport
841 instead - this is what is used when a message is actually being sent. If
842 there is no remote transport (possible with a router that sets up host
843 addresses), $smtp_active_hostname is used.
845 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
846 tweaks were necessary in order to get it to work (see also 21 below):
847 (a) The code assumed that strncpy() returns a negative number on buffer
848 overflow, which isn't the case. Replaced with Exim's string_format()
850 (b) There were several signed/unsigned issues. I just did the minimum
851 hacking in of casts. There is scope for a larger refactoring.
852 (c) The code used strcasecmp() which is not a standard C function.
853 Replaced with Exim's strcmpic() function.
854 (d) The code set only $1; it now sets $auth1 as well.
855 (e) A simple test gave the error "authentication client didn't specify
856 service in request". It would seem that Dovecot has changed its
857 interface. Fortunately there's a specification; I followed it and
858 changed what the client sends and it appears to be working now.
860 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
863 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
864 address A is aliased to B and C, where B exists and C does not. Without
865 -v the output is "A verified" because verification stops after a
866 successful redirection if more than one address is generated. However,
867 with -v the child addresses are also verified. Exim was outputting "A
868 failed to verify" and then showing the successful verification for C,
869 with its parentage. It now outputs "B failed to verify", showing B's
870 parentage before showing the successful verification of C.
872 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
873 look up a TXT record in a specific list after matching in a combined
876 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
877 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
878 they consult the DNS. I had assumed they would set it the way they
879 wanted; and indeed my experiments on Linux seem to show that in some
880 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
881 To be on the safe side, however, I have now made the interface to
882 host_find_byname() similar to host_find_bydns(), with an argument
883 containing the DNS resolver options. The host_find_byname() function now
884 sets these options at its start, just as host_find_bydns() does. The smtp
885 transport options dns_qualify_single and dns_search_parents are passed to
886 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
887 of host_find_byname() use the default settings of RES_DEFNAMES
888 (qualify_single) but not RES_DNSRCH (search_parents).
890 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
891 spool_read_header() do less string testing, by means of a preliminary
892 switch on the second character of optional "-foo" lines. (This is
893 overdue, caused by the large number of possibilities that now exist.
894 Originally there were few.) While I was there, I also converted the
895 str(n)cmp tests so they don't re-test the leading "-" and the first
896 character, in the hope this might squeeze out yet more improvement.
898 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
899 flag allowing group syntax was set by the header_syntax check but not
900 turned off, possible causing trouble later; (2) The flag was not being
901 set at all for the header_verify test, causing "group"-style headers to
902 be rejected. I have now set it in this case, and also caused header_
903 verify to ignore an empty address taken from a group. While doing this, I
904 came across some other cases where the code for allowing group syntax
905 while scanning a header line wasn't quite right (mostly, not resetting
906 the flag correctly in the right place). These bugs could have caused
907 trouble for malformed header lines. I hope it is now all correct.
909 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
910 with the "reply" argument non-NULL. The code, however (which originally
911 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
912 but it didn't always do it. This confused somebody who was copying the
913 code for some other use. I have removed all the tests.
915 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
916 feature that was used to support insecure browsers during the U.S. crypto
917 embargo. It requires special client support, and Exim is probably the
918 only MTA that supported it -- and would never use it because real RSA is
919 always available. This code has been removed, because it had the bad
920 effect of slowing Exim down by computing (never used) parameters for the
921 RSA_EXPORT functionality.
923 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
924 authenticator to fail if there's a tab character in the incoming data
925 (there should never be unless someone is messing about, as it's supposed
926 to be base64-encoded). Also added, on Timo's advice, the "secured" option
927 if the connection is using TLS or if the remote IP is the same as the
928 local IP, and the "valid-client-cert option" if a client certificate has
931 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
932 authenticators. This can be used for authorization after authentication
933 succeeds. (In the case of plaintext, it servers for both authentication
936 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
937 if any retry times were supplied.
939 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
940 connection, where there is no client IP address to check. In this
941 situation, the verify now always succeeds.
943 PH/25 Applied John Jetmore's -Mset patch.
945 PH/26 Added -bem to be like -Mset, but loading a message from a file.
947 PH/27 In a string expansion for a processed (not raw) header when multiple
948 headers of the same name were present, leading whitespace was being
949 removed from all of them, but trailing whitespace was being removed only
950 from the last one. Now trailing whitespace is removed from each header
951 before concatenation. Completely empty headers in a concatenation (as
954 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
955 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
957 PH/29 [Removed. This was a change that I later backed out, and forgot to
958 correct the ChangeLog entry (that I had efficiently created) before
959 committing the later change.]
961 PH/30 Exim was sometimes attempting to deliver messages that had suffered
962 address errors (4xx response to RCPT) over the same connection as other
963 messages routed to the same hosts. Such deliveries are always "forced",
964 so retry times are not inspected. This resulted in far too many retries
965 for the affected addresses. The effect occurred only when there were more
966 hosts than the hosts_max_try setting in the smtp transport when it had
967 the 4xx errors. Those hosts that it had tried were not added to the list
968 of hosts for which the message was waiting, so if all were tried, there
969 was no problem. Two fixes have been applied:
971 (i) If there are any address or message errors in an SMTP delivery, none
972 of the hosts (tried or untried) are now added to the list of hosts
973 for which the message is waiting, so the message should not be a
974 candidate for sending over the same connection that was used for a
975 successful delivery of some other message. This seems entirely
976 reasonable: after all the message is NOT "waiting for some host".
977 This is so "obvious" that I'm not sure why it wasn't done
978 previously. Hope I haven't missed anything, but it can't do any
979 harm, as the worst effect is to miss an optimization.
981 (ii) If, despite (i), such a delivery is accidentally attempted, the
982 routing retry time is respected, so at least it doesn't keep
983 hammering the server.
985 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
986 in ${readsocket because some servers need this prod.
988 PH/32 Added some extra debug output when updating a wait-xxx database.
990 PH/33 The hint "could be header name not terminated by colon", which has been
991 given for certain expansion errors for a long time, was not being given
992 for the ${if def:h_colon_omitted{... case.
994 PH/34 The spec says: "With one important exception, whenever a domain list is
995 being scanned, $domain contains the subject domain." There was at least
996 one case where this was not true.
998 PH/35 The error "getsockname() failed: connection reset by peer" was being
999 written to the panic log as well as the main log, but it isn't really
1000 panic-worthy as it just means the connection died rather early on. I have
1001 removed the panic log writing for the ECONNRESET error when getsockname()
1004 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1005 runs only) independently of the message's sender address. This meant
1006 that, if the 4xx error was in fact related to the sender, a different
1007 message to the same recipient with a different sender could confuse
1008 things. In particualar, this can happen when sending to a greylisting
1009 server, but other circumstances could also provoke similar problems.
1010 I have changed the default so that the retry time for these errors is now
1011 based a combination of the sender and recipient addresses. This change
1012 can be overridden by setting address_retry_include_sender=false in the
1015 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1016 remote server are returned as part of bounce messages. This was not
1017 happening for LMTP over a pipe (the lmtp transport), but now it is the
1018 same for both kinds of LMTP.
1020 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1021 in header lines that were in fact CNAMEs. This is no longer the case.
1023 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1024 and queue runs started by the daemon processed all messages. This has
1025 been fixed so that -R and -S can now usefully be given with -q<time>.
1027 PH/40 Import PCRE release 6.7 (fixes some bugs).
1029 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1031 PH/42 Give an error if -q is specified more than once.
1033 PH/43 Renamed the variables $interface_address and $interface_port as
1034 $received_ip_address and $received_port, to make it clear that these
1035 values apply to message reception, and not to the outgoing interface when
1036 a message is delivered. (The old names remain recognized, of course.)
1038 PH/44 There was no timeout on the connect() call when using a Unix domain
1039 socket in the ${readsocket expansion. There now is.
1041 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1042 be meaningful with "accept".
1044 SC/01 Eximstats V1.43
1045 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1047 SC/02 Eximstats V1.44
1048 Use a glob alias rather than an array ref in the generated
1049 parser. This improves both readability and performance.
1051 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1052 Collect SpamAssassin and rejection statistics.
1053 Don't display local sender or destination tables unless
1054 there is data to show.
1055 Added average volumes into the top table text output.
1057 SC/04 Eximstats V1.46
1058 Collect data on the number of addresses (recipients)
1059 as well as the number of messages.
1061 SC/05 Eximstats V1.47
1062 Added 'Message too big' to the list of mail rejection
1063 reasons (thanks to Marco Gaiarin).
1065 SC/06 Eximstats V1.48
1066 Mainlog lines which have GMT offsets and are too short to
1067 have a flag are now skipped.
1069 SC/07 Eximstats V1.49 (Alain Williams)
1070 Added the -emptyok flag.
1072 SC/08 Eximstats V1.50
1073 Fixes for obtaining the IP address from reject messages.
1075 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1076 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1077 whitesspace changes from 4.64-PH/27
1079 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1082 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1083 are found, allow negative numbers in numeric criteria)
1085 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1087 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1088 to match changes made in 4.64-PH/43
1090 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1092 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1093 "domain" is actually an IP address, and force a failure. This locks out
1094 those revolvers/nameservers that support "A-for-A" lookups, in
1095 contravention of the specifications.
1097 PH/48 When a host name was looked up from an IP address, and the subsequent
1098 forward lookup of the name timed out, the host name was left in
1099 $sender_host_name, contrary to the specification.
1101 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1102 restricted to single-key lookups, Exim was not diagnosing an error if
1103 * or *@ was used with a query-style lookup.
1105 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1107 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1108 long ago, but noone interested enough thought of it. Let's just say that
1109 the "1.1" means that there are some new functions that weren't there at
1110 some point in the past.
1112 PH/51 Error processing for expansion failure of helo_data from an smtp
1113 transport during callout processing was broken.
1115 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1116 tested/used via the -bh/-bhc/-bs options.
1118 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1119 bug, fixed in subsequent PCRE releases).
1121 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1122 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1124 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1130 SC/01 Use a glob alias rather than an array ref in eximstats generated
1131 parser. This improves both readability and performance.
1133 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1134 Don't display local sender or destination tables in eximstats unless
1135 there is data to show.
1136 Added average volumes into the eximstats top table text output.
1138 SC/03 Collect data on the number of addresses (recipients) as well
1139 as the number of messages in eximstats.
1141 TF/01 Correct an error in the documentation for the redirect router. Exim
1142 does (usually) call initgroups() when daemonizing.
1144 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1145 with consistent privilege compared to when running as a daemon.
1147 TF/03 Note in the spec that $authenticated_id is not set for local
1148 submissions from trusted users.
1150 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1151 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1153 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1154 by adding some example configuration directives to the default
1155 configuration file. A little bit of work is required to uncomment the
1156 directives and define how usernames and passwords are checked, but
1157 there is now a framework to start from.
1159 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1160 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1161 without this. I don't know how relevant this is to other LDAP libraries.
1163 PH/02 Add the verb name to the "unknown ACL verb" error.
1165 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1167 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1169 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1170 directory not expanded when it should be if an expanded home directory
1171 was set for the address (which is overridden by the transport).
1173 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1176 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1177 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1178 because it is too late at that time, and has no effect.
1180 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1181 security issue with \' (bugzilla #107). I could not use the
1182 PQescapeStringConn() function, because it needs a PGconn value as one of
1185 PH/08 When testing addresses using -bt, indicate those final addresses that
1186 are duplicates that would not cause an additional delivery. At least one
1187 person was confused, thinking that -bt output corresponded to deliveries.
1188 (Suppressing duplicates isn't a good idea as you lose the information
1189 about possibly different redirections that led to the duplicates.)
1191 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1192 systems where poll() doesn't work, in particular OS X.
1194 PH/10 Added more information to debugging output for retry time not reached.
1196 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1197 operations in malware.c.
1199 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1202 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1203 syslog_duplication set false, log lines that would normally be written
1204 both the the main log and to the reject log were not written to syslog at
1207 PH/14 In the default configuration, change the use of "message" in ACL warn
1208 statements to "add_header".
1210 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1211 not followed by a command (e.g. "seen endif").
1213 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1214 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1217 PH/17 Added extra conditions to the default value of delay_warning_condition
1221 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1222 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1223 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1226 The Auto-Submitted: and various List- headers are standardised, whereas I
1227 don't think Precedence: ever was.
1229 PH/18 Refactored debugging code in route_finduser() to show more information,
1230 in particular, the error code if getpwnam() issues one.
1232 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1233 This is apparently needed in addition to the PH/07 change above to avoid
1234 any possible encoding problems.
1236 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1237 but not after initializing Perl.
1239 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1240 output them only if debugging. By default they are written stderr,
1241 apparently, which is not desirable.
1243 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1246 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1249 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1251 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1252 authenticated or an ident call has been made. Suppress the default
1253 values for $authenticated_id and $authenticated_sender (but permit -oMai
1254 and -oMas) when testing with -bh.
1256 PH/24 Re-jigged the order of the tests in the default configuration so that the
1257 tests for valid domains and recipients precede the DNS black list and CSA
1258 tests, on the grounds that those ones are more expensive.
1260 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1261 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1262 This bug exists in every version of Exim that I still have, right back to
1265 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1266 However, an attempt to turn on case-sensitivity in a regex key by
1267 including (?-i) didn't work because the subject string was already
1268 lowercased, and the effects were non-intuitive. It turns out that a
1269 one-line patch can be used to allow (?-i) to work as expected.
1275 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1276 other effects) broke the use of negated acl sub-conditions.
1278 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1281 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1282 "Deny" causes Exim to reject the incoming connection with a 554 error.
1283 Unfortunately, if there is a major crisis, such as a disk failure,
1284 tcp-wrappers gives "deny", whereas what one would like would be some
1285 kind of temporary error. A kludge has been added to help with this.
1286 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1287 554 error is used if errno is still zero or contains ENOENT (which occurs
1288 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1291 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1293 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1294 errors. Otherwise a message that provokes a temporary error (when other
1295 messages do not) can cause a whole host to time out.
1297 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1298 addresses were routed directly to files or pipes from a redirect router.
1299 File deliveries just didn't batch; pipe deliveries might have suffered
1302 PH/06 A failure to get a lock for a hints database would erroneously always say
1303 "Failed to get write lock", even when it was really a read lock.
1305 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1306 of 0600. This has been changed to use the value of the lockfile_mode
1307 option (which defaults to 0600).
1309 PH/08 Applied small patch from the Sieve maintainer.
1311 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1312 folder from quota calculations, a direct delivery into this folder messed
1313 up the contents of the maildirsize file. This was because the regex was
1314 used only to exclude .Trash (or whatever) when the size of the mailbox
1315 was calculated. There was no check that a delivery was happening into an
1316 excluded directory. This bug has been fixed by ignoring all quota
1317 processing for deliveries into excluded directories.
1319 PH/10 Added the maildirfolder_create_regex option to appendfile.
1325 PH/01 The code for finding all the local interface addresses on a FreeBSD
1326 system running IPv6 was broken. This may well have applied to all BSD
1327 systems, as well as to others that have similar system calls. The broken
1328 code found IPv4 interfaces correctly, but gave incorrect values for the
1329 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1330 that it would not match correctly against @[] and not recognize the IPv6
1333 PH/02 The ipliteral router was not recognizing addresses of the form user@
1334 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1336 PH/03 Added disable_ipv6.
1338 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1339 decoded form, because it is most often used to construct To: headers
1340 lines in autoreplies, and the decoded form may well be syntactically
1341 invalid. However, $reply_address has leading white space removed, and all
1342 newlines turned into spaces so that the autoreply transport does not
1345 PH/05 If group was specified without a user on a router, and no group or user
1346 was specified on a transport, the group from the router was ignored.
1348 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1349 for visible compile-time settings that can be used to change these
1350 numbers, for those that want even more. Backwards compatibility with old
1351 spool files has been maintained. However, going back to a previous Exim
1352 release will lost any variables that are in spool files.
1354 PH/07 Two small changes when running in the test harness: increase delay when
1355 passing a TCP/IP connection to a new process, in case the original
1356 process has to generate a bounce, and remove special handling of
1357 127.0.0.2 (sic), which is no longer necessary.
1359 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1360 be the same on different OS.
1362 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1365 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1366 whether --show-vars was specified or not
1368 JJ/02 exipick: Added support for new ACL variable spool format introduced
1371 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1372 syntactically invalid From: or Reply-to: line, and a filter used this to
1373 generate an autoreply, and therefore failed to obtain an address for the
1374 autoreply, Exim could try to deliver to a non-existent relative file
1375 name, causing unrelated and misleading errors. What now happens is that
1376 it logs this as a hard delivery error, but does not attempt to create a
1379 PH/11 The exinext utility has a -C option for testing purposes, but although
1380 the given file was scanned by exinext itself; it wasn't being passed on
1381 when Exim was called.
1383 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1384 an end-of-file indication when reading a command response.
1386 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1387 compiled. In many other places in Exim, IPv6 addresses are always
1388 recognized, so I have changed this. It also means that IPv4 domain
1389 literals of the form [IPV4:n.n.n.n] are now always recognized.
1391 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1392 used if the router is not running as root, for example, when verifying at
1393 ACL time, or when using -bh. The debugging output from this situation was
1394 non-existent - all you got was a failure to exec. I have made two
1397 (a) Failures to set uid/gid, the current directory, or a process leader
1398 in a subprocess such as that created by queryprogram now generate
1399 suitable debugging ouput when -d is set.
1401 (b) The queryprogram router detects when it is not running as root,
1402 outputs suitable debugging information if -d is set, and then runs
1403 the subprocess without attempting to change uid/gid.
1405 PH/15 Minor change to Makefile for building test_host (undocumented testing
1408 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1409 additional section of a DNS packet that returns MX or SRV records.
1410 Instead, it always explicitly searches for A/AAAA records. This avoids
1411 major problems that occur when a DNS server includes only records of one
1412 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1413 fixed another bug: if SRV records were looked up and the corresponding
1414 address records were *not* found in the additional section, the port
1415 values from the SRV records were lost.
1417 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1418 using the correct key (the original address) when searching the retry
1419 rules in order to find which one to use for generating the retry hint.
1421 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1422 adding the default one. Similarly, if it contains a Reply-To: header, the
1423 errors_reply_to option, if set, is not used.
1425 PH/19 When calculating a retry time, Exim used to measure the "time since
1426 failure" by looking at the "first failed" field in the retry record. Now
1427 it does not use this if it is later than than the arrival time of the
1428 message. Instead it uses the arrival time. This makes for better
1429 behaviour in cases where some deliveries succeed, thus re-setting the
1430 "first failed" field. An example is a quota failure for a huge message
1431 when small messages continue to be delivered. Without this change, the
1432 "time since failure" will always be short, possible causing more frequent
1433 delivery attempts for the huge message than are intended.
1434 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1436 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1437 $1, $2, $3) because the numerical variables can be reset during some
1438 expansion items (e.g. "match"), thereby losing the authentication data.
1440 PH/21 Make -bV show the size of off_t variables so that the test suite can
1441 decide whether to run tests for quotas > 2G.
1443 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1444 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1445 filecount value is greater than 2G or if a quota value is greater than 2G
1446 on a system where the size of off_t is not greater than 4, a panic error
1449 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1450 never match. The debug and -bh output now contains an explicit error
1451 message indicating a malformed IPv4 address or mask.
1453 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1454 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1455 PH/23 above applies.
1457 PH/25 Do not write to syslog when running in the test harness. The only
1458 occasion when this arises is a failure to open the main or panic logs
1459 (for which there is an explicit test).
1461 PH/26 Added the /no_tell option to "control=freeze".
1463 PH/27 If a host name lookup failed very early in a connection, for example, if
1464 the IP address matched host_lookup and the reverse lookup yielded a name
1465 that did not have a forward lookup, an error message of the form "no IP
1466 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1467 could be logged. Now it outputs the IP address instead of "NULL".
1469 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1470 allows the sender and the authenticated sender to be set when
1471 submitting a message from within Exim. Since child_open_exim() is
1472 documented for local_scan(), the new function should be too.
1474 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1475 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1476 results in an empty string is now treated as unset.
1478 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1480 PH/31 Added sender_verify_fail logging option.
1482 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1483 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1484 tidied the source and removed it altogether.
1486 PH/33 When a queue run was abandoned because the load average was too high, a
1487 log line was always written; now it is written only if the queue_run log
1488 selector is set. In addition, the log line for abandonment now contains
1489 information about the queue run such as the pid. This is always present
1490 in "start" and "stop" lines but was omitted from the "abandon" line.
1492 PH/34 Omit spaces between a header name and the colon in the error message that
1493 is given when verify = headers_syntax fails (if there are lots of them,
1494 the message gets confusing).
1496 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1497 names, as there are now some PTR records that contain slashes. This check
1498 is only to protect against broken name servers that fall over on strange
1499 characters, so the fact that it applies to all lookups doesn't matter.
1501 PH/36 Now that the new test suite is complete, we can remove some of the
1502 special code in Exim that was needed for the old test suite. For example,
1503 sorting DNS records because real resolvers return them in an arbitrary
1504 order. The new test suite's fake resolver always returns records in the
1507 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1508 bounces) except when queue_only is set, to avoid logging races between
1509 the different processes.
1511 PH/38 Panic-die if .include specifies a non-absolute path.
1513 PH/39 A tweak to the "H" retry rule from its user.
1515 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1516 a label. They prevented compilation on older perls.
1518 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1519 a warning to be raised on newish perls.
1521 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1522 on queue. Changes to match documented behaviour of showing count of
1523 messages matching specified criteria.
1525 PH/40 Changed the default ident timeout from 30s to 5s.
1527 PH/41 Added support for the use of login_cap features, on those BSD systems
1528 that have them, for controlling the resources used by pipe deliveries.
1530 PH/42 The content-scanning code uses fopen() to create files in which to put
1531 message data. Previously it was not paying any attention to the mode of
1532 the files. Exim runs with umask(0) because the rest of the code creates
1533 files with open(), and sets the required mode explicitly. Thus, these
1534 files were ending up world-writeable. This was not a big issue, because,
1535 being within the spool directory, they were not world-accessible. I have
1536 created a function called modefopen, which takes an additional mode
1537 argument. It sets umask(777), creates the file, chmods it to the required
1538 mode, then resets the umask. All the relevant calls to fopen() in the
1539 content scanning code have been changed to use this function.
1541 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1542 to 24 hours. This avoids potential overflow problems when processing G
1543 and H retry rules. I suspect nobody ever tinkers with this value.
1545 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1547 PH/45 When the plaintext authenticator is running as a client, the server's
1548 challenges are checked to ensure they are valid base64 strings. By
1549 default, the authentication attempt is cancelled if an invalid string is
1550 received. Setting client_ignore_invalid_base64 true ignores these errors.
1551 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1552 they are received. Thus, the responses can be made to depend on the
1553 challenges. If an invalid string is ignored, an empty string is placed in
1556 PH/46 Messages that are created by the autoreply transport now contains a
1557 References: header, in accordance with RFCs 2822 and 3834.
1559 PH/47 Added authenticated_sender_force to the smtp transport.
1561 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1563 PH/49 Installed latest patch from the Sieve maintainer.
1565 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1566 was also set, the appendfile transport was unnecessarily scanning a
1567 directory of message files (e.g. for maildir delivery) to find the count
1568 of files (along with the size), even though it did not need this
1569 information. It now does the scan only if it needs to find either the
1570 size of the count of files.
1572 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1574 PH/52 Two bugs concerned with error handling when the smtp transport is
1577 (i) Exim was not creating retry information for temporary errors given
1578 for individual recipients after the DATA command when the smtp transport
1579 was used in LMTP mode. This meant that they could be retried too
1580 frequently, and not timed out correctly.
1582 (ii) Exim was setting the flag that allows error details to be returned
1583 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1584 recipients that were returned after the DATA command.
1586 PH/53 This is related to PH/52, but is more general: for any failing address,
1587 when detailed error information was permitted to be returned to the
1588 sender, but the error was temporary, then after the final timeout, only
1589 "retry timeout exceeded" was returned. Now it returns the full error as
1590 well as "retry timeout exceeded".
1592 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1593 do this, and (what is worse) MTAs that accept it.
1595 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1596 will now be deprecated.
1598 PH/56 New os.c-cygwin from the Cygwin maintainer.
1600 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1601 formats (previously only available in exim formats via -bpr, -bpru,
1602 and -bpra. Now also available in native and exiqgrep formats)
1604 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1605 with very large, slow to parse queues
1607 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1609 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1611 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1612 responses to authentication challenges, though it was showing the
1613 challenges; (ii) I've removed the CR characters from the debug output for
1616 PH/58 Allow for the insertion of a newline as well as a space when a string
1617 is turned into more than one encoded-word during RFC 2047 encoding. The
1618 Sieve code now uses this.
1620 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1621 data_4xx, lost_connection, tls_required.
1623 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1624 message was being sent as an SMTP response.
1626 PH/61 Add -l and -k options to exicyclog.
1628 PH/62 When verifying, if an address was redirected to one new address, so that
1629 verification continued, and the new address failed or deferred after
1630 having set something in $address_data, the value of $address_data was not
1631 passed back to the ACL. This was different to the case when no
1632 redirection occurred. The value is now passed back in both cases.
1634 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1635 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1636 use login_cap.h, so on its own it isn't the distinguishing feature. The
1637 new name refers directly to the setclassresources() function.
1639 PH/65 Added configuration files for NetBSD3.
1641 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1643 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1644 is preferred over IPv4.
1646 PH/68 The bounce_return_message and bounce_return_body options were not being
1647 honoured for bounces generated during the reception of non-SMTP messages.
1648 In particular, this applied to messages rejected by the ACL. This bug has
1649 been fixed. However, if bounce_return_message is true and bounce_return_
1650 body is false, the headers that are returned for a non-SMTP message
1651 include only those that have been read before the error was detected.
1652 (In the case of an ACL rejection, they have all been read.)
1654 PH/69 The HTML version of the specification is now built in a directory called
1655 spec_html instead of spec.html, because the latter looks like a path with
1656 a MIME-type, and this confuses some software.
1658 PH/70 Catch two compiler warnings in sieve.c.
1660 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1661 function verify_get_ident() calls ip_connect() to connect a socket, but
1662 if the "connect()" function timed out, ip_connect() used to close the
1663 socket. However, verify_get_ident() also closes the socket later, and in
1664 between Exim writes to the log, which may get opened at this point. When
1665 the socket was closed in ip_connect(), the log could get the same file
1666 descriptor number as the socket. This naturally causes chaos. The fix is
1667 not to close the socket in ip_connect(); the socket should be closed by
1668 the function that creates it. There was only one place in the code where
1669 this was missing, in the iplookup router, which I don't think anybody now
1670 uses, but I've fixed it anyway.
1672 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1673 well as to direct DNS lookups. Otherwise the handling of names in host
1674 lists is inconsistent and therefore confusing.
1680 PH/01 Two changes to the default runtime configuration:
1682 (1) Move the checks for relay_from_hosts and authenticated clients from
1683 after to before the (commented out) DNS black list checks.
1685 (2) Add control=submission to the relay_from_hosts and authenticated
1686 clients checks, on the grounds that messages accepted by these
1687 statements are most likely to be submissions.
1689 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1691 (1) Generate an error if the third argument for the ${prvs expansion is
1694 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1697 (3) Reset the variables that are obtained from the first argument of
1698 ${prvscheck and used in the second argument before leaving the code,
1699 because their memory is reclaimed, so using them afterwards may do
1702 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1703 one (it's much easier than Tom thought :-).
1705 (5) Because of (4), we can now allow for the use of $prvscheck_result
1706 inside the third argument.
1708 PH/03 For some reason, the default setting of PATH when running a command from
1709 a pipe transport was just "/usr/bin". I have changed it to
1712 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1713 anything to be listed in the output from -bV.
1715 PH/05 When a filter generated an autoreply, the entire To: header line was
1716 quoted in the delivery log line, like this:
1718 => >A.N.Other <ano@some.domain> <original@ddress> ...
1720 This has been changed so that it extracts the operative address. There
1721 may be more than one such address. If so, they are comma-separated, like
1724 => >ano@some.domain,ona@other.domain <original@ddress> ...
1726 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1727 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1728 not being looked up in the rDNS to get a host name, Exim was showing the
1729 IP address twice in Received: lines, even though the IP addresses were
1730 identical. For example:
1732 Received: from [1.2.3.4] (helo=[1.2.3.4])
1734 However, if the real host name was known, it was omitting the HELO data
1735 if it matched the actual IP address. This has been tidied up so that it
1736 doesn't show the same IP address twice.
1738 PH/07 When both +timestamp and +memory debugging was on, the value given by
1739 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1740 called by the debug printing, thereby overwriting the timestamp buffer.
1741 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1743 PH/08 When the original message was included in an autoreply transport, it
1744 always said "this is a copy of the message, including all the headers",
1745 even if body_only or headers_only was set. It now gives an appropriate
1748 PH/09 Applied a patch from the Sieve maintainer which:
1750 o fixes some comments
1751 o adds the (disabled) notify extension core
1752 o adds some debug output for the result of if/elsif tests
1753 o points to the current vacation draft in the documentation
1754 and documents the missing references header update
1758 o fixes a bug in processing the envelope test (when testing
1759 multiple envelope elements, the last element determinted the
1762 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1763 Electronic Mail") by including:
1765 Auto-submitted: auto-generated
1767 in the messages that it generates (bounce messages and others, such as
1768 warnings). In the case of bounce messages for non-SMTP mesages, there was
1769 also a typo: it was using "Auto_submitted" (underscore instead of
1770 hyphen). Since every message generated by Exim is necessarily in response
1771 to another message, thes have all been changed to:
1773 Auto-Submitted: auto-replied
1775 in accordance with these statements in the RFC:
1777 The auto-replied keyword:
1779 - SHOULD be used on messages sent in direct response to another
1780 message by an automatic process,
1782 - MUST NOT be used on manually-generated messages,
1784 - MAY be used on Delivery Status Notifications (DSNs) and Message
1785 Disposition Notifications (MDNs),
1787 - MUST NOT be used on messages generated by automatic or periodic
1788 processes, except for messages which are automatic responses to
1791 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1792 to the default Received: header definition.
1794 PH/12 Added log selector acl_warn_skipped (default on).
1796 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1797 variables because (a) they are in the wrong storage pool and (b) even if
1798 they were copied, it wouldn't work properly because of the caching.
1800 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1801 checking when decoding. Apparently there are clients that generate
1802 overlong encoded strings. Why am I not surprised?
1804 PH/15 If the first argument of "${if match_address" was not empty, but did not
1805 contain an "@" character, Exim crashed. Now it writes a panic log message
1806 and treats the condition as false.
1808 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1810 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1811 extension "envelope-auth". The code is finished and in agreement with
1812 other implementations, but there is no documentation so far and in fact,
1813 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1814 not changing the active code.
1816 Print executed "if" and "elsif" statements when debugging is used. This
1817 helps a great deal to understand what a filter does.
1819 Document more things not specified clearly in RFC3028. I had all this
1820 sorted out, when out of a sudden new issues came to my mind. Oops."
1822 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1825 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1826 canonical form (as documented). However, after a host name lookup from
1827 the IP address, check_host() was doing a simple string comparison with
1828 addresses acquired from the DNS when checking that the found name did
1829 have the original IP as one of its addresses. Since any found IPv6
1830 addresses are likely to be in abbreviated form, the comparison could
1831 fail. Luckily, there already exists a function for doing the comparison
1832 by converting both addresses to binary, so now that is used instead of
1833 the text comparison.
1835 PH/20 There was another similar case to PH/19, when a complete host name was
1836 given in a host list; looking up its IP address could give an abbreviated
1837 form, whereas the current host's name might or might not be abbreviated.
1838 The same fix has been applied.
1844 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1845 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1848 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1849 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1851 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1853 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1854 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1855 Also, for "Auto-Submitted", treat anything other than "no" as
1856 non-personal, in accordance with RFC 3834. (Previously it treated
1857 anything starting "auto-" as non-personal.)
1859 TF/01 The control=submission/name=... option had a problem with syntax
1860 errors if the name included a slash character. The /name= option
1861 now slurps the rest of the string, so it can include any characters
1862 but it must come last in the list of options (after /sender_retain
1865 PH/05 Some modifications to the interface to the fake nameserver for the new
1873 TK/01 Added the "success_on_redirect" address verification option. See
1874 NewStuff for rationale and an example.
1876 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1878 PH/02 Patch to exigrep to allow it to work on syslog lines.
1880 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1881 fread() to skip over the body file's header line, because in Cygwin the
1882 header line is locked and is inaccessible.
1884 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1885 co-exist for some time) to make it clear that it is the Exim ID that is
1886 referenced, not the Message-ID: header line.
1888 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1889 string_format() function, because snprintf() does not exist on all
1892 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1893 ${stat: expansion item.
1895 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1896 protocol synchronization error", to keep the pedants happy.
1898 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1899 well as for IRIX systems, when gcc is being used. See the host.c source
1902 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1904 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1907 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1908 transport and to the smtp transport in LMTP mode.
1910 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1912 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1913 run a filter in a subprocess. This could lead to confusion in subsequent
1914 lookups in the parent process. There should also be a search_tidyup() at
1915 the end of the subprocess.
1917 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1918 only if the host matched helo_try_verify_hosts, which caused the
1919 verification to occur when the EHLO/HELO command was issued. The ACL just
1920 tested the remembered result. Now, if a previous verification attempt has
1921 not happened, "verify = helo" does it there and then.
1923 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1925 TK/03 Fix log output including CR from clamd.
1927 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1928 exist provoked a memory error which could cause a segfault.
1930 PH/15 Installed PCRE 6.2
1932 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1934 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1935 of the problem. Specifically, suggested +O2 rather than +O1 for the
1938 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1940 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1941 record, the port was not correctly logged when the outgoing_port log
1942 selector was set (it logged the transort's default port).
1944 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1945 fallback_hosts, and "hosts" in the smtp transport.
1947 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1948 host errors such as "Connection refused".
1950 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1951 authentication with radiusclient 0.4.9:
1953 - Error returned from rc_read_config was caught wrongly
1954 - Username/password not passed on to radius server due to wrong length.
1956 The presumption is that some radiusclient API changes for 4.51/PH/17
1957 were not taken care of correctly. The code is still untested by me (my
1958 Linux distribution still has 0.3.2 of radiusclient), but it was
1959 contributed by a Radius user.
1961 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1962 expanding the "port" option of the smtp transport.
1964 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1965 while reading a MIME header. Thanks to Tom Hughes for a patch.
1967 PH/24 Include config.h inside local_scan.h so that configuration settings are
1970 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1971 that in an ACL for RCPT (for example), you can examine exactly what was
1974 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1975 commands, but it was not correctly comparing the address with the actual
1976 client host address. Thus, it would show the EHLO address in Received:
1977 header lines when this was not necessary.
1979 PH/27 Added the % operator to ${eval:}.
1981 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1982 it should be ignoring failures (because with -C, for example, it has lost
1983 privilege). It wasn't ignoring creation failures other than "already
1986 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1989 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1990 because an input line was too long, either on its own, or by virtue of
1991 too many continuations, the temporary file was not being removed, and the
1992 return code was incorrect.
1994 PH/31 Missing "BOOL" in function definition in filtertest.c.
1996 PH/32 Applied Sieve patches from the maintainer.
1998 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2000 PH/33 Added "verify = not_blind".
2002 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2003 Local/Makefile (with some defaults set). These are used in built scripts
2004 such as exicyclog, but they have never been used in the exim_install
2005 script (though there are many overriding facilities there). I have
2006 arranged that the exim_install script now takes note of these two
2009 PH/35 Installed configuration files for Dragonfly.
2011 PH/36 When a locally submitted message by a trusted user did not contain a
2012 From: header, and the sender address was obtained from -f or from an SMTP
2013 MAIL command, and the trusted user did not use -F to supply a sender
2014 name, $originator_name was incorrectly used when constructing a From:
2015 header. Furthermore, $originator_name was used for submission mode
2016 messages from external hosts without From: headers in a similar way,
2017 which is clearly wrong.
2019 PH/37 Added control=suppress_local_fixups.
2021 PH/38 When log_selector = +received_sender was set, and the addition of the
2022 sender made the log line's construction buffer exactly full, or one byte
2023 less than full, an overflow happened when the terminating "\n" was
2026 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2027 when the result of a list match is failure because a DNS lookup failed.
2029 PH/40 RM_COMMAND is now used in the building process.
2031 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2032 the "build-* directories that it finds.
2034 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2035 address in a domain literal was a prefix of an interface address.
2037 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2038 when verifying a sender address, unless rewrite_headers is false.
2040 PH/44 Wrote a long comment about why errors_to addresses are verified as
2041 recipients, not senders.
2043 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2044 the ratelimit ACL was added.
2046 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2048 PH/46 Added extra information about PostgreSQL errors to the error string.
2050 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2051 avoiding the need to install special zones in a real server. This is
2052 backwards compatible; if it can't find the fake resolver, it drops back.
2053 Thus, both old and new test suites can be run.
2055 TF/02 Added util/ratelimit.pl
2057 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2058 clock is set back in time.
2060 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2061 Candler <B.Candler@pobox.com>.
2063 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2064 OK for addresses that are the result of redirections.
2066 PH/48 A number of further additions for the benefit of the new test suite,
2067 including a fake gethostbyname() that interfaces to the fake DNS resolver
2070 TF/06 The fix for widen_domains has also been applied to qualify_single and
2071 search_parents which are the other dnslookup options that can cause
2074 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2077 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2079 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2080 service" if it encountered an MX record with an empty target hostname.
2081 The message is now "an MX or SRV record indicated no SMTP service".
2083 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2084 if the DNS of the sending site is misconfigured. This is quite a
2085 common situation. This change restores the behaviour of treating a
2086 helo verification defer as a failure.
2088 PH/51 If self=fail was set on a router, the bounce message did not include the
2089 actual error message.
2095 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2097 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2098 command itself ended in error, the underlying message about the transport
2099 filter timeout was being overwritten with the pipe command error. Now the
2100 underlying error message should be appended to the second error message.
2102 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2104 PH/02 Reduce the amount of output that "make" produces by default. Full output
2105 can still be requested.
2107 PH/03 The warning log line about a condition test deferring for a "warn" verb
2108 was being output only once per connection, rather than after each
2109 occurrence (because it was using the same function as for successful
2110 "warn" verbs). This seems wrong, so I have changed it.
2112 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2113 it should not have, which might have caused a crash in the right
2114 circumstances, but probably never did.
2116 PH/04 Installed a modified version of Tony Finch's patch to make submission
2117 mode fix the return path as well as the Sender: header line, and to
2118 add a /name= option so that you can make the user's friendly name appear
2121 TF/03 Added the control = fakedefer ACL modifier.
2123 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2124 Mark Lowes for thorough testing.
2126 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2128 TK/03 Merged latest SRS patch from Miles Wilton.
2130 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2131 with the definition in sysexits.h (which is #included earlier).
2132 Fortunately, Exim does not actually use EX_OK. The code used to try to
2133 preserve the sysexits.h value, by assumimg that macro definitions were
2134 scanned for macro replacements. I have been disabused of this notion,
2135 so now the code just undefines EX_OK before #including unistd.h.
2137 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2138 in the smtp transport. When a block could not be written in a single
2139 write() function, the timeout was being re-applied to each part-write.
2140 This seems wrong - if the receiver was accepting one byte at a time it
2141 would take for ever. The timeout is now adjusted when this happens. It
2142 doesn't have to be particularly precise.
2144 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2145 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2147 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2148 without a domain if the check to <postmaster@domain> fails.
2150 SC/01 Eximstats: added -xls and the ability to specify output files
2151 (patch written by Frank Heydlauf).
2153 SC/02 Eximstats: use FileHandles for outputing results.
2155 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2157 SC/04 Eximstats: fixed display of large numbers with -nvr option
2159 SC/05 Eximstats: fixed merging of reports with empty tables.
2161 SC/06 Eximstats: added the -include_original_destination flag
2163 SC/07 Eximstats: removed tabs and trailing whitespace.
2165 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2167 TK/06 MBOX spool code: Add real "From " MBOX separator line
2168 so the .eml file is really in mbox format (even though
2169 most programs do not really care). Patch from Alex Miller.
2171 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2172 The latter is generated from $received_to and is only set if the
2173 message has one envelope recipient. SA can use these headers,
2174 obviously out-of-the-box. Patch from Alex Miller.
2176 PH/08 The ${def test on a variable was returning false if the variable's
2177 value was "0", contrary to what the specification has always said!
2178 The result should be true unless the variable is empty.
2180 PH/09 The syntax error of a character other than { following "${if
2181 def:variable_name" (after optional whitespace) was not being diagnosed.
2182 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2183 accidental colon was present, for example, could give incorrect results.
2185 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2186 result is used (not including appendfile, where other changes are about
2189 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2190 This involved changing a lot of size variables from int to off_t. It
2191 should work with maildirs and everything.
2193 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2194 spamd dying while we are connected to it.
2196 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2199 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2200 function for BDB 4.3.
2202 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2204 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2205 the sources for PCRE have been re-arranged and more files are now
2208 PH/15 The code I had for printing potentially long long variables in PH/11
2209 above was not the best (it lost precision). The length of off_t variables
2210 is now inspected at build time, and an appropriate printing format (%ld
2211 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2212 to be "long long int" or "long int". This is needed for the internal
2213 formatting function string_vformat().
2215 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2216 the configuration file to be ":syslog", then the script "guesses" where
2217 the logs files are, rather than using the compiled in default. In our
2218 case the guess is not the same as the compiled default, so the script
2219 suddenly stopped working when I started to use syslog. The patch checks
2220 to see if log_file_path is "". If so, it attempts to read it from exim
2221 with no configuration file to get the compiled in version, before it
2222 falls back to the previous guessing code."
2224 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2225 implementing BATV in an Exim configuration. See NewStuff for the gory
2228 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2229 Makefile that are specific to HP-UX.
2231 PH/18 If the "use_postmaster" option was set for a recipient callout together
2232 with the "random" option, the postmaster address was used as the MAIL
2233 FROM address for the random test, but not for the subsequent recipient
2234 test. It is now used for both.
2236 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2237 patch removes a few documentation additions to RFC 3028, because the
2238 latest draft now contains them. It adds the new en;ascii-case comparator
2239 and a new error check for 8bit text in MIME parts. Comparator and
2240 require names are now matched exactly. I enabled the subaddress
2241 extension, but it is not well tested yet (read: it works for me)."
2243 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2244 rework some of the code of TK/09 above to avoid the hardwired use of
2245 "%lld" and "long long". Replaced the call to snprintf() with a call to
2248 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2249 records point to non-existent hosts", "retry timeout exceeded", and
2250 "retry time not reached for any host after a long failure period".
2252 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2253 experimental DomainKeys support:
2255 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2256 (2) On an error such as an illegally used "control", the wrong name for
2257 the control was given.
2259 These problems did NOT occur unless DomainKeys support was compiled.
2261 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2263 PH/24 Added ${if match_ip condition.
2265 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2266 databases so that it will be absolutely obvious if a crash occurs in the
2267 DB library. This is a regular occurrence (often caused by mis-matched
2270 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2271 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2272 picked up on a user's system that detects such things. There doesn't seem
2273 to be a gcc warning option for this - only an attribute that has to be
2274 put on the function's prototype. It seems that in Fedora Core 4 they have
2275 set this on a number of new functions. No doubt there will be more in due
2278 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2279 specify a transport. However, if an address that was verified by such a
2280 router was the subject of a callout, Exim crashed because it tried to
2281 read the rcpt_include_affixes from the non-existent transport. Now it
2282 just assumes that the setting of that option is false. This bug was
2283 introduced by 4.51/PH/31.
2285 PH/28 Changed -d+all to exclude +memory, because that information is very
2286 rarely of interest, but it makes the output a lot bigger. People tend to
2287 do -d+all out of habit.
2289 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2290 code in os-type was giving problems when libc.so lives in lib64, like on
2293 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2294 aren't the modern standard, and it seems that some systems' include files
2295 don't always have them. Exim was already checking for some of the newer
2296 ones like T_AAAA, and defining it itself. I've added checks for all the
2297 record types that Exim uses.
2299 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2300 not automatically generating a new one, as it is supposed to. This
2301 prevented TLS from working. If the file did exist, but contained invalid
2302 data, a new version was generated, as expected. It was only the case of a
2303 non-existent file that was broken.
2305 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2306 with a change in libdomainkeys > 0.64.
2308 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2309 from DNS. If the selector record carries the flag, it now has
2310 precedence over the domain-wide flag.
2312 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2314 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2315 the use of a port name that isn't defined in /etc/services) occurred, the
2316 message was deferred as in a normal delivery, and thus remained on the
2317 spool, instead of being failed because of the mua_wrapper setting. This
2318 is now fixed, and I tidied up some of the mua_wrapper messages at the
2321 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2322 the messages in a hash of arrays rather than using individual hashes.
2323 This is a bit cleaner and results in dramatic memory savings, albeit
2324 at a slight CPU cost.
2326 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2327 as requested by Marc Sherman.
2329 SC/10 Eximstats: added histograms for user specified patterns as requested
2332 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2334 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2335 fopen() in the content-scanning modules that did not already have it.
2341 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2342 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2344 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2346 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2348 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2349 to test Sieve filters that use "vacation".
2351 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2352 that changes the way the GnuTLS parameters are stored in the cache file.
2353 The new format can be generated externally. For backward compatibility,
2354 if the data in the cache doesn't make sense, Exim assumes it has read an
2355 old-format file, and it generates new data and writes a new file. This
2356 means that you can't go back to an older release without removing the
2359 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2360 work if there are any delivery delays because "one_time" forces the
2361 parent to be marked "delivered", so its unseen clone is never tried
2362 again. For this reason, Exim now forbids the simultaneous setting of
2365 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2366 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2367 entry if you want to know the details. The fix, however, neglected to
2368 consider the case when local delivery batching is involved. The test for
2369 "previously delivered" was not happening when checking to see if an
2370 address could be batched with a previous (undelivered) one; under
2371 certain circumstances this could lead to multiple deliveries to the same
2374 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2375 in its include files, and this causes problems building Exim.
2377 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2378 header_syntax) but Exim was just ignoring anything given after a slash.
2379 In particular, this caused confusion with an attempt to use "verify =
2380 reverse_host_lookup/defer_ok". An error is now given when options are
2381 supplied for verify items that do not have them. (Maybe reverse_host_
2382 lookup should have a defer_ok option, but that's a different point.)
2384 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2385 defined by RFC 821) to 2048, because there were problems with some AUTH
2386 commands, and RFC 1869 says the size should be increased for extended
2387 SMTP commands that take arguments.
2389 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2392 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2393 "unknown" error; now it says that the functionality isn't in the binary.
2395 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2396 an address' error message when a string expansion fails (syntax or
2397 whatever). Otherwise the password may appear in the log. Following change
2398 PH/42 below, there is no longer a chance of it appearing in a bounce
2401 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2403 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2404 of its addresses was ever tried. (Bugzilla bug #2.)
2406 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2407 the result incorrectly in the debug output. (It correctly added a newline
2408 to what was transported.)
2410 TF/01 Added $received_time.
2412 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2413 commented out examples of how to interface to a virus scanner and to
2414 SpamAssassin. Also added commented examples of av_scanner and
2415 spamd_address settings.
2417 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2418 and controls are allowed in which ACLs. There were a couple of minor
2419 errors. Some of the entries in the conditions table (which is a table of
2420 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2421 negation of where the condition IS allowed.
2423 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2425 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2426 header file does not have a version number, so I've had to invent a new
2427 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2428 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2429 radiusclient), but it was contributed by a Radius user.
2431 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2432 files or directories, for OpenSSL.
2434 PH/19 When an Exim process that is running as root has to create an Exim log
2435 file, it does so in a subprocess that runs as exim:exim so as to get the
2436 ownership right at creation (otherwise, other Exim processes might see
2437 the file with the wrong ownership). There was no test for failure of this
2438 fork() call, which would lead to the process getting stuck as it waited
2439 for a non-existent subprocess. Forks do occasionally fail when resources
2440 run out. I reviewed all the other calls to fork(); they all seem to check
2443 PH/20 When checking for unexpected SMTP input at connect time (before writing
2444 the banner), Exim was not dealing correctly with a non-positive return
2445 from the read() function. If the client had disconnected by this time,
2446 the result was a log entry for a synchronization error with an empty
2447 string after "input=" when read() returned zero. If read() returned -1
2448 (an event I could not check), uninitialized data bytes were printed.
2449 There were reports of junk text (parts of files, etc) appearing after
2452 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2454 PH/22 Added support for macro redefinition, and (re)definition in between
2455 driver and ACL definitions.
2457 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2458 forgetting to use the resulting value; it was using the unexpanded value.
2460 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2461 hadn't been configured. The fix is from Juergen Kreileder, who
2462 understands it better than I do:
2464 "Here's what I see happening with three configured cyrus_sasl
2465 authenticators configured (plain, login, cram-md5):
2467 On startup auth_cyrus_sasl_init() gets called for each of these.
2468 This means three calls to sasl_listmech() without a specified mech_list.
2469 => SASL tests which mechs of all available mechs actually work
2470 => three warnings about OTP not working
2471 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2473 With the patch, sasl_listmech() also gets called three times. But now
2474 SASL's mech_list option is set to the server_mech specified in the the
2475 authenticator. Or in other words, the answer from sasl_listmech()
2476 gets limited to just the mech you're testing for (which is different
2478 => the return list contains just 'plain' or 'login', 'cram-md5' or
2479 nothing depending on the value of ob->server_mech.
2481 I've just tested the patch: Authentication still works fine,
2482 unavailable mechs specified in the exim configuration are still
2483 caught, and the auth.log warnings about OTP are gone."
2485 PH/25 When debugging is enabled, the contents of the command line are added
2486 to the debugging output, even when log_selector=+arguments is not
2489 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2490 answer is "GNU", and only if the return is "GNU/something" is the answer
2493 PH/27 $acl_verify_message is now set immediately after the failure of a
2494 verification in an ACL, and so is available in subsequent modifiers. In
2495 particular, the message can be preserved by coding like this:
2497 warn !verify = sender
2498 set acl_m0 = $acl_verify_message
2500 Previously, $acl_verify_message was set only while expanding "message"
2501 and "log_message" when a very denied access.
2503 PH/28 Modified OS/os.c-Linux with
2505 -#ifndef OS_LOAD_AVERAGE
2506 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2508 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2509 nomenclature these days.)
2511 PH/29 Installed patch from the Sieve maintainer that adds the options
2512 sieve_useraddress and sieve_subaddress to the redirect router.
2514 PH/30 In these circumstances:
2515 . Two addresses routed to the same list of hosts;
2516 . First host does not offer TLS;
2517 . First host accepts first address;
2518 . First host gives temporary error to second address;
2519 . Second host offers TLS and a TLS session is established;
2520 . Second host accepts second address.
2521 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2522 and peerdn, if requested) that were in fact used only for the second
2525 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2526 attention to any local part prefix or suffix that was matched by the
2527 router that accepted the address. It now behaves in the same way as it
2528 does for delivery: the affixes are removed from the local part unless
2529 rcpt_include_affixes is set on the transport.
2531 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2532 timeout during the DATA phase of an incoming message.
2534 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2535 applied a patch sanctioned by the Sieve maintainer.
2537 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2538 the uid or gid is negative. A case of a negative gid caused this to be
2539 noticed. The fix allows for either to be negative.
2541 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2542 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2545 PH/36 The change PH/12 above was broken. Fixed it.
2547 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2548 the grounds that routing the same address twice would always produce the
2549 same answer. This might have been true once, but it is certainly no
2550 longer true now. Routing a child address may depend on the previous
2551 routing that produced that child. Some complicated redirection strategies
2552 went wrong when messages had multiple recipients, and made Exim's
2553 behaviour dependent on the order in which the addresses were given.
2555 I have moved the duplicate checking until after the routing is complete.
2556 Exim scans the addresses that are assigned to local and remote
2557 transports, and removes any duplicates. This means that more work will be
2558 done, as duplicates will always all be routed, but duplicates are
2559 presumably rare, so I don't expect this is of any significance.
2561 For deliveries to pipes, files, and autoreplies, the duplicate checking
2562 still happens during the routing process, since they are not going to be
2565 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2566 It corrects a timeout issue with spamd. This is Ian's comment: "The
2567 background is that sometimes spamd either never reads data from a
2568 connection it has accepted, or it never writes response data. The exiscan
2569 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2570 blindly assumes that writes won't block so it may never time out."
2572 PH/39 Allow G after quota size as well as K and M.
2574 PH/40 The value set for $authenticated_id in an authenticator may not contain
2575 binary zeroes or newlines because the value is written to log lines and
2576 to spool files. There was no check on this. Now the value is run through
2577 the string_printing() function so that such characters are converted to
2578 printable escape sequences.
2580 PH/41 $message_linecount is a new variable that contains the total number of
2581 lines in the message. Compare $body_linecount, which is the count for the
2584 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2585 bounce and delay warning messages, except in certain special cases, which
2588 (a) An SMTP error message from a remote host;
2589 (b) A message specified in a :fail: redirection;
2590 (c) A message specified in a "fail" command in a system filter;
2591 (d) A message specified in a FAIL return from the queryprogram router;
2592 (e) A message specified by the cannot_route_message router option.
2594 In these cases only, Exim does include the error details in bounce and
2595 warning messages. There are also a few cases where bland messages such
2596 as "unrouteable address" or "local delivery error" are given.
2598 PH/43 $value is now also set for the "else" part of a ${run expansion.
2600 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2601 being worked on, but at least Exim now implements the latest version to
2604 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2605 process to complete was treated as a delivery failure, a timeout while
2606 writing the message to the pipe was logged, but erroneously treated as a
2607 successful delivery. Such timeouts include transport filter timeouts. For
2608 consistency with the overall process timeout, these timeouts are now
2609 treated as errors, giving rise to delivery failures by default. However,
2610 there is now a new Boolean option for the pipe transport called
2611 timeout_defer, which, if set TRUE, converts the failures into defers for
2612 both kinds of timeout. A transport filter timeout is now identified in
2615 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2616 systems where "make" and "gmake" are different, calling "gmake" at top
2617 level broke things. I've arranged for the value of $(MAKE) to be passed
2618 from the Makefile to this script so that it can call the same version of
2622 A note about Exim versions 4.44 and 4.50
2623 ----------------------------------------
2625 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2626 changes of various kinds. As a consequence, a big documentation update was
2627 needed. This delayed the release for rather longer than seemed good, especially
2628 in the light of a couple of (minor) security issues. Therefore, the changes
2629 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2630 release. So 4.44 and 4.50 are in effect two different branches that both start
2633 I have left the 4.50 change log unchanged; it contains all the changes since
2634 4.43. The change log for 4.44 is below; many of its items are identical to
2635 those for 4.50. This seems to be the most sensible way to preserve the
2636 historical information.
2642 1. Minor wording change to the doc/README.SIEVE file.
2644 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2645 computation of the current number of files was incorrect.
2647 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2648 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2651 4. Give more explanation in the error message when the command for a transport
2652 filter fails to execute.
2654 5. There are several places where Exim runs a non-Exim command in a
2655 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2656 was being done only for the command run by the queryprogram router. It is
2657 now done for all such subprocesses. The other cases are: ${run, transport
2658 filters, and the commands run by the lmtp and pipe transports.
2660 6. Added CONFIGURE_GROUP build-time option.
2662 7. Some older OS have a limit of 256 on the maximum number of file
2663 descriptors. Exim was using setrlimit() to set 1000 as a large value
2664 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2665 systems. I've change it so that if it can't get 1000, it tries for 256.
2667 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2668 was an oversight, and furthermore, ever since the addition of extra
2669 controls (e.g. 4.43/32), the checks on when to allow different forms of
2670 "control" were broken. There should now be diagnostics for all cases when a
2671 control that does not make sense is encountered.
2673 9. Added the /retain_sender option to "control=submission".
2675 10. $recipients is now available in the predata ACL (oversight).
2677 11. Tidy the search cache before the fork to do a delivery from a message
2678 received from the command line. Otherwise the child will trigger a lookup
2679 failure and thereby defer the delivery if it tries to use (for example) a
2680 cached ldap connection that the parent has called unbind on.
2682 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2683 of $address_data from the recipient verification was clobbered by the
2684 sender verification.
2686 13. The value of address_data from a sender verification is now available in
2687 $sender_address_data in subsequent conditions in the ACL statement.
2689 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2691 15. Added a new option "connect=<time>" to callout options, to set a different
2694 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2695 was its contents. (It was OK if the option was not defined at all.)
2697 17. A "Completed" log line is now written for messages that are removed from
2698 the spool by the -Mrm option.
2700 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2701 information about exactly what failed.
2703 19. Added -dd to debug only the daemon process.
2705 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2706 handles timeouts, both on the server side and network timeouts. Renamed the
2707 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2709 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2710 It is now set to "smtps".
2712 22. $host_address is now set to the target address during the checking of
2713 ignore_target_hosts.
2715 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2716 being passed; this would have caused $sender_host_name to have been used if
2717 matching the list had actually called for a host name (not very likely,
2718 since this list is usually IP addresses). A host name is now passed as
2721 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2722 code that specifies a non-restarting handler (typically sigaction() in
2723 modern systems) in an attempt to fix a rare and obscure crash bug.
2725 25. Narrowed the window for a race in the daemon that could cause it to ignore
2726 SIGCHLD signals. This is not a major problem, because they are used only to
2727 wake it up if nothing else does.
2729 26. A malformed maildirsize file could cause Exim to calculate negative values
2730 for the mailbox size or file count. Odd effects could occur as a result.
2731 The maildirsize information is now recalculated if the size or filecount
2734 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2735 support for a long time. Removed HAVE_SYS_VFS_H.
2737 28. Installed the latest version of exipick from John Jetmore.
2739 29. In an address list, if the pattern was not a regular expression, an empty
2740 subject address (from a bounce message) matched only if the pattern was an
2741 empty string. Non-empty patterns were not even tested. This was the wrong
2742 because it is perfectly reasonable to use an empty address as part of a
2743 database query. An empty address is now tested by patterns that are
2744 lookups. However, all the other forms of pattern expect the subject to
2745 contain a local part and a domain, and therefore, for them, an empty
2746 address still always fails if the pattern is not itself empty.
2748 30. Exim went into a mad DNS loop when attempting to do a callout where the
2749 host was specified on an smtp transport, and looking it up yielded more
2750 than one IP address.
2752 31. Re-factored the code for checking spool and log partition space into a
2753 function that finds that data and another that does the check. The former
2754 is then used to implement four new variables: $spool_space, $log_space,
2755 $spool_inodes, and $log_inodes.
2757 32. The RFC2047 encoding function was originally intended for short strings
2758 such as real names; it was not keeping to the 75-character limit for
2759 encoded words that the RFC imposes. It now respects the limit, and
2760 generates multiple encoded words if necessary. To be on the safe side, I
2761 have increased the buffer size for the ${rfc2047: expansion operator from
2764 33. It is now permitted to omit both strings after an "if" condition; if the
2765 condition is true, the result is "true". As before, when the second string
2766 is omitted, a false condition yields an empty string. This makes it less
2767 cumbersome to write custom ACL and router conditions.
2769 34. Failure to deliver a bounce message always caused it to be frozen, even if
2770 there was an errors_to setting on the router. The errors_to setting is now
2773 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2774 canonical form (fully expanded) before being placed in
2775 $sender_host_address.
2777 36. The table in the code that translates DNS record types into text (T_A to
2778 "A" for instance) was missing entries for NS and CNAME. It is just possible
2779 that this could have caused confusion if both these types were looked up
2780 for the same domain, because the text type is used as part of Exim's
2781 per-process caching. But the chance of anyone hitting this buglet seems
2784 37. The dnsdb lookup has been extended in a number of ways.
2786 (1) There is a new type, "zns", which walks up the domain tree until it
2787 finds some nameserver records. It should be used with care.
2789 (2) There is a new type, "mxh", which is like "mx" except that it returns
2790 just the host names, not the priorities.
2792 (3) It is now possible to give a list of domains (or IP addresses) to be
2793 looked up. The behaviour when one of the lookups defers can be
2794 controlled by a keyword.
2796 (4) It is now possible to specify the separator character for use when
2797 multiple records are returned.
2799 38. The dnslists ACL condition has been extended: it is now possible to supply
2800 a list of IP addresses and/or domains to be looked up in a particular DNS
2803 39. Added log_selector=+queue_time_overall.
2805 40. When running the queue in the test harness, wait just a tad after forking a
2806 delivery process, to get repeatability of debugging output.
2808 41. Include certificate and key file names in error message when GnuTLS fails
2809 to set them up, because the GnuTLS error message doesn't include the name
2810 of the failing file when there is a problem reading it.
2812 42. Allow both -bf and -bF in the same test run.
2814 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2816 44. The "Exiscan patch" is now merged into the mainline Exim source.
2818 45. Sometimes the final signoff response after QUIT could fail to get
2819 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2820 < 0 before doing a fflush(). This bug looks as though it goes back to the
2821 introduction of TLS in release 3.20, but "sometimes" must have been rare
2822 because the tests only now provoked it.
2824 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2825 (this can affect the format of dates).
2827 47. exim_tidydb, when checking for the continued existence of a message for
2828 which it has found a message-specific retry record, was not finding
2829 messages that were in split spool directories. Consequently, it was
2830 deleting retry records that should have stayed in existence.
2832 48. Steve fixed some bugs in eximstats.
2834 49. The SPA authentication driver was not abandoning authentication and moving
2835 on to the next authenticator when an expansion was forced to fail,
2836 contradicting the general specification for all authenticators. Instead it
2837 was generating a temporary error. It now behaves as specified.
2839 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2840 (the order specifies the preference for clients). The order is now AES256,
2841 AES128, 3DES, ARCFOUR128.
2843 51. Small patch to Sieve code - explicitly set From: when generating an
2846 52. Exim crashed if a remote delivery caused a very long error message to be
2847 recorded - for instance if somebody sent an entire SpamAssassin report back
2848 as a large number of 550 error lines. This bug was coincidentally fixed by
2849 increasing the size of one of Exim's internal buffers (big_buffer) that
2850 happened as part of the Exiscan merge. However, to be on the safe side, I
2851 have made the code more robust (and fixed the comments that describe what
2854 53. Now that there can be additional text after "Completed" in log lines (if
2855 the queue_time_overall log selector is set), a one-byte patch to exigrep
2856 was needed to allow it to recognize "Completed" as not the last thing in
2859 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2860 patch that reportedly fixes this has been added. I am not expert enough to
2861 create a test for it. This is what the patch creator wrote:
2863 "I found a little strange behaviour of ldap code when working with
2864 Windows 2003 AD Domain, where users was placed in more than one
2865 Organization Units. When I tried to give exim partial DN, the exit code
2866 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2867 But simultaneously result of request was absolutely normal ldap result,
2868 so I produce this patch..."
2870 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2871 REFERENCE, so I have modified the code to exclude the patch when that macro
2874 55. Some experimental protocols are using DNS PTR records for new purposes. The
2875 keys for these records are domain names, not reversed IP addresses. The
2876 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2877 leaves it alone. Component reversal etc. now happens only for IP addresses.
2880 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2882 57. Double the size of the debug message buffer (to 2048) so that more of very
2883 long debug lines gets shown.
2885 58. The exicyclog utility now does better if the number of log files to keep
2886 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2888 59. Two changes related to the smtp_active_hostname option:
2890 (1) $smtp_active_hostname is now available as a variable.
2891 (2) The default for smtp_banner uses $smtp_active_hostname instead
2892 of $primary_hostname.
2894 60. The host_aton() function is supposed to be passed a string that is known
2895 to be a valid IP address. However, in the case of IPv6 addresses, it was
2896 not checking this. This is a hostage to fortune. Exim now panics and dies
2897 if the condition is not met. A case was found where this could be provoked
2898 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2899 components; fortuitously, this particular loophole had already been fixed
2900 by change 4.50/55 above.
2902 If there are any other similar loopholes, the new check in host_aton()
2903 itself should stop them being exploited. The report I received stated that
2904 data on the command line could provoke the exploit when Exim was running as
2905 exim, but did not say which command line option was involved. All I could
2906 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2907 running as the user.
2910 61. There was a buffer overflow vulnerability in the SPA authentication code
2911 (which came originally from the Samba project). I have added a test to the
2912 spa_base64_to_bits() function which I hope fixes it.
2915 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2916 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2918 63. The daemon start-up calls getloadavg() while still root for those OS that
2919 need the first call to be done as root, but it missed one case: when
2920 deliver_queue_load_max is set with deliver_drop_privilege. This is
2921 necessary for the benefit of the queue runner, because there is no re-exec
2922 when deliver_drop_privilege is set.
2924 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2925 This has been fixed.
2927 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2928 was in use, was not putting the data itself into the right store pool;
2929 consequently, it could be overwritten for a subsequent message in the same
2930 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2933 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2935 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2936 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2937 the function were treating the return as a boolean value, which happened to
2938 work because 0=false and not-0=true, but is not correct code.
2940 68. The host_aton() function was not handling scoped IPv6 addresses (those
2941 with, for example, "%eth0" on the end) correctly.
2943 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2944 negated items (that is, ~something) in unsigned ints. Some compilers
2945 apparently mutter when there is no cast.
2947 70. If an address verification called from an ACL failed, and did not produce a
2948 user-specific message (i.e. there was only a "system" message), nothing was
2949 put in $acl_verify_message. In this situation, it now puts the system
2952 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2953 session; change 4.31/43 added the unwanted input to the log line - except
2954 that it did not do this in the start of session case. It now does.
2956 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2957 This is wrong and can cause the other end to generate a synchronization
2958 error if it is another Exim or anything else that does the synchronization
2959 check. A QUIT command is no longer sent after a timeout.
2961 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2962 during host lookups.
2964 74. The defer_ok option of callout verification was not working if it was used
2965 when verifying addresses in header lines, that is, for this case:
2967 verify = header_sender/callout=defer_ok
2969 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2970 those file descriptors could be used for SMTP connections. If anything
2971 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2972 could be sent to the SMTP client, causing chaos. The daemon now opens
2973 stdin, stdout, and stderr to /dev/null when it puts itself into the
2976 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2977 log by default. The user can override this with suitable Perl magic.
2979 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2980 the log message when discard triggers, was not working for the DATA ACL or
2981 for the non-SMTP ACL.
2983 78. Error message wording change in sieve.c.
2985 79. If smtp_accept_max_per_host was set, the number of connections could be
2986 restricted to fewer than expected, because the daemon was trying to set up
2987 a new connection before checking whether the processes handling previous
2988 connections had finished. The check for completed processes is now done
2989 earlier. On busy systems, this bug wouldn't be noticed because something
2990 else would have woken the daemon, and it would have reaped the completed
2993 80. If a message was submitted locally by a user whose login name contained one
2994 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2995 It caused a spool format error. I have fixed the spool reading code. A
2996 related problem was that the "from" clause in the Received: line became
2997 illegal because of the space(s). It is now covered by ${quote_local_part.
2999 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3002 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3004 83. Patch from Sieve maintainer to fix unterminated string problem in
3005 "vacation" handling.
3007 84. Some minor changes to the Linux configuration files to help with other
3008 OS variants using glibc.
3010 85. One more patch for Sieve to update vacation handling to latest spec.
3013 ----------------------------------------------------
3014 See the note above about the 4.44 and 4.50 releases.
3015 ----------------------------------------------------
3021 1. Change 4.43/35 introduced a bug that caused file counts to be
3022 incorrectly computed when quota_filecount was set in an appendfile
3025 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3026 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3029 3. Give more explanation in the error message when the command for a transport
3030 filter fails to execute.
3032 4. There are several places where Exim runs a non-Exim command in a
3033 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3034 was being done only for the command run by the queryprogram router. It is
3035 now done for all such subprocesses. The other cases are: ${run, transport
3036 filters, and the commands run by the lmtp and pipe transports.
3038 5. Some older OS have a limit of 256 on the maximum number of file
3039 descriptors. Exim was using setrlimit() to set 1000 as a large value
3040 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3041 systems. I've change it so that if it can't get 1000, it tries for 256.
3043 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3044 was an oversight, and furthermore, ever since the addition of extra
3045 controls (e.g. 4.43/32), the checks on when to allow different forms of
3046 "control" were broken. There should now be diagnostics for all cases when a
3047 control that does not make sense is encountered.
3049 7. $recipients is now available in the predata ACL (oversight).
3051 8. Tidy the search cache before the fork to do a delivery from a message
3052 received from the command line. Otherwise the child will trigger a lookup
3053 failure and thereby defer the delivery if it tries to use (for example) a
3054 cached ldap connection that the parent has called unbind on.
3056 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3057 of $address_data from the recipient verification was clobbered by the
3058 sender verification.
3060 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3061 was its contents. (It was OK if the option was not defined at all.)
3063 11. A "Completed" log line is now written for messages that are removed from
3064 the spool by the -Mrm option.
3066 12. $host_address is now set to the target address during the checking of
3067 ignore_target_hosts.
3069 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3070 being passed; this would have caused $sender_host_name to have been used if
3071 matching the list had actually called for a host name (not very likely,
3072 since this list is usually IP addresses). A host name is now passed as
3075 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3076 code that specifies a non-restarting handler (typically sigaction() in
3077 modern systems) in an attempt to fix a rare and obscure crash bug.
3079 15. Narrowed the window for a race in the daemon that could cause it to ignore
3080 SIGCHLD signals. This is not a major problem, because they are used only to
3081 wake it up if nothing else does.
3083 16. A malformed maildirsize file could cause Exim to calculate negative values
3084 for the mailbox size or file count. Odd effects could occur as a result.
3085 The maildirsize information is now recalculated if the size or filecount
3088 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3089 support for a long time. Removed HAVE_SYS_VFS_H.
3091 18. Updated exipick to current release from John Jetmore.
3093 19. Allow an empty sender to be matched against a lookup in an address list.
3094 Previously the only cases considered were a regular expression, or an
3097 20. Exim went into a mad DNS lookup loop when doing a callout where the
3098 host was specified on the transport, if the DNS lookup yielded more than
3101 21. The RFC2047 encoding function was originally intended for short strings
3102 such as real names; it was not keeping to the 75-character limit for
3103 encoded words that the RFC imposes. It now respects the limit, and
3104 generates multiple encoded words if necessary. To be on the safe side, I
3105 have increased the buffer size for the ${rfc2047: expansion operator from
3108 22. Failure to deliver a bounce message always caused it to be frozen, even if
3109 there was an errors_to setting on the router. The errors_to setting is now
3112 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3113 canonical form (fully expanded) before being placed in
3114 $sender_host_address.
3116 24. Updated eximstats to version 1.33
3118 25. Include certificate and key file names in error message when GnuTLS fails
3119 to set them up, because the GnuTLS error message doesn't include the name
3120 of the failing file when there is a problem reading it.
3122 26. Expand error message when OpenSSL has problems setting up cert/key files.
3125 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3126 (this can affect the format of dates).
3128 28. exim_tidydb, when checking for the continued existence of a message for
3129 which it has found a message-specific retry record, was not finding
3130 messages that were in split spool directories. Consequently, it was
3131 deleting retry records that should have stayed in existence.
3133 29. eximstats updated to version 1.35
3134 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3135 1.35 - bugfix such that pie charts by volume are generated correctly
3137 30. The SPA authentication driver was not abandoning authentication and moving
3138 on to the next authenticator when an expansion was forced to fail,
3139 contradicting the general specification for all authenticators. Instead it
3140 was generating a temporary error. It now behaves as specified.
3142 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3143 (the order specifies the preference for clients). The order is now AES256,
3144 AES128, 3DES, ARCFOUR128.
3146 31. Small patch to Sieve code - explicitly set From: when generating an
3149 32. Exim crashed if a remote delivery caused a very long error message to be
3150 recorded - for instance if somebody sent an entire SpamAssassin report back
3151 as a large number of 550 error lines. This bug was coincidentally fixed by
3152 increasing the size of one of Exim's internal buffers (big_buffer) that
3153 happened as part of the Exiscan merge. However, to be on the safe side, I
3154 have made the code more robust (and fixed the comments that describe what
3157 33. Some experimental protocols are using DNS PTR records for new purposes. The
3158 keys for these records are domain names, not reversed IP addresses. The
3159 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3160 leaves it alone. Component reversal etc. now happens only for IP addresses.
3163 34. The host_aton() function is supposed to be passed a string that is known
3164 to be a valid IP address. However, in the case of IPv6 addresses, it was
3165 not checking this. This is a hostage to fortune. Exim now panics and dies
3166 if the condition is not met. A case was found where this could be provoked
3167 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3168 components; fortuitously, this particular loophole had already been fixed
3169 by change 4.50/55 or 4.44/33 above.
3171 If there are any other similar loopholes, the new check in host_aton()
3172 itself should stop them being exploited. The report I received stated that
3173 data on the command line could provoke the exploit when Exim was running as
3174 exim, but did not say which command line option was involved. All I could
3175 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3176 running as the user.
3179 35. There was a buffer overflow vulnerability in the SPA authentication code
3180 (which came originally from the Samba project). I have added a test to the
3181 spa_base64_to_bits() function which I hope fixes it.
3184 36. The daemon start-up calls getloadavg() while still root for those OS that
3185 need the first call to be done as root, but it missed one case: when
3186 deliver_queue_load_max is set with deliver_drop_privilege. This is
3187 necessary for the benefit of the queue runner, because there is no re-exec
3188 when deliver_drop_privilege is set.
3190 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3191 was in use, was not putting the data itself into the right store pool;
3192 consequently, it could be overwritten for a subsequent message in the same
3193 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3196 38. Sometimes the final signoff response after QUIT could fail to get
3197 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3198 < 0 before doing a fflush(). This bug looks as though it goes back to the
3199 introduction of TLS in release 3.20, but "sometimes" must have been rare
3200 because the tests only now provoked it.
3206 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3207 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3208 Now it returns an integer. A number of calls were still expecting a T/F
3209 return. Fortuitously, in all cases, the tests worked in OK situations,
3210 which is the norm. However, things would have gone wrong on any write
3211 failures on the smtp file descriptor. This function is used when sending
3212 messages over SMTP and also when doing verify callouts.
3214 2. When Exim is called to do synchronous delivery of a locally submitted
3215 message (the -odf or -odi options), it no longer closes stderr before doing
3218 3. Implemented the mua_wrapper option.
3220 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3222 5. Implemented the functions header_remove(), header_testname(),
3223 header_add_at_position(), and receive_remove_recipient(), and exported them
3226 6. If an ACL "warn" statement specified the addition of headers, Exim already
3227 inserted X-ACL-Warn: at the start if there was no header name. However, it
3228 was not making this test for the second and subsequent header lines if
3229 there were newlines in the string. This meant that an invalid header could
3230 be inserted if Exim was badly configured.
3232 7. Allow an ACL "warn" statement to add header lines at the start or after all
3233 the Received: headers, as well as at the end.
3235 8. Added the rcpt_4xx retry error code.
3237 9. Added postmaster_mailfrom=xxx to callout verification option.
3239 10. Added mailfrom=xxxx to the callout verification option, for verify=
3242 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3243 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3245 12. Inserted some casts to stop certain compilers warning when using pointer
3246 differences as field lengths or precisions in printf-type calls (mostly
3247 affecting debugging statements).
3249 13. Added optional readline() support for -be (dynamically loaded).
3251 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3252 same clock tick as a message's arrival, so that its received time was the
3253 same as the "first fail" time on the retry record, and that message
3254 remained on the queue past the ultimate address timeout, every queue runner
3255 would try a delivery (because it was past the ultimate address timeout) but
3256 after another failure, the ultimate address timeout, which should have then
3257 bounced the address, did not kick in. This was a "< instead of <=" error;
3258 in most cases the first failure would have been in the next clock tick
3259 after the received time, and all would be well.
3261 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3262 being recognized when the domain list was tested by the match_domain
3263 condition in an expansion string.
3265 16. Added the ${str2b64: operator.
3267 17. Exim was always calling setrlimit() to set a large limit for the number of
3268 processes, without checking whether the existing limit was already
3269 adequate. (It did check for the limit on file descriptors.) Furthermore,
3270 errors from getrlimit() and setrlimit() were being ignored. Now they are
3271 logged to the main and panic logs, but Exim does carry on, to try to do its
3272 job under whatever limits there are.
3274 18. Imported PCRE 5.0.
3276 19. Trivial typo in log message " temporarily refused connection" (the leading
3279 20. If the log selector return_path_on_delivery was set and an address was
3280 redirected to /dev/null, the delivery process crashed because it assumed
3281 that a return path would always be set for a "successful" delivery. In this
3282 case, the whole delivery is bypassed as an optimization, and therefore no
3285 21. Internal re-arrangement: the function for sending a challenge and reading
3286 a response while authentication was assuming a zero-terminated challenge
3287 string. It's now changed to take a pointer and a length, to allow for
3288 binary data in such strings.
3290 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3292 23. Exim was not respecting finduser_retries when seeking the login of the
3293 uid under which it was called; it was always trying 10 times. (The default
3294 setting of finduser_retries is zero.) Also, it was sleeping after the final
3295 failure, which is pointless.
3297 24. Implemented tls_on_connect_ports.
3299 25. Implemented acl_smtp_predata.
3301 26. If the domain in control=submission is set empty, Exim assumes that the
3302 authenticated id is a complete email address when it generates From: or
3303 Sender: header lines.
3305 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3306 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3307 chown and chgrp in /bin and hostname in /usr/bin.
3309 28. Exim was keeping the "process log" file open after each use, just as it
3310 does for the main log. This opens the possibility of it remaining open for
3311 long periods when the USR1 signal hits a daemon. Occasional processlog
3312 errors were reported, that could have been caused by this. Anyway, it seems
3313 much more sensible not to leave this file open at all, so that is what now
3316 29. The long-running daemon process does not normally write to the log once it
3317 has entered its main loop, and it closes the log before doing so. This is
3318 so that log files can straightforwardly be renamed and moved. However,
3319 there are a couple of unusual error situations where the daemon does write
3320 log entries, and I had neglected to close the log afterwards.
3322 30. The text of an SMTP error response that was received during a remote
3323 delivery was being truncated at 512 bytes. This is too short for some of
3324 the long messages that one sometimes sees. I've increased the limit to
3327 31. It is now possible to make retry rules that apply only when a message has a
3328 specific sender, in particular, an empty sender.
3330 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3331 it possible to be selective about when SMTP synchronization is enforced.
3333 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3335 32. Implemented hosts_connection_nolog.
3337 33. Added an ACL for QUIT.
3339 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3342 35. Added mailbox_size and mailbox_filecount to appendfile.
3344 36. Added control = no_multiline_responses to ACLs.
3346 37. There was a bug in the logic of the code that waits for the clock to tick
3347 in the case where the clock went backwards by a substantial amount such
3348 that the microsecond fraction of "now" was more than the microsecond
3349 fraction of "then" (but the whole seconds number was less).
3351 38. Added support for the libradius Radius client library this is found on
3352 FreeBSD (previously only the radiusclient library was supported).
3358 1. When certain lookups returned multiple values in the form name=value, the
3359 quoting of the values was not always being done properly. Specifically:
3360 (a) If the value started with a double quote, but contained no whitespace,
3362 (b) If the value contained whitespace other than a space character (i.e.
3363 tabs or newlines or carriage returns) it was not quoted.
3364 This fix has been applied to the mysql and pgsql lookups by writing a
3365 separate quoting function and calling it from the lookup code. The fix
3366 should probably also be applied to nisplus, ibase and oracle lookups, but
3367 since I cannot test any of those, I have not disturbed their existing code.
3369 2. A hit in the callout cache for a specific address caused a log line with no
3370 reason for rejecting RCPT. Now it says "Previous (cached) callout
3371 verification failure".
3373 3. There was an off-by-one bug in the queryprogram router. An over-long
3374 return line was truncated at 256 instead of 255 characters, thereby
3375 overflowing its buffer with the terminating zero. As well as fixing this, I
3376 have increased the buffer size to 1024 (and made a note to document this).
3378 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3379 when Exim is waiting for an SMTP response from a remote server, Exim
3380 restarts its select() call on the socket, thereby resetting its timeout.
3381 This is not a problem when such interrupts are rare. Somebody set up a cron
3382 job to run exiwhat every 2 minutes, which is less than the normal select()
3383 timeout (5 or 10 minutes). This meant that the select() timeout never
3384 kicked in because it was always reset. I have fixed this by comparing the
3385 time when an interrupt arrives with the time at the start of the first call
3386 to select(). If more time than the timeout has elapsed, the interrupt is
3387 treated as a timeout.
3389 5. Some internal re-factoring in preparation for the addition of Sieve
3390 extensions (by MH). In particular, the "personal" test is moved to a
3391 separate function, and given an option for scanning Cc: and Bcc: (which is
3392 not set for Exim filters).
3394 6. When Exim created an email address using the login of the caller as the
3395 local part (e.g. when creating a From: or Sender: header line), it was not
3396 quoting the local part when it contained special characters such as @.
3398 7. Installed new OpenBSD configuration files.
3400 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3401 try to make them clearer.
3403 9. Callout options, other than the timeout value, were being ignored when
3404 verifying sender addresses in header lines. For example, when using
3406 verify = header_sender/callout=no_cache
3408 the cache was (incorrectly) being used.
3410 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3411 only the Cygwin environment.
3413 11. When return_path_on_delivery was set as a log selector, if different remote
3414 addresses in the same message used different return paths and parallel
3415 remote delivery occurred, the wrong values would sometimes be logged.
3416 (Whenever a remote delivery process finished, the return path value from
3417 the most recently started remote delivery process was logged.)
3419 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3420 lines when AUTH and/or TLS are in use. This is the "received protocol"
3421 field. Exim used to use "asmtp" for authenticated SMTP, without any
3422 indication (in the protocol name) for TLS use. Now it follows the RFC and
3423 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3424 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3425 names appear in log lines as well as in Received: header lines.
3427 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3428 extensions, and comparison tests, and to fix some bugs.
3430 14. Changes to the "personal" filter test:
3432 (1) The test was buggy in that it was just doing the equivalent of
3433 "contains" tests on header lines. For example, if a user's address was
3434 anne@some.where, the "personal" test would incorrectly be true for
3436 To: susanne@some.where
3438 This test is now done by extracting each address from the header in turn,
3439 and checking the entire address. Other tests that are part of "personal"
3440 are now done using regular expressions (for example, to check local parts
3441 of addresses in From: header lines).
3443 (2) The list of non-personal local parts in From: addresses has been
3444 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3445 taken from the Sieve specification recommendations.
3447 (3) If the message contains any header line starting with "List-" it is
3448 treated as non-personal.
3450 (4) The test for "circular" in the Subject: header line has been removed
3451 because it now seems ill-conceived.
3453 15. Minor typos in src/EDITME comments corrected.
3455 16. Installed latest exipick from John Jetmore.
3457 17. If headers_add on a router specified a text string that was too long for
3458 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3459 of string_sprintf() is now avoided.
3461 18. $message_body_size was not set (it was always zero) when running the DATA
3462 ACL and the local_scan() function.
3464 19. For the "mail" command in an Exim filter, no default was being set for
3465 the once_repeat time, causing a random time value to be used if "once" was
3466 specified. (If the value happened to be <= 0, no repeat happened.) The
3467 default is now 0s, meaning "never repeat". The "vacation" command was OK
3468 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3469 (I found it when inspecting the code).
3471 20. There is now an overall timeout for performing a callout verification. It
3472 defaults to 4 times the callout timeout, which applies to individual SMTP
3473 commands during the callout. The overall timeout applies when there is more
3474 than one host that can be tried. The timeout is checked before trying the
3475 next host. This prevents very long delays if there are a large number of
3476 hosts and all are timing out (e.g. when the network connections are timing
3477 out). The value of the overall timeout can be changed by specifying an
3478 additional sub-option for "callout", called "maxwait". For example:
3480 verify = sender/callout=5s,maxwait=20s
3482 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3483 to the end before writing, but this should make it even safer).
3485 22. Exim was forgetting that it had advertised PIPELINING for the second and
3486 subsequent messages on an SMTP connection. It was also not resetting its
3487 memory on STARTTLS and an internal HELO.
3489 23. When Exim logs an SMTP synchronization error within a session, it now
3490 records whether PIPELINING has been advertised or not.
3492 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3493 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3494 rather than long int.
3496 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3498 26. Added the never_mail option to autoreply.
3504 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3505 crash if the getsockname() call failed; this can happen if a connection is
3506 closed very soon after it is established. The problem was simply in the
3507 order in which certain operations were done, causing Exim to try to write
3508 to the SMTP stream before it had set up the file descriptor. The bug has
3509 been fixed by making things happen in the correct order.
3515 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3516 before the connection was closed, thus losing the rejection response.
3518 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3519 some early Solaris releases, but causes trouble in current releases where
3520 socklen_t is defined.
3522 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3525 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3528 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3529 that releasing the top of it at the end releases what was used for sub-
3530 expansions (unless the block got too big). However, discard this block if
3531 the first thing is a variable or header, so that we can use its block when
3532 it is dynamic (useful for very large $message_headers, for example).
3534 6. Lookups now cache *every* query, not just the most recent. A new, separate
3535 store pool is used for this. It can be recovered when all lookup caches are
3536 flushed. Lookups now release memory at the end of their result strings.
3537 This has involved some general refactoring of the lookup sources.
3539 7. Some code has been added to the store_xxx() functions to reduce the amount
3540 of flapping under certain conditions.
3542 8. log_incoming_interface used to affect only the <= reception log lines. Now
3543 it causes the local interface and port to be added to several more SMTP log
3544 lines, for example "SMTP connection from", and rejection lines.
3546 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3548 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3550 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3551 could be overwritten at the end of the current message (or the start of a
3552 new message if it was set in a HELO ACL). The value is now preserved for
3553 the duration of the SMTP connection.
3555 12. If a transport had a headers_rewrite setting, and a matching header line
3556 contained an unqualified address, that address was qualified, even if it
3557 did not match any rewriting rules. The underlying bug was that the values
3558 of the flags that permit the existence of unqualified sender and recipient
3559 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3560 non-local messages, and by -bnq for local messages) were not being
3561 preserved with the message after it was received.
3563 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3564 "next input=" as part of the text comprising the host identity instead of
3565 the correct text. The code was using the same buffer for two different
3566 strings. However, depending on which order the printing function evaluated
3567 its arguments, the bug did not always show up. Under Linux, for example, my
3568 test suite worked just fine.
3570 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3571 doesn't work with some older versions of Perl. It has been changed to "my",
3572 which in any case is probably the better facility to use.
3574 15. A really picky compiler found some instances of statements for creating
3575 error messages that either had too many or two few arguments for the format
3578 16. The size of the buffer for calls to the DNS resolver has been increased
3579 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3580 for addresses that have a lot of PTR records. This alleviates a problem; it
3581 does not fully solve it.
3583 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3584 buffer now truncates the list and logs the incident, which is the same
3585 action as happens when Exim is looking up a host name and its aliases.
3586 Previously in this situation something unpredictable would happen;
3587 sometimes it was "internal error: store_reset failed".
3589 18. If a server dropped the connection unexpectedly when an Exim client was
3590 using GnuTLS and trying to read a response, the client delivery process
3591 crashed while trying to generate an error log message.
3593 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3594 string, for example:
3596 warn message = H1: something\nH2: something
3598 the text was added as a single header line from Exim's point of view
3599 though it ended up OK in the delivered message. However, searching for the
3600 second and subsequent header lines using $h_h2: did not work. This has been
3601 fixed. Similarly, if a system filter added multiple headers in this way,
3602 the routers could not see them.
3604 20. Expanded the error message when iplsearch is called with an invalid key to
3605 suggest using net-iplsearch in a host list.
3607 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3608 ACLs are no longer actually imposed (and a message to that effect is
3611 22. If a "gecos" field in a passwd entry contained escaped characters, in
3612 particular, if it contained a \" sequence, Exim got it wrong when building
3613 a From: or a Sender: header from that name. A second bug also caused
3614 incorrect handling when an unquoted " was present following a character
3615 that needed quoting.
3617 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3618 was not being matched caselessly.
3620 24. Arranged for all hyphens in the exim.8 source to be escaped with
3623 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3624 properly. Recipient callout cache records were still being keyed to include
3625 the sender, even when use_sender was set false. This led to far more
3626 callouts that were necessary. The sender is no longer included in the key
3627 when use_sender is false.
3629 26. Added "control = submission" modifier to ACLs.
3631 27. Added the ${base62d: operator to decode base 62 numbers.
3633 28. dnsdb lookups can now access SRV records.
3635 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3636 the configuration file.
3638 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3639 (-v) mode. This makes the output for a verbose queue run more intelligible.
3641 31. Added a use_postmaster feature to recipient callouts.
3643 32. Added the $body_zerocount variable, containing the number of binary zero
3644 bytes in the message body.
3646 33. The time of last modification of the "new" subdirectory is now used as the
3647 "mailbox time last read" when there is a quota error for a maildir
3650 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3652 35. Added +ignore_unknown as a special item in host lists.
3654 36. Code for decoding IPv6 addresses in host lists is now included, even if
3655 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3656 address was recognized as an IP address, but was then not correctly decoded
3657 into binary, causing unexpected and incorrect effects when compared with
3664 1. Very minor rewording of debugging text in manualroute to say "list of
3665 hosts" instead of "hostlist".
3667 2. If verify=header_syntax was set, and a header line with an unqualified
3668 address (no domain) and a large number of spaces between the end of the
3669 name and the colon was received, the reception process suffered a buffer
3670 overflow, and (when I tested it) crashed. This was caused by some obsolete
3671 code that should have been removed. The fix is to remove it!
3673 3. When running in the test harness, delay a bit after writing a bounce
3674 message to get a bit more predictability in the log output.
3676 4. Added a call to search_tidyup() just before forking a reception process. In
3677 theory, someone could use a lookup in the expansion of smtp_accept_max_
3678 per_host which, without the tidyup, could leave open a database connection.
3680 5. Added the variables $recipient_data and $sender_data which get set from a
3681 lookup success in an ACL "recipients" or "senders" condition, or a router
3682 "senders" option, similar to $domain_data and $local_part_data.
3684 6. Moved the writing of debug_print from before to after the "senders" test
3687 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3688 problems for message scanning, either using a data ACL, or using
3689 local_scan() because the Received: header was not generated till after they
3690 were called (in order to set the time as the time of reception completion).
3691 I have revised the way this works. The header is now generated after the
3692 body is received, but before the ACL or local_scan() are called. After they
3693 are run, the timestamp in the header is updated.
3699 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3700 before starting a queue runner without re-exec. This happened only when
3701 deliver_drop_privilege was set or when the Exim user was set to root. The
3702 effect of the bug was that timeouts during subsequent deliveries caused
3703 crashes instead of being properly handled. The handler is now left at its
3704 default (and expected) setting.
3706 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3707 message, again when deliver_drop_privilege is set or Exim is run as root.
3708 The bug described in (1) was not present in this case, but the tidying up
3709 of the other signals was missing. I have made the two cases consistent.
3711 3. The ignore_target_hosts setting on a manualroute router was being ignored
3712 for hosts that were looked up using the /MX notation.
3714 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3717 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3718 operated on the sender address. After changing the $sender_address to <>
3719 for the sender address verify, Exim was re-instated it as the original
3720 (before rewriting) address, but remembering that it had rewritten it, so it
3721 wasn't rewriting it again. This bug also had the effect of breaking the
3722 sender address verification caching when the sender address was rewritten.
3724 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3725 This has been changed so that if the ip literal address matches
3726 ignore_target_hosts, the router declines.
3728 7. Added expansion conditions match_domain, match_address, and match_local_
3729 part (NOT match_host).
3731 8. The placeholder for the Received: header didn't have a length field set.
3733 9. Added code to Exim itself and to exim_lock to test for a specific race
3734 condition that could lead to file corruption when using MBX delivery. The
3735 issue is with the lockfile that is created in /tmp. If this file is removed
3736 after a process has opened it but before that process has acquired a lock,
3737 there is the potential for a second process to recreate the file and also
3738 acquire a lock. This could lead to two Exim processes writing to the file
3739 at the same time. The added code performs the same test as UW imapd; it
3740 checks after acquiring the lock that its file descriptor still refers to
3741 the same named file.
3743 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3744 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3747 11. Added the smtp_active_hostname option. If used, this will typically be made
3748 to depend on the incoming interface address. Because $interface_address is
3749 not set up until the daemon has forked a reception process, error responses
3750 that can happen earlier (such as "too many connections") no longer contain
3753 12. If an expansion in a condition on a "warn" statement fails because a lookup
3754 defers, the "warn" statement is abandoned, and the next ACL statement is
3755 processed. Previously this caused the whole ACL to be aborted.
3757 13. Added the iplsearch lookup type.
3759 14. Added ident_timeout as a log selector.
3761 15. Added tls_certificate_verified as a log selector.
3763 16. Added a global option tls_require_ciphers (compare the smtp transport
3764 option of the same name). This controls incoming TLS connections.
3766 17. I finally figured out how to make tls_require_ciphers do a similar thing
3767 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3768 before starting the TLS session.
3770 18. Tabs are now shown as \t in -bP output.
3772 19. If the log selector return_path_on_delivery was set, Exim crashed when
3773 bouncing a message because it had too many Received: header lines.
3775 20. If two routers both had headers_remove settings, and the first one included
3776 a superfluous trailing colon, the final name in the first list and the
3777 first name in the second list were incorrectly joined into one item (with a
3778 colon in the middle).
3784 1. Added -C and -D options to the exinext utility, mainly to make it easier
3785 to include in the automated testing, but these could be helpful when
3786 multiple configurations are in use.
3788 2. The exinext utility was not formatting the output nicely when there was
3789 an alternate port involved in the retry record key, nor when there was a
3790 message id as well (for retries that were specific to a specific message
3791 and a specific host). It was also confused by IPv6 addresses, because of
3792 the additional colons they contain. I have fixed the IPv4 problem, and
3793 patched it up to do a reasonable job for IPv6.
3795 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3796 delivery, the log line now contains "pipelined" if PIPELINING was used.
3798 4. An SMTP transport process used to panic and die if the bind() call to set
3799 an explicit outgoing interface failed. This has been changed; it is now
3800 treated in the same way as a connect() failure.
3802 5. A reference to $sender_host_name in the part of a conditional expansion
3803 that was being skipped was still causing a DNS lookup. This no longer
3806 6. The def: expansion condition was not recognizing references to header lines
3807 that used bh_ and bheader_.
3809 7. Added the _cache feature to named lists.
3811 8. The code for checking quota_filecount in the appendfile transport was
3812 allowing one more file than it should have been.
3814 9. For compatibility with Sendmail, the command line option
3822 and sets the incoming protocol and host name (for trusted callers). The
3823 host name and its colon can be omitted when only the protocol is to be set.
3824 Note the Exim already has two private options, -pd and -ps, that refer to
3825 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3826 "s", but I don't think that's a major issue.
3828 10. A number of refactoring changes to the code, none of which should affect
3831 (a) The number of logging options was getting close to filling up the
3832 32-bit word that was used as a bit map. I have split them into two classes:
3833 those that are passed in the argument to log_write(), and those that are
3834 only ever tested independently outside of that function. These are now in
3835 separate 32-bit words, so there is plenty of room for expansion again.
3836 There is no change in the user interface or the logging behaviour.
3838 (b) When building, for example, log lines, the code previously used a
3839 macro that called string_cat() twice, in order to add two strings. This is
3840 not really sufficiently general. Furthermore, there was one instance where
3841 it was actually wrong because one of the argument was used twice, and in
3842 one call a function was used. (As it happened, calling the function twice
3843 did not affect the overall behaviour.) The macro has been replaced by a
3844 function that can join an arbitrary number of extra strings onto a growing
3847 (c) The code for expansion conditions now uses a table and a binary chop
3848 instead of a serial search (which was left over from when there were very
3849 few conditions). Also, it now recognizes conditions like "pam" even when
3850 the relevant support is not compiled in: a suitably worded error message is
3851 given if an attempt is made to use such a condition.
3853 11. Added ${time_interval:xxxxx}.
3855 12. A bug was causing one of the ddress fields not to be passed back correctly
3856 from remote delivery subprocesses. The field in question was not being
3857 subsequently used, so this caused to problems in practice.
3859 13. Added new log selectors queue_time and deliver_time.
3861 14. Might have fixed a bug in maildirsizefile handling that threw up
3862 "unexpected character" debug warnings, and recalculated the data
3863 unnecessarily. In any case, I expanded the warning message to give more
3866 15. Added the message "Restricted characters in address" to the statements in
3867 the default ACL that block characters like @ and % in local parts.
3869 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3870 Three changes have been made:
3872 (a) There was a serious bug; a negative response to MAIL caused the whole
3873 recipient domain to be cached as invalid, thereby blocking all messages
3874 to all local parts at the same domain, from all senders. This bug has
3875 been fixed. The domain is no longer cached after a negative response to
3876 MAIL if the sender used is not empty.
3878 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3881 (c) A new callout option, "use_sender" has been added for people who want
3882 the modified behaviour.
3888 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3891 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3892 indeed breaks things for older releases.
3894 3. Added additional logging to the case where there is a problem reading data
3895 from a filter that is running in a subprocess using a pipe, in order to
3896 try to track down a specific problem.
3898 4. Testing facility fudge: when running in the test harness and attempting
3899 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3900 getting "No route to host". Convert this to a timeout.
3902 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3905 6. Some OS don't have socklen_t but use size_t instead. This affects the
3906 fifth argument of getsockopt() amongst other things. This is now
3907 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3908 can be set for individual OS. I have set it for SunOS5, OSF1, and
3909 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3910 some earlier ones do not.
3912 7. Change 4.30/15 was not doing the test caselessly.
3914 8. The standard form for an IPv6 address literal was being rejected by address
3915 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3916 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3917 this, as well as the form without the "IPv6" on the front (but only when
3918 address literals are enabled, of course).
3920 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3922 10. Exim crashed if a message with an empty sender address specified by -f
3923 encountered a router with an errors_to setting. This could be provoked only
3924 by a command such as
3928 where an empty string was supplied; "<>" did not hit this bug.
3930 11. Installed PCRE release 4.5.
3932 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3933 remained set. It is now erased.
3935 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3936 times from message ids (which are base 36 rather than the normal 62).
3938 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3939 were being counted as actual protocol errors, and logged if the log
3940 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3941 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3942 and DATA following a set of rejected RCPTs do not count as protocol errors.
3943 In other words, Exim assumes they were pipelined, though this may not
3944 actually be the case. Of course, in all cases the client gets an
3945 appropriate error code.
3947 15. If a lookup fails in an ACL condition, a message about the failure may
3948 be available; it is used if testing the ACL cannot continue, because most
3949 such messages specify what the cause of the deferral is. However, some
3950 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3951 that caused an old message to be retained and used if a later statement
3952 caused a defer, replacing the real cause of the deferral.
3954 16. If an IP address had so many PTR records that the DNS lookup buffer
3955 was not large enough to hold them, Exim could crash while trying to process
3956 the truncated data. It now detects and logs this case.
3958 17. Further to 4.21/58, another change has been made: if (and only if) the
3959 first line of a message (the first header line) ends with CRLF, a bare LF
3960 in a subsequent header line has a space inserted after it, so as not to
3961 terminate the header.
3963 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3964 unnecessarily, used atoi() instead of strtol(), and didn't check the
3965 termination when getting file sizes from file names by regex.
3967 19. Completely re-implemented the support for maildirsize files, in the light
3968 of a number of problems with the previous contributed implementation
3969 (4.30/29). In particular:
3971 . If the quota is zero, the maildirsize file is maintained, but no quota is
3974 . If the maildir directory does not exist, it is created before any attempt
3975 to write a maildirsize file.
3977 . The quota value in the file is just a cache; if the quota is changed in
3978 the transport, the new value overrides.
3980 . A regular expression is available for excluding directories from the
3983 20. The autoreply transport checks the characters in options that define the
3984 message's headers; it allows continued headers, but it was checking with
3985 isspace() after an embedded newline instead of explicitly looking for a
3988 21. If all the "regular" hosts to which an address was routed had passed their
3989 expiry times, and had not reached their retry times, the address was
3990 bounced, even if fallback hosts were defined. Now Exim should go on to try
3993 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3994 equivalent code in the SMTP transport. Some hosts send humungous responses
3995 to HELO/EHLO, more than 1024 it seems.
3997 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3998 gives compiler warnings in some environments. I've now done it "properly",
4001 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4002 (because of problems with the built-in one) was declared to return uschar *
4003 instead of char *, causing compiler failure.
4005 25. Fixed a file descriptor leak when processing alias/forward files.
4007 26. Fixed a minor format string issue in dbfn.c.
4009 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4011 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4012 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4015 29. When a sender address is verified, it is cached, to save repeating the test
4016 when there is more than one recipient in a message. However, when the
4017 verification involves a callout, it is possible for different callout
4018 options to be set for different recipients. It is too complicated to keep
4019 track of this in the cache, so now Exim always runs a verification when a
4020 callout is required, relying on the callout cache for the optimization.
4021 The overhead is duplication of the address routing, but this should not be
4024 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4025 to be verified with callout=postmaster, and the main callout worked but the
4026 postmaster check failed, the verification correctly failed. However, if a
4027 subsequent RCPT command asked for sender verification *without* the
4028 postmaster check, incorrect caching caused this verification also to fail,
4031 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4032 it was not caching the DNS options (qualify_single, search_parents) that
4033 were used when the lookup failed. A subsequent lookup with different
4034 options therefore always gave the same answer, though there were cases
4035 where it should not have. (Example: a "domains = !$mx_any" option on a
4036 dnslookup router: the "domains" option is always processed without any
4037 widening, but the router might have qualify_single set.) Now Exim uses the
4038 cached value only when the same options are set.
4040 32. Added John Jetmore's "exipick" utility to the distribution.
4042 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4043 than a timeout (e.g. a certificate is required, and is not provided), an
4044 Exim server now closes the connection immediately. Previously it waited for
4045 the client to close - but if the client is SSL, it seems that they each
4046 wait for each other, leading to a delay before one of them times out.
4048 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4049 maintained 0.8.x compatibility because I don't think many are using it, and
4050 it is clearly obsolete.
4052 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4055 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4056 client certificate was expired. A simple patch fixes this, though I don't
4057 understand the full logic of why the verify callback is called multiple
4060 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4061 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4062 which causes problems with some clients (such as the Certicom SSL Plus
4063 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4064 disables the coutermeasure allowing Eudora to connect."
4066 38. Exim was not checking that a write() to a log file succeeded. This could
4067 lead to Bad Things if a log got too big, in particular if it hit a file
4068 size limit. Exim now panics and dies if it cannot write to a log file, just
4069 as it does if it cannot open a log file.
4071 39. Modified OS/Makefile-Linux so that it now contains
4073 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4075 The two -D definitions ensure that Exim is compiled with large file
4076 support, which makes it possible to handle log files that are bigger than
4079 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4080 instance) a domain was checked against a named list that involved a lookup,
4081 causing $domain_data to be set, then another domain was checked against the
4082 same list, then the first domain was re-checked, the value of $domain_data
4083 after the final check could be wrong. In particular, if the second check
4084 failed, it could be set empty. This bug probably also applied to
4087 41. The strip_trailing_dot option was not being applied to the address given
4088 with the -f command-line option.
4090 42. The code for reading a message's header from the spool was incrementing
4091 $received_count, but never initializing it. This meant that the value was
4092 incorrect (doubled) while delivering a message in the same process in which
4093 it was received. In the most common configuration of Exim, this never
4094 happens - a fresh exec is done - but it can happen when
4095 deliver_drop_privilege is set.
4097 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4098 it now includes up to 150 characters of the unexpected data in the log
4101 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4102 and building data strings. The size of both of these buffers was 10 000
4103 bytes - far larger than anybody would *ever* want, thought I. Needless to
4104 say, somebody hit the limit. I have increased the maximum line length to
4105 20 000 and the maximum data length of concatenated lines to 100 000. I have
4106 also fixed two bugs, because there was no checking on these buffers. Tsk,
4107 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4108 buffer is too small.
4110 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4111 lsearch lookups. Now it does.
4113 46. When parsing a route_list item in a manualroute router, a fixed-length
4114 buffer was used for the list of hosts. I made this 1024 bytes long,
4115 thinking that nobody would ever have a list of hosts that long. Wrong.
4116 Somebody had a whole pile of complicated expansion conditions, and the
4117 string was silently truncated, leading to an expansion error. It turns out
4118 that it is easier to change to an unlimited length (owing to other changes
4119 that have happened since this code was originally written) than to build
4120 structure for giving a limitation error. The length of the item that
4121 expands into the list of hosts is now unlimited.
4123 47. The lsearch lookup could not handle data where the length of text line was
4124 more than 4095 characters. Such lines were truncated, leading to shortened
4125 data being returned. It should now handle lines of any length.
4127 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4128 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4131 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4132 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4133 no longer happens in comment lines. A list of replacements is now placed
4134 at the head of all of the source files, except those whose only change is
4135 to replace PERL_COMMAND in the very first #! line.
4137 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4138 messages on the queue, with a bottom-up merge sort, using code contributed
4139 by Michael Haardt. This should make operations like -bp somewhat faster on
4140 large queues. It won't affect queue runners, except when queue_run_in_order
4143 51. Installed eximstats 1.31 in the distribution.
4145 52. Added support for SRV lookups to the dnslookup router.
4147 53. If an ACL referred to $message_body or $message_body_end, the value was not
4148 reset for any messages that followed in the same SMTP session.
4150 54. The store-handling optimization for building very long strings was not
4151 differentiating between the different store pools. I don't think this
4152 actually made any difference in practice, but I've tidied it.
4154 55. While running the routers to verify a sender address, $sender_address
4155 was still set to the sender address. This is wrong, because when routing to
4156 send a bounce to the sender, it would be empty. Therefore, I have changed
4157 it so that, while verifying a sender address, $sender_address is set to <>.
4158 (There is no change to what happens when verifying a recipient address.)
4160 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4161 target A or AAAA records (if not already returned) without resetting the
4162 qualify_single or search_parents options of the DNS resolver. These are
4163 inappropriate in this case because the targets of MX and SRV records must
4164 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4165 target that, when qualified, matched something in the local domain. These
4166 two options are now turned off when doing these lookups.
4168 57. It seems that at least some releases of Reiserfs (which does not have the
4169 concept of a fixed number of inodes) returns zero and not -1 for the
4170 number of available inodes. This interacted badly with check_spool_inodes,
4171 which assumed that -1 was the "no such thing" setting. What I have done is
4172 to check that the total number of inodes is greater than zero before doing
4173 the test of how many are available.
4175 58. When a "warn" ACL statement has a log_message modifier, the message is
4176 remembered, and not repeated. This is to avoid a lot of repetition when a
4177 message has many recipients that cause the same warning to be written.
4178 Howewer, Exim was preserving the list of already written lines for an
4179 entire SMTP session, which doesn't seem right. The memory is now reset if a
4180 new message is started.
4182 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4183 debugging output unless log_rewrite was also set.
4185 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4186 of (int)(handle) when we know that handle contains (void *)(-1).
4188 61. The Exim daemon panic-logs an error return when it closes the incoming
4189 connection. However "connection reset by peer" seems to be common, and
4190 isn't really an error worthy of noting specially, so that particular error
4193 62. When Exim is trying to find all the local interfaces, it used to panic and
4194 die if the ioctl to get the interface flags failed. However, it seems that
4195 on at least one OS (Solaris 9) it is possible to have an interface that is
4196 included in the list of interfaces, but for which you get a failure error
4197 for this call. This happens when the interface is not "plumbed" into a
4198 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4199 failure of the "get flags" call assumes that the interface is down.
4201 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4202 makes life easier for people who are doing arithmetic on fields extracted
4203 from dates, where you often get leading zeros that should not be
4204 interpreted as octal.
4206 64. Added qualify_domain to the redirect router, to override the global
4209 65. If a pathologically long header line contained very many addresses (the
4210 report of this problem mentioned 10 000) and each of them was rewritten,
4211 Exim could use up a very large amount of memory. (It kept on making new
4212 copies of the header line as it rewrote, and never released the old ones.)
4213 At the expense of a bit more processing, the header rewriting function has
4214 been changed so that it no longer eats memory in this way.
4216 66. The generation of the Received: header has been moved from the time that a
4217 message starts to be received, to the time that it finishes. The timestamp
4218 in the Received: header should now be very close to that of the <= log
4219 line. There are two side-effects of this change:
4221 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4222 logged header lines no longer include the local Received: line, because
4223 it has not yet been created. The same applies to a copy of the message
4224 that is returned to a non-SMTP sender when a message is rejected.
4226 (b) When a filter file is tested using -bf, no additional Received: header
4227 is added to the test message. After some thought, I decided that this
4230 This change does not affect the value of $received_for. It is still set
4231 after address rewriting, but before local_scan() is called.
4233 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4235 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4236 gave an unhelpful panic error message, and a defer error. I have managed to
4237 change this behaviour so that it now rejects any supplied certificate,
4238 which seems right, as the list of acceptable certificates is empty.
4240 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4241 gave an unhelpful defer error. I have not managed to make this reject any
4242 supplied certificates, but the error message it gives is "no certificate
4243 supplied", which is not helpful.
4245 70. exigrep's output now also includes lines that are not associated with any
4246 message, but which match the given pattern. Implemented by a patch from
4247 Martin Sluka, which also tidied up the Perl a bit.
4249 71. Recipient callout verification, like sender verification, was using <> in
4250 the MAIL FROM command. This isn't really the right thing, since the actual
4251 sender may affect whether the remote host accepts the recipient or not. I
4252 have changed it to use the actual sender in the callout; this means that
4253 the cache record is now keyed on a recipient/sender pair, not just the
4254 recipient address. There doesn't seem to be a real danger of callout loops,
4255 since a callout by the remote host to check the sender would use <>.
4256 [SEE ABOVE: changed after hitting problems.]
4258 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4259 temporary errors. However, in the case of such a code being given after
4260 the end of a data transmission (i.e. after ".") Exim was failing to write
4261 a retry record for the message. (Yes, there was some broken host that was
4262 actually sending 8xx at this point.)
4264 73. An unknown lookup type in a host list could cause Exim to panic-die when
4265 the list was checked. (An example that provoked this was putting <; in the
4266 middle of a list instead of at the start.) If this happened during a DATA
4267 ACL check, a -D file could be left lying around. This kind of configuration
4268 error no longer causes Exim to die; instead it causes a defer errror. The
4269 incident is still logged to the main and panic logs.
4271 74. Buglet left over from Exim 3 conversion. The message "too many messages
4272 in one connection" was written to the rejectlog but not the mainlog, except
4273 when address rewriting (yes!) was being logged.
4275 75. Added write_rejectlog option.
4277 76. When a system filter was run not as root (that is, when system_filter_user
4278 was set), the values of the $n variables were not being returned to the
4279 main process; thus, they were not subsequently available in the $sn
4282 77. Added +return_path_on_delivery log selector.
4284 78. A connection timeout was being treated differently from recipients deferred
4285 when testing hosts_max_try with a message that was older than the host's
4286 retry timeout. (The host should not be counted, thus allowing all hosts to
4287 be tried at least once before bouncing.) This may have been the cause of an
4288 occasionally reported bug whereby a message would remain on the queue
4289 longer than the retry timeout, but would be bounced if a delivery was
4290 forced. I say "may" because I never totally pinned down the problem;
4291 setting up timeout/retry tests is difficult. See also the next item.
4293 79. The ultimate address timeout was not being applied to errors that involved
4294 a combination of host plus message (for example, a timeout on a MAIL
4295 command). When an address resolved to a number of possible hosts, and they
4296 were not all tried for each delivery (e.g. because of hosts_max_try), a
4297 message could remain on the queue longer than the retry timeout.
4299 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4302 81. Fixed an obscure SMTP outgoing bug which required at least the following
4303 conditions: (a) there was another message waiting for the same server;
4304 (b) the server returned 5xx to all RCPT commands in the first message so
4305 that the message was not completed; (c) the server dropped the connection
4306 or gave a negative response to the RSET that Exim sends to abort the
4307 transaction. The observed case was a dropped connection after DATA that had
4308 been sent in pipelining mode. That is, the server had advertised PIPELINING
4309 but was not implementing it correctly. The effect of the bug was incorrect
4310 behaviour, such as trying another host, and this could lead to a crash.
4316 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4317 and daemon.c were passed as pointers to ints; they should have been
4318 pointers to socklen_t variables (which are typically unsigned ints).
4320 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4323 3. Fixed a really odd bug that affected only the testing scheme; patching a
4324 certain fixed string in the binary changed the value of another string that
4325 happened to be identical to the end of the original first string.
4327 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4328 name", it returns that address as the IP address. On some operating
4329 systems (e.g. Solaris), it also passes back the IP address string as the
4330 "host name". However, on others (e.g. Linux), it passes back an empty
4331 string. Exim wasn't checking for this, and was changing the host name to an
4332 empty string, assuming it had been canonicized.
4334 5. Although rare, it is permitted to have more than one PTR record for a given
4335 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4336 all the names associated with an address, because they do in Solaris.
4337 However, it seems that they do not in Linux for data that comes from the
4338 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4339 I found this out when I moved to a new Linux workstation and tried to run
4340 the Exim test suite.
4342 To get round this problem I have changed the code so that it now does its
4343 own call to the DNS to look up PTR records when searching for a host name.
4344 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4345 addresses that are only in /etc/hosts are still found.
4347 This behaviour is, however, controlled by an option called host_lookup_
4348 order, which defaults to "bydns:byaddr". If people want to use the other
4349 order, or indeed, just use one or the other means of lookup, they can
4350 specify it in this variable.
4352 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4353 some operating systems, this comes back from gethostbyaddr() as an empty
4354 string, and this is what Exim used to test for. However, it seems that in
4355 other systems, "." is yielded. Exim now tests for this case too.
4357 7. The values of check_spool_space and check_log_space are now held internally
4358 as a number of kilobytes instead of an absolute number of bytes. If a
4359 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4360 kilobyte. This means that much larger values can be stored.
4362 8. Exim monitor: an attempt to get the action menu when not actually pointing
4363 at a message produces an empty menu entitled "No message selected". This
4364 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4365 no entries in it ("Shell widget menu has zero width and/or height"). So I
4366 have added a single, blank menu entry in this case.
4368 9. Added ${quote_local_part.
4370 10. MIME decoding is now applied to the contents of Subject: header lines when
4373 11. Now that a reference to $sender_host_address automatically causes a reverse
4374 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4375 host lookup before query-style lookups in lists that might use this
4376 variable. This has therefore been abolished, and the "net-" prefix is no
4377 longer necessary for query-style lookups.
4379 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4380 have been a typo for LFLAGS, so it has been changed.
4382 13. The install script calls Exim with "-C /dev/null" in order to find the
4383 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4384 to be output. Howeve, since Exim outputs its version number before the
4385 error, it didn't break the script. It just looked ugly. I fixed this by
4386 always allowing "-C /dev/null" if the caller is root.
4388 14. Ignore overlarge ACL variable number when reading spool file - insurance
4389 against a later release with more variables having written the file.
4391 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4392 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4393 this, as well as the form without the "IPv6" on the front.
4395 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4396 OS/Makefile-Darwin file.
4398 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4399 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4401 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4402 variable. However, some OS use ints for inodes. I've added cast to long int
4403 to get rid of the compiler warning.
4405 19. I had forgotten to lock out "/../" in configuration file names when
4406 ALT_CONFIG_PREFIX was set.
4408 20. Routers used for verification do not need to specify transports. However,
4409 if such a router generated a host list, and callout was configured, Exim
4410 crashed, because it could not find a port number from the (non-existent)
4411 transport. It now assumes port 25 in this circumstance.
4413 21. Added the -t option to exigrep.
4415 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4416 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4417 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4418 them is set, LOOKUP_LSEARCH is forced.
4420 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4422 24. Added sender and host information to the "rejected by local_scan()" log
4423 line; previously there was no indication of these.
4425 25. Added .include_if_exists.
4427 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4428 Linux. It turns out that not all file systems support this. Apparently some
4429 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4430 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4431 not supported on the file descriptor, is now ignored when Exim is trying to
4432 sync a directory. This applies only to Linux.
4434 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4436 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4437 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4440 29. Added support for maildirsize files from supplied patch (modified a bit).
4442 30. The use of :fail: followed by an empty string could lead Exim to respond to
4443 sender verification failures with (e.g.):
4445 550 Verification failed for <xxx>
4446 550 Sender verify failed
4448 where the first response line was missing the '-' that indicates it is not
4449 the final line of the response.
4451 31. The loop for finding the name of the user that called Exim had a hardwired
4452 limit of 10; it now uses the value of finduser_retries, which is used for
4453 all other user lookups.
4455 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4458 33. Exim was neglecting to zero errno before one call of strtol() when
4459 expanding a string and expecting an integer value. On some systems this
4460 resulted in spurious "integer overflow" errors. Also, it was casting the
4461 result into an int without checking.
4463 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4464 did not work. The code looks as if it has *never* worked, though it appears
4465 to have been documented since at least releast 1.62. I have made it work.
4467 35. The "timeout_DNS" error in retry rules, also documented since at least
4468 1.62, also never worked. As it isn't clear exactly what this means, and
4469 clearly it isn't a major issue, I have abolished the feature by treating it
4470 as "timeout", and writing a warning to the main and panic logs.
4472 36. The display of retry rules for -brt wasn't always showing the error code
4475 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4476 timeout_connect_A, timeout_connect_MX.
4478 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4479 to the empty sender.
4481 39. The daemon was not analysing the content of -oX till after it had closed
4482 stderr and disconnected from the controlling terminal. This meant that any
4483 syntax errors were only noted on the panic log, and the return code from
4484 the command was 0. By re-arranging the code a little, I've made the
4485 decoding happen first, so such errors now appear on stderr, and the return
4486 code is 1. However, the actual setting up of the sockets still happens in
4487 the disconnected process, so errors there are still only recorded on the
4490 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4491 connections (as happens on some IP stacks) was logged at start up time as
4492 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4493 it from "IPv6 and IPv4", which means that two separate sockets are being
4496 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4497 says whether AF_INET or AF_INET6 was passed as an argument.
4499 42. Exiwhat output was messed up when time zones were included in log
4502 43. Exiwhat now gives more information about the daemon's listening ports,
4503 and whether -tls-on-connect was used.
4505 44. The "port" option of the smtp transport is now expanded.
4507 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4508 silently ignored. Now an error message is written to the main and panic
4511 46. There's a new ACL modifier called "logwrite" which writes to a log file
4512 as soon as it is encountered.
4514 47. Added $local_user_uid and $local_user_gid at routing time.
4516 48. Exim crashed when trying to verify a sender address that was being
4519 49. Exim was recognizing only a space character after ".include". It now also
4520 recognizes a tab character.
4522 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4523 extracting the relevant information from the specification. The man page no
4524 longer contains scrambled data for the -d option, and I've added a section
4525 at the front about calling Exim under different names.
4527 51. Added "extra_headers" argument to the "mail" command in filter files.
4529 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4532 53. Installed eximstats 1.29.
4534 54. Added transport_filter_timeout as a generic transport option.
4536 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4537 Cc: header lines. This was required by RFC 822, but it not required by RFC
4540 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4541 incoming messages that did not have them. Now it does so only if the
4542 message originates locally, that is, if there is no associated remote host
4543 address. When Resent- header lines are present, this applies to the Resent-
4544 lines rather than the non-Resent- lines.
4546 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4547 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4549 58. Messages for configuration errors now include the name of the main
4550 configuration file - useful now that there may be more than one file in a
4551 list (.included file names were always shown).
4553 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4554 for those rare installations that do not start the daemon as root or run it
4555 setuid root. I've cut out the call to initgroups() if the daemon is not
4558 60. The Exim user and group can now be bound into the binary as text strings
4559 that are looked up at the start of Exim's processing.
4561 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4563 62. Added $mailstore_basename variable.
4565 63. Installed patch to sieve.c from Michael Haardt.
4567 64. When Exim failed to open the panic log after failing to open the main log,
4568 the original message it was trying to log was written to stderr and debug
4569 output, but if they were not available (the usual case in production), it
4570 was lost. Now it is written to syslog before the two lines that record the
4571 failures to open the logs.
4573 65. Users' Exim filters run in subprocesses under the user's uid. It is
4574 possible for a "deliver" command or an alias in a "personal" command to
4575 provoke an address rewrite. If logging of address rewriting is configured,
4576 this fails because the process is not running as root or exim. There may be
4577 a better way of dealing with this, but for the moment (because 4.30 needs
4578 to be released), I have disabled address rewrite logging when running a
4579 filter in a non-root, non-exim process.
4585 1. The buildconfig auxiliary program wasn't quoting the value set for
4586 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4587 not defined. This bug was masked in 4.22 by the effect that was fixed in
4590 2. Some messages that were rejected after a message id was allocated were
4591 shown as "incomplete" by exigrep. It no longer does this for messages that
4592 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4594 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4595 have allow_domain_literals set, the ID did not get logged in the <= line.
4596 Domain literals are now always recognized in Message-ID: header lines.
4598 4. The first argument for a ${extract expansion item is the key name or field
4599 number. Leading and trailing spaces in this item were not being ignored,
4600 causing some misleading effects.
4602 5. When deliver_drop_privilege was set, single queue runner processes started
4603 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4604 same command in the process it spins off) were not dropping privilege.
4606 6. When the daemon running as "exim" started a queue runner, it always
4607 re-executed Exim in the spun-off process. This is a waste of effort when
4608 deliver_drop_privilege is set. The new process now just calls the
4609 queue-runner function directly.
4615 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4618 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4619 ignored. Though the use of -oP was forcing the writing of a pid file, it
4620 was always written to the default place.
4622 3. If the message "no IP address found for host xxxx" is generated during
4623 incoming verification, it is now followed by identification of the incoming
4624 connection (so you can more easily find what provoked it).
4626 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4628 5. Added some features to "harden" Exim a bit more against certain attacks:
4630 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4631 be put in Local/Makefile. This is like the never_users runtime option,
4632 but it cannot be overridden. The default setting is "root".
4634 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4635 prefix string with which any file named in a -C command line option
4638 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4639 is retained for -C and -D only if the caller of Exim is root. Without
4640 it, the exim user may also use -C and -D and retain privilege.
4642 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4643 command line option is disabled.
4645 6. Macro names set by the -D option must start with an upper case letter, just
4646 like macro names defined in the configuration file.
4648 7. Added "dereference=" facility to LDAP.
4650 8. Two instances of the typo "uknown" in the source files are fixed.
4652 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4653 the Configure-Makefile script screwed up while processing it.
4655 10. Incorporated PCRE 4.4.
4657 11. The SMTP synchronization check was not operating right at the start of an
4658 SMTP session. For example, it could not catch a HELO sent before the client
4659 waited for the greeting. There is now a check for outstanding input at the
4660 point when the greeting is written. Because of the duplex, asynchronous
4661 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4662 way, but not yet received, when the check is performed.
4664 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4665 on TCP/IP sockets, because this apparently causes some broken clients to
4668 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4669 unchanged) from the Cygwin maintainer.
4671 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4672 was supported instead of testing for mbx. Effectively a typo.
4674 15. The spa authenticator server code was not checking that the input it
4675 received was valid base64.
4677 16. The debug output line for the "set" modifier in ACLs was not showing the
4678 name of the variable that was being set.
4680 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4682 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4683 lookup on its own. Something else was needed to trigger the lookup. For
4684 example, a match in host_lookup or the need for a host name in a host list.
4685 Now, if $sender_host_name is referenced and the host name has not yet been
4686 looked up, a lookup is performed. If the lookup fails, the variable remains
4687 empty, and $host_lookup_failed is set to "1".
4689 19. Added "eqi" as a case-independent comparison operator.
4691 20. The saslauthd authentication condition could segfault if neither service
4692 nor realm was specified.
4694 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4695 error message that was logged was misleading, and incoming SMTP
4696 connections were dropped. The message is now more accurate, and temporary
4697 errors are given to SMTP connections.
4699 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4700 (except RSET) with a 421 error, until QUIT is received. However, it was
4701 failing to send a response to QUIT.
4703 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4704 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4705 ACL is now run after the helo_try_verify_hosts code.
4707 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4708 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4709 case-independent, so other case variants are also recognized). Apparently
4710 some systems use these upper case variants.
4712 25. If more than two messages were waiting for the same host, and a transport
4713 filter was specified for the transport, Exim sent two messages over the
4714 same TCP/IP connection, and then failed with "socket operation on non-
4715 socket" when it tried to send the third.
4717 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4719 27. The extern definition of crypt16() in expand.c was not being excluded when
4720 the OS had its own crypt16() function.
4722 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4723 as a preferred synonym for return_size_limit, both as an option and as an
4726 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4728 30. Changed the default configuration ACL to relax the local part checking rule
4729 for addresses that are not in any local domains. For these addresses,
4730 slashes and pipe symbols are allowed within local parts, but the sequence
4731 /../ is explicitly forbidden.
4733 31. SPA server authentication was not clearing the challenge buffer before
4736 32. log_message in a "warn" ACL statement was writing to the reject log as
4737 well as to the main log, which contradicts the documentation and doesn't
4738 seem right (because no rejection is happening). So I have stopped it.
4740 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4741 However, I am unable to do any testing of this.
4743 34. Fixed an infelicity in the appendfile transport. When checking directories
4744 for a mailbox, to see if any needed to be created, it was accidentally
4745 using path names with one or more superfluous leading slashes; tracing
4746 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4748 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4749 contents are added to the log line that is written for every discarded
4750 recipient. (Previously a log_message setting was ignored.)
4752 36. The ${quote: operator now quotes the string if it is empty.
4754 37. The install script runs exim in order to find its version number. If for
4755 some reason other than non-existence or emptiness, which it checks, it
4756 could not run './exim', it was installing it with an empty version number,
4757 i.e. as "exim-". This error state is now caught, and the installation is
4760 38. An argument was missing from the function that creates an error message
4761 when Exim fails to connect to the socket for saslauthd authentication.
4762 This could cause Exim to crash, or give a corrupted message.
4764 39. Added isip, isip4, and isip6 to ${if conditions.
4766 40. The ACL variables $acl_xx are now saved with the message, and can be
4767 accessed later in routers, transports, and filters.
4769 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4770 strings in the file are not string-expanded.
4772 42. If a MAIL command specified a SIZE value that was too large to fit into an
4773 int variable, the check against message_size_limit failed. Such values are
4774 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4775 day this will have to be increased, but I don't think I want to be around
4776 when emails are that large.
4783 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4784 iconv() is not standard in FreeBSD.
4786 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4787 IPv6 enabled. The observed symptom was a segmentation fault on return from
4788 the function os_common_find_running_interfaces() in src/os.c.
4790 3. In the check_special_case() function in daemon.c I had used "errno" as an
4791 argument name, which causes warnings on some systems. This was basically a
4792 typo, since it was named "eno" in the comments!
4794 4. The code that waits for the clock to tick (at a resolution of some fraction
4795 of a second) so as to ensure message-id uniqueness was always waiting for
4796 at least one whole tick, when it could have waited for less. [This is
4797 almost certainly not relevant at current processor speeds, where it is
4798 unlikely to ever wait at all. But we try to future-proof.]
4800 5. The function that sleeps for a time interval that includes fractions of a
4801 second contained a race. It did not block SIGALRM between setting the
4802 timer, and suspending (a couple of lines later). If the interval was short
4803 and the sigsuspend() was delayed until after it had expired, the suspension
4804 never ended. On busy systems this could lead to processes getting stuck for
4807 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4808 process, before it forks any delivery processes. The open lookup caching
4809 mechanism meant that the open file or database connection was passed into
4810 the delivery process. The problem was that delivery processes always tidy
4811 up cached lookup data. This could cause a problem for the next delivery
4812 process started by the queue runner, because the external queue runner
4813 process does not know about the closure. So the next delivery process
4814 still has data in the lookup cache. In the case of a file lookup, there was
4815 no problem because closing a file descriptor in a subprocess doesn't affect
4816 the parent. However, if the lookup was caching a connection to a database,
4817 the connection was closed, and the second delivery process was likely to
4818 see errors such as "PGSQL: query failed: server closed the connection
4819 unexpectedly". The problem has been fixed by closing all cached lookups
4820 in a queue runner before running a delivery process.
4822 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4823 seem to have the "const" qualifier which it has on other OS. I've
4826 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4827 *of the same type* (client or server) with the same public name that an
4828 error should be diagnosed.
4830 9. When Exim looked up a host name for an IP address, but failed to find the
4831 original IP address when looking up the host name (a safety check), it
4832 output the message "<ip address> does not match any IP for NULL", which was
4833 confusing, to say the least. The bug was that the host name should have
4834 appeared instead of "NULL".
4836 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4837 user that is built into the binary, and the -C or -D options is used, root
4838 privilege is dropped before the configuration file is read. In addition,
4839 logging is switched to stderr instead of the normal log files. If the
4840 configuration then re-defines the Exim user, the unprivileged environment
4841 is probably not what is expected, so Exim logs a panic warning message (but
4844 However, if deliver_drop_privilege is set, the unprivileged state may well
4845 be exactly what is intended, so the warning has been cut out in that case,
4846 and Exim is allowed to try to write to its normal log files.
4852 1. smtp_return_error_details was not giving details for temporary sender
4853 or receiver verification errors.
4855 2. Diagnose a configuration error if two authenticators have the same public
4858 3. Exim used not to create the message log file for a message until the first
4859 delivery attempt. This could be confusing when incoming messages were held
4860 for policy or load reasons. The message log file is now created at the time
4861 the message is received, and an initial "Received" line is written to it.
4863 4. The automatically generated man page for command line options had a minor
4864 bug that caused no ill effects; however, a more serious problem was that
4865 the procedure for building the man page automatically didn't always
4866 operate. Consequently, release 4.20 contains an out-of-date version. This
4867 shouldn't happen again.
4869 5. When building Exim with embedded Perl support, the script that builds the
4870 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4871 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4873 6. The freeze_tell option was not being used for messages that were frozen on
4874 arrival, either by an ACL or by local_scan().
4876 7. Added the smtp_incomplete_transaction log selector.
4878 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4879 was accepting AUTH without a new EHLO.
4881 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4882 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4885 10. Logging of TCP/IP connections (when configured) now happens in the main
4886 daemon process instead of the child process, so that the TCP/IP connection
4887 count is more accurate (but it can never be perfect).
4889 11. The use of "drop" in a nested ACL was not being handled correctly in the
4890 outer ACL. Now, if condition failure induced by the nested "drop" causes
4891 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4892 or "require"), the connection is dropped.
4894 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4895 that yield "discard" can now be used with an "accept" or a "discard" verb,
4896 but an error is generated for any others (because I can't see a useful way
4897 to define what should happen).
4899 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4900 are now processed in the same way as lines in the Exim configuration file.
4901 In particular, continuation lines are supported.
4903 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4905 15. Added -ti meaning -t -i.
4907 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4908 domains, and warn by logging if others are found.
4910 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4911 returned from GIFCONF when Exim is trying to find the list of interfaces on
4912 a host. The code in os.c has been modified to copy each ifreq to an aligned
4913 structure in all cases.
4915 Also, in some cases, the returned ifreq's were being copied to a 'struct
4916 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4917 means the last couple of bytes of an IPv6 address could be chopped if the
4918 ifreq contained only a normal sockaddr (14 bytes storage).
4920 18. Named domain lists were not supported in the hosts_treat_as_local option.
4921 An entry such as +xxxx was not recognized, and was treated as a literal
4924 19. Ensure that header lines added by a DATA ACL are included in the reject log
4925 if the ACL subsequently rejects the message.
4927 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4928 MD5 (which is deprecated).
4930 21. When testing a filter file using -bf, Exim was writing a message when it
4931 took the sender from a "From " line in the message, but it was not doing so
4932 when it took $return_path from a Return-Path: header line. It now does.
4934 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4935 with a valid header line field name (a series of printing characters
4936 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4938 23. Changed "disc" in the source to "disk" to conform to the documentation and
4939 the book and for uniformity.
4941 24. Ignore Sendmail's -Ooption=value command line item.
4943 25. When execve() failed while trying to run a command in a pipe transport,
4944 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4945 could be confused with a return value of 69 from the command itself. This
4946 has been changed to 127, the value the shell returns if it is asked to run
4947 a non-existent command. The wording for the related log line suggests a
4948 non-existent command as the problem.
4950 26. If received_header_text expands to an empty string, do not add a Received:
4951 header line to the message. (Well, it adds a token one on the spool, but
4952 marks it "old" so that it doesn't get used or transmitted.)
4954 27. Installed eximstats 1.28 (addition of -nt option).
4956 28. There was no check for failure on the call to getsockname() in the daemon
4957 code. This can fail if there is a shortage of resources on the system, with
4958 ENOMEM, for example. A temporary error is now given on failure.
4960 29. Contrary to the C standard, it seems that in some environments, the
4961 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4962 program. Exim now does this explicitly; it affects the formatting of
4963 timestamps using strftime().
4965 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4966 complaints. I've now initialized all the variables, to avoid this.
4968 32. Header lines added by a system filter were not being "seen" during
4969 transport-time rewrites.
4971 33. The info_callback() function passed to OpenSSL is set up with type void
4972 (*)(SSL *, int, int), as described somewhere. However, when calling the
4973 function (actually a macro) that sets it up, the type void(*)() is
4974 expected. I've put in a cast to prevent warnings from picky compilers.
4976 34. If a DNS black list lookup found a CNAME record, but there were no A
4977 records associated with the domain it pointed at, Exim crashed.
4979 35. If a DNS black list lookup returned more than one A record, Exim ignored
4980 all but the first. It now scans all returned addresses if a particular IP
4981 value is being sought. In this situation, the contents of the
4982 $dnslist_value variable are a list of all the addresses, separated by a
4985 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4986 to accept a host name and all its aliases if the forward lookup for any of
4987 them yielded the IP address of the incoming connection. Now it accepts only
4988 those names whose forward lookup yields the correct IP address. Any other
4989 names are discarded. This closes a loophole whereby a rogue DNS
4990 administrator could create reverse DNS records to break through a
4991 wildcarded host restriction in an ACL.
4993 37. If a user filter or a system filter that ran in a subprocess used any of
4994 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4995 the wrong values were passed to the pipe command ($thisaddress had the
4996 value of $0, $0 had the value of $1, etc). This bug was introduced by
4997 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4999 38. Improved the line breaking for long SMTP error messages from ACLs.
5000 Previously, if there was no break point between 40 and 75 characters, Exim
5001 left the rest of the message alone. Two changes have been made: (a) I've
5002 reduced the minimum length to 35 characters; (b) if it can't find a break
5003 point between 35 and 75 characters, it looks ahead and uses the first one
5004 that it finds. This may give the occasional overlong line, but at least the
5005 remaining text gets split now.
5007 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5008 file descriptors might be low, and that setting 1000 would always raise it.
5009 It turns out that in some environments, the limit is already over 1000 and
5010 that lowering it causes trouble. So now Exim takes care not to decrease it.
5012 40. When delivering a message, the value of $return_path is set to $sender_
5013 address at the start of routing (routers may change the value). By an
5014 oversight, this default was not being set up when an address was tested by
5015 -bt or -bv, which affected the outcome if any router or filter referred to
5018 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5019 log only when "message" or "log_message" are set. However, if one of the
5020 conditions was an address verification, or a call to a nested ACL, the
5021 messages generated by the underlying test were being passed through. This
5022 no longer happens. The underlying message is available in $acl_verify_
5023 message for both "message" and "log_message" expansions, so it can be
5024 passed through if needed.
5026 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5027 new expansion $bh_ to give the encoded byte string without charset
5028 translation. Translation happens only if iconv() is available; HAVE_ICONV
5029 indicates this at build time. HEADERS_CHARSET gives the charset to
5030 translate to; headers_charset can change it in the configuration, and
5031 "headers charset" can change it in an individual filter file.
5033 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5034 that creates RFC 2047 encoded "words" labels them as that charset instead
5035 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5036 expansion operator; (ii) when Exim creates a From: line for a local
5037 message; (iii) when a header line is rewritten to include a "phrase" part.
5039 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5040 buggy, causing it to skip the first lines of messages whose message ID
5041 ended in 'D'. This would not have bitten before Exim release 4.14, because
5042 message IDs were unlikely to end in 'D' before then. The effect was to have
5043 incorrect size information for certain domains.
5045 45. #include "config.h" was missing at the start of the crypt16.c module. This
5046 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5049 46. If there was a timeout during a "random" callout check, Exim treated it as
5050 a failure of the random address, and carried on sending RSET and the real
5051 address. If the delay was just some slowness somewhere, the response to the
5052 original RCPT would be taken as a response to RSET and so on, causing
5053 mayhem of various kinds.
5055 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5056 when I implemented it. It didn't allow for the fact that some option values
5057 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5058 the right test for positive values.
5060 48. Domain names in DNS records are case-independent. Exim always looks them up
5061 in lower case. Some resolvers return domain names in exactly the case they
5062 appear in the zone file, that is, they may contain uppercase letters. Not
5063 all resolvers do this - some return always lower case. Exim was treating a
5064 change of case by a resolver as a change of domain, similar to a widening
5065 of a domain abbreviation. This triggered its re-routing code and so it was
5066 trying to route what was effectively the same domain again. This normally
5067 caused routing to fail (because the router wouldn't handle the domain
5068 twice). Now Exim checks for this case specially, and just changes the
5069 casing of the domain that it ultimately uses when it transmits the message
5072 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5075 50. If a filter generated a file delivery with a non-absolute name (possible if
5076 no home directory exists for the router), the forbid_file option was not
5079 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5080 the existing equality matching.
5082 52. Exim was using ints instead of ino_t variables in some places where it was
5083 dealing with inode numbers.
5085 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5086 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5087 called TMPDIR, and if it finds it is different, it changes its value.
5089 54. The smtp_printf() function is now made available to local_scan() so
5090 additional output lines can be written before returning. There is also an
5091 smtp_fflush() function to enable the detection of a dropped connection.
5092 The variables smtp_input and smtp_batched_input are exported to
5095 55. Changed the default runtime configuration: the message "Unknown user"
5096 has been removed from the ACL, and instead placed on the localuser router,
5097 using the cannot_route_message feature. This means that any verification
5098 failures that generate their own messages won't get overridden. Similarly,
5099 the "Unrouteable address" message that was in the ACL for unverifiable
5100 relay addresses has also been removed.
5102 56. Added hosts_avoid_esmtp to the smtp transport.
5104 57. The exicyclog script was not checking for the esoteric option
5105 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5106 will work only if exicyclog is run under the appropriate euid.
5108 58. Following a discussion on the list, the rules by which Exim recognises line
5109 endings on incoming messages have been changed. The -dropcr and drop_cr
5110 options are now no-ops, retained only for backwards compatibility. The
5111 following line terminators are recognized: LF CRLF CR. However, special
5112 processing applies to CR:
5114 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5115 nor a local message in the state where . is a terminator.
5117 (ii) If a bare CR is encountered in a header line, an extra space is added
5118 after the line terminator so as not to end the header. The reasoning
5119 behind this is that bare CRs in header lines are most likely either
5120 to be mistakes, or people trying to play silly games.
5122 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5123 was being incorrectly given as 18 bytes larger than it should have been.
5124 This is a VOB (very old bug).
5126 60. This may never have affected anything current, but just in case it has:
5127 When the local host is found other than at the start of a list of hosts,
5128 the local host, those with the same MX, and any that follow, are discarded.
5129 When the list in question was part of a longer list of hosts, the following
5130 hosts (not currently being processed) were also being discarded. This no
5131 longer happens. I'm not sure if this situation could ever has previously
5134 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5137 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5138 header. This is something that is recommended in a new Internet Draft, and
5139 is something that is documented as being done by Sendmail. There are two
5140 possible values. For messages generated by the autoreply transport, Exim
5143 Auto-Submitted: auto-replied
5145 whereas for all other generated messages (e.g. bounces) it adds
5147 Auto-Submitted: auto-generated
5149 63. The "personal" condition in filters now includes a test for the
5150 Auto-Submitted: header. If it contains the string "auto-" the message it
5151 not considered personal.
5153 64. Added rcpt_include_affixes as a generic transport option.
5155 65. Added queue_only_override (default true).
5157 66. Added the syslog_duplication option.
5159 67. If what should have been the first header line of a message consisted of
5160 a space followed by a colon, Exim was mis-interpreting it as a header line.
5161 It isn't of course - it is syntactically invalid and should therefore be
5162 treated as the start of the message body. The misbehaviour could have
5163 caused a number of strange effects, including loss of data in subsequent
5164 header lines, and spool format errors.
5166 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5167 client host had authenticated. This control can now be exercised by an ACL
5168 for more flexibility.
5170 69. By default, callouts do not happen when testing with -bh. There is now a
5171 variant, -bhc, which does actually run the callout code, including
5172 consulting and updating the callout cache.
5174 70. Added support for saslauthd authentication, courtesy of Alexander
5177 71. If statvfs() failed on the spool or log directories while checking their
5178 size for availability, Exim confusingly gave the error "space shortage".
5179 Furthermore, in debugging mode it crashed with a floating point exception.
5180 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5181 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5182 serious problem, Exim now writes to the main and panic logs when this
5183 happens, with details of the failure. It then refuses to accept the
5184 incoming message, giving the message "spool directory problem" or "log
5185 directory problem" with a 421 code for SMTP messages.
5187 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5188 0, 1, and 2 exist, because some OS complain for execs without them (see
5189 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5190 descriptors. However, the code omitted to check that the open succeeded,
5191 causing mysterious errors if for some reason the permissions on /dev/null
5192 got screwed. Now Exim writes a message to the main and panic logs, and
5193 bombs out if it can't open /dev/null.
5195 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5196 interact so that it is all more flexible. It is supposed to remain
5197 backwards compatible. Also added extra_local_interfaces.
5199 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5200 to bomb out with an assertion failure - to the client this appears as a
5201 connection drop. This problem occurs in the part of the code that was taken
5202 from the Samba project. Fortunately, the assertion is in a very simple
5203 function, so I have fixed this by reproducing the function inline in the
5204 one place where it is called, and arranging for authentication to fail
5205 instead of killing the process with assert().
5207 75. The SPA client code was not working when the server requested OEM rather
5208 than Unicode encoding.
5210 76. Added code to make require_files with a specific uid setting more usable in
5211 the case where statting the file as root fails - usually a non-root-mounted
5212 NFS file system. When this happens and the failure is EACCES, Exim now
5213 forks a subprocess and does the per-uid checking as the relevant uid.
5215 77. Added process_log_path.
5217 78. If log_file_path was not explicitly set, a setting of check_log_space or
5218 check_log_inodes was ignored.
5220 79. If a space check for the spool or log partitions fails, the incident is now
5221 logged. Of course, in the latter case the data may get lost...
5223 80. Added the %p formatting code to string_format() so that it can be used to
5224 print addresses in debug_print(). Adjusted all the address printing in the
5225 debugging in store.c to use %p rather than %d.
5227 81. There was a concern that a line of code in smtp_in.c could overflow a
5228 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5229 initially expressed, the concern was not well-founded, because trailing
5230 spaces are removed early. However, if the trailing spaces were followed by
5231 a NULL, they did not get removed, so the overflow was possible. Two fixes
5234 (a) I re-wrote the offending code in a cleaner fashion.
5235 (b) If an incoming SMTP command contains a NULL character, it is rejected
5238 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5239 runs initgroups(), so that if the Exim user is in any additional groups,
5240 they will be used during message reception.
5246 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob