1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
44 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
45 Patch from Jakob Hirsch.
47 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
48 SQL string expansion failure details.
49 Patch from Andrey Oktyabrski.
51 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
52 Patch from Simon Arlott.
54 PP/09 New lookups functionality failed to compile on old gcc which rejects
55 extern declarations in function scope.
56 Patch from Oliver Fleischmann
58 PP/10 Use sig_atomic_t for flags set from signal handlers.
59 Check getgroups() return and improve debugging.
60 Fixed developed for diagnosis in bug 927 (which turned out to be
63 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
64 Patch from Mark Zealey.
70 TF/01 Failure to get a lock on a hints database can have serious
71 consequences so log it to the panic log.
73 TF/02 Log LMTP confirmation messages in the same way as SMTP,
74 controlled using the smtp_confirmation log selector.
76 TF/03 Include the error message when we fail to unlink a spool file.
78 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
79 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
80 for maintaining out-of-tree patches for some time.
82 PP/01 Bugzilla 139: Documentation and portability issues.
83 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
84 Handle per-OS dynamic-module compilation flags.
86 PP/02 Let /dev/null have normal permissions.
87 The 4.73 fixes were a little too stringent and complained about the
88 permissions on /dev/null. Exempt it from some checks.
89 Reported by Andreas M. Kirchwitz.
91 PP/03 Report version information for many libraries, including
92 Exim version information for dynamically loaded libraries. Created
93 version.h, now support a version extension string for distributors
94 who patch heavily. Dynamic module ABI change.
96 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
97 privilege escalation vulnerability whereby the Exim run-time user
98 can cause root to append content of the attacker's choosing to
101 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
104 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
105 If dropping privileges for untrusted macros, we disabled normal logging
106 on the basis that it would fail; for the Exim run-time user, this is not
107 the case, and it resulted in successful deliveries going unlogged.
108 Fixed. Reported by Andreas Metzler.
114 PP/01 Date: & Message-Id: revert to normally being appended to a message,
115 only prepend for the Resent-* case. Fixes regression introduced in
116 Exim 4.70 by NM/22 for Bugzilla 607.
118 PP/02 Include check_rfc2047_length in configure.default because we're seeing
119 increasing numbers of administrators be bitten by this.
121 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
123 PP/03 Bugzilla 994: added openssl_options main configuration option.
125 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
127 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
129 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
131 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
132 without a peer certificate, leading to a segfault because of an
133 assumption that peers always have certificates. Be a little more
134 paranoid. Problem reported by Martin Tscholak.
136 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
137 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
138 NB: ClamAV planning to remove STREAM in "middle of 2010".
139 CL also introduces -bmalware, various -d+acl logging additions and
140 more caution in buffer sizes.
142 PP/09 Implemented reverse_ip expansion operator.
144 PP/10 Bugzilla 937: provide a "debug" ACL control.
146 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
148 PP/12 Bugzilla 973: Implement --version.
150 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
152 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
154 PP/15 Bugzilla 816: support multiple condition rules on Routers.
156 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
157 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
158 ignore trailing whitespace.
160 JJ/02 prevent non-panic DKIM error from being sent to paniclog
162 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
165 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
166 Notification from Dr Andrew Aitchison.
168 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
169 ExtendedDetectionInfo response format.
170 Notification from John Horne.
172 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
175 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
176 XSL and documented dependency on system catalogs, with examples of how
179 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
182 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
183 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
184 configuration file which is writeable by the Exim user or group.
186 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
187 of configuration files to cover files specified with the -C option if
188 they are going to be used with root privileges, not just the default
191 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
192 option (effectively making it always true).
194 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
195 files to be used while preserving root privileges.
197 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
198 that rogue child processes cannot use them.
200 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
201 run-time user, instead of root.
203 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
204 Exim run-time user without dropping privileges.
206 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
207 result string, instead of calling string_vformat() twice with the same
210 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
211 for other users. Others should always drop root privileges if they use
212 -C on the command line, even for a whitelisted configure file.
214 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
216 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
222 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
223 $data_path, and $header_path variables; fixed documentation bugs and
226 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
227 exipick to access non-standard spools, including the "frozen" queue
230 NM/01 Bugzilla 965: Support mysql stored procedures.
231 Patch from Alain Williams
233 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
235 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
236 Patch from Andreas Metzler
238 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
239 Patch from Kirill Miazine
241 NM/05 Bugzilla 671: Added umask to procmail example.
243 JJ/03 installed exipick 20100323.0, fixing doc bug
245 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
246 directory. Notification and patch from Dan Rosenberg.
248 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
250 TK/02 Improve log output when DKIM signing operation fails.
252 MH/01 Treat the transport option dkim_domain as a colon separated
253 list, not as a single string, and sign the message with each element,
254 omitting multiple occurences of the same signer.
256 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
257 Bugzilla 985, 986. Patch by Simon Arlott
259 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
260 Patch by Simon Arlott
262 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
263 MBX locking. Notification from Dan Rosenberg.
269 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
271 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
273 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
275 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
277 NM/04 Bugzilla 915: Flag broken perl installation during build.
283 TK/01 Added patch by Johannes Berg that expands the main option
284 "spamd_address" if it starts with a dollar sign.
286 TK/02 Write list of recipients to X-Envelope-Sender header when building
287 the mbox-format spool file for content scanning (suggested by Jakob
290 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
291 (http://www.dcc-servers.net/) support via dccifd. Activated by
292 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
294 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
295 by Mark Daniel Reidel <mr@df.eu>.
297 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
298 When building exim an external PCRE library is now needed -
299 PCRE is a system library on the majority of modern systems.
300 See entry on PCRE_LIBS in EDITME file.
302 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
303 conversation. Added nologin parameter to request.
304 Patch contributed by Kirill Miazine.
306 TF/01 Do not log submission mode rewrites if they do not change the address.
308 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
310 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
311 log files in place. Contributed by Roberto Lima.
313 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
315 TF/03 Bugzilla 615: When checking the local_parts router precondition
316 after a local_part_suffix or local_part_prefix option, Exim now
317 does not use the address's named list lookup cache, since this
318 contains cached lookups for the whole local part.
320 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
321 Robert Millan. Documentation is in experimental-spec.txt.
323 TF/04 Bugzilla 668: Fix parallel build (make -j).
325 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
327 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
328 Patch provided by Jan Srzednicki.
330 TF/05 Leading white space used to be stripped from $spam_report which
331 wrecked the formatting. Now it is preserved.
333 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
334 that they are available at delivery time.
336 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
338 TF/08 TLS error reporting now respects the incoming_interface and
339 incoming_port log selectors.
341 TF/09 Produce a more useful error message if an SMTP transport's hosts
342 setting expands to an empty string.
344 NM/06 Bugzilla 744: EXPN did not work under TLS.
345 Patch provided by Phil Pennock.
347 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
348 Patch provided by Richard Godbee.
350 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
351 acl_smtp_notquit, added index entry.
353 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
354 Patch provided by Eugene Bujak.
356 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
357 accept(). Patch provided by Maxim Dounin.
359 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
360 Patch provided by Phil Pennock.
362 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
364 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
365 Patch provided by Brad "anomie" Jorsch.
367 NM/14 Bugzilla 622: Added timeout setting to transport filter.
368 Patch provided by Dean Brooks.
370 TK/05 Add native DKIM support (does not depend on external libraries).
372 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
373 Patch provided by Graeme Fowler.
375 NM/16 Bugzilla 851: Documentation example syntax fix.
377 NM/17 Changed NOTICE file to remove references to embedded PCRE.
379 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
382 NM/19 Bugzilla 745: TLS version reporting.
383 Patch provided by Phil Pennock.
385 NM/20 Bugzilla 167: bool: condition support.
386 Patch provided by Phil Pennock.
388 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
389 clients. Patch provided by Phil Pennock.
391 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
392 Patch provided by Brad "anomie" Jorsch.
394 NM/23 Bugzilla 687: Fix misparses in eximstats.
395 Patch provided by Heiko Schlittermann.
397 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
398 Patch provided by Heiko Schlittermann.
400 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
401 plus update to original patch.
403 NM/26 Bugzilla 799: Documentation correction for ratelimit.
405 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
406 Patch provided by David Brownlee.
408 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
410 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
412 NM/30 Bugzilla 888: TLS documentation bugfixes.
414 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
416 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
417 Unlike the original bugzilla I have changed all shell scripts in src tree.
419 NM/33 Bugzilla 898: Transport filter timeout fix.
420 Patch by Todd Rinaldo.
422 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
423 Patch by Serge Demonchaux.
425 NM/35 Bugzilla 39: Base64 decode bug fixes.
426 Patch by Jakob Hirsch.
428 NM/36 Bugzilla 909: Correct connect() call in dcc code.
430 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
432 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
434 NM/39 Bugzilla 911: Fixed MakeLinks build script.
440 TK/01 Add preliminary DKIM support. Currently requires a forked version of
441 ALT-N's libdkim that I have put here:
442 http://duncanthrax.net/exim-experimental/
444 Note to Michael Haardt: I had to rename some vars in sieve.c. They
445 were called 'true' and it seems that C99 defines that as a reserved
446 keyword to be used with 'bool' variable types. That means you could
447 not include C99-style headers which use bools without triggering
448 build errors in sieve.c.
450 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
451 as mailq or other aliases. Changed the --help handling significantly
452 to do whats expected. exim_usage() emits usage/help information.
454 SC/01 Added the -bylocaldomain option to eximstats.
456 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
458 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
460 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
466 PH/01 Another patch from the Sieve maintainer.
468 PH/02 When an IPv6 address is converted to a string for single-key lookup
469 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
470 dots are used instead of colons so that keys in lsearch files need not
471 contain colons. This was done some time before quoting was made available
472 in lsearch files. However, iplsearch files do require colons in IPv6 keys
473 (notated using the quote facility) so as to distinguish them from IPv4
474 keys. This meant that lookups for IP addresses in host lists did not work
475 for iplsearch lookups.
477 This has been fixed by arranging for IPv6 addresses to be expressed with
478 colons if the lookup type is iplsearch. This is not incompatible, because
479 previously such lookups could never work.
481 The situation is now rather anomolous, since one *can* have colons in
482 ordinary lsearch keys. However, making the change in all cases is
483 incompatible and would probably break a number of configurations.
485 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
488 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
489 conversion specification without a maximum field width, thereby enabling
490 a rogue spamd server to cause a buffer overflow. While nobody in their
491 right mind would setup Exim to query an untrusted spamd server, an
492 attacker that gains access to a server running spamd could potentially
493 exploit this vulnerability to run arbitrary code as the Exim user.
495 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
496 $primary_hostname instead of what libspf2 thinks the hosts name is.
498 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
499 a directory entry by the name of the lookup key. Previously, if a
500 symlink pointed to a non-existing file or a file in a directory that
501 Exim lacked permissions to read, a lookup for a key matching that
502 symlink would fail. Now it is enough that a matching directory entry
503 exists, symlink or not. (Bugzilla 503.)
505 PH/03 The body_linecount and body_zerocount variables are now exported in the
508 PH/04 Added the $dnslist_matched variable.
510 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
511 This means they are set thereafter only if the connection becomes
514 PH/06 Added the client_condition to authenticators so that some can be skipped
515 by clients under certain conditions.
517 PH/07 The error message for a badly-placed control=no_multiline_responses left
518 "_responses" off the end of the name.
520 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
522 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
523 (without spaces) instead of just copying the configuration text.
525 PH/10 Added the /noupdate option to the ratelimit ACL condition.
527 PH/11 Added $max_received_linelength.
529 PH/12 Added +ignore_defer and +include_defer to host lists.
531 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
532 way in which PCRE > 7.0 is built.
534 PH/14 Implemented queue_only_load_latch.
536 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
537 MAIL command. The effect was to mangle the value on 64-bit systems.
539 PH/16 Another patch from the Sieve maintainer.
541 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
543 PH/18 If a system quota error occurred while trying to create the file for
544 a maildir delivery, the message "Mailbox is full" was not appended to the
545 bounce if the delivery eventually timed out. Change 4.67/27 below applied
546 only to a quota excession during the actual writing of the file.
548 PH/19 It seems that peer DN values may contain newlines (and other non-printing
549 characters?) which causes problems in log lines. The DN values are now
550 passed through string_printing() before being added to log lines.
552 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
553 and InterBase are left for another time.)
555 PH/21 Added message_body_newlines option.
557 PH/22 Guard against possible overflow in moan_check_errorcopy().
559 PH/23 POSIX allows open() to be a macro; guard against that.
561 PH/24 If the recipient of an error message contained an @ in the local part
562 (suitably quoted, of course), incorrect values were put in $domain and
563 $local_part during the evaluation of errors_copy.
569 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
570 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
573 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
574 issue a MAIL command.
576 PH/02 In an ACL statement such as
578 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
580 if a client was not listed at all, or was listed with a value other than
581 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
582 the condition was not true (as it should be), so access was not denied.
583 The bug was that the ! inversion was incorrectly passed on to the second
584 item. This has been fixed.
586 PH/03 Added additional dnslists conditions == and =& which are different from
587 = and & when the dns lookup returns more than one IP address.
589 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
590 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
592 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
593 FSYNC, which compiles an option called disable_fsync that allows for
594 bypassing fsync(). The documentation is heavily laced with warnings.
596 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
598 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
599 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
600 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
601 including adding "make clean"; (3) Added -fPIC when compiling the test
602 dynamically loaded module, to get rid of a warning.
604 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
605 message fails, move_frozen_messages = true and ignore_bounce_errors_after
606 = 0s. The bug is otherwise harmless.
608 PH/07 There was a bug in the dovecot authenticator such that the value of
609 $auth1 could be overwritten, and so not correctly preserved, after a
610 successful authentication. This usually meant that the value preserved by
611 the server_setid option was incorrect.
613 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
615 PH/09 Installed PCRE release 7.0.
617 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
618 run for batched SMTP input. It is now run at the start of every message
619 in the batch. While fixing this I discovered that the process information
620 (output by running exiwhat) was not always getting set for -bs and -bS
621 input. This is fixed, and it now also says "batched" for BSMTP.
623 PH/11 Added control=no_pipelining.
625 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
626 patch, slightly modified), and move the expansion of helo_data till after
627 the connection is made in the smtp transport (so it can use these
630 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
632 PH/14 Added log_selector = +pid.
634 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
636 PH/16 Add ${if forany and ${if forall.
638 PH/17 Added dsn_from option to vary the From: line in DSNs.
640 PH/18 Flush SMTP output before performing a callout, unless control =
641 no_callout_flush is set.
643 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
644 was true (the default) a successful delivery failed to delete the retry
645 item, thus causing premature timeout of the address. The bug is now
648 PH/20 Added hosts_avoid_pipelining to the smtp transport.
650 PH/21 Long custom messages for fakedefer and fakereject are now split up
651 into multiline reponses in the same way that messages for "deny" and
652 other ACL rejections are.
654 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
655 with slight modification.
657 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
658 draft, changing the syntax and factoring some duplicate code".
660 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
661 for deliveries of the second and subsequent messages over the same SMTP
664 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
665 ${reduce, with only minor "tidies".
667 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
669 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
670 expansion side effects.
672 PH/27 When a message times out after an over-quota error from an Exim-imposed
673 quota, the bounce message says "mailbox is full". This message was not
674 being given when it was a system quota that was exceeded. It now should
677 MH/03 Made $recipients available in local_scan(). local_scan() already has
678 better access to the recipient list through recipients_list[], but
679 $recipients can be useful in postmaster-provided expansion strings.
681 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
682 in the case of a MAIL command with additional options following the
683 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
684 were accidentally chopped off.
686 PH/29 SMTP synchronization checks are implemented when a command is read -
687 there is a check that no more input is waiting when there shouldn't be
688 any. However, for some commands, a delay in an ACL can mean that it is
689 some time before the response is written. In this time, more input might
690 arrive, invalidly. So now there are extra checks after an ACL has run for
691 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
692 pipelining has not been advertised.
694 PH/30 MH's patch to allow iscntrl() characters to be list separators.
696 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
697 returned in the SMTP response when smtp_return_error_details was false.
700 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
701 instead of the C I/O that was originally supplied, because problems were
704 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
705 Exim which did not show up earlier: it was assuming that a call to
706 SSL_CTX_set_info_callback() might give an error value. In fact, there is
707 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
708 was a macro that became an assignment, so it seemed to work. This has
709 changed to a proper function call with a void return, hence the compile
710 error. Exim's code has been fixed.
712 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
715 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
717 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
719 PH/37 If a message is not accepted after it has had an id assigned (e.g.
720 because it turns out to be too big or there is a timeout) there is no
721 "Completed" line in the log. When some messages of this type were
722 selected by exigrep, they were listed as "not completed". Others were
723 picked up by some special patterns. I have improved the selection
724 criteria to be more general.
726 PH/38 The host_find_failed option in the manualroute router can now be set
727 to "ignore", to completely ignore a host whose IP address cannot be
728 found. If all hosts are ignored, the behaviour is controlled by the new
729 host_all_ignored option.
731 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
732 homing or because of multiple MX records with /mx) generated more than
733 one IP address, and the following item turned out to be the local host,
734 all the secondary addresses of the first item were incorrectly removed
735 from the list, along with the local host and any following hosts (which
736 is what is supposed to happen).
738 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
739 whoever called Exim into the -H file. In the case of the daemon it was
740 behaving confusingly. When first started, it used values for whoever
741 started the daemon, but after a SIGHUP it used the Exim user (because it
742 calls itself on a restart). I have changed the code so that it now always
745 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
746 message are rejected with the same error (e.g. no authentication or bad
747 sender address), and a DATA command is nevertheless sent (as can happen
748 with PIPELINING or a stupid MUA), the error message that was given to the
749 RCPT commands is included in the rejection of the DATA command. This is
750 intended to be helpful for MUAs that show only the final error to their
753 PH/42 Another patch from the Sieve maintainer.
755 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
756 Eximstats - Fixed some broken HTML links and added missing column headers
758 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
759 columns for Rejects, Temp Rejects, Ham, and Spam rows.
761 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
763 PH/43 Yet another patch from the Sieve maintainer.
765 PH/44 I found a way to check for a TCP/IP connection going away before sending
766 the response to the final '.' that terminates a message, but only in the
767 case where the client has not sent further data following the '.'
768 (unfortunately, this is allowed). However, in many cases there won't be
769 any further data because there won't be any more messages to send. A call
770 to select() can be used: if it shows that the input is "ready", there is
771 either input waiting, or the socket has been closed. An attempt to read
772 the next input character can distinguish the two cases. Previously, Exim
773 would have sent an OK response which the client would never have see.
774 This could lead to message repetition. This fix should cure that, at
775 least in a lot of common cases.
777 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
778 advertised in response to EHLO.
784 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
785 fixed by 4.65/MH/01 (is this a record?) are fixed:
787 (i) An empty string was always treated as zero by the numeric comparison
788 operators. This behaviour has been restored.
790 (ii) It is documented that the numeric comparison operators always treat
791 their arguments as decimal numbers. This was broken in that numbers
792 starting with 0 were being interpreted as octal.
794 While fixing these problems I realized that there was another issue that
795 hadn't been noticed. Values of message_size_limit (both the global option
796 and the transport option) were treated as octal if they started with 0.
797 The documentation was vague. These values are now always treated as
798 decimal, and I will make that clear in the documentation.
804 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
805 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
808 MH/01 Don't check that the operands of numeric comparison operators are
809 integers when their expansion is in "skipping" mode (fixes bug
810 introduced by 4.64-PH/07).
812 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
813 child addresses, Exim now panics and dies. Previously, because the count
814 is held in a short int, deliveries were likely to be lost. As such a
815 large number of recipients for a single message is ridiculous
816 (performance will be very, very poor), I have chosen to impose a limit
817 rather than extend the field.
823 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
824 leftover -K file (the existence of which was triggered by #402).
825 While we were at it, introduced process PID as part of the -K
826 filename. This should rule out race conditions when creating
829 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
830 processing considerably. Previous code took too long for large mails,
831 triggering a timeout which in turn triggers #401.
833 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
834 in the DK code in transports.c. sendfile() is not really portable,
835 hence the _LINUX specificness.
837 TF/01 In the add_headers option to the mail command in an Exim filter,
838 there was a bug that Exim would claim a syntax error in any
839 header after the first one which had an odd number of characters
842 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
843 callout verification, Exim cached a "reject" for the entire domain. This
844 is correct for most verifications, but it is not correct for a recipient
845 verification with use_sender or use_postmaster set, because in that case
846 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
847 case of MAIL FROM:<> rejection from other early rejections (e.g.
848 rejection of HELO). When verifying a recipient using a non-null MAIL
849 address, the cache is ignored if it shows MAIL FROM:<> rejection.
850 Whatever the result of the callout, the value of the domain cache is
851 left unchanged (for any other kind of callout, getting as far as trying
852 RCPT means that the domain itself is ok).
854 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
857 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
858 manifest itself as EPIPE rather than ECONNECT. When tidying away a
859 session, the daemon ignores ECONNECT errors and logs others; it now
860 ignores EPIPE as well.
862 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
863 (quoted-printable decoding).
865 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
866 later the small subsequent patch to fix an introduced bug.
868 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
870 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
872 PH/08 An error is now given if message_size_limit is specified negative.
874 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
875 to be given (somewhat) arbitrary names.
877 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
880 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
881 miscellaneous code fixes
883 PH/10 Added the log_reject_target ACL modifier to specify where to log
886 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
887 hostname. This is wrong, because it relates to the incoming message (and
888 probably the interface on which it is arriving) and not to the outgoing
889 callout (which could be using a different interface). This has been
890 changed to use the value of the helo_data option from the smtp transport
891 instead - this is what is used when a message is actually being sent. If
892 there is no remote transport (possible with a router that sets up host
893 addresses), $smtp_active_hostname is used.
895 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
896 tweaks were necessary in order to get it to work (see also 21 below):
897 (a) The code assumed that strncpy() returns a negative number on buffer
898 overflow, which isn't the case. Replaced with Exim's string_format()
900 (b) There were several signed/unsigned issues. I just did the minimum
901 hacking in of casts. There is scope for a larger refactoring.
902 (c) The code used strcasecmp() which is not a standard C function.
903 Replaced with Exim's strcmpic() function.
904 (d) The code set only $1; it now sets $auth1 as well.
905 (e) A simple test gave the error "authentication client didn't specify
906 service in request". It would seem that Dovecot has changed its
907 interface. Fortunately there's a specification; I followed it and
908 changed what the client sends and it appears to be working now.
910 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
913 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
914 address A is aliased to B and C, where B exists and C does not. Without
915 -v the output is "A verified" because verification stops after a
916 successful redirection if more than one address is generated. However,
917 with -v the child addresses are also verified. Exim was outputting "A
918 failed to verify" and then showing the successful verification for C,
919 with its parentage. It now outputs "B failed to verify", showing B's
920 parentage before showing the successful verification of C.
922 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
923 look up a TXT record in a specific list after matching in a combined
926 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
927 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
928 they consult the DNS. I had assumed they would set it the way they
929 wanted; and indeed my experiments on Linux seem to show that in some
930 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
931 To be on the safe side, however, I have now made the interface to
932 host_find_byname() similar to host_find_bydns(), with an argument
933 containing the DNS resolver options. The host_find_byname() function now
934 sets these options at its start, just as host_find_bydns() does. The smtp
935 transport options dns_qualify_single and dns_search_parents are passed to
936 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
937 of host_find_byname() use the default settings of RES_DEFNAMES
938 (qualify_single) but not RES_DNSRCH (search_parents).
940 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
941 spool_read_header() do less string testing, by means of a preliminary
942 switch on the second character of optional "-foo" lines. (This is
943 overdue, caused by the large number of possibilities that now exist.
944 Originally there were few.) While I was there, I also converted the
945 str(n)cmp tests so they don't re-test the leading "-" and the first
946 character, in the hope this might squeeze out yet more improvement.
948 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
949 flag allowing group syntax was set by the header_syntax check but not
950 turned off, possible causing trouble later; (2) The flag was not being
951 set at all for the header_verify test, causing "group"-style headers to
952 be rejected. I have now set it in this case, and also caused header_
953 verify to ignore an empty address taken from a group. While doing this, I
954 came across some other cases where the code for allowing group syntax
955 while scanning a header line wasn't quite right (mostly, not resetting
956 the flag correctly in the right place). These bugs could have caused
957 trouble for malformed header lines. I hope it is now all correct.
959 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
960 with the "reply" argument non-NULL. The code, however (which originally
961 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
962 but it didn't always do it. This confused somebody who was copying the
963 code for some other use. I have removed all the tests.
965 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
966 feature that was used to support insecure browsers during the U.S. crypto
967 embargo. It requires special client support, and Exim is probably the
968 only MTA that supported it -- and would never use it because real RSA is
969 always available. This code has been removed, because it had the bad
970 effect of slowing Exim down by computing (never used) parameters for the
971 RSA_EXPORT functionality.
973 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
974 authenticator to fail if there's a tab character in the incoming data
975 (there should never be unless someone is messing about, as it's supposed
976 to be base64-encoded). Also added, on Timo's advice, the "secured" option
977 if the connection is using TLS or if the remote IP is the same as the
978 local IP, and the "valid-client-cert option" if a client certificate has
981 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
982 authenticators. This can be used for authorization after authentication
983 succeeds. (In the case of plaintext, it servers for both authentication
986 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
987 if any retry times were supplied.
989 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
990 connection, where there is no client IP address to check. In this
991 situation, the verify now always succeeds.
993 PH/25 Applied John Jetmore's -Mset patch.
995 PH/26 Added -bem to be like -Mset, but loading a message from a file.
997 PH/27 In a string expansion for a processed (not raw) header when multiple
998 headers of the same name were present, leading whitespace was being
999 removed from all of them, but trailing whitespace was being removed only
1000 from the last one. Now trailing whitespace is removed from each header
1001 before concatenation. Completely empty headers in a concatenation (as
1002 before) are ignored.
1004 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1005 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1007 PH/29 [Removed. This was a change that I later backed out, and forgot to
1008 correct the ChangeLog entry (that I had efficiently created) before
1009 committing the later change.]
1011 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1012 address errors (4xx response to RCPT) over the same connection as other
1013 messages routed to the same hosts. Such deliveries are always "forced",
1014 so retry times are not inspected. This resulted in far too many retries
1015 for the affected addresses. The effect occurred only when there were more
1016 hosts than the hosts_max_try setting in the smtp transport when it had
1017 the 4xx errors. Those hosts that it had tried were not added to the list
1018 of hosts for which the message was waiting, so if all were tried, there
1019 was no problem. Two fixes have been applied:
1021 (i) If there are any address or message errors in an SMTP delivery, none
1022 of the hosts (tried or untried) are now added to the list of hosts
1023 for which the message is waiting, so the message should not be a
1024 candidate for sending over the same connection that was used for a
1025 successful delivery of some other message. This seems entirely
1026 reasonable: after all the message is NOT "waiting for some host".
1027 This is so "obvious" that I'm not sure why it wasn't done
1028 previously. Hope I haven't missed anything, but it can't do any
1029 harm, as the worst effect is to miss an optimization.
1031 (ii) If, despite (i), such a delivery is accidentally attempted, the
1032 routing retry time is respected, so at least it doesn't keep
1033 hammering the server.
1035 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1036 in ${readsocket because some servers need this prod.
1038 PH/32 Added some extra debug output when updating a wait-xxx database.
1040 PH/33 The hint "could be header name not terminated by colon", which has been
1041 given for certain expansion errors for a long time, was not being given
1042 for the ${if def:h_colon_omitted{... case.
1044 PH/34 The spec says: "With one important exception, whenever a domain list is
1045 being scanned, $domain contains the subject domain." There was at least
1046 one case where this was not true.
1048 PH/35 The error "getsockname() failed: connection reset by peer" was being
1049 written to the panic log as well as the main log, but it isn't really
1050 panic-worthy as it just means the connection died rather early on. I have
1051 removed the panic log writing for the ECONNRESET error when getsockname()
1054 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1055 runs only) independently of the message's sender address. This meant
1056 that, if the 4xx error was in fact related to the sender, a different
1057 message to the same recipient with a different sender could confuse
1058 things. In particualar, this can happen when sending to a greylisting
1059 server, but other circumstances could also provoke similar problems.
1060 I have changed the default so that the retry time for these errors is now
1061 based a combination of the sender and recipient addresses. This change
1062 can be overridden by setting address_retry_include_sender=false in the
1065 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1066 remote server are returned as part of bounce messages. This was not
1067 happening for LMTP over a pipe (the lmtp transport), but now it is the
1068 same for both kinds of LMTP.
1070 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1071 in header lines that were in fact CNAMEs. This is no longer the case.
1073 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1074 and queue runs started by the daemon processed all messages. This has
1075 been fixed so that -R and -S can now usefully be given with -q<time>.
1077 PH/40 Import PCRE release 6.7 (fixes some bugs).
1079 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1081 PH/42 Give an error if -q is specified more than once.
1083 PH/43 Renamed the variables $interface_address and $interface_port as
1084 $received_ip_address and $received_port, to make it clear that these
1085 values apply to message reception, and not to the outgoing interface when
1086 a message is delivered. (The old names remain recognized, of course.)
1088 PH/44 There was no timeout on the connect() call when using a Unix domain
1089 socket in the ${readsocket expansion. There now is.
1091 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1092 be meaningful with "accept".
1094 SC/01 Eximstats V1.43
1095 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1097 SC/02 Eximstats V1.44
1098 Use a glob alias rather than an array ref in the generated
1099 parser. This improves both readability and performance.
1101 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1102 Collect SpamAssassin and rejection statistics.
1103 Don't display local sender or destination tables unless
1104 there is data to show.
1105 Added average volumes into the top table text output.
1107 SC/04 Eximstats V1.46
1108 Collect data on the number of addresses (recipients)
1109 as well as the number of messages.
1111 SC/05 Eximstats V1.47
1112 Added 'Message too big' to the list of mail rejection
1113 reasons (thanks to Marco Gaiarin).
1115 SC/06 Eximstats V1.48
1116 Mainlog lines which have GMT offsets and are too short to
1117 have a flag are now skipped.
1119 SC/07 Eximstats V1.49 (Alain Williams)
1120 Added the -emptyok flag.
1122 SC/08 Eximstats V1.50
1123 Fixes for obtaining the IP address from reject messages.
1125 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1126 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1127 whitesspace changes from 4.64-PH/27
1129 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1132 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1133 are found, allow negative numbers in numeric criteria)
1135 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1137 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1138 to match changes made in 4.64-PH/43
1140 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1142 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1143 "domain" is actually an IP address, and force a failure. This locks out
1144 those revolvers/nameservers that support "A-for-A" lookups, in
1145 contravention of the specifications.
1147 PH/48 When a host name was looked up from an IP address, and the subsequent
1148 forward lookup of the name timed out, the host name was left in
1149 $sender_host_name, contrary to the specification.
1151 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1152 restricted to single-key lookups, Exim was not diagnosing an error if
1153 * or *@ was used with a query-style lookup.
1155 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1157 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1158 long ago, but noone interested enough thought of it. Let's just say that
1159 the "1.1" means that there are some new functions that weren't there at
1160 some point in the past.
1162 PH/51 Error processing for expansion failure of helo_data from an smtp
1163 transport during callout processing was broken.
1165 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1166 tested/used via the -bh/-bhc/-bs options.
1168 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1169 bug, fixed in subsequent PCRE releases).
1171 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1172 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1174 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1180 SC/01 Use a glob alias rather than an array ref in eximstats generated
1181 parser. This improves both readability and performance.
1183 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1184 Don't display local sender or destination tables in eximstats unless
1185 there is data to show.
1186 Added average volumes into the eximstats top table text output.
1188 SC/03 Collect data on the number of addresses (recipients) as well
1189 as the number of messages in eximstats.
1191 TF/01 Correct an error in the documentation for the redirect router. Exim
1192 does (usually) call initgroups() when daemonizing.
1194 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1195 with consistent privilege compared to when running as a daemon.
1197 TF/03 Note in the spec that $authenticated_id is not set for local
1198 submissions from trusted users.
1200 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1201 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1203 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1204 by adding some example configuration directives to the default
1205 configuration file. A little bit of work is required to uncomment the
1206 directives and define how usernames and passwords are checked, but
1207 there is now a framework to start from.
1209 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1210 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1211 without this. I don't know how relevant this is to other LDAP libraries.
1213 PH/02 Add the verb name to the "unknown ACL verb" error.
1215 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1217 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1219 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1220 directory not expanded when it should be if an expanded home directory
1221 was set for the address (which is overridden by the transport).
1223 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1226 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1227 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1228 because it is too late at that time, and has no effect.
1230 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1231 security issue with \' (bugzilla #107). I could not use the
1232 PQescapeStringConn() function, because it needs a PGconn value as one of
1235 PH/08 When testing addresses using -bt, indicate those final addresses that
1236 are duplicates that would not cause an additional delivery. At least one
1237 person was confused, thinking that -bt output corresponded to deliveries.
1238 (Suppressing duplicates isn't a good idea as you lose the information
1239 about possibly different redirections that led to the duplicates.)
1241 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1242 systems where poll() doesn't work, in particular OS X.
1244 PH/10 Added more information to debugging output for retry time not reached.
1246 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1247 operations in malware.c.
1249 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1252 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1253 syslog_duplication set false, log lines that would normally be written
1254 both the the main log and to the reject log were not written to syslog at
1257 PH/14 In the default configuration, change the use of "message" in ACL warn
1258 statements to "add_header".
1260 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1261 not followed by a command (e.g. "seen endif").
1263 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1264 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1267 PH/17 Added extra conditions to the default value of delay_warning_condition
1271 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1272 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1273 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1276 The Auto-Submitted: and various List- headers are standardised, whereas I
1277 don't think Precedence: ever was.
1279 PH/18 Refactored debugging code in route_finduser() to show more information,
1280 in particular, the error code if getpwnam() issues one.
1282 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1283 This is apparently needed in addition to the PH/07 change above to avoid
1284 any possible encoding problems.
1286 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1287 but not after initializing Perl.
1289 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1290 output them only if debugging. By default they are written stderr,
1291 apparently, which is not desirable.
1293 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1296 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1299 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1301 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1302 authenticated or an ident call has been made. Suppress the default
1303 values for $authenticated_id and $authenticated_sender (but permit -oMai
1304 and -oMas) when testing with -bh.
1306 PH/24 Re-jigged the order of the tests in the default configuration so that the
1307 tests for valid domains and recipients precede the DNS black list and CSA
1308 tests, on the grounds that those ones are more expensive.
1310 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1311 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1312 This bug exists in every version of Exim that I still have, right back to
1315 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1316 However, an attempt to turn on case-sensitivity in a regex key by
1317 including (?-i) didn't work because the subject string was already
1318 lowercased, and the effects were non-intuitive. It turns out that a
1319 one-line patch can be used to allow (?-i) to work as expected.
1325 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1326 other effects) broke the use of negated acl sub-conditions.
1328 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1331 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1332 "Deny" causes Exim to reject the incoming connection with a 554 error.
1333 Unfortunately, if there is a major crisis, such as a disk failure,
1334 tcp-wrappers gives "deny", whereas what one would like would be some
1335 kind of temporary error. A kludge has been added to help with this.
1336 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1337 554 error is used if errno is still zero or contains ENOENT (which occurs
1338 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1341 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1343 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1344 errors. Otherwise a message that provokes a temporary error (when other
1345 messages do not) can cause a whole host to time out.
1347 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1348 addresses were routed directly to files or pipes from a redirect router.
1349 File deliveries just didn't batch; pipe deliveries might have suffered
1352 PH/06 A failure to get a lock for a hints database would erroneously always say
1353 "Failed to get write lock", even when it was really a read lock.
1355 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1356 of 0600. This has been changed to use the value of the lockfile_mode
1357 option (which defaults to 0600).
1359 PH/08 Applied small patch from the Sieve maintainer.
1361 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1362 folder from quota calculations, a direct delivery into this folder messed
1363 up the contents of the maildirsize file. This was because the regex was
1364 used only to exclude .Trash (or whatever) when the size of the mailbox
1365 was calculated. There was no check that a delivery was happening into an
1366 excluded directory. This bug has been fixed by ignoring all quota
1367 processing for deliveries into excluded directories.
1369 PH/10 Added the maildirfolder_create_regex option to appendfile.
1375 PH/01 The code for finding all the local interface addresses on a FreeBSD
1376 system running IPv6 was broken. This may well have applied to all BSD
1377 systems, as well as to others that have similar system calls. The broken
1378 code found IPv4 interfaces correctly, but gave incorrect values for the
1379 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1380 that it would not match correctly against @[] and not recognize the IPv6
1383 PH/02 The ipliteral router was not recognizing addresses of the form user@
1384 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1386 PH/03 Added disable_ipv6.
1388 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1389 decoded form, because it is most often used to construct To: headers
1390 lines in autoreplies, and the decoded form may well be syntactically
1391 invalid. However, $reply_address has leading white space removed, and all
1392 newlines turned into spaces so that the autoreply transport does not
1395 PH/05 If group was specified without a user on a router, and no group or user
1396 was specified on a transport, the group from the router was ignored.
1398 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1399 for visible compile-time settings that can be used to change these
1400 numbers, for those that want even more. Backwards compatibility with old
1401 spool files has been maintained. However, going back to a previous Exim
1402 release will lost any variables that are in spool files.
1404 PH/07 Two small changes when running in the test harness: increase delay when
1405 passing a TCP/IP connection to a new process, in case the original
1406 process has to generate a bounce, and remove special handling of
1407 127.0.0.2 (sic), which is no longer necessary.
1409 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1410 be the same on different OS.
1412 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1415 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1416 whether --show-vars was specified or not
1418 JJ/02 exipick: Added support for new ACL variable spool format introduced
1421 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1422 syntactically invalid From: or Reply-to: line, and a filter used this to
1423 generate an autoreply, and therefore failed to obtain an address for the
1424 autoreply, Exim could try to deliver to a non-existent relative file
1425 name, causing unrelated and misleading errors. What now happens is that
1426 it logs this as a hard delivery error, but does not attempt to create a
1429 PH/11 The exinext utility has a -C option for testing purposes, but although
1430 the given file was scanned by exinext itself; it wasn't being passed on
1431 when Exim was called.
1433 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1434 an end-of-file indication when reading a command response.
1436 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1437 compiled. In many other places in Exim, IPv6 addresses are always
1438 recognized, so I have changed this. It also means that IPv4 domain
1439 literals of the form [IPV4:n.n.n.n] are now always recognized.
1441 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1442 used if the router is not running as root, for example, when verifying at
1443 ACL time, or when using -bh. The debugging output from this situation was
1444 non-existent - all you got was a failure to exec. I have made two
1447 (a) Failures to set uid/gid, the current directory, or a process leader
1448 in a subprocess such as that created by queryprogram now generate
1449 suitable debugging ouput when -d is set.
1451 (b) The queryprogram router detects when it is not running as root,
1452 outputs suitable debugging information if -d is set, and then runs
1453 the subprocess without attempting to change uid/gid.
1455 PH/15 Minor change to Makefile for building test_host (undocumented testing
1458 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1459 additional section of a DNS packet that returns MX or SRV records.
1460 Instead, it always explicitly searches for A/AAAA records. This avoids
1461 major problems that occur when a DNS server includes only records of one
1462 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1463 fixed another bug: if SRV records were looked up and the corresponding
1464 address records were *not* found in the additional section, the port
1465 values from the SRV records were lost.
1467 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1468 using the correct key (the original address) when searching the retry
1469 rules in order to find which one to use for generating the retry hint.
1471 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1472 adding the default one. Similarly, if it contains a Reply-To: header, the
1473 errors_reply_to option, if set, is not used.
1475 PH/19 When calculating a retry time, Exim used to measure the "time since
1476 failure" by looking at the "first failed" field in the retry record. Now
1477 it does not use this if it is later than than the arrival time of the
1478 message. Instead it uses the arrival time. This makes for better
1479 behaviour in cases where some deliveries succeed, thus re-setting the
1480 "first failed" field. An example is a quota failure for a huge message
1481 when small messages continue to be delivered. Without this change, the
1482 "time since failure" will always be short, possible causing more frequent
1483 delivery attempts for the huge message than are intended.
1484 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1486 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1487 $1, $2, $3) because the numerical variables can be reset during some
1488 expansion items (e.g. "match"), thereby losing the authentication data.
1490 PH/21 Make -bV show the size of off_t variables so that the test suite can
1491 decide whether to run tests for quotas > 2G.
1493 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1494 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1495 filecount value is greater than 2G or if a quota value is greater than 2G
1496 on a system where the size of off_t is not greater than 4, a panic error
1499 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1500 never match. The debug and -bh output now contains an explicit error
1501 message indicating a malformed IPv4 address or mask.
1503 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1504 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1505 PH/23 above applies.
1507 PH/25 Do not write to syslog when running in the test harness. The only
1508 occasion when this arises is a failure to open the main or panic logs
1509 (for which there is an explicit test).
1511 PH/26 Added the /no_tell option to "control=freeze".
1513 PH/27 If a host name lookup failed very early in a connection, for example, if
1514 the IP address matched host_lookup and the reverse lookup yielded a name
1515 that did not have a forward lookup, an error message of the form "no IP
1516 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1517 could be logged. Now it outputs the IP address instead of "NULL".
1519 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1520 allows the sender and the authenticated sender to be set when
1521 submitting a message from within Exim. Since child_open_exim() is
1522 documented for local_scan(), the new function should be too.
1524 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1525 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1526 results in an empty string is now treated as unset.
1528 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1530 PH/31 Added sender_verify_fail logging option.
1532 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1533 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1534 tidied the source and removed it altogether.
1536 PH/33 When a queue run was abandoned because the load average was too high, a
1537 log line was always written; now it is written only if the queue_run log
1538 selector is set. In addition, the log line for abandonment now contains
1539 information about the queue run such as the pid. This is always present
1540 in "start" and "stop" lines but was omitted from the "abandon" line.
1542 PH/34 Omit spaces between a header name and the colon in the error message that
1543 is given when verify = headers_syntax fails (if there are lots of them,
1544 the message gets confusing).
1546 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1547 names, as there are now some PTR records that contain slashes. This check
1548 is only to protect against broken name servers that fall over on strange
1549 characters, so the fact that it applies to all lookups doesn't matter.
1551 PH/36 Now that the new test suite is complete, we can remove some of the
1552 special code in Exim that was needed for the old test suite. For example,
1553 sorting DNS records because real resolvers return them in an arbitrary
1554 order. The new test suite's fake resolver always returns records in the
1557 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1558 bounces) except when queue_only is set, to avoid logging races between
1559 the different processes.
1561 PH/38 Panic-die if .include specifies a non-absolute path.
1563 PH/39 A tweak to the "H" retry rule from its user.
1565 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1566 a label. They prevented compilation on older perls.
1568 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1569 a warning to be raised on newish perls.
1571 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1572 on queue. Changes to match documented behaviour of showing count of
1573 messages matching specified criteria.
1575 PH/40 Changed the default ident timeout from 30s to 5s.
1577 PH/41 Added support for the use of login_cap features, on those BSD systems
1578 that have them, for controlling the resources used by pipe deliveries.
1580 PH/42 The content-scanning code uses fopen() to create files in which to put
1581 message data. Previously it was not paying any attention to the mode of
1582 the files. Exim runs with umask(0) because the rest of the code creates
1583 files with open(), and sets the required mode explicitly. Thus, these
1584 files were ending up world-writeable. This was not a big issue, because,
1585 being within the spool directory, they were not world-accessible. I have
1586 created a function called modefopen, which takes an additional mode
1587 argument. It sets umask(777), creates the file, chmods it to the required
1588 mode, then resets the umask. All the relevant calls to fopen() in the
1589 content scanning code have been changed to use this function.
1591 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1592 to 24 hours. This avoids potential overflow problems when processing G
1593 and H retry rules. I suspect nobody ever tinkers with this value.
1595 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1597 PH/45 When the plaintext authenticator is running as a client, the server's
1598 challenges are checked to ensure they are valid base64 strings. By
1599 default, the authentication attempt is cancelled if an invalid string is
1600 received. Setting client_ignore_invalid_base64 true ignores these errors.
1601 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1602 they are received. Thus, the responses can be made to depend on the
1603 challenges. If an invalid string is ignored, an empty string is placed in
1606 PH/46 Messages that are created by the autoreply transport now contains a
1607 References: header, in accordance with RFCs 2822 and 3834.
1609 PH/47 Added authenticated_sender_force to the smtp transport.
1611 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1613 PH/49 Installed latest patch from the Sieve maintainer.
1615 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1616 was also set, the appendfile transport was unnecessarily scanning a
1617 directory of message files (e.g. for maildir delivery) to find the count
1618 of files (along with the size), even though it did not need this
1619 information. It now does the scan only if it needs to find either the
1620 size of the count of files.
1622 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1624 PH/52 Two bugs concerned with error handling when the smtp transport is
1627 (i) Exim was not creating retry information for temporary errors given
1628 for individual recipients after the DATA command when the smtp transport
1629 was used in LMTP mode. This meant that they could be retried too
1630 frequently, and not timed out correctly.
1632 (ii) Exim was setting the flag that allows error details to be returned
1633 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1634 recipients that were returned after the DATA command.
1636 PH/53 This is related to PH/52, but is more general: for any failing address,
1637 when detailed error information was permitted to be returned to the
1638 sender, but the error was temporary, then after the final timeout, only
1639 "retry timeout exceeded" was returned. Now it returns the full error as
1640 well as "retry timeout exceeded".
1642 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1643 do this, and (what is worse) MTAs that accept it.
1645 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1646 will now be deprecated.
1648 PH/56 New os.c-cygwin from the Cygwin maintainer.
1650 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1651 formats (previously only available in exim formats via -bpr, -bpru,
1652 and -bpra. Now also available in native and exiqgrep formats)
1654 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1655 with very large, slow to parse queues
1657 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1659 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1661 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1662 responses to authentication challenges, though it was showing the
1663 challenges; (ii) I've removed the CR characters from the debug output for
1666 PH/58 Allow for the insertion of a newline as well as a space when a string
1667 is turned into more than one encoded-word during RFC 2047 encoding. The
1668 Sieve code now uses this.
1670 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1671 data_4xx, lost_connection, tls_required.
1673 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1674 message was being sent as an SMTP response.
1676 PH/61 Add -l and -k options to exicyclog.
1678 PH/62 When verifying, if an address was redirected to one new address, so that
1679 verification continued, and the new address failed or deferred after
1680 having set something in $address_data, the value of $address_data was not
1681 passed back to the ACL. This was different to the case when no
1682 redirection occurred. The value is now passed back in both cases.
1684 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1685 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1686 use login_cap.h, so on its own it isn't the distinguishing feature. The
1687 new name refers directly to the setclassresources() function.
1689 PH/65 Added configuration files for NetBSD3.
1691 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1693 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1694 is preferred over IPv4.
1696 PH/68 The bounce_return_message and bounce_return_body options were not being
1697 honoured for bounces generated during the reception of non-SMTP messages.
1698 In particular, this applied to messages rejected by the ACL. This bug has
1699 been fixed. However, if bounce_return_message is true and bounce_return_
1700 body is false, the headers that are returned for a non-SMTP message
1701 include only those that have been read before the error was detected.
1702 (In the case of an ACL rejection, they have all been read.)
1704 PH/69 The HTML version of the specification is now built in a directory called
1705 spec_html instead of spec.html, because the latter looks like a path with
1706 a MIME-type, and this confuses some software.
1708 PH/70 Catch two compiler warnings in sieve.c.
1710 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1711 function verify_get_ident() calls ip_connect() to connect a socket, but
1712 if the "connect()" function timed out, ip_connect() used to close the
1713 socket. However, verify_get_ident() also closes the socket later, and in
1714 between Exim writes to the log, which may get opened at this point. When
1715 the socket was closed in ip_connect(), the log could get the same file
1716 descriptor number as the socket. This naturally causes chaos. The fix is
1717 not to close the socket in ip_connect(); the socket should be closed by
1718 the function that creates it. There was only one place in the code where
1719 this was missing, in the iplookup router, which I don't think anybody now
1720 uses, but I've fixed it anyway.
1722 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1723 well as to direct DNS lookups. Otherwise the handling of names in host
1724 lists is inconsistent and therefore confusing.
1730 PH/01 Two changes to the default runtime configuration:
1732 (1) Move the checks for relay_from_hosts and authenticated clients from
1733 after to before the (commented out) DNS black list checks.
1735 (2) Add control=submission to the relay_from_hosts and authenticated
1736 clients checks, on the grounds that messages accepted by these
1737 statements are most likely to be submissions.
1739 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1741 (1) Generate an error if the third argument for the ${prvs expansion is
1744 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1747 (3) Reset the variables that are obtained from the first argument of
1748 ${prvscheck and used in the second argument before leaving the code,
1749 because their memory is reclaimed, so using them afterwards may do
1752 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1753 one (it's much easier than Tom thought :-).
1755 (5) Because of (4), we can now allow for the use of $prvscheck_result
1756 inside the third argument.
1758 PH/03 For some reason, the default setting of PATH when running a command from
1759 a pipe transport was just "/usr/bin". I have changed it to
1762 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1763 anything to be listed in the output from -bV.
1765 PH/05 When a filter generated an autoreply, the entire To: header line was
1766 quoted in the delivery log line, like this:
1768 => >A.N.Other <ano@some.domain> <original@ddress> ...
1770 This has been changed so that it extracts the operative address. There
1771 may be more than one such address. If so, they are comma-separated, like
1774 => >ano@some.domain,ona@other.domain <original@ddress> ...
1776 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1777 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1778 not being looked up in the rDNS to get a host name, Exim was showing the
1779 IP address twice in Received: lines, even though the IP addresses were
1780 identical. For example:
1782 Received: from [1.2.3.4] (helo=[1.2.3.4])
1784 However, if the real host name was known, it was omitting the HELO data
1785 if it matched the actual IP address. This has been tidied up so that it
1786 doesn't show the same IP address twice.
1788 PH/07 When both +timestamp and +memory debugging was on, the value given by
1789 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1790 called by the debug printing, thereby overwriting the timestamp buffer.
1791 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1793 PH/08 When the original message was included in an autoreply transport, it
1794 always said "this is a copy of the message, including all the headers",
1795 even if body_only or headers_only was set. It now gives an appropriate
1798 PH/09 Applied a patch from the Sieve maintainer which:
1800 o fixes some comments
1801 o adds the (disabled) notify extension core
1802 o adds some debug output for the result of if/elsif tests
1803 o points to the current vacation draft in the documentation
1804 and documents the missing references header update
1808 o fixes a bug in processing the envelope test (when testing
1809 multiple envelope elements, the last element determinted the
1812 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1813 Electronic Mail") by including:
1815 Auto-submitted: auto-generated
1817 in the messages that it generates (bounce messages and others, such as
1818 warnings). In the case of bounce messages for non-SMTP mesages, there was
1819 also a typo: it was using "Auto_submitted" (underscore instead of
1820 hyphen). Since every message generated by Exim is necessarily in response
1821 to another message, thes have all been changed to:
1823 Auto-Submitted: auto-replied
1825 in accordance with these statements in the RFC:
1827 The auto-replied keyword:
1829 - SHOULD be used on messages sent in direct response to another
1830 message by an automatic process,
1832 - MUST NOT be used on manually-generated messages,
1834 - MAY be used on Delivery Status Notifications (DSNs) and Message
1835 Disposition Notifications (MDNs),
1837 - MUST NOT be used on messages generated by automatic or periodic
1838 processes, except for messages which are automatic responses to
1841 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1842 to the default Received: header definition.
1844 PH/12 Added log selector acl_warn_skipped (default on).
1846 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1847 variables because (a) they are in the wrong storage pool and (b) even if
1848 they were copied, it wouldn't work properly because of the caching.
1850 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1851 checking when decoding. Apparently there are clients that generate
1852 overlong encoded strings. Why am I not surprised?
1854 PH/15 If the first argument of "${if match_address" was not empty, but did not
1855 contain an "@" character, Exim crashed. Now it writes a panic log message
1856 and treats the condition as false.
1858 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1860 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1861 extension "envelope-auth". The code is finished and in agreement with
1862 other implementations, but there is no documentation so far and in fact,
1863 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1864 not changing the active code.
1866 Print executed "if" and "elsif" statements when debugging is used. This
1867 helps a great deal to understand what a filter does.
1869 Document more things not specified clearly in RFC3028. I had all this
1870 sorted out, when out of a sudden new issues came to my mind. Oops."
1872 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1875 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1876 canonical form (as documented). However, after a host name lookup from
1877 the IP address, check_host() was doing a simple string comparison with
1878 addresses acquired from the DNS when checking that the found name did
1879 have the original IP as one of its addresses. Since any found IPv6
1880 addresses are likely to be in abbreviated form, the comparison could
1881 fail. Luckily, there already exists a function for doing the comparison
1882 by converting both addresses to binary, so now that is used instead of
1883 the text comparison.
1885 PH/20 There was another similar case to PH/19, when a complete host name was
1886 given in a host list; looking up its IP address could give an abbreviated
1887 form, whereas the current host's name might or might not be abbreviated.
1888 The same fix has been applied.
1894 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1895 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1898 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1899 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1901 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1903 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1904 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1905 Also, for "Auto-Submitted", treat anything other than "no" as
1906 non-personal, in accordance with RFC 3834. (Previously it treated
1907 anything starting "auto-" as non-personal.)
1909 TF/01 The control=submission/name=... option had a problem with syntax
1910 errors if the name included a slash character. The /name= option
1911 now slurps the rest of the string, so it can include any characters
1912 but it must come last in the list of options (after /sender_retain
1915 PH/05 Some modifications to the interface to the fake nameserver for the new
1923 TK/01 Added the "success_on_redirect" address verification option. See
1924 NewStuff for rationale and an example.
1926 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1928 PH/02 Patch to exigrep to allow it to work on syslog lines.
1930 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1931 fread() to skip over the body file's header line, because in Cygwin the
1932 header line is locked and is inaccessible.
1934 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1935 co-exist for some time) to make it clear that it is the Exim ID that is
1936 referenced, not the Message-ID: header line.
1938 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1939 string_format() function, because snprintf() does not exist on all
1942 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1943 ${stat: expansion item.
1945 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1946 protocol synchronization error", to keep the pedants happy.
1948 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1949 well as for IRIX systems, when gcc is being used. See the host.c source
1952 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1954 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1957 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1958 transport and to the smtp transport in LMTP mode.
1960 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1962 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1963 run a filter in a subprocess. This could lead to confusion in subsequent
1964 lookups in the parent process. There should also be a search_tidyup() at
1965 the end of the subprocess.
1967 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1968 only if the host matched helo_try_verify_hosts, which caused the
1969 verification to occur when the EHLO/HELO command was issued. The ACL just
1970 tested the remembered result. Now, if a previous verification attempt has
1971 not happened, "verify = helo" does it there and then.
1973 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1975 TK/03 Fix log output including CR from clamd.
1977 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1978 exist provoked a memory error which could cause a segfault.
1980 PH/15 Installed PCRE 6.2
1982 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1984 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1985 of the problem. Specifically, suggested +O2 rather than +O1 for the
1988 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1990 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1991 record, the port was not correctly logged when the outgoing_port log
1992 selector was set (it logged the transort's default port).
1994 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1995 fallback_hosts, and "hosts" in the smtp transport.
1997 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1998 host errors such as "Connection refused".
2000 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2001 authentication with radiusclient 0.4.9:
2003 - Error returned from rc_read_config was caught wrongly
2004 - Username/password not passed on to radius server due to wrong length.
2006 The presumption is that some radiusclient API changes for 4.51/PH/17
2007 were not taken care of correctly. The code is still untested by me (my
2008 Linux distribution still has 0.3.2 of radiusclient), but it was
2009 contributed by a Radius user.
2011 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2012 expanding the "port" option of the smtp transport.
2014 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2015 while reading a MIME header. Thanks to Tom Hughes for a patch.
2017 PH/24 Include config.h inside local_scan.h so that configuration settings are
2020 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2021 that in an ACL for RCPT (for example), you can examine exactly what was
2024 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2025 commands, but it was not correctly comparing the address with the actual
2026 client host address. Thus, it would show the EHLO address in Received:
2027 header lines when this was not necessary.
2029 PH/27 Added the % operator to ${eval:}.
2031 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2032 it should be ignoring failures (because with -C, for example, it has lost
2033 privilege). It wasn't ignoring creation failures other than "already
2036 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2039 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2040 because an input line was too long, either on its own, or by virtue of
2041 too many continuations, the temporary file was not being removed, and the
2042 return code was incorrect.
2044 PH/31 Missing "BOOL" in function definition in filtertest.c.
2046 PH/32 Applied Sieve patches from the maintainer.
2048 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2050 PH/33 Added "verify = not_blind".
2052 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2053 Local/Makefile (with some defaults set). These are used in built scripts
2054 such as exicyclog, but they have never been used in the exim_install
2055 script (though there are many overriding facilities there). I have
2056 arranged that the exim_install script now takes note of these two
2059 PH/35 Installed configuration files for Dragonfly.
2061 PH/36 When a locally submitted message by a trusted user did not contain a
2062 From: header, and the sender address was obtained from -f or from an SMTP
2063 MAIL command, and the trusted user did not use -F to supply a sender
2064 name, $originator_name was incorrectly used when constructing a From:
2065 header. Furthermore, $originator_name was used for submission mode
2066 messages from external hosts without From: headers in a similar way,
2067 which is clearly wrong.
2069 PH/37 Added control=suppress_local_fixups.
2071 PH/38 When log_selector = +received_sender was set, and the addition of the
2072 sender made the log line's construction buffer exactly full, or one byte
2073 less than full, an overflow happened when the terminating "\n" was
2076 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2077 when the result of a list match is failure because a DNS lookup failed.
2079 PH/40 RM_COMMAND is now used in the building process.
2081 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2082 the "build-* directories that it finds.
2084 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2085 address in a domain literal was a prefix of an interface address.
2087 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2088 when verifying a sender address, unless rewrite_headers is false.
2090 PH/44 Wrote a long comment about why errors_to addresses are verified as
2091 recipients, not senders.
2093 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2094 the ratelimit ACL was added.
2096 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2098 PH/46 Added extra information about PostgreSQL errors to the error string.
2100 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2101 avoiding the need to install special zones in a real server. This is
2102 backwards compatible; if it can't find the fake resolver, it drops back.
2103 Thus, both old and new test suites can be run.
2105 TF/02 Added util/ratelimit.pl
2107 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2108 clock is set back in time.
2110 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2111 Candler <B.Candler@pobox.com>.
2113 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2114 OK for addresses that are the result of redirections.
2116 PH/48 A number of further additions for the benefit of the new test suite,
2117 including a fake gethostbyname() that interfaces to the fake DNS resolver
2120 TF/06 The fix for widen_domains has also been applied to qualify_single and
2121 search_parents which are the other dnslookup options that can cause
2124 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2127 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2129 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2130 service" if it encountered an MX record with an empty target hostname.
2131 The message is now "an MX or SRV record indicated no SMTP service".
2133 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2134 if the DNS of the sending site is misconfigured. This is quite a
2135 common situation. This change restores the behaviour of treating a
2136 helo verification defer as a failure.
2138 PH/51 If self=fail was set on a router, the bounce message did not include the
2139 actual error message.
2145 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2147 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2148 command itself ended in error, the underlying message about the transport
2149 filter timeout was being overwritten with the pipe command error. Now the
2150 underlying error message should be appended to the second error message.
2152 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2154 PH/02 Reduce the amount of output that "make" produces by default. Full output
2155 can still be requested.
2157 PH/03 The warning log line about a condition test deferring for a "warn" verb
2158 was being output only once per connection, rather than after each
2159 occurrence (because it was using the same function as for successful
2160 "warn" verbs). This seems wrong, so I have changed it.
2162 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2163 it should not have, which might have caused a crash in the right
2164 circumstances, but probably never did.
2166 PH/04 Installed a modified version of Tony Finch's patch to make submission
2167 mode fix the return path as well as the Sender: header line, and to
2168 add a /name= option so that you can make the user's friendly name appear
2171 TF/03 Added the control = fakedefer ACL modifier.
2173 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2174 Mark Lowes for thorough testing.
2176 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2178 TK/03 Merged latest SRS patch from Miles Wilton.
2180 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2181 with the definition in sysexits.h (which is #included earlier).
2182 Fortunately, Exim does not actually use EX_OK. The code used to try to
2183 preserve the sysexits.h value, by assumimg that macro definitions were
2184 scanned for macro replacements. I have been disabused of this notion,
2185 so now the code just undefines EX_OK before #including unistd.h.
2187 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2188 in the smtp transport. When a block could not be written in a single
2189 write() function, the timeout was being re-applied to each part-write.
2190 This seems wrong - if the receiver was accepting one byte at a time it
2191 would take for ever. The timeout is now adjusted when this happens. It
2192 doesn't have to be particularly precise.
2194 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2195 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2197 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2198 without a domain if the check to <postmaster@domain> fails.
2200 SC/01 Eximstats: added -xls and the ability to specify output files
2201 (patch written by Frank Heydlauf).
2203 SC/02 Eximstats: use FileHandles for outputing results.
2205 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2207 SC/04 Eximstats: fixed display of large numbers with -nvr option
2209 SC/05 Eximstats: fixed merging of reports with empty tables.
2211 SC/06 Eximstats: added the -include_original_destination flag
2213 SC/07 Eximstats: removed tabs and trailing whitespace.
2215 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2217 TK/06 MBOX spool code: Add real "From " MBOX separator line
2218 so the .eml file is really in mbox format (even though
2219 most programs do not really care). Patch from Alex Miller.
2221 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2222 The latter is generated from $received_to and is only set if the
2223 message has one envelope recipient. SA can use these headers,
2224 obviously out-of-the-box. Patch from Alex Miller.
2226 PH/08 The ${def test on a variable was returning false if the variable's
2227 value was "0", contrary to what the specification has always said!
2228 The result should be true unless the variable is empty.
2230 PH/09 The syntax error of a character other than { following "${if
2231 def:variable_name" (after optional whitespace) was not being diagnosed.
2232 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2233 accidental colon was present, for example, could give incorrect results.
2235 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2236 result is used (not including appendfile, where other changes are about
2239 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2240 This involved changing a lot of size variables from int to off_t. It
2241 should work with maildirs and everything.
2243 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2244 spamd dying while we are connected to it.
2246 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2249 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2250 function for BDB 4.3.
2252 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2254 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2255 the sources for PCRE have been re-arranged and more files are now
2258 PH/15 The code I had for printing potentially long long variables in PH/11
2259 above was not the best (it lost precision). The length of off_t variables
2260 is now inspected at build time, and an appropriate printing format (%ld
2261 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2262 to be "long long int" or "long int". This is needed for the internal
2263 formatting function string_vformat().
2265 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2266 the configuration file to be ":syslog", then the script "guesses" where
2267 the logs files are, rather than using the compiled in default. In our
2268 case the guess is not the same as the compiled default, so the script
2269 suddenly stopped working when I started to use syslog. The patch checks
2270 to see if log_file_path is "". If so, it attempts to read it from exim
2271 with no configuration file to get the compiled in version, before it
2272 falls back to the previous guessing code."
2274 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2275 implementing BATV in an Exim configuration. See NewStuff for the gory
2278 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2279 Makefile that are specific to HP-UX.
2281 PH/18 If the "use_postmaster" option was set for a recipient callout together
2282 with the "random" option, the postmaster address was used as the MAIL
2283 FROM address for the random test, but not for the subsequent recipient
2284 test. It is now used for both.
2286 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2287 patch removes a few documentation additions to RFC 3028, because the
2288 latest draft now contains them. It adds the new en;ascii-case comparator
2289 and a new error check for 8bit text in MIME parts. Comparator and
2290 require names are now matched exactly. I enabled the subaddress
2291 extension, but it is not well tested yet (read: it works for me)."
2293 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2294 rework some of the code of TK/09 above to avoid the hardwired use of
2295 "%lld" and "long long". Replaced the call to snprintf() with a call to
2298 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2299 records point to non-existent hosts", "retry timeout exceeded", and
2300 "retry time not reached for any host after a long failure period".
2302 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2303 experimental DomainKeys support:
2305 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2306 (2) On an error such as an illegally used "control", the wrong name for
2307 the control was given.
2309 These problems did NOT occur unless DomainKeys support was compiled.
2311 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2313 PH/24 Added ${if match_ip condition.
2315 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2316 databases so that it will be absolutely obvious if a crash occurs in the
2317 DB library. This is a regular occurrence (often caused by mis-matched
2320 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2321 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2322 picked up on a user's system that detects such things. There doesn't seem
2323 to be a gcc warning option for this - only an attribute that has to be
2324 put on the function's prototype. It seems that in Fedora Core 4 they have
2325 set this on a number of new functions. No doubt there will be more in due
2328 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2329 specify a transport. However, if an address that was verified by such a
2330 router was the subject of a callout, Exim crashed because it tried to
2331 read the rcpt_include_affixes from the non-existent transport. Now it
2332 just assumes that the setting of that option is false. This bug was
2333 introduced by 4.51/PH/31.
2335 PH/28 Changed -d+all to exclude +memory, because that information is very
2336 rarely of interest, but it makes the output a lot bigger. People tend to
2337 do -d+all out of habit.
2339 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2340 code in os-type was giving problems when libc.so lives in lib64, like on
2343 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2344 aren't the modern standard, and it seems that some systems' include files
2345 don't always have them. Exim was already checking for some of the newer
2346 ones like T_AAAA, and defining it itself. I've added checks for all the
2347 record types that Exim uses.
2349 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2350 not automatically generating a new one, as it is supposed to. This
2351 prevented TLS from working. If the file did exist, but contained invalid
2352 data, a new version was generated, as expected. It was only the case of a
2353 non-existent file that was broken.
2355 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2356 with a change in libdomainkeys > 0.64.
2358 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2359 from DNS. If the selector record carries the flag, it now has
2360 precedence over the domain-wide flag.
2362 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2364 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2365 the use of a port name that isn't defined in /etc/services) occurred, the
2366 message was deferred as in a normal delivery, and thus remained on the
2367 spool, instead of being failed because of the mua_wrapper setting. This
2368 is now fixed, and I tidied up some of the mua_wrapper messages at the
2371 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2372 the messages in a hash of arrays rather than using individual hashes.
2373 This is a bit cleaner and results in dramatic memory savings, albeit
2374 at a slight CPU cost.
2376 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2377 as requested by Marc Sherman.
2379 SC/10 Eximstats: added histograms for user specified patterns as requested
2382 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2384 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2385 fopen() in the content-scanning modules that did not already have it.
2391 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2392 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2394 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2396 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2398 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2399 to test Sieve filters that use "vacation".
2401 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2402 that changes the way the GnuTLS parameters are stored in the cache file.
2403 The new format can be generated externally. For backward compatibility,
2404 if the data in the cache doesn't make sense, Exim assumes it has read an
2405 old-format file, and it generates new data and writes a new file. This
2406 means that you can't go back to an older release without removing the
2409 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2410 work if there are any delivery delays because "one_time" forces the
2411 parent to be marked "delivered", so its unseen clone is never tried
2412 again. For this reason, Exim now forbids the simultaneous setting of
2415 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2416 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2417 entry if you want to know the details. The fix, however, neglected to
2418 consider the case when local delivery batching is involved. The test for
2419 "previously delivered" was not happening when checking to see if an
2420 address could be batched with a previous (undelivered) one; under
2421 certain circumstances this could lead to multiple deliveries to the same
2424 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2425 in its include files, and this causes problems building Exim.
2427 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2428 header_syntax) but Exim was just ignoring anything given after a slash.
2429 In particular, this caused confusion with an attempt to use "verify =
2430 reverse_host_lookup/defer_ok". An error is now given when options are
2431 supplied for verify items that do not have them. (Maybe reverse_host_
2432 lookup should have a defer_ok option, but that's a different point.)
2434 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2435 defined by RFC 821) to 2048, because there were problems with some AUTH
2436 commands, and RFC 1869 says the size should be increased for extended
2437 SMTP commands that take arguments.
2439 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2442 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2443 "unknown" error; now it says that the functionality isn't in the binary.
2445 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2446 an address' error message when a string expansion fails (syntax or
2447 whatever). Otherwise the password may appear in the log. Following change
2448 PH/42 below, there is no longer a chance of it appearing in a bounce
2451 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2453 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2454 of its addresses was ever tried. (Bugzilla bug #2.)
2456 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2457 the result incorrectly in the debug output. (It correctly added a newline
2458 to what was transported.)
2460 TF/01 Added $received_time.
2462 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2463 commented out examples of how to interface to a virus scanner and to
2464 SpamAssassin. Also added commented examples of av_scanner and
2465 spamd_address settings.
2467 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2468 and controls are allowed in which ACLs. There were a couple of minor
2469 errors. Some of the entries in the conditions table (which is a table of
2470 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2471 negation of where the condition IS allowed.
2473 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2475 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2476 header file does not have a version number, so I've had to invent a new
2477 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2478 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2479 radiusclient), but it was contributed by a Radius user.
2481 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2482 files or directories, for OpenSSL.
2484 PH/19 When an Exim process that is running as root has to create an Exim log
2485 file, it does so in a subprocess that runs as exim:exim so as to get the
2486 ownership right at creation (otherwise, other Exim processes might see
2487 the file with the wrong ownership). There was no test for failure of this
2488 fork() call, which would lead to the process getting stuck as it waited
2489 for a non-existent subprocess. Forks do occasionally fail when resources
2490 run out. I reviewed all the other calls to fork(); they all seem to check
2493 PH/20 When checking for unexpected SMTP input at connect time (before writing
2494 the banner), Exim was not dealing correctly with a non-positive return
2495 from the read() function. If the client had disconnected by this time,
2496 the result was a log entry for a synchronization error with an empty
2497 string after "input=" when read() returned zero. If read() returned -1
2498 (an event I could not check), uninitialized data bytes were printed.
2499 There were reports of junk text (parts of files, etc) appearing after
2502 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2504 PH/22 Added support for macro redefinition, and (re)definition in between
2505 driver and ACL definitions.
2507 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2508 forgetting to use the resulting value; it was using the unexpanded value.
2510 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2511 hadn't been configured. The fix is from Juergen Kreileder, who
2512 understands it better than I do:
2514 "Here's what I see happening with three configured cyrus_sasl
2515 authenticators configured (plain, login, cram-md5):
2517 On startup auth_cyrus_sasl_init() gets called for each of these.
2518 This means three calls to sasl_listmech() without a specified mech_list.
2519 => SASL tests which mechs of all available mechs actually work
2520 => three warnings about OTP not working
2521 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2523 With the patch, sasl_listmech() also gets called three times. But now
2524 SASL's mech_list option is set to the server_mech specified in the the
2525 authenticator. Or in other words, the answer from sasl_listmech()
2526 gets limited to just the mech you're testing for (which is different
2528 => the return list contains just 'plain' or 'login', 'cram-md5' or
2529 nothing depending on the value of ob->server_mech.
2531 I've just tested the patch: Authentication still works fine,
2532 unavailable mechs specified in the exim configuration are still
2533 caught, and the auth.log warnings about OTP are gone."
2535 PH/25 When debugging is enabled, the contents of the command line are added
2536 to the debugging output, even when log_selector=+arguments is not
2539 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2540 answer is "GNU", and only if the return is "GNU/something" is the answer
2543 PH/27 $acl_verify_message is now set immediately after the failure of a
2544 verification in an ACL, and so is available in subsequent modifiers. In
2545 particular, the message can be preserved by coding like this:
2547 warn !verify = sender
2548 set acl_m0 = $acl_verify_message
2550 Previously, $acl_verify_message was set only while expanding "message"
2551 and "log_message" when a very denied access.
2553 PH/28 Modified OS/os.c-Linux with
2555 -#ifndef OS_LOAD_AVERAGE
2556 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2558 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2559 nomenclature these days.)
2561 PH/29 Installed patch from the Sieve maintainer that adds the options
2562 sieve_useraddress and sieve_subaddress to the redirect router.
2564 PH/30 In these circumstances:
2565 . Two addresses routed to the same list of hosts;
2566 . First host does not offer TLS;
2567 . First host accepts first address;
2568 . First host gives temporary error to second address;
2569 . Second host offers TLS and a TLS session is established;
2570 . Second host accepts second address.
2571 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2572 and peerdn, if requested) that were in fact used only for the second
2575 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2576 attention to any local part prefix or suffix that was matched by the
2577 router that accepted the address. It now behaves in the same way as it
2578 does for delivery: the affixes are removed from the local part unless
2579 rcpt_include_affixes is set on the transport.
2581 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2582 timeout during the DATA phase of an incoming message.
2584 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2585 applied a patch sanctioned by the Sieve maintainer.
2587 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2588 the uid or gid is negative. A case of a negative gid caused this to be
2589 noticed. The fix allows for either to be negative.
2591 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2592 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2595 PH/36 The change PH/12 above was broken. Fixed it.
2597 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2598 the grounds that routing the same address twice would always produce the
2599 same answer. This might have been true once, but it is certainly no
2600 longer true now. Routing a child address may depend on the previous
2601 routing that produced that child. Some complicated redirection strategies
2602 went wrong when messages had multiple recipients, and made Exim's
2603 behaviour dependent on the order in which the addresses were given.
2605 I have moved the duplicate checking until after the routing is complete.
2606 Exim scans the addresses that are assigned to local and remote
2607 transports, and removes any duplicates. This means that more work will be
2608 done, as duplicates will always all be routed, but duplicates are
2609 presumably rare, so I don't expect this is of any significance.
2611 For deliveries to pipes, files, and autoreplies, the duplicate checking
2612 still happens during the routing process, since they are not going to be
2615 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2616 It corrects a timeout issue with spamd. This is Ian's comment: "The
2617 background is that sometimes spamd either never reads data from a
2618 connection it has accepted, or it never writes response data. The exiscan
2619 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2620 blindly assumes that writes won't block so it may never time out."
2622 PH/39 Allow G after quota size as well as K and M.
2624 PH/40 The value set for $authenticated_id in an authenticator may not contain
2625 binary zeroes or newlines because the value is written to log lines and
2626 to spool files. There was no check on this. Now the value is run through
2627 the string_printing() function so that such characters are converted to
2628 printable escape sequences.
2630 PH/41 $message_linecount is a new variable that contains the total number of
2631 lines in the message. Compare $body_linecount, which is the count for the
2634 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2635 bounce and delay warning messages, except in certain special cases, which
2638 (a) An SMTP error message from a remote host;
2639 (b) A message specified in a :fail: redirection;
2640 (c) A message specified in a "fail" command in a system filter;
2641 (d) A message specified in a FAIL return from the queryprogram router;
2642 (e) A message specified by the cannot_route_message router option.
2644 In these cases only, Exim does include the error details in bounce and
2645 warning messages. There are also a few cases where bland messages such
2646 as "unrouteable address" or "local delivery error" are given.
2648 PH/43 $value is now also set for the "else" part of a ${run expansion.
2650 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2651 being worked on, but at least Exim now implements the latest version to
2654 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2655 process to complete was treated as a delivery failure, a timeout while
2656 writing the message to the pipe was logged, but erroneously treated as a
2657 successful delivery. Such timeouts include transport filter timeouts. For
2658 consistency with the overall process timeout, these timeouts are now
2659 treated as errors, giving rise to delivery failures by default. However,
2660 there is now a new Boolean option for the pipe transport called
2661 timeout_defer, which, if set TRUE, converts the failures into defers for
2662 both kinds of timeout. A transport filter timeout is now identified in
2665 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2666 systems where "make" and "gmake" are different, calling "gmake" at top
2667 level broke things. I've arranged for the value of $(MAKE) to be passed
2668 from the Makefile to this script so that it can call the same version of
2672 A note about Exim versions 4.44 and 4.50
2673 ----------------------------------------
2675 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2676 changes of various kinds. As a consequence, a big documentation update was
2677 needed. This delayed the release for rather longer than seemed good, especially
2678 in the light of a couple of (minor) security issues. Therefore, the changes
2679 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2680 release. So 4.44 and 4.50 are in effect two different branches that both start
2683 I have left the 4.50 change log unchanged; it contains all the changes since
2684 4.43. The change log for 4.44 is below; many of its items are identical to
2685 those for 4.50. This seems to be the most sensible way to preserve the
2686 historical information.
2692 1. Minor wording change to the doc/README.SIEVE file.
2694 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2695 computation of the current number of files was incorrect.
2697 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2698 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2701 4. Give more explanation in the error message when the command for a transport
2702 filter fails to execute.
2704 5. There are several places where Exim runs a non-Exim command in a
2705 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2706 was being done only for the command run by the queryprogram router. It is
2707 now done for all such subprocesses. The other cases are: ${run, transport
2708 filters, and the commands run by the lmtp and pipe transports.
2710 6. Added CONFIGURE_GROUP build-time option.
2712 7. Some older OS have a limit of 256 on the maximum number of file
2713 descriptors. Exim was using setrlimit() to set 1000 as a large value
2714 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2715 systems. I've change it so that if it can't get 1000, it tries for 256.
2717 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2718 was an oversight, and furthermore, ever since the addition of extra
2719 controls (e.g. 4.43/32), the checks on when to allow different forms of
2720 "control" were broken. There should now be diagnostics for all cases when a
2721 control that does not make sense is encountered.
2723 9. Added the /retain_sender option to "control=submission".
2725 10. $recipients is now available in the predata ACL (oversight).
2727 11. Tidy the search cache before the fork to do a delivery from a message
2728 received from the command line. Otherwise the child will trigger a lookup
2729 failure and thereby defer the delivery if it tries to use (for example) a
2730 cached ldap connection that the parent has called unbind on.
2732 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2733 of $address_data from the recipient verification was clobbered by the
2734 sender verification.
2736 13. The value of address_data from a sender verification is now available in
2737 $sender_address_data in subsequent conditions in the ACL statement.
2739 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2741 15. Added a new option "connect=<time>" to callout options, to set a different
2744 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2745 was its contents. (It was OK if the option was not defined at all.)
2747 17. A "Completed" log line is now written for messages that are removed from
2748 the spool by the -Mrm option.
2750 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2751 information about exactly what failed.
2753 19. Added -dd to debug only the daemon process.
2755 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2756 handles timeouts, both on the server side and network timeouts. Renamed the
2757 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2759 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2760 It is now set to "smtps".
2762 22. $host_address is now set to the target address during the checking of
2763 ignore_target_hosts.
2765 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2766 being passed; this would have caused $sender_host_name to have been used if
2767 matching the list had actually called for a host name (not very likely,
2768 since this list is usually IP addresses). A host name is now passed as
2771 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2772 code that specifies a non-restarting handler (typically sigaction() in
2773 modern systems) in an attempt to fix a rare and obscure crash bug.
2775 25. Narrowed the window for a race in the daemon that could cause it to ignore
2776 SIGCHLD signals. This is not a major problem, because they are used only to
2777 wake it up if nothing else does.
2779 26. A malformed maildirsize file could cause Exim to calculate negative values
2780 for the mailbox size or file count. Odd effects could occur as a result.
2781 The maildirsize information is now recalculated if the size or filecount
2784 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2785 support for a long time. Removed HAVE_SYS_VFS_H.
2787 28. Installed the latest version of exipick from John Jetmore.
2789 29. In an address list, if the pattern was not a regular expression, an empty
2790 subject address (from a bounce message) matched only if the pattern was an
2791 empty string. Non-empty patterns were not even tested. This was the wrong
2792 because it is perfectly reasonable to use an empty address as part of a
2793 database query. An empty address is now tested by patterns that are
2794 lookups. However, all the other forms of pattern expect the subject to
2795 contain a local part and a domain, and therefore, for them, an empty
2796 address still always fails if the pattern is not itself empty.
2798 30. Exim went into a mad DNS loop when attempting to do a callout where the
2799 host was specified on an smtp transport, and looking it up yielded more
2800 than one IP address.
2802 31. Re-factored the code for checking spool and log partition space into a
2803 function that finds that data and another that does the check. The former
2804 is then used to implement four new variables: $spool_space, $log_space,
2805 $spool_inodes, and $log_inodes.
2807 32. The RFC2047 encoding function was originally intended for short strings
2808 such as real names; it was not keeping to the 75-character limit for
2809 encoded words that the RFC imposes. It now respects the limit, and
2810 generates multiple encoded words if necessary. To be on the safe side, I
2811 have increased the buffer size for the ${rfc2047: expansion operator from
2814 33. It is now permitted to omit both strings after an "if" condition; if the
2815 condition is true, the result is "true". As before, when the second string
2816 is omitted, a false condition yields an empty string. This makes it less
2817 cumbersome to write custom ACL and router conditions.
2819 34. Failure to deliver a bounce message always caused it to be frozen, even if
2820 there was an errors_to setting on the router. The errors_to setting is now
2823 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2824 canonical form (fully expanded) before being placed in
2825 $sender_host_address.
2827 36. The table in the code that translates DNS record types into text (T_A to
2828 "A" for instance) was missing entries for NS and CNAME. It is just possible
2829 that this could have caused confusion if both these types were looked up
2830 for the same domain, because the text type is used as part of Exim's
2831 per-process caching. But the chance of anyone hitting this buglet seems
2834 37. The dnsdb lookup has been extended in a number of ways.
2836 (1) There is a new type, "zns", which walks up the domain tree until it
2837 finds some nameserver records. It should be used with care.
2839 (2) There is a new type, "mxh", which is like "mx" except that it returns
2840 just the host names, not the priorities.
2842 (3) It is now possible to give a list of domains (or IP addresses) to be
2843 looked up. The behaviour when one of the lookups defers can be
2844 controlled by a keyword.
2846 (4) It is now possible to specify the separator character for use when
2847 multiple records are returned.
2849 38. The dnslists ACL condition has been extended: it is now possible to supply
2850 a list of IP addresses and/or domains to be looked up in a particular DNS
2853 39. Added log_selector=+queue_time_overall.
2855 40. When running the queue in the test harness, wait just a tad after forking a
2856 delivery process, to get repeatability of debugging output.
2858 41. Include certificate and key file names in error message when GnuTLS fails
2859 to set them up, because the GnuTLS error message doesn't include the name
2860 of the failing file when there is a problem reading it.
2862 42. Allow both -bf and -bF in the same test run.
2864 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2866 44. The "Exiscan patch" is now merged into the mainline Exim source.
2868 45. Sometimes the final signoff response after QUIT could fail to get
2869 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2870 < 0 before doing a fflush(). This bug looks as though it goes back to the
2871 introduction of TLS in release 3.20, but "sometimes" must have been rare
2872 because the tests only now provoked it.
2874 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2875 (this can affect the format of dates).
2877 47. exim_tidydb, when checking for the continued existence of a message for
2878 which it has found a message-specific retry record, was not finding
2879 messages that were in split spool directories. Consequently, it was
2880 deleting retry records that should have stayed in existence.
2882 48. Steve fixed some bugs in eximstats.
2884 49. The SPA authentication driver was not abandoning authentication and moving
2885 on to the next authenticator when an expansion was forced to fail,
2886 contradicting the general specification for all authenticators. Instead it
2887 was generating a temporary error. It now behaves as specified.
2889 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2890 (the order specifies the preference for clients). The order is now AES256,
2891 AES128, 3DES, ARCFOUR128.
2893 51. Small patch to Sieve code - explicitly set From: when generating an
2896 52. Exim crashed if a remote delivery caused a very long error message to be
2897 recorded - for instance if somebody sent an entire SpamAssassin report back
2898 as a large number of 550 error lines. This bug was coincidentally fixed by
2899 increasing the size of one of Exim's internal buffers (big_buffer) that
2900 happened as part of the Exiscan merge. However, to be on the safe side, I
2901 have made the code more robust (and fixed the comments that describe what
2904 53. Now that there can be additional text after "Completed" in log lines (if
2905 the queue_time_overall log selector is set), a one-byte patch to exigrep
2906 was needed to allow it to recognize "Completed" as not the last thing in
2909 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2910 patch that reportedly fixes this has been added. I am not expert enough to
2911 create a test for it. This is what the patch creator wrote:
2913 "I found a little strange behaviour of ldap code when working with
2914 Windows 2003 AD Domain, where users was placed in more than one
2915 Organization Units. When I tried to give exim partial DN, the exit code
2916 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2917 But simultaneously result of request was absolutely normal ldap result,
2918 so I produce this patch..."
2920 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2921 REFERENCE, so I have modified the code to exclude the patch when that macro
2924 55. Some experimental protocols are using DNS PTR records for new purposes. The
2925 keys for these records are domain names, not reversed IP addresses. The
2926 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2927 leaves it alone. Component reversal etc. now happens only for IP addresses.
2930 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2932 57. Double the size of the debug message buffer (to 2048) so that more of very
2933 long debug lines gets shown.
2935 58. The exicyclog utility now does better if the number of log files to keep
2936 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2938 59. Two changes related to the smtp_active_hostname option:
2940 (1) $smtp_active_hostname is now available as a variable.
2941 (2) The default for smtp_banner uses $smtp_active_hostname instead
2942 of $primary_hostname.
2944 60. The host_aton() function is supposed to be passed a string that is known
2945 to be a valid IP address. However, in the case of IPv6 addresses, it was
2946 not checking this. This is a hostage to fortune. Exim now panics and dies
2947 if the condition is not met. A case was found where this could be provoked
2948 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2949 components; fortuitously, this particular loophole had already been fixed
2950 by change 4.50/55 above.
2952 If there are any other similar loopholes, the new check in host_aton()
2953 itself should stop them being exploited. The report I received stated that
2954 data on the command line could provoke the exploit when Exim was running as
2955 exim, but did not say which command line option was involved. All I could
2956 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2957 running as the user.
2960 61. There was a buffer overflow vulnerability in the SPA authentication code
2961 (which came originally from the Samba project). I have added a test to the
2962 spa_base64_to_bits() function which I hope fixes it.
2965 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2966 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2968 63. The daemon start-up calls getloadavg() while still root for those OS that
2969 need the first call to be done as root, but it missed one case: when
2970 deliver_queue_load_max is set with deliver_drop_privilege. This is
2971 necessary for the benefit of the queue runner, because there is no re-exec
2972 when deliver_drop_privilege is set.
2974 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2975 This has been fixed.
2977 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2978 was in use, was not putting the data itself into the right store pool;
2979 consequently, it could be overwritten for a subsequent message in the same
2980 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2983 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2985 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2986 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2987 the function were treating the return as a boolean value, which happened to
2988 work because 0=false and not-0=true, but is not correct code.
2990 68. The host_aton() function was not handling scoped IPv6 addresses (those
2991 with, for example, "%eth0" on the end) correctly.
2993 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2994 negated items (that is, ~something) in unsigned ints. Some compilers
2995 apparently mutter when there is no cast.
2997 70. If an address verification called from an ACL failed, and did not produce a
2998 user-specific message (i.e. there was only a "system" message), nothing was
2999 put in $acl_verify_message. In this situation, it now puts the system
3002 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3003 session; change 4.31/43 added the unwanted input to the log line - except
3004 that it did not do this in the start of session case. It now does.
3006 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3007 This is wrong and can cause the other end to generate a synchronization
3008 error if it is another Exim or anything else that does the synchronization
3009 check. A QUIT command is no longer sent after a timeout.
3011 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3012 during host lookups.
3014 74. The defer_ok option of callout verification was not working if it was used
3015 when verifying addresses in header lines, that is, for this case:
3017 verify = header_sender/callout=defer_ok
3019 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3020 those file descriptors could be used for SMTP connections. If anything
3021 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3022 could be sent to the SMTP client, causing chaos. The daemon now opens
3023 stdin, stdout, and stderr to /dev/null when it puts itself into the
3026 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3027 log by default. The user can override this with suitable Perl magic.
3029 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3030 the log message when discard triggers, was not working for the DATA ACL or
3031 for the non-SMTP ACL.
3033 78. Error message wording change in sieve.c.
3035 79. If smtp_accept_max_per_host was set, the number of connections could be
3036 restricted to fewer than expected, because the daemon was trying to set up
3037 a new connection before checking whether the processes handling previous
3038 connections had finished. The check for completed processes is now done
3039 earlier. On busy systems, this bug wouldn't be noticed because something
3040 else would have woken the daemon, and it would have reaped the completed
3043 80. If a message was submitted locally by a user whose login name contained one
3044 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3045 It caused a spool format error. I have fixed the spool reading code. A
3046 related problem was that the "from" clause in the Received: line became
3047 illegal because of the space(s). It is now covered by ${quote_local_part.
3049 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3052 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3054 83. Patch from Sieve maintainer to fix unterminated string problem in
3055 "vacation" handling.
3057 84. Some minor changes to the Linux configuration files to help with other
3058 OS variants using glibc.
3060 85. One more patch for Sieve to update vacation handling to latest spec.
3063 ----------------------------------------------------
3064 See the note above about the 4.44 and 4.50 releases.
3065 ----------------------------------------------------
3071 1. Change 4.43/35 introduced a bug that caused file counts to be
3072 incorrectly computed when quota_filecount was set in an appendfile
3075 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3076 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3079 3. Give more explanation in the error message when the command for a transport
3080 filter fails to execute.
3082 4. There are several places where Exim runs a non-Exim command in a
3083 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3084 was being done only for the command run by the queryprogram router. It is
3085 now done for all such subprocesses. The other cases are: ${run, transport
3086 filters, and the commands run by the lmtp and pipe transports.
3088 5. Some older OS have a limit of 256 on the maximum number of file
3089 descriptors. Exim was using setrlimit() to set 1000 as a large value
3090 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3091 systems. I've change it so that if it can't get 1000, it tries for 256.
3093 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3094 was an oversight, and furthermore, ever since the addition of extra
3095 controls (e.g. 4.43/32), the checks on when to allow different forms of
3096 "control" were broken. There should now be diagnostics for all cases when a
3097 control that does not make sense is encountered.
3099 7. $recipients is now available in the predata ACL (oversight).
3101 8. Tidy the search cache before the fork to do a delivery from a message
3102 received from the command line. Otherwise the child will trigger a lookup
3103 failure and thereby defer the delivery if it tries to use (for example) a
3104 cached ldap connection that the parent has called unbind on.
3106 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3107 of $address_data from the recipient verification was clobbered by the
3108 sender verification.
3110 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3111 was its contents. (It was OK if the option was not defined at all.)
3113 11. A "Completed" log line is now written for messages that are removed from
3114 the spool by the -Mrm option.
3116 12. $host_address is now set to the target address during the checking of
3117 ignore_target_hosts.
3119 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3120 being passed; this would have caused $sender_host_name to have been used if
3121 matching the list had actually called for a host name (not very likely,
3122 since this list is usually IP addresses). A host name is now passed as
3125 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3126 code that specifies a non-restarting handler (typically sigaction() in
3127 modern systems) in an attempt to fix a rare and obscure crash bug.
3129 15. Narrowed the window for a race in the daemon that could cause it to ignore
3130 SIGCHLD signals. This is not a major problem, because they are used only to
3131 wake it up if nothing else does.
3133 16. A malformed maildirsize file could cause Exim to calculate negative values
3134 for the mailbox size or file count. Odd effects could occur as a result.
3135 The maildirsize information is now recalculated if the size or filecount
3138 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3139 support for a long time. Removed HAVE_SYS_VFS_H.
3141 18. Updated exipick to current release from John Jetmore.
3143 19. Allow an empty sender to be matched against a lookup in an address list.
3144 Previously the only cases considered were a regular expression, or an
3147 20. Exim went into a mad DNS lookup loop when doing a callout where the
3148 host was specified on the transport, if the DNS lookup yielded more than
3151 21. The RFC2047 encoding function was originally intended for short strings
3152 such as real names; it was not keeping to the 75-character limit for
3153 encoded words that the RFC imposes. It now respects the limit, and
3154 generates multiple encoded words if necessary. To be on the safe side, I
3155 have increased the buffer size for the ${rfc2047: expansion operator from
3158 22. Failure to deliver a bounce message always caused it to be frozen, even if
3159 there was an errors_to setting on the router. The errors_to setting is now
3162 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3163 canonical form (fully expanded) before being placed in
3164 $sender_host_address.
3166 24. Updated eximstats to version 1.33
3168 25. Include certificate and key file names in error message when GnuTLS fails
3169 to set them up, because the GnuTLS error message doesn't include the name
3170 of the failing file when there is a problem reading it.
3172 26. Expand error message when OpenSSL has problems setting up cert/key files.
3175 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3176 (this can affect the format of dates).
3178 28. exim_tidydb, when checking for the continued existence of a message for
3179 which it has found a message-specific retry record, was not finding
3180 messages that were in split spool directories. Consequently, it was
3181 deleting retry records that should have stayed in existence.
3183 29. eximstats updated to version 1.35
3184 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3185 1.35 - bugfix such that pie charts by volume are generated correctly
3187 30. The SPA authentication driver was not abandoning authentication and moving
3188 on to the next authenticator when an expansion was forced to fail,
3189 contradicting the general specification for all authenticators. Instead it
3190 was generating a temporary error. It now behaves as specified.
3192 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3193 (the order specifies the preference for clients). The order is now AES256,
3194 AES128, 3DES, ARCFOUR128.
3196 31. Small patch to Sieve code - explicitly set From: when generating an
3199 32. Exim crashed if a remote delivery caused a very long error message to be
3200 recorded - for instance if somebody sent an entire SpamAssassin report back
3201 as a large number of 550 error lines. This bug was coincidentally fixed by
3202 increasing the size of one of Exim's internal buffers (big_buffer) that
3203 happened as part of the Exiscan merge. However, to be on the safe side, I
3204 have made the code more robust (and fixed the comments that describe what
3207 33. Some experimental protocols are using DNS PTR records for new purposes. The
3208 keys for these records are domain names, not reversed IP addresses. The
3209 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3210 leaves it alone. Component reversal etc. now happens only for IP addresses.
3213 34. The host_aton() function is supposed to be passed a string that is known
3214 to be a valid IP address. However, in the case of IPv6 addresses, it was
3215 not checking this. This is a hostage to fortune. Exim now panics and dies
3216 if the condition is not met. A case was found where this could be provoked
3217 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3218 components; fortuitously, this particular loophole had already been fixed
3219 by change 4.50/55 or 4.44/33 above.
3221 If there are any other similar loopholes, the new check in host_aton()
3222 itself should stop them being exploited. The report I received stated that
3223 data on the command line could provoke the exploit when Exim was running as
3224 exim, but did not say which command line option was involved. All I could
3225 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3226 running as the user.
3229 35. There was a buffer overflow vulnerability in the SPA authentication code
3230 (which came originally from the Samba project). I have added a test to the
3231 spa_base64_to_bits() function which I hope fixes it.
3234 36. The daemon start-up calls getloadavg() while still root for those OS that
3235 need the first call to be done as root, but it missed one case: when
3236 deliver_queue_load_max is set with deliver_drop_privilege. This is
3237 necessary for the benefit of the queue runner, because there is no re-exec
3238 when deliver_drop_privilege is set.
3240 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3241 was in use, was not putting the data itself into the right store pool;
3242 consequently, it could be overwritten for a subsequent message in the same
3243 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3246 38. Sometimes the final signoff response after QUIT could fail to get
3247 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3248 < 0 before doing a fflush(). This bug looks as though it goes back to the
3249 introduction of TLS in release 3.20, but "sometimes" must have been rare
3250 because the tests only now provoked it.
3256 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3257 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3258 Now it returns an integer. A number of calls were still expecting a T/F
3259 return. Fortuitously, in all cases, the tests worked in OK situations,
3260 which is the norm. However, things would have gone wrong on any write
3261 failures on the smtp file descriptor. This function is used when sending
3262 messages over SMTP and also when doing verify callouts.
3264 2. When Exim is called to do synchronous delivery of a locally submitted
3265 message (the -odf or -odi options), it no longer closes stderr before doing
3268 3. Implemented the mua_wrapper option.
3270 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3272 5. Implemented the functions header_remove(), header_testname(),
3273 header_add_at_position(), and receive_remove_recipient(), and exported them
3276 6. If an ACL "warn" statement specified the addition of headers, Exim already
3277 inserted X-ACL-Warn: at the start if there was no header name. However, it
3278 was not making this test for the second and subsequent header lines if
3279 there were newlines in the string. This meant that an invalid header could
3280 be inserted if Exim was badly configured.
3282 7. Allow an ACL "warn" statement to add header lines at the start or after all
3283 the Received: headers, as well as at the end.
3285 8. Added the rcpt_4xx retry error code.
3287 9. Added postmaster_mailfrom=xxx to callout verification option.
3289 10. Added mailfrom=xxxx to the callout verification option, for verify=
3292 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3293 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3295 12. Inserted some casts to stop certain compilers warning when using pointer
3296 differences as field lengths or precisions in printf-type calls (mostly
3297 affecting debugging statements).
3299 13. Added optional readline() support for -be (dynamically loaded).
3301 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3302 same clock tick as a message's arrival, so that its received time was the
3303 same as the "first fail" time on the retry record, and that message
3304 remained on the queue past the ultimate address timeout, every queue runner
3305 would try a delivery (because it was past the ultimate address timeout) but
3306 after another failure, the ultimate address timeout, which should have then
3307 bounced the address, did not kick in. This was a "< instead of <=" error;
3308 in most cases the first failure would have been in the next clock tick
3309 after the received time, and all would be well.
3311 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3312 being recognized when the domain list was tested by the match_domain
3313 condition in an expansion string.
3315 16. Added the ${str2b64: operator.
3317 17. Exim was always calling setrlimit() to set a large limit for the number of
3318 processes, without checking whether the existing limit was already
3319 adequate. (It did check for the limit on file descriptors.) Furthermore,
3320 errors from getrlimit() and setrlimit() were being ignored. Now they are
3321 logged to the main and panic logs, but Exim does carry on, to try to do its
3322 job under whatever limits there are.
3324 18. Imported PCRE 5.0.
3326 19. Trivial typo in log message " temporarily refused connection" (the leading
3329 20. If the log selector return_path_on_delivery was set and an address was
3330 redirected to /dev/null, the delivery process crashed because it assumed
3331 that a return path would always be set for a "successful" delivery. In this
3332 case, the whole delivery is bypassed as an optimization, and therefore no
3335 21. Internal re-arrangement: the function for sending a challenge and reading
3336 a response while authentication was assuming a zero-terminated challenge
3337 string. It's now changed to take a pointer and a length, to allow for
3338 binary data in such strings.
3340 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3342 23. Exim was not respecting finduser_retries when seeking the login of the
3343 uid under which it was called; it was always trying 10 times. (The default
3344 setting of finduser_retries is zero.) Also, it was sleeping after the final
3345 failure, which is pointless.
3347 24. Implemented tls_on_connect_ports.
3349 25. Implemented acl_smtp_predata.
3351 26. If the domain in control=submission is set empty, Exim assumes that the
3352 authenticated id is a complete email address when it generates From: or
3353 Sender: header lines.
3355 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3356 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3357 chown and chgrp in /bin and hostname in /usr/bin.
3359 28. Exim was keeping the "process log" file open after each use, just as it
3360 does for the main log. This opens the possibility of it remaining open for
3361 long periods when the USR1 signal hits a daemon. Occasional processlog
3362 errors were reported, that could have been caused by this. Anyway, it seems
3363 much more sensible not to leave this file open at all, so that is what now
3366 29. The long-running daemon process does not normally write to the log once it
3367 has entered its main loop, and it closes the log before doing so. This is
3368 so that log files can straightforwardly be renamed and moved. However,
3369 there are a couple of unusual error situations where the daemon does write
3370 log entries, and I had neglected to close the log afterwards.
3372 30. The text of an SMTP error response that was received during a remote
3373 delivery was being truncated at 512 bytes. This is too short for some of
3374 the long messages that one sometimes sees. I've increased the limit to
3377 31. It is now possible to make retry rules that apply only when a message has a
3378 specific sender, in particular, an empty sender.
3380 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3381 it possible to be selective about when SMTP synchronization is enforced.
3383 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3385 32. Implemented hosts_connection_nolog.
3387 33. Added an ACL for QUIT.
3389 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3392 35. Added mailbox_size and mailbox_filecount to appendfile.
3394 36. Added control = no_multiline_responses to ACLs.
3396 37. There was a bug in the logic of the code that waits for the clock to tick
3397 in the case where the clock went backwards by a substantial amount such
3398 that the microsecond fraction of "now" was more than the microsecond
3399 fraction of "then" (but the whole seconds number was less).
3401 38. Added support for the libradius Radius client library this is found on
3402 FreeBSD (previously only the radiusclient library was supported).
3408 1. When certain lookups returned multiple values in the form name=value, the
3409 quoting of the values was not always being done properly. Specifically:
3410 (a) If the value started with a double quote, but contained no whitespace,
3412 (b) If the value contained whitespace other than a space character (i.e.
3413 tabs or newlines or carriage returns) it was not quoted.
3414 This fix has been applied to the mysql and pgsql lookups by writing a
3415 separate quoting function and calling it from the lookup code. The fix
3416 should probably also be applied to nisplus, ibase and oracle lookups, but
3417 since I cannot test any of those, I have not disturbed their existing code.
3419 2. A hit in the callout cache for a specific address caused a log line with no
3420 reason for rejecting RCPT. Now it says "Previous (cached) callout
3421 verification failure".
3423 3. There was an off-by-one bug in the queryprogram router. An over-long
3424 return line was truncated at 256 instead of 255 characters, thereby
3425 overflowing its buffer with the terminating zero. As well as fixing this, I
3426 have increased the buffer size to 1024 (and made a note to document this).
3428 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3429 when Exim is waiting for an SMTP response from a remote server, Exim
3430 restarts its select() call on the socket, thereby resetting its timeout.
3431 This is not a problem when such interrupts are rare. Somebody set up a cron
3432 job to run exiwhat every 2 minutes, which is less than the normal select()
3433 timeout (5 or 10 minutes). This meant that the select() timeout never
3434 kicked in because it was always reset. I have fixed this by comparing the
3435 time when an interrupt arrives with the time at the start of the first call
3436 to select(). If more time than the timeout has elapsed, the interrupt is
3437 treated as a timeout.
3439 5. Some internal re-factoring in preparation for the addition of Sieve
3440 extensions (by MH). In particular, the "personal" test is moved to a
3441 separate function, and given an option for scanning Cc: and Bcc: (which is
3442 not set for Exim filters).
3444 6. When Exim created an email address using the login of the caller as the
3445 local part (e.g. when creating a From: or Sender: header line), it was not
3446 quoting the local part when it contained special characters such as @.
3448 7. Installed new OpenBSD configuration files.
3450 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3451 try to make them clearer.
3453 9. Callout options, other than the timeout value, were being ignored when
3454 verifying sender addresses in header lines. For example, when using
3456 verify = header_sender/callout=no_cache
3458 the cache was (incorrectly) being used.
3460 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3461 only the Cygwin environment.
3463 11. When return_path_on_delivery was set as a log selector, if different remote
3464 addresses in the same message used different return paths and parallel
3465 remote delivery occurred, the wrong values would sometimes be logged.
3466 (Whenever a remote delivery process finished, the return path value from
3467 the most recently started remote delivery process was logged.)
3469 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3470 lines when AUTH and/or TLS are in use. This is the "received protocol"
3471 field. Exim used to use "asmtp" for authenticated SMTP, without any
3472 indication (in the protocol name) for TLS use. Now it follows the RFC and
3473 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3474 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3475 names appear in log lines as well as in Received: header lines.
3477 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3478 extensions, and comparison tests, and to fix some bugs.
3480 14. Changes to the "personal" filter test:
3482 (1) The test was buggy in that it was just doing the equivalent of
3483 "contains" tests on header lines. For example, if a user's address was
3484 anne@some.where, the "personal" test would incorrectly be true for
3486 To: susanne@some.where
3488 This test is now done by extracting each address from the header in turn,
3489 and checking the entire address. Other tests that are part of "personal"
3490 are now done using regular expressions (for example, to check local parts
3491 of addresses in From: header lines).
3493 (2) The list of non-personal local parts in From: addresses has been
3494 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3495 taken from the Sieve specification recommendations.
3497 (3) If the message contains any header line starting with "List-" it is
3498 treated as non-personal.
3500 (4) The test for "circular" in the Subject: header line has been removed
3501 because it now seems ill-conceived.
3503 15. Minor typos in src/EDITME comments corrected.
3505 16. Installed latest exipick from John Jetmore.
3507 17. If headers_add on a router specified a text string that was too long for
3508 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3509 of string_sprintf() is now avoided.
3511 18. $message_body_size was not set (it was always zero) when running the DATA
3512 ACL and the local_scan() function.
3514 19. For the "mail" command in an Exim filter, no default was being set for
3515 the once_repeat time, causing a random time value to be used if "once" was
3516 specified. (If the value happened to be <= 0, no repeat happened.) The
3517 default is now 0s, meaning "never repeat". The "vacation" command was OK
3518 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3519 (I found it when inspecting the code).
3521 20. There is now an overall timeout for performing a callout verification. It
3522 defaults to 4 times the callout timeout, which applies to individual SMTP
3523 commands during the callout. The overall timeout applies when there is more
3524 than one host that can be tried. The timeout is checked before trying the
3525 next host. This prevents very long delays if there are a large number of
3526 hosts and all are timing out (e.g. when the network connections are timing
3527 out). The value of the overall timeout can be changed by specifying an
3528 additional sub-option for "callout", called "maxwait". For example:
3530 verify = sender/callout=5s,maxwait=20s
3532 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3533 to the end before writing, but this should make it even safer).
3535 22. Exim was forgetting that it had advertised PIPELINING for the second and
3536 subsequent messages on an SMTP connection. It was also not resetting its
3537 memory on STARTTLS and an internal HELO.
3539 23. When Exim logs an SMTP synchronization error within a session, it now
3540 records whether PIPELINING has been advertised or not.
3542 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3543 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3544 rather than long int.
3546 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3548 26. Added the never_mail option to autoreply.
3554 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3555 crash if the getsockname() call failed; this can happen if a connection is
3556 closed very soon after it is established. The problem was simply in the
3557 order in which certain operations were done, causing Exim to try to write
3558 to the SMTP stream before it had set up the file descriptor. The bug has
3559 been fixed by making things happen in the correct order.
3565 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3566 before the connection was closed, thus losing the rejection response.
3568 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3569 some early Solaris releases, but causes trouble in current releases where
3570 socklen_t is defined.
3572 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3575 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3578 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3579 that releasing the top of it at the end releases what was used for sub-
3580 expansions (unless the block got too big). However, discard this block if
3581 the first thing is a variable or header, so that we can use its block when
3582 it is dynamic (useful for very large $message_headers, for example).
3584 6. Lookups now cache *every* query, not just the most recent. A new, separate
3585 store pool is used for this. It can be recovered when all lookup caches are
3586 flushed. Lookups now release memory at the end of their result strings.
3587 This has involved some general refactoring of the lookup sources.
3589 7. Some code has been added to the store_xxx() functions to reduce the amount
3590 of flapping under certain conditions.
3592 8. log_incoming_interface used to affect only the <= reception log lines. Now
3593 it causes the local interface and port to be added to several more SMTP log
3594 lines, for example "SMTP connection from", and rejection lines.
3596 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3598 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3600 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3601 could be overwritten at the end of the current message (or the start of a
3602 new message if it was set in a HELO ACL). The value is now preserved for
3603 the duration of the SMTP connection.
3605 12. If a transport had a headers_rewrite setting, and a matching header line
3606 contained an unqualified address, that address was qualified, even if it
3607 did not match any rewriting rules. The underlying bug was that the values
3608 of the flags that permit the existence of unqualified sender and recipient
3609 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3610 non-local messages, and by -bnq for local messages) were not being
3611 preserved with the message after it was received.
3613 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3614 "next input=" as part of the text comprising the host identity instead of
3615 the correct text. The code was using the same buffer for two different
3616 strings. However, depending on which order the printing function evaluated
3617 its arguments, the bug did not always show up. Under Linux, for example, my
3618 test suite worked just fine.
3620 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3621 doesn't work with some older versions of Perl. It has been changed to "my",
3622 which in any case is probably the better facility to use.
3624 15. A really picky compiler found some instances of statements for creating
3625 error messages that either had too many or two few arguments for the format
3628 16. The size of the buffer for calls to the DNS resolver has been increased
3629 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3630 for addresses that have a lot of PTR records. This alleviates a problem; it
3631 does not fully solve it.
3633 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3634 buffer now truncates the list and logs the incident, which is the same
3635 action as happens when Exim is looking up a host name and its aliases.
3636 Previously in this situation something unpredictable would happen;
3637 sometimes it was "internal error: store_reset failed".
3639 18. If a server dropped the connection unexpectedly when an Exim client was
3640 using GnuTLS and trying to read a response, the client delivery process
3641 crashed while trying to generate an error log message.
3643 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3644 string, for example:
3646 warn message = H1: something\nH2: something
3648 the text was added as a single header line from Exim's point of view
3649 though it ended up OK in the delivered message. However, searching for the
3650 second and subsequent header lines using $h_h2: did not work. This has been
3651 fixed. Similarly, if a system filter added multiple headers in this way,
3652 the routers could not see them.
3654 20. Expanded the error message when iplsearch is called with an invalid key to
3655 suggest using net-iplsearch in a host list.
3657 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3658 ACLs are no longer actually imposed (and a message to that effect is
3661 22. If a "gecos" field in a passwd entry contained escaped characters, in
3662 particular, if it contained a \" sequence, Exim got it wrong when building
3663 a From: or a Sender: header from that name. A second bug also caused
3664 incorrect handling when an unquoted " was present following a character
3665 that needed quoting.
3667 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3668 was not being matched caselessly.
3670 24. Arranged for all hyphens in the exim.8 source to be escaped with
3673 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3674 properly. Recipient callout cache records were still being keyed to include
3675 the sender, even when use_sender was set false. This led to far more
3676 callouts that were necessary. The sender is no longer included in the key
3677 when use_sender is false.
3679 26. Added "control = submission" modifier to ACLs.
3681 27. Added the ${base62d: operator to decode base 62 numbers.
3683 28. dnsdb lookups can now access SRV records.
3685 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3686 the configuration file.
3688 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3689 (-v) mode. This makes the output for a verbose queue run more intelligible.
3691 31. Added a use_postmaster feature to recipient callouts.
3693 32. Added the $body_zerocount variable, containing the number of binary zero
3694 bytes in the message body.
3696 33. The time of last modification of the "new" subdirectory is now used as the
3697 "mailbox time last read" when there is a quota error for a maildir
3700 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3702 35. Added +ignore_unknown as a special item in host lists.
3704 36. Code for decoding IPv6 addresses in host lists is now included, even if
3705 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3706 address was recognized as an IP address, but was then not correctly decoded
3707 into binary, causing unexpected and incorrect effects when compared with
3714 1. Very minor rewording of debugging text in manualroute to say "list of
3715 hosts" instead of "hostlist".
3717 2. If verify=header_syntax was set, and a header line with an unqualified
3718 address (no domain) and a large number of spaces between the end of the
3719 name and the colon was received, the reception process suffered a buffer
3720 overflow, and (when I tested it) crashed. This was caused by some obsolete
3721 code that should have been removed. The fix is to remove it!
3723 3. When running in the test harness, delay a bit after writing a bounce
3724 message to get a bit more predictability in the log output.
3726 4. Added a call to search_tidyup() just before forking a reception process. In
3727 theory, someone could use a lookup in the expansion of smtp_accept_max_
3728 per_host which, without the tidyup, could leave open a database connection.
3730 5. Added the variables $recipient_data and $sender_data which get set from a
3731 lookup success in an ACL "recipients" or "senders" condition, or a router
3732 "senders" option, similar to $domain_data and $local_part_data.
3734 6. Moved the writing of debug_print from before to after the "senders" test
3737 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3738 problems for message scanning, either using a data ACL, or using
3739 local_scan() because the Received: header was not generated till after they
3740 were called (in order to set the time as the time of reception completion).
3741 I have revised the way this works. The header is now generated after the
3742 body is received, but before the ACL or local_scan() are called. After they
3743 are run, the timestamp in the header is updated.
3749 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3750 before starting a queue runner without re-exec. This happened only when
3751 deliver_drop_privilege was set or when the Exim user was set to root. The
3752 effect of the bug was that timeouts during subsequent deliveries caused
3753 crashes instead of being properly handled. The handler is now left at its
3754 default (and expected) setting.
3756 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3757 message, again when deliver_drop_privilege is set or Exim is run as root.
3758 The bug described in (1) was not present in this case, but the tidying up
3759 of the other signals was missing. I have made the two cases consistent.
3761 3. The ignore_target_hosts setting on a manualroute router was being ignored
3762 for hosts that were looked up using the /MX notation.
3764 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3767 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3768 operated on the sender address. After changing the $sender_address to <>
3769 for the sender address verify, Exim was re-instated it as the original
3770 (before rewriting) address, but remembering that it had rewritten it, so it
3771 wasn't rewriting it again. This bug also had the effect of breaking the
3772 sender address verification caching when the sender address was rewritten.
3774 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3775 This has been changed so that if the ip literal address matches
3776 ignore_target_hosts, the router declines.
3778 7. Added expansion conditions match_domain, match_address, and match_local_
3779 part (NOT match_host).
3781 8. The placeholder for the Received: header didn't have a length field set.
3783 9. Added code to Exim itself and to exim_lock to test for a specific race
3784 condition that could lead to file corruption when using MBX delivery. The
3785 issue is with the lockfile that is created in /tmp. If this file is removed
3786 after a process has opened it but before that process has acquired a lock,
3787 there is the potential for a second process to recreate the file and also
3788 acquire a lock. This could lead to two Exim processes writing to the file
3789 at the same time. The added code performs the same test as UW imapd; it
3790 checks after acquiring the lock that its file descriptor still refers to
3791 the same named file.
3793 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3794 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3797 11. Added the smtp_active_hostname option. If used, this will typically be made
3798 to depend on the incoming interface address. Because $interface_address is
3799 not set up until the daemon has forked a reception process, error responses
3800 that can happen earlier (such as "too many connections") no longer contain
3803 12. If an expansion in a condition on a "warn" statement fails because a lookup
3804 defers, the "warn" statement is abandoned, and the next ACL statement is
3805 processed. Previously this caused the whole ACL to be aborted.
3807 13. Added the iplsearch lookup type.
3809 14. Added ident_timeout as a log selector.
3811 15. Added tls_certificate_verified as a log selector.
3813 16. Added a global option tls_require_ciphers (compare the smtp transport
3814 option of the same name). This controls incoming TLS connections.
3816 17. I finally figured out how to make tls_require_ciphers do a similar thing
3817 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3818 before starting the TLS session.
3820 18. Tabs are now shown as \t in -bP output.
3822 19. If the log selector return_path_on_delivery was set, Exim crashed when
3823 bouncing a message because it had too many Received: header lines.
3825 20. If two routers both had headers_remove settings, and the first one included
3826 a superfluous trailing colon, the final name in the first list and the
3827 first name in the second list were incorrectly joined into one item (with a
3828 colon in the middle).
3834 1. Added -C and -D options to the exinext utility, mainly to make it easier
3835 to include in the automated testing, but these could be helpful when
3836 multiple configurations are in use.
3838 2. The exinext utility was not formatting the output nicely when there was
3839 an alternate port involved in the retry record key, nor when there was a
3840 message id as well (for retries that were specific to a specific message
3841 and a specific host). It was also confused by IPv6 addresses, because of
3842 the additional colons they contain. I have fixed the IPv4 problem, and
3843 patched it up to do a reasonable job for IPv6.
3845 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3846 delivery, the log line now contains "pipelined" if PIPELINING was used.
3848 4. An SMTP transport process used to panic and die if the bind() call to set
3849 an explicit outgoing interface failed. This has been changed; it is now
3850 treated in the same way as a connect() failure.
3852 5. A reference to $sender_host_name in the part of a conditional expansion
3853 that was being skipped was still causing a DNS lookup. This no longer
3856 6. The def: expansion condition was not recognizing references to header lines
3857 that used bh_ and bheader_.
3859 7. Added the _cache feature to named lists.
3861 8. The code for checking quota_filecount in the appendfile transport was
3862 allowing one more file than it should have been.
3864 9. For compatibility with Sendmail, the command line option
3872 and sets the incoming protocol and host name (for trusted callers). The
3873 host name and its colon can be omitted when only the protocol is to be set.
3874 Note the Exim already has two private options, -pd and -ps, that refer to
3875 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3876 "s", but I don't think that's a major issue.
3878 10. A number of refactoring changes to the code, none of which should affect
3881 (a) The number of logging options was getting close to filling up the
3882 32-bit word that was used as a bit map. I have split them into two classes:
3883 those that are passed in the argument to log_write(), and those that are
3884 only ever tested independently outside of that function. These are now in
3885 separate 32-bit words, so there is plenty of room for expansion again.
3886 There is no change in the user interface or the logging behaviour.
3888 (b) When building, for example, log lines, the code previously used a
3889 macro that called string_cat() twice, in order to add two strings. This is
3890 not really sufficiently general. Furthermore, there was one instance where
3891 it was actually wrong because one of the argument was used twice, and in
3892 one call a function was used. (As it happened, calling the function twice
3893 did not affect the overall behaviour.) The macro has been replaced by a
3894 function that can join an arbitrary number of extra strings onto a growing
3897 (c) The code for expansion conditions now uses a table and a binary chop
3898 instead of a serial search (which was left over from when there were very
3899 few conditions). Also, it now recognizes conditions like "pam" even when
3900 the relevant support is not compiled in: a suitably worded error message is
3901 given if an attempt is made to use such a condition.
3903 11. Added ${time_interval:xxxxx}.
3905 12. A bug was causing one of the ddress fields not to be passed back correctly
3906 from remote delivery subprocesses. The field in question was not being
3907 subsequently used, so this caused to problems in practice.
3909 13. Added new log selectors queue_time and deliver_time.
3911 14. Might have fixed a bug in maildirsizefile handling that threw up
3912 "unexpected character" debug warnings, and recalculated the data
3913 unnecessarily. In any case, I expanded the warning message to give more
3916 15. Added the message "Restricted characters in address" to the statements in
3917 the default ACL that block characters like @ and % in local parts.
3919 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3920 Three changes have been made:
3922 (a) There was a serious bug; a negative response to MAIL caused the whole
3923 recipient domain to be cached as invalid, thereby blocking all messages
3924 to all local parts at the same domain, from all senders. This bug has
3925 been fixed. The domain is no longer cached after a negative response to
3926 MAIL if the sender used is not empty.
3928 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3931 (c) A new callout option, "use_sender" has been added for people who want
3932 the modified behaviour.
3938 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3941 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3942 indeed breaks things for older releases.
3944 3. Added additional logging to the case where there is a problem reading data
3945 from a filter that is running in a subprocess using a pipe, in order to
3946 try to track down a specific problem.
3948 4. Testing facility fudge: when running in the test harness and attempting
3949 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3950 getting "No route to host". Convert this to a timeout.
3952 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3955 6. Some OS don't have socklen_t but use size_t instead. This affects the
3956 fifth argument of getsockopt() amongst other things. This is now
3957 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3958 can be set for individual OS. I have set it for SunOS5, OSF1, and
3959 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3960 some earlier ones do not.
3962 7. Change 4.30/15 was not doing the test caselessly.
3964 8. The standard form for an IPv6 address literal was being rejected by address
3965 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3966 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3967 this, as well as the form without the "IPv6" on the front (but only when
3968 address literals are enabled, of course).
3970 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3972 10. Exim crashed if a message with an empty sender address specified by -f
3973 encountered a router with an errors_to setting. This could be provoked only
3974 by a command such as
3978 where an empty string was supplied; "<>" did not hit this bug.
3980 11. Installed PCRE release 4.5.
3982 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3983 remained set. It is now erased.
3985 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3986 times from message ids (which are base 36 rather than the normal 62).
3988 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3989 were being counted as actual protocol errors, and logged if the log
3990 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3991 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3992 and DATA following a set of rejected RCPTs do not count as protocol errors.
3993 In other words, Exim assumes they were pipelined, though this may not
3994 actually be the case. Of course, in all cases the client gets an
3995 appropriate error code.
3997 15. If a lookup fails in an ACL condition, a message about the failure may
3998 be available; it is used if testing the ACL cannot continue, because most
3999 such messages specify what the cause of the deferral is. However, some
4000 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4001 that caused an old message to be retained and used if a later statement
4002 caused a defer, replacing the real cause of the deferral.
4004 16. If an IP address had so many PTR records that the DNS lookup buffer
4005 was not large enough to hold them, Exim could crash while trying to process
4006 the truncated data. It now detects and logs this case.
4008 17. Further to 4.21/58, another change has been made: if (and only if) the
4009 first line of a message (the first header line) ends with CRLF, a bare LF
4010 in a subsequent header line has a space inserted after it, so as not to
4011 terminate the header.
4013 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4014 unnecessarily, used atoi() instead of strtol(), and didn't check the
4015 termination when getting file sizes from file names by regex.
4017 19. Completely re-implemented the support for maildirsize files, in the light
4018 of a number of problems with the previous contributed implementation
4019 (4.30/29). In particular:
4021 . If the quota is zero, the maildirsize file is maintained, but no quota is
4024 . If the maildir directory does not exist, it is created before any attempt
4025 to write a maildirsize file.
4027 . The quota value in the file is just a cache; if the quota is changed in
4028 the transport, the new value overrides.
4030 . A regular expression is available for excluding directories from the
4033 20. The autoreply transport checks the characters in options that define the
4034 message's headers; it allows continued headers, but it was checking with
4035 isspace() after an embedded newline instead of explicitly looking for a
4038 21. If all the "regular" hosts to which an address was routed had passed their
4039 expiry times, and had not reached their retry times, the address was
4040 bounced, even if fallback hosts were defined. Now Exim should go on to try
4043 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4044 equivalent code in the SMTP transport. Some hosts send humungous responses
4045 to HELO/EHLO, more than 1024 it seems.
4047 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4048 gives compiler warnings in some environments. I've now done it "properly",
4051 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4052 (because of problems with the built-in one) was declared to return uschar *
4053 instead of char *, causing compiler failure.
4055 25. Fixed a file descriptor leak when processing alias/forward files.
4057 26. Fixed a minor format string issue in dbfn.c.
4059 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4061 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4062 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4065 29. When a sender address is verified, it is cached, to save repeating the test
4066 when there is more than one recipient in a message. However, when the
4067 verification involves a callout, it is possible for different callout
4068 options to be set for different recipients. It is too complicated to keep
4069 track of this in the cache, so now Exim always runs a verification when a
4070 callout is required, relying on the callout cache for the optimization.
4071 The overhead is duplication of the address routing, but this should not be
4074 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4075 to be verified with callout=postmaster, and the main callout worked but the
4076 postmaster check failed, the verification correctly failed. However, if a
4077 subsequent RCPT command asked for sender verification *without* the
4078 postmaster check, incorrect caching caused this verification also to fail,
4081 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4082 it was not caching the DNS options (qualify_single, search_parents) that
4083 were used when the lookup failed. A subsequent lookup with different
4084 options therefore always gave the same answer, though there were cases
4085 where it should not have. (Example: a "domains = !$mx_any" option on a
4086 dnslookup router: the "domains" option is always processed without any
4087 widening, but the router might have qualify_single set.) Now Exim uses the
4088 cached value only when the same options are set.
4090 32. Added John Jetmore's "exipick" utility to the distribution.
4092 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4093 than a timeout (e.g. a certificate is required, and is not provided), an
4094 Exim server now closes the connection immediately. Previously it waited for
4095 the client to close - but if the client is SSL, it seems that they each
4096 wait for each other, leading to a delay before one of them times out.
4098 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4099 maintained 0.8.x compatibility because I don't think many are using it, and
4100 it is clearly obsolete.
4102 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4105 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4106 client certificate was expired. A simple patch fixes this, though I don't
4107 understand the full logic of why the verify callback is called multiple
4110 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4111 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4112 which causes problems with some clients (such as the Certicom SSL Plus
4113 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4114 disables the coutermeasure allowing Eudora to connect."
4116 38. Exim was not checking that a write() to a log file succeeded. This could
4117 lead to Bad Things if a log got too big, in particular if it hit a file
4118 size limit. Exim now panics and dies if it cannot write to a log file, just
4119 as it does if it cannot open a log file.
4121 39. Modified OS/Makefile-Linux so that it now contains
4123 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4125 The two -D definitions ensure that Exim is compiled with large file
4126 support, which makes it possible to handle log files that are bigger than
4129 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4130 instance) a domain was checked against a named list that involved a lookup,
4131 causing $domain_data to be set, then another domain was checked against the
4132 same list, then the first domain was re-checked, the value of $domain_data
4133 after the final check could be wrong. In particular, if the second check
4134 failed, it could be set empty. This bug probably also applied to
4137 41. The strip_trailing_dot option was not being applied to the address given
4138 with the -f command-line option.
4140 42. The code for reading a message's header from the spool was incrementing
4141 $received_count, but never initializing it. This meant that the value was
4142 incorrect (doubled) while delivering a message in the same process in which
4143 it was received. In the most common configuration of Exim, this never
4144 happens - a fresh exec is done - but it can happen when
4145 deliver_drop_privilege is set.
4147 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4148 it now includes up to 150 characters of the unexpected data in the log
4151 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4152 and building data strings. The size of both of these buffers was 10 000
4153 bytes - far larger than anybody would *ever* want, thought I. Needless to
4154 say, somebody hit the limit. I have increased the maximum line length to
4155 20 000 and the maximum data length of concatenated lines to 100 000. I have
4156 also fixed two bugs, because there was no checking on these buffers. Tsk,
4157 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4158 buffer is too small.
4160 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4161 lsearch lookups. Now it does.
4163 46. When parsing a route_list item in a manualroute router, a fixed-length
4164 buffer was used for the list of hosts. I made this 1024 bytes long,
4165 thinking that nobody would ever have a list of hosts that long. Wrong.
4166 Somebody had a whole pile of complicated expansion conditions, and the
4167 string was silently truncated, leading to an expansion error. It turns out
4168 that it is easier to change to an unlimited length (owing to other changes
4169 that have happened since this code was originally written) than to build
4170 structure for giving a limitation error. The length of the item that
4171 expands into the list of hosts is now unlimited.
4173 47. The lsearch lookup could not handle data where the length of text line was
4174 more than 4095 characters. Such lines were truncated, leading to shortened
4175 data being returned. It should now handle lines of any length.
4177 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4178 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4181 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4182 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4183 no longer happens in comment lines. A list of replacements is now placed
4184 at the head of all of the source files, except those whose only change is
4185 to replace PERL_COMMAND in the very first #! line.
4187 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4188 messages on the queue, with a bottom-up merge sort, using code contributed
4189 by Michael Haardt. This should make operations like -bp somewhat faster on
4190 large queues. It won't affect queue runners, except when queue_run_in_order
4193 51. Installed eximstats 1.31 in the distribution.
4195 52. Added support for SRV lookups to the dnslookup router.
4197 53. If an ACL referred to $message_body or $message_body_end, the value was not
4198 reset for any messages that followed in the same SMTP session.
4200 54. The store-handling optimization for building very long strings was not
4201 differentiating between the different store pools. I don't think this
4202 actually made any difference in practice, but I've tidied it.
4204 55. While running the routers to verify a sender address, $sender_address
4205 was still set to the sender address. This is wrong, because when routing to
4206 send a bounce to the sender, it would be empty. Therefore, I have changed
4207 it so that, while verifying a sender address, $sender_address is set to <>.
4208 (There is no change to what happens when verifying a recipient address.)
4210 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4211 target A or AAAA records (if not already returned) without resetting the
4212 qualify_single or search_parents options of the DNS resolver. These are
4213 inappropriate in this case because the targets of MX and SRV records must
4214 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4215 target that, when qualified, matched something in the local domain. These
4216 two options are now turned off when doing these lookups.
4218 57. It seems that at least some releases of Reiserfs (which does not have the
4219 concept of a fixed number of inodes) returns zero and not -1 for the
4220 number of available inodes. This interacted badly with check_spool_inodes,
4221 which assumed that -1 was the "no such thing" setting. What I have done is
4222 to check that the total number of inodes is greater than zero before doing
4223 the test of how many are available.
4225 58. When a "warn" ACL statement has a log_message modifier, the message is
4226 remembered, and not repeated. This is to avoid a lot of repetition when a
4227 message has many recipients that cause the same warning to be written.
4228 Howewer, Exim was preserving the list of already written lines for an
4229 entire SMTP session, which doesn't seem right. The memory is now reset if a
4230 new message is started.
4232 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4233 debugging output unless log_rewrite was also set.
4235 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4236 of (int)(handle) when we know that handle contains (void *)(-1).
4238 61. The Exim daemon panic-logs an error return when it closes the incoming
4239 connection. However "connection reset by peer" seems to be common, and
4240 isn't really an error worthy of noting specially, so that particular error
4243 62. When Exim is trying to find all the local interfaces, it used to panic and
4244 die if the ioctl to get the interface flags failed. However, it seems that
4245 on at least one OS (Solaris 9) it is possible to have an interface that is
4246 included in the list of interfaces, but for which you get a failure error
4247 for this call. This happens when the interface is not "plumbed" into a
4248 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4249 failure of the "get flags" call assumes that the interface is down.
4251 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4252 makes life easier for people who are doing arithmetic on fields extracted
4253 from dates, where you often get leading zeros that should not be
4254 interpreted as octal.
4256 64. Added qualify_domain to the redirect router, to override the global
4259 65. If a pathologically long header line contained very many addresses (the
4260 report of this problem mentioned 10 000) and each of them was rewritten,
4261 Exim could use up a very large amount of memory. (It kept on making new
4262 copies of the header line as it rewrote, and never released the old ones.)
4263 At the expense of a bit more processing, the header rewriting function has
4264 been changed so that it no longer eats memory in this way.
4266 66. The generation of the Received: header has been moved from the time that a
4267 message starts to be received, to the time that it finishes. The timestamp
4268 in the Received: header should now be very close to that of the <= log
4269 line. There are two side-effects of this change:
4271 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4272 logged header lines no longer include the local Received: line, because
4273 it has not yet been created. The same applies to a copy of the message
4274 that is returned to a non-SMTP sender when a message is rejected.
4276 (b) When a filter file is tested using -bf, no additional Received: header
4277 is added to the test message. After some thought, I decided that this
4280 This change does not affect the value of $received_for. It is still set
4281 after address rewriting, but before local_scan() is called.
4283 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4285 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4286 gave an unhelpful panic error message, and a defer error. I have managed to
4287 change this behaviour so that it now rejects any supplied certificate,
4288 which seems right, as the list of acceptable certificates is empty.
4290 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4291 gave an unhelpful defer error. I have not managed to make this reject any
4292 supplied certificates, but the error message it gives is "no certificate
4293 supplied", which is not helpful.
4295 70. exigrep's output now also includes lines that are not associated with any
4296 message, but which match the given pattern. Implemented by a patch from
4297 Martin Sluka, which also tidied up the Perl a bit.
4299 71. Recipient callout verification, like sender verification, was using <> in
4300 the MAIL FROM command. This isn't really the right thing, since the actual
4301 sender may affect whether the remote host accepts the recipient or not. I
4302 have changed it to use the actual sender in the callout; this means that
4303 the cache record is now keyed on a recipient/sender pair, not just the
4304 recipient address. There doesn't seem to be a real danger of callout loops,
4305 since a callout by the remote host to check the sender would use <>.
4306 [SEE ABOVE: changed after hitting problems.]
4308 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4309 temporary errors. However, in the case of such a code being given after
4310 the end of a data transmission (i.e. after ".") Exim was failing to write
4311 a retry record for the message. (Yes, there was some broken host that was
4312 actually sending 8xx at this point.)
4314 73. An unknown lookup type in a host list could cause Exim to panic-die when
4315 the list was checked. (An example that provoked this was putting <; in the
4316 middle of a list instead of at the start.) If this happened during a DATA
4317 ACL check, a -D file could be left lying around. This kind of configuration
4318 error no longer causes Exim to die; instead it causes a defer errror. The
4319 incident is still logged to the main and panic logs.
4321 74. Buglet left over from Exim 3 conversion. The message "too many messages
4322 in one connection" was written to the rejectlog but not the mainlog, except
4323 when address rewriting (yes!) was being logged.
4325 75. Added write_rejectlog option.
4327 76. When a system filter was run not as root (that is, when system_filter_user
4328 was set), the values of the $n variables were not being returned to the
4329 main process; thus, they were not subsequently available in the $sn
4332 77. Added +return_path_on_delivery log selector.
4334 78. A connection timeout was being treated differently from recipients deferred
4335 when testing hosts_max_try with a message that was older than the host's
4336 retry timeout. (The host should not be counted, thus allowing all hosts to
4337 be tried at least once before bouncing.) This may have been the cause of an
4338 occasionally reported bug whereby a message would remain on the queue
4339 longer than the retry timeout, but would be bounced if a delivery was
4340 forced. I say "may" because I never totally pinned down the problem;
4341 setting up timeout/retry tests is difficult. See also the next item.
4343 79. The ultimate address timeout was not being applied to errors that involved
4344 a combination of host plus message (for example, a timeout on a MAIL
4345 command). When an address resolved to a number of possible hosts, and they
4346 were not all tried for each delivery (e.g. because of hosts_max_try), a
4347 message could remain on the queue longer than the retry timeout.
4349 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4352 81. Fixed an obscure SMTP outgoing bug which required at least the following
4353 conditions: (a) there was another message waiting for the same server;
4354 (b) the server returned 5xx to all RCPT commands in the first message so
4355 that the message was not completed; (c) the server dropped the connection
4356 or gave a negative response to the RSET that Exim sends to abort the
4357 transaction. The observed case was a dropped connection after DATA that had
4358 been sent in pipelining mode. That is, the server had advertised PIPELINING
4359 but was not implementing it correctly. The effect of the bug was incorrect
4360 behaviour, such as trying another host, and this could lead to a crash.
4366 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4367 and daemon.c were passed as pointers to ints; they should have been
4368 pointers to socklen_t variables (which are typically unsigned ints).
4370 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4373 3. Fixed a really odd bug that affected only the testing scheme; patching a
4374 certain fixed string in the binary changed the value of another string that
4375 happened to be identical to the end of the original first string.
4377 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4378 name", it returns that address as the IP address. On some operating
4379 systems (e.g. Solaris), it also passes back the IP address string as the
4380 "host name". However, on others (e.g. Linux), it passes back an empty
4381 string. Exim wasn't checking for this, and was changing the host name to an
4382 empty string, assuming it had been canonicized.
4384 5. Although rare, it is permitted to have more than one PTR record for a given
4385 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4386 all the names associated with an address, because they do in Solaris.
4387 However, it seems that they do not in Linux for data that comes from the
4388 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4389 I found this out when I moved to a new Linux workstation and tried to run
4390 the Exim test suite.
4392 To get round this problem I have changed the code so that it now does its
4393 own call to the DNS to look up PTR records when searching for a host name.
4394 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4395 addresses that are only in /etc/hosts are still found.
4397 This behaviour is, however, controlled by an option called host_lookup_
4398 order, which defaults to "bydns:byaddr". If people want to use the other
4399 order, or indeed, just use one or the other means of lookup, they can
4400 specify it in this variable.
4402 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4403 some operating systems, this comes back from gethostbyaddr() as an empty
4404 string, and this is what Exim used to test for. However, it seems that in
4405 other systems, "." is yielded. Exim now tests for this case too.
4407 7. The values of check_spool_space and check_log_space are now held internally
4408 as a number of kilobytes instead of an absolute number of bytes. If a
4409 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4410 kilobyte. This means that much larger values can be stored.
4412 8. Exim monitor: an attempt to get the action menu when not actually pointing
4413 at a message produces an empty menu entitled "No message selected". This
4414 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4415 no entries in it ("Shell widget menu has zero width and/or height"). So I
4416 have added a single, blank menu entry in this case.
4418 9. Added ${quote_local_part.
4420 10. MIME decoding is now applied to the contents of Subject: header lines when
4423 11. Now that a reference to $sender_host_address automatically causes a reverse
4424 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4425 host lookup before query-style lookups in lists that might use this
4426 variable. This has therefore been abolished, and the "net-" prefix is no
4427 longer necessary for query-style lookups.
4429 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4430 have been a typo for LFLAGS, so it has been changed.
4432 13. The install script calls Exim with "-C /dev/null" in order to find the
4433 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4434 to be output. Howeve, since Exim outputs its version number before the
4435 error, it didn't break the script. It just looked ugly. I fixed this by
4436 always allowing "-C /dev/null" if the caller is root.
4438 14. Ignore overlarge ACL variable number when reading spool file - insurance
4439 against a later release with more variables having written the file.
4441 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4442 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4443 this, as well as the form without the "IPv6" on the front.
4445 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4446 OS/Makefile-Darwin file.
4448 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4449 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4451 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4452 variable. However, some OS use ints for inodes. I've added cast to long int
4453 to get rid of the compiler warning.
4455 19. I had forgotten to lock out "/../" in configuration file names when
4456 ALT_CONFIG_PREFIX was set.
4458 20. Routers used for verification do not need to specify transports. However,
4459 if such a router generated a host list, and callout was configured, Exim
4460 crashed, because it could not find a port number from the (non-existent)
4461 transport. It now assumes port 25 in this circumstance.
4463 21. Added the -t option to exigrep.
4465 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4466 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4467 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4468 them is set, LOOKUP_LSEARCH is forced.
4470 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4472 24. Added sender and host information to the "rejected by local_scan()" log
4473 line; previously there was no indication of these.
4475 25. Added .include_if_exists.
4477 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4478 Linux. It turns out that not all file systems support this. Apparently some
4479 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4480 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4481 not supported on the file descriptor, is now ignored when Exim is trying to
4482 sync a directory. This applies only to Linux.
4484 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4486 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4487 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4490 29. Added support for maildirsize files from supplied patch (modified a bit).
4492 30. The use of :fail: followed by an empty string could lead Exim to respond to
4493 sender verification failures with (e.g.):
4495 550 Verification failed for <xxx>
4496 550 Sender verify failed
4498 where the first response line was missing the '-' that indicates it is not
4499 the final line of the response.
4501 31. The loop for finding the name of the user that called Exim had a hardwired
4502 limit of 10; it now uses the value of finduser_retries, which is used for
4503 all other user lookups.
4505 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4508 33. Exim was neglecting to zero errno before one call of strtol() when
4509 expanding a string and expecting an integer value. On some systems this
4510 resulted in spurious "integer overflow" errors. Also, it was casting the
4511 result into an int without checking.
4513 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4514 did not work. The code looks as if it has *never* worked, though it appears
4515 to have been documented since at least releast 1.62. I have made it work.
4517 35. The "timeout_DNS" error in retry rules, also documented since at least
4518 1.62, also never worked. As it isn't clear exactly what this means, and
4519 clearly it isn't a major issue, I have abolished the feature by treating it
4520 as "timeout", and writing a warning to the main and panic logs.
4522 36. The display of retry rules for -brt wasn't always showing the error code
4525 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4526 timeout_connect_A, timeout_connect_MX.
4528 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4529 to the empty sender.
4531 39. The daemon was not analysing the content of -oX till after it had closed
4532 stderr and disconnected from the controlling terminal. This meant that any
4533 syntax errors were only noted on the panic log, and the return code from
4534 the command was 0. By re-arranging the code a little, I've made the
4535 decoding happen first, so such errors now appear on stderr, and the return
4536 code is 1. However, the actual setting up of the sockets still happens in
4537 the disconnected process, so errors there are still only recorded on the
4540 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4541 connections (as happens on some IP stacks) was logged at start up time as
4542 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4543 it from "IPv6 and IPv4", which means that two separate sockets are being
4546 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4547 says whether AF_INET or AF_INET6 was passed as an argument.
4549 42. Exiwhat output was messed up when time zones were included in log
4552 43. Exiwhat now gives more information about the daemon's listening ports,
4553 and whether -tls-on-connect was used.
4555 44. The "port" option of the smtp transport is now expanded.
4557 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4558 silently ignored. Now an error message is written to the main and panic
4561 46. There's a new ACL modifier called "logwrite" which writes to a log file
4562 as soon as it is encountered.
4564 47. Added $local_user_uid and $local_user_gid at routing time.
4566 48. Exim crashed when trying to verify a sender address that was being
4569 49. Exim was recognizing only a space character after ".include". It now also
4570 recognizes a tab character.
4572 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4573 extracting the relevant information from the specification. The man page no
4574 longer contains scrambled data for the -d option, and I've added a section
4575 at the front about calling Exim under different names.
4577 51. Added "extra_headers" argument to the "mail" command in filter files.
4579 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4582 53. Installed eximstats 1.29.
4584 54. Added transport_filter_timeout as a generic transport option.
4586 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4587 Cc: header lines. This was required by RFC 822, but it not required by RFC
4590 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4591 incoming messages that did not have them. Now it does so only if the
4592 message originates locally, that is, if there is no associated remote host
4593 address. When Resent- header lines are present, this applies to the Resent-
4594 lines rather than the non-Resent- lines.
4596 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4597 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4599 58. Messages for configuration errors now include the name of the main
4600 configuration file - useful now that there may be more than one file in a
4601 list (.included file names were always shown).
4603 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4604 for those rare installations that do not start the daemon as root or run it
4605 setuid root. I've cut out the call to initgroups() if the daemon is not
4608 60. The Exim user and group can now be bound into the binary as text strings
4609 that are looked up at the start of Exim's processing.
4611 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4613 62. Added $mailstore_basename variable.
4615 63. Installed patch to sieve.c from Michael Haardt.
4617 64. When Exim failed to open the panic log after failing to open the main log,
4618 the original message it was trying to log was written to stderr and debug
4619 output, but if they were not available (the usual case in production), it
4620 was lost. Now it is written to syslog before the two lines that record the
4621 failures to open the logs.
4623 65. Users' Exim filters run in subprocesses under the user's uid. It is
4624 possible for a "deliver" command or an alias in a "personal" command to
4625 provoke an address rewrite. If logging of address rewriting is configured,
4626 this fails because the process is not running as root or exim. There may be
4627 a better way of dealing with this, but for the moment (because 4.30 needs
4628 to be released), I have disabled address rewrite logging when running a
4629 filter in a non-root, non-exim process.
4635 1. The buildconfig auxiliary program wasn't quoting the value set for
4636 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4637 not defined. This bug was masked in 4.22 by the effect that was fixed in
4640 2. Some messages that were rejected after a message id was allocated were
4641 shown as "incomplete" by exigrep. It no longer does this for messages that
4642 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4644 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4645 have allow_domain_literals set, the ID did not get logged in the <= line.
4646 Domain literals are now always recognized in Message-ID: header lines.
4648 4. The first argument for a ${extract expansion item is the key name or field
4649 number. Leading and trailing spaces in this item were not being ignored,
4650 causing some misleading effects.
4652 5. When deliver_drop_privilege was set, single queue runner processes started
4653 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4654 same command in the process it spins off) were not dropping privilege.
4656 6. When the daemon running as "exim" started a queue runner, it always
4657 re-executed Exim in the spun-off process. This is a waste of effort when
4658 deliver_drop_privilege is set. The new process now just calls the
4659 queue-runner function directly.
4665 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4668 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4669 ignored. Though the use of -oP was forcing the writing of a pid file, it
4670 was always written to the default place.
4672 3. If the message "no IP address found for host xxxx" is generated during
4673 incoming verification, it is now followed by identification of the incoming
4674 connection (so you can more easily find what provoked it).
4676 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4678 5. Added some features to "harden" Exim a bit more against certain attacks:
4680 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4681 be put in Local/Makefile. This is like the never_users runtime option,
4682 but it cannot be overridden. The default setting is "root".
4684 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4685 prefix string with which any file named in a -C command line option
4688 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4689 is retained for -C and -D only if the caller of Exim is root. Without
4690 it, the exim user may also use -C and -D and retain privilege.
4692 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4693 command line option is disabled.
4695 6. Macro names set by the -D option must start with an upper case letter, just
4696 like macro names defined in the configuration file.
4698 7. Added "dereference=" facility to LDAP.
4700 8. Two instances of the typo "uknown" in the source files are fixed.
4702 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4703 the Configure-Makefile script screwed up while processing it.
4705 10. Incorporated PCRE 4.4.
4707 11. The SMTP synchronization check was not operating right at the start of an
4708 SMTP session. For example, it could not catch a HELO sent before the client
4709 waited for the greeting. There is now a check for outstanding input at the
4710 point when the greeting is written. Because of the duplex, asynchronous
4711 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4712 way, but not yet received, when the check is performed.
4714 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4715 on TCP/IP sockets, because this apparently causes some broken clients to
4718 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4719 unchanged) from the Cygwin maintainer.
4721 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4722 was supported instead of testing for mbx. Effectively a typo.
4724 15. The spa authenticator server code was not checking that the input it
4725 received was valid base64.
4727 16. The debug output line for the "set" modifier in ACLs was not showing the
4728 name of the variable that was being set.
4730 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4732 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4733 lookup on its own. Something else was needed to trigger the lookup. For
4734 example, a match in host_lookup or the need for a host name in a host list.
4735 Now, if $sender_host_name is referenced and the host name has not yet been
4736 looked up, a lookup is performed. If the lookup fails, the variable remains
4737 empty, and $host_lookup_failed is set to "1".
4739 19. Added "eqi" as a case-independent comparison operator.
4741 20. The saslauthd authentication condition could segfault if neither service
4742 nor realm was specified.
4744 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4745 error message that was logged was misleading, and incoming SMTP
4746 connections were dropped. The message is now more accurate, and temporary
4747 errors are given to SMTP connections.
4749 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4750 (except RSET) with a 421 error, until QUIT is received. However, it was
4751 failing to send a response to QUIT.
4753 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4754 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4755 ACL is now run after the helo_try_verify_hosts code.
4757 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4758 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4759 case-independent, so other case variants are also recognized). Apparently
4760 some systems use these upper case variants.
4762 25. If more than two messages were waiting for the same host, and a transport
4763 filter was specified for the transport, Exim sent two messages over the
4764 same TCP/IP connection, and then failed with "socket operation on non-
4765 socket" when it tried to send the third.
4767 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4769 27. The extern definition of crypt16() in expand.c was not being excluded when
4770 the OS had its own crypt16() function.
4772 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4773 as a preferred synonym for return_size_limit, both as an option and as an
4776 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4778 30. Changed the default configuration ACL to relax the local part checking rule
4779 for addresses that are not in any local domains. For these addresses,
4780 slashes and pipe symbols are allowed within local parts, but the sequence
4781 /../ is explicitly forbidden.
4783 31. SPA server authentication was not clearing the challenge buffer before
4786 32. log_message in a "warn" ACL statement was writing to the reject log as
4787 well as to the main log, which contradicts the documentation and doesn't
4788 seem right (because no rejection is happening). So I have stopped it.
4790 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4791 However, I am unable to do any testing of this.
4793 34. Fixed an infelicity in the appendfile transport. When checking directories
4794 for a mailbox, to see if any needed to be created, it was accidentally
4795 using path names with one or more superfluous leading slashes; tracing
4796 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4798 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4799 contents are added to the log line that is written for every discarded
4800 recipient. (Previously a log_message setting was ignored.)
4802 36. The ${quote: operator now quotes the string if it is empty.
4804 37. The install script runs exim in order to find its version number. If for
4805 some reason other than non-existence or emptiness, which it checks, it
4806 could not run './exim', it was installing it with an empty version number,
4807 i.e. as "exim-". This error state is now caught, and the installation is
4810 38. An argument was missing from the function that creates an error message
4811 when Exim fails to connect to the socket for saslauthd authentication.
4812 This could cause Exim to crash, or give a corrupted message.
4814 39. Added isip, isip4, and isip6 to ${if conditions.
4816 40. The ACL variables $acl_xx are now saved with the message, and can be
4817 accessed later in routers, transports, and filters.
4819 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4820 strings in the file are not string-expanded.
4822 42. If a MAIL command specified a SIZE value that was too large to fit into an
4823 int variable, the check against message_size_limit failed. Such values are
4824 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4825 day this will have to be increased, but I don't think I want to be around
4826 when emails are that large.
4833 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4834 iconv() is not standard in FreeBSD.
4836 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4837 IPv6 enabled. The observed symptom was a segmentation fault on return from
4838 the function os_common_find_running_interfaces() in src/os.c.
4840 3. In the check_special_case() function in daemon.c I had used "errno" as an
4841 argument name, which causes warnings on some systems. This was basically a
4842 typo, since it was named "eno" in the comments!
4844 4. The code that waits for the clock to tick (at a resolution of some fraction
4845 of a second) so as to ensure message-id uniqueness was always waiting for
4846 at least one whole tick, when it could have waited for less. [This is
4847 almost certainly not relevant at current processor speeds, where it is
4848 unlikely to ever wait at all. But we try to future-proof.]
4850 5. The function that sleeps for a time interval that includes fractions of a
4851 second contained a race. It did not block SIGALRM between setting the
4852 timer, and suspending (a couple of lines later). If the interval was short
4853 and the sigsuspend() was delayed until after it had expired, the suspension
4854 never ended. On busy systems this could lead to processes getting stuck for
4857 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4858 process, before it forks any delivery processes. The open lookup caching
4859 mechanism meant that the open file or database connection was passed into
4860 the delivery process. The problem was that delivery processes always tidy
4861 up cached lookup data. This could cause a problem for the next delivery
4862 process started by the queue runner, because the external queue runner
4863 process does not know about the closure. So the next delivery process
4864 still has data in the lookup cache. In the case of a file lookup, there was
4865 no problem because closing a file descriptor in a subprocess doesn't affect
4866 the parent. However, if the lookup was caching a connection to a database,
4867 the connection was closed, and the second delivery process was likely to
4868 see errors such as "PGSQL: query failed: server closed the connection
4869 unexpectedly". The problem has been fixed by closing all cached lookups
4870 in a queue runner before running a delivery process.
4872 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4873 seem to have the "const" qualifier which it has on other OS. I've
4876 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4877 *of the same type* (client or server) with the same public name that an
4878 error should be diagnosed.
4880 9. When Exim looked up a host name for an IP address, but failed to find the
4881 original IP address when looking up the host name (a safety check), it
4882 output the message "<ip address> does not match any IP for NULL", which was
4883 confusing, to say the least. The bug was that the host name should have
4884 appeared instead of "NULL".
4886 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4887 user that is built into the binary, and the -C or -D options is used, root
4888 privilege is dropped before the configuration file is read. In addition,
4889 logging is switched to stderr instead of the normal log files. If the
4890 configuration then re-defines the Exim user, the unprivileged environment
4891 is probably not what is expected, so Exim logs a panic warning message (but
4894 However, if deliver_drop_privilege is set, the unprivileged state may well
4895 be exactly what is intended, so the warning has been cut out in that case,
4896 and Exim is allowed to try to write to its normal log files.
4902 1. smtp_return_error_details was not giving details for temporary sender
4903 or receiver verification errors.
4905 2. Diagnose a configuration error if two authenticators have the same public
4908 3. Exim used not to create the message log file for a message until the first
4909 delivery attempt. This could be confusing when incoming messages were held
4910 for policy or load reasons. The message log file is now created at the time
4911 the message is received, and an initial "Received" line is written to it.
4913 4. The automatically generated man page for command line options had a minor
4914 bug that caused no ill effects; however, a more serious problem was that
4915 the procedure for building the man page automatically didn't always
4916 operate. Consequently, release 4.20 contains an out-of-date version. This
4917 shouldn't happen again.
4919 5. When building Exim with embedded Perl support, the script that builds the
4920 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4921 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4923 6. The freeze_tell option was not being used for messages that were frozen on
4924 arrival, either by an ACL or by local_scan().
4926 7. Added the smtp_incomplete_transaction log selector.
4928 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4929 was accepting AUTH without a new EHLO.
4931 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4932 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4935 10. Logging of TCP/IP connections (when configured) now happens in the main
4936 daemon process instead of the child process, so that the TCP/IP connection
4937 count is more accurate (but it can never be perfect).
4939 11. The use of "drop" in a nested ACL was not being handled correctly in the
4940 outer ACL. Now, if condition failure induced by the nested "drop" causes
4941 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4942 or "require"), the connection is dropped.
4944 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4945 that yield "discard" can now be used with an "accept" or a "discard" verb,
4946 but an error is generated for any others (because I can't see a useful way
4947 to define what should happen).
4949 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4950 are now processed in the same way as lines in the Exim configuration file.
4951 In particular, continuation lines are supported.
4953 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4955 15. Added -ti meaning -t -i.
4957 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4958 domains, and warn by logging if others are found.
4960 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4961 returned from GIFCONF when Exim is trying to find the list of interfaces on
4962 a host. The code in os.c has been modified to copy each ifreq to an aligned
4963 structure in all cases.
4965 Also, in some cases, the returned ifreq's were being copied to a 'struct
4966 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4967 means the last couple of bytes of an IPv6 address could be chopped if the
4968 ifreq contained only a normal sockaddr (14 bytes storage).
4970 18. Named domain lists were not supported in the hosts_treat_as_local option.
4971 An entry such as +xxxx was not recognized, and was treated as a literal
4974 19. Ensure that header lines added by a DATA ACL are included in the reject log
4975 if the ACL subsequently rejects the message.
4977 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4978 MD5 (which is deprecated).
4980 21. When testing a filter file using -bf, Exim was writing a message when it
4981 took the sender from a "From " line in the message, but it was not doing so
4982 when it took $return_path from a Return-Path: header line. It now does.
4984 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4985 with a valid header line field name (a series of printing characters
4986 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4988 23. Changed "disc" in the source to "disk" to conform to the documentation and
4989 the book and for uniformity.
4991 24. Ignore Sendmail's -Ooption=value command line item.
4993 25. When execve() failed while trying to run a command in a pipe transport,
4994 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4995 could be confused with a return value of 69 from the command itself. This
4996 has been changed to 127, the value the shell returns if it is asked to run
4997 a non-existent command. The wording for the related log line suggests a
4998 non-existent command as the problem.
5000 26. If received_header_text expands to an empty string, do not add a Received:
5001 header line to the message. (Well, it adds a token one on the spool, but
5002 marks it "old" so that it doesn't get used or transmitted.)
5004 27. Installed eximstats 1.28 (addition of -nt option).
5006 28. There was no check for failure on the call to getsockname() in the daemon
5007 code. This can fail if there is a shortage of resources on the system, with
5008 ENOMEM, for example. A temporary error is now given on failure.
5010 29. Contrary to the C standard, it seems that in some environments, the
5011 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5012 program. Exim now does this explicitly; it affects the formatting of
5013 timestamps using strftime().
5015 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5016 complaints. I've now initialized all the variables, to avoid this.
5018 32. Header lines added by a system filter were not being "seen" during
5019 transport-time rewrites.
5021 33. The info_callback() function passed to OpenSSL is set up with type void
5022 (*)(SSL *, int, int), as described somewhere. However, when calling the
5023 function (actually a macro) that sets it up, the type void(*)() is
5024 expected. I've put in a cast to prevent warnings from picky compilers.
5026 34. If a DNS black list lookup found a CNAME record, but there were no A
5027 records associated with the domain it pointed at, Exim crashed.
5029 35. If a DNS black list lookup returned more than one A record, Exim ignored
5030 all but the first. It now scans all returned addresses if a particular IP
5031 value is being sought. In this situation, the contents of the
5032 $dnslist_value variable are a list of all the addresses, separated by a
5035 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5036 to accept a host name and all its aliases if the forward lookup for any of
5037 them yielded the IP address of the incoming connection. Now it accepts only
5038 those names whose forward lookup yields the correct IP address. Any other
5039 names are discarded. This closes a loophole whereby a rogue DNS
5040 administrator could create reverse DNS records to break through a
5041 wildcarded host restriction in an ACL.
5043 37. If a user filter or a system filter that ran in a subprocess used any of
5044 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5045 the wrong values were passed to the pipe command ($thisaddress had the
5046 value of $0, $0 had the value of $1, etc). This bug was introduced by
5047 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5049 38. Improved the line breaking for long SMTP error messages from ACLs.
5050 Previously, if there was no break point between 40 and 75 characters, Exim
5051 left the rest of the message alone. Two changes have been made: (a) I've
5052 reduced the minimum length to 35 characters; (b) if it can't find a break
5053 point between 35 and 75 characters, it looks ahead and uses the first one
5054 that it finds. This may give the occasional overlong line, but at least the
5055 remaining text gets split now.
5057 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5058 file descriptors might be low, and that setting 1000 would always raise it.
5059 It turns out that in some environments, the limit is already over 1000 and
5060 that lowering it causes trouble. So now Exim takes care not to decrease it.
5062 40. When delivering a message, the value of $return_path is set to $sender_
5063 address at the start of routing (routers may change the value). By an
5064 oversight, this default was not being set up when an address was tested by
5065 -bt or -bv, which affected the outcome if any router or filter referred to
5068 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5069 log only when "message" or "log_message" are set. However, if one of the
5070 conditions was an address verification, or a call to a nested ACL, the
5071 messages generated by the underlying test were being passed through. This
5072 no longer happens. The underlying message is available in $acl_verify_
5073 message for both "message" and "log_message" expansions, so it can be
5074 passed through if needed.
5076 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5077 new expansion $bh_ to give the encoded byte string without charset
5078 translation. Translation happens only if iconv() is available; HAVE_ICONV
5079 indicates this at build time. HEADERS_CHARSET gives the charset to
5080 translate to; headers_charset can change it in the configuration, and
5081 "headers charset" can change it in an individual filter file.
5083 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5084 that creates RFC 2047 encoded "words" labels them as that charset instead
5085 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5086 expansion operator; (ii) when Exim creates a From: line for a local
5087 message; (iii) when a header line is rewritten to include a "phrase" part.
5089 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5090 buggy, causing it to skip the first lines of messages whose message ID
5091 ended in 'D'. This would not have bitten before Exim release 4.14, because
5092 message IDs were unlikely to end in 'D' before then. The effect was to have
5093 incorrect size information for certain domains.
5095 45. #include "config.h" was missing at the start of the crypt16.c module. This
5096 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5099 46. If there was a timeout during a "random" callout check, Exim treated it as
5100 a failure of the random address, and carried on sending RSET and the real
5101 address. If the delay was just some slowness somewhere, the response to the
5102 original RCPT would be taken as a response to RSET and so on, causing
5103 mayhem of various kinds.
5105 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5106 when I implemented it. It didn't allow for the fact that some option values
5107 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5108 the right test for positive values.
5110 48. Domain names in DNS records are case-independent. Exim always looks them up
5111 in lower case. Some resolvers return domain names in exactly the case they
5112 appear in the zone file, that is, they may contain uppercase letters. Not
5113 all resolvers do this - some return always lower case. Exim was treating a
5114 change of case by a resolver as a change of domain, similar to a widening
5115 of a domain abbreviation. This triggered its re-routing code and so it was
5116 trying to route what was effectively the same domain again. This normally
5117 caused routing to fail (because the router wouldn't handle the domain
5118 twice). Now Exim checks for this case specially, and just changes the
5119 casing of the domain that it ultimately uses when it transmits the message
5122 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5125 50. If a filter generated a file delivery with a non-absolute name (possible if
5126 no home directory exists for the router), the forbid_file option was not
5129 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5130 the existing equality matching.
5132 52. Exim was using ints instead of ino_t variables in some places where it was
5133 dealing with inode numbers.
5135 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5136 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5137 called TMPDIR, and if it finds it is different, it changes its value.
5139 54. The smtp_printf() function is now made available to local_scan() so
5140 additional output lines can be written before returning. There is also an
5141 smtp_fflush() function to enable the detection of a dropped connection.
5142 The variables smtp_input and smtp_batched_input are exported to
5145 55. Changed the default runtime configuration: the message "Unknown user"
5146 has been removed from the ACL, and instead placed on the localuser router,
5147 using the cannot_route_message feature. This means that any verification
5148 failures that generate their own messages won't get overridden. Similarly,
5149 the "Unrouteable address" message that was in the ACL for unverifiable
5150 relay addresses has also been removed.
5152 56. Added hosts_avoid_esmtp to the smtp transport.
5154 57. The exicyclog script was not checking for the esoteric option
5155 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5156 will work only if exicyclog is run under the appropriate euid.
5158 58. Following a discussion on the list, the rules by which Exim recognises line
5159 endings on incoming messages have been changed. The -dropcr and drop_cr
5160 options are now no-ops, retained only for backwards compatibility. The
5161 following line terminators are recognized: LF CRLF CR. However, special
5162 processing applies to CR:
5164 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5165 nor a local message in the state where . is a terminator.
5167 (ii) If a bare CR is encountered in a header line, an extra space is added
5168 after the line terminator so as not to end the header. The reasoning
5169 behind this is that bare CRs in header lines are most likely either
5170 to be mistakes, or people trying to play silly games.
5172 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5173 was being incorrectly given as 18 bytes larger than it should have been.
5174 This is a VOB (very old bug).
5176 60. This may never have affected anything current, but just in case it has:
5177 When the local host is found other than at the start of a list of hosts,
5178 the local host, those with the same MX, and any that follow, are discarded.
5179 When the list in question was part of a longer list of hosts, the following
5180 hosts (not currently being processed) were also being discarded. This no
5181 longer happens. I'm not sure if this situation could ever has previously
5184 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5187 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5188 header. This is something that is recommended in a new Internet Draft, and
5189 is something that is documented as being done by Sendmail. There are two
5190 possible values. For messages generated by the autoreply transport, Exim
5193 Auto-Submitted: auto-replied
5195 whereas for all other generated messages (e.g. bounces) it adds
5197 Auto-Submitted: auto-generated
5199 63. The "personal" condition in filters now includes a test for the
5200 Auto-Submitted: header. If it contains the string "auto-" the message it
5201 not considered personal.
5203 64. Added rcpt_include_affixes as a generic transport option.
5205 65. Added queue_only_override (default true).
5207 66. Added the syslog_duplication option.
5209 67. If what should have been the first header line of a message consisted of
5210 a space followed by a colon, Exim was mis-interpreting it as a header line.
5211 It isn't of course - it is syntactically invalid and should therefore be
5212 treated as the start of the message body. The misbehaviour could have
5213 caused a number of strange effects, including loss of data in subsequent
5214 header lines, and spool format errors.
5216 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5217 client host had authenticated. This control can now be exercised by an ACL
5218 for more flexibility.
5220 69. By default, callouts do not happen when testing with -bh. There is now a
5221 variant, -bhc, which does actually run the callout code, including
5222 consulting and updating the callout cache.
5224 70. Added support for saslauthd authentication, courtesy of Alexander
5227 71. If statvfs() failed on the spool or log directories while checking their
5228 size for availability, Exim confusingly gave the error "space shortage".
5229 Furthermore, in debugging mode it crashed with a floating point exception.
5230 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5231 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5232 serious problem, Exim now writes to the main and panic logs when this
5233 happens, with details of the failure. It then refuses to accept the
5234 incoming message, giving the message "spool directory problem" or "log
5235 directory problem" with a 421 code for SMTP messages.
5237 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5238 0, 1, and 2 exist, because some OS complain for execs without them (see
5239 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5240 descriptors. However, the code omitted to check that the open succeeded,
5241 causing mysterious errors if for some reason the permissions on /dev/null
5242 got screwed. Now Exim writes a message to the main and panic logs, and
5243 bombs out if it can't open /dev/null.
5245 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5246 interact so that it is all more flexible. It is supposed to remain
5247 backwards compatible. Also added extra_local_interfaces.
5249 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5250 to bomb out with an assertion failure - to the client this appears as a
5251 connection drop. This problem occurs in the part of the code that was taken
5252 from the Samba project. Fortunately, the assertion is in a very simple
5253 function, so I have fixed this by reproducing the function inline in the
5254 one place where it is called, and arranging for authentication to fail
5255 instead of killing the process with assert().
5257 75. The SPA client code was not working when the server requested OEM rather
5258 than Unicode encoding.
5260 76. Added code to make require_files with a specific uid setting more usable in
5261 the case where statting the file as root fails - usually a non-root-mounted
5262 NFS file system. When this happens and the failure is EACCES, Exim now
5263 forks a subprocess and does the per-uid checking as the relevant uid.
5265 77. Added process_log_path.
5267 78. If log_file_path was not explicitly set, a setting of check_log_space or
5268 check_log_inodes was ignored.
5270 79. If a space check for the spool or log partitions fails, the incident is now
5271 logged. Of course, in the latter case the data may get lost...
5273 80. Added the %p formatting code to string_format() so that it can be used to
5274 print addresses in debug_print(). Adjusted all the address printing in the
5275 debugging in store.c to use %p rather than %d.
5277 81. There was a concern that a line of code in smtp_in.c could overflow a
5278 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5279 initially expressed, the concern was not well-founded, because trailing
5280 spaces are removed early. However, if the trailing spaces were followed by
5281 a NULL, they did not get removed, so the overflow was possible. Two fixes
5284 (a) I re-wrote the offending code in a cleaner fashion.
5285 (b) If an incoming SMTP command contains a NULL character, it is rejected
5288 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5289 runs initgroups(), so that if the Exim user is in any additional groups,
5290 they will be used during message reception.
5296 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob