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
62 TF/01 Failure to get a lock on a hints database can have serious
63 consequences so log it to the panic log.
65 TF/02 Log LMTP confirmation messages in the same way as SMTP,
66 controlled using the smtp_confirmation log selector.
68 TF/03 Include the error message when we fail to unlink a spool file.
70 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
71 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
72 for maintaining out-of-tree patches for some time.
74 PP/01 Bugzilla 139: Documentation and portability issues.
75 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
76 Handle per-OS dynamic-module compilation flags.
78 PP/02 Let /dev/null have normal permissions.
79 The 4.73 fixes were a little too stringent and complained about the
80 permissions on /dev/null. Exempt it from some checks.
81 Reported by Andreas M. Kirchwitz.
83 PP/03 Report version information for many libraries, including
84 Exim version information for dynamically loaded libraries. Created
85 version.h, now support a version extension string for distributors
86 who patch heavily. Dynamic module ABI change.
88 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
89 privilege escalation vulnerability whereby the Exim run-time user
90 can cause root to append content of the attacker's choosing to
93 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
96 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
97 If dropping privileges for untrusted macros, we disabled normal logging
98 on the basis that it would fail; for the Exim run-time user, this is not
99 the case, and it resulted in successful deliveries going unlogged.
100 Fixed. Reported by Andreas Metzler.
106 PP/01 Date: & Message-Id: revert to normally being appended to a message,
107 only prepend for the Resent-* case. Fixes regression introduced in
108 Exim 4.70 by NM/22 for Bugzilla 607.
110 PP/02 Include check_rfc2047_length in configure.default because we're seeing
111 increasing numbers of administrators be bitten by this.
113 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
115 PP/03 Bugzilla 994: added openssl_options main configuration option.
117 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
119 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
121 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
123 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
124 without a peer certificate, leading to a segfault because of an
125 assumption that peers always have certificates. Be a little more
126 paranoid. Problem reported by Martin Tscholak.
128 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
129 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
130 NB: ClamAV planning to remove STREAM in "middle of 2010".
131 CL also introduces -bmalware, various -d+acl logging additions and
132 more caution in buffer sizes.
134 PP/09 Implemented reverse_ip expansion operator.
136 PP/10 Bugzilla 937: provide a "debug" ACL control.
138 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
140 PP/12 Bugzilla 973: Implement --version.
142 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
144 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
146 PP/15 Bugzilla 816: support multiple condition rules on Routers.
148 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
149 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
150 ignore trailing whitespace.
152 JJ/02 prevent non-panic DKIM error from being sent to paniclog
154 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
157 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
158 Notification from Dr Andrew Aitchison.
160 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
161 ExtendedDetectionInfo response format.
162 Notification from John Horne.
164 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
167 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
168 XSL and documented dependency on system catalogs, with examples of how
171 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
174 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
175 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
176 configuration file which is writeable by the Exim user or group.
178 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
179 of configuration files to cover files specified with the -C option if
180 they are going to be used with root privileges, not just the default
183 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
184 option (effectively making it always true).
186 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
187 files to be used while preserving root privileges.
189 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
190 that rogue child processes cannot use them.
192 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
193 run-time user, instead of root.
195 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
196 Exim run-time user without dropping privileges.
198 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
199 result string, instead of calling string_vformat() twice with the same
202 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
203 for other users. Others should always drop root privileges if they use
204 -C on the command line, even for a whitelisted configure file.
206 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
208 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
214 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
215 $data_path, and $header_path variables; fixed documentation bugs and
218 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
219 exipick to access non-standard spools, including the "frozen" queue
222 NM/01 Bugzilla 965: Support mysql stored procedures.
223 Patch from Alain Williams
225 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
227 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
228 Patch from Andreas Metzler
230 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
231 Patch from Kirill Miazine
233 NM/05 Bugzilla 671: Added umask to procmail example.
235 JJ/03 installed exipick 20100323.0, fixing doc bug
237 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
238 directory. Notification and patch from Dan Rosenberg.
240 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
242 TK/02 Improve log output when DKIM signing operation fails.
244 MH/01 Treat the transport option dkim_domain as a colon separated
245 list, not as a single string, and sign the message with each element,
246 omitting multiple occurences of the same signer.
248 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
249 Bugzilla 985, 986. Patch by Simon Arlott
251 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
252 Patch by Simon Arlott
254 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
255 MBX locking. Notification from Dan Rosenberg.
261 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
263 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
265 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
267 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
269 NM/04 Bugzilla 915: Flag broken perl installation during build.
275 TK/01 Added patch by Johannes Berg that expands the main option
276 "spamd_address" if it starts with a dollar sign.
278 TK/02 Write list of recipients to X-Envelope-Sender header when building
279 the mbox-format spool file for content scanning (suggested by Jakob
282 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
283 (http://www.dcc-servers.net/) support via dccifd. Activated by
284 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
286 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
287 by Mark Daniel Reidel <mr@df.eu>.
289 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
290 When building exim an external PCRE library is now needed -
291 PCRE is a system library on the majority of modern systems.
292 See entry on PCRE_LIBS in EDITME file.
294 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
295 conversation. Added nologin parameter to request.
296 Patch contributed by Kirill Miazine.
298 TF/01 Do not log submission mode rewrites if they do not change the address.
300 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
302 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
303 log files in place. Contributed by Roberto Lima.
305 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
307 TF/03 Bugzilla 615: When checking the local_parts router precondition
308 after a local_part_suffix or local_part_prefix option, Exim now
309 does not use the address's named list lookup cache, since this
310 contains cached lookups for the whole local part.
312 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
313 Robert Millan. Documentation is in experimental-spec.txt.
315 TF/04 Bugzilla 668: Fix parallel build (make -j).
317 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
319 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
320 Patch provided by Jan Srzednicki.
322 TF/05 Leading white space used to be stripped from $spam_report which
323 wrecked the formatting. Now it is preserved.
325 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
326 that they are available at delivery time.
328 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
330 TF/08 TLS error reporting now respects the incoming_interface and
331 incoming_port log selectors.
333 TF/09 Produce a more useful error message if an SMTP transport's hosts
334 setting expands to an empty string.
336 NM/06 Bugzilla 744: EXPN did not work under TLS.
337 Patch provided by Phil Pennock.
339 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
340 Patch provided by Richard Godbee.
342 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
343 acl_smtp_notquit, added index entry.
345 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
346 Patch provided by Eugene Bujak.
348 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
349 accept(). Patch provided by Maxim Dounin.
351 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
352 Patch provided by Phil Pennock.
354 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
356 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
357 Patch provided by Brad "anomie" Jorsch.
359 NM/14 Bugzilla 622: Added timeout setting to transport filter.
360 Patch provided by Dean Brooks.
362 TK/05 Add native DKIM support (does not depend on external libraries).
364 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
365 Patch provided by Graeme Fowler.
367 NM/16 Bugzilla 851: Documentation example syntax fix.
369 NM/17 Changed NOTICE file to remove references to embedded PCRE.
371 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
374 NM/19 Bugzilla 745: TLS version reporting.
375 Patch provided by Phil Pennock.
377 NM/20 Bugzilla 167: bool: condition support.
378 Patch provided by Phil Pennock.
380 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
381 clients. Patch provided by Phil Pennock.
383 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
384 Patch provided by Brad "anomie" Jorsch.
386 NM/23 Bugzilla 687: Fix misparses in eximstats.
387 Patch provided by Heiko Schlittermann.
389 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
390 Patch provided by Heiko Schlittermann.
392 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
393 plus update to original patch.
395 NM/26 Bugzilla 799: Documentation correction for ratelimit.
397 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
398 Patch provided by David Brownlee.
400 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
402 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
404 NM/30 Bugzilla 888: TLS documentation bugfixes.
406 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
408 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
409 Unlike the original bugzilla I have changed all shell scripts in src tree.
411 NM/33 Bugzilla 898: Transport filter timeout fix.
412 Patch by Todd Rinaldo.
414 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
415 Patch by Serge Demonchaux.
417 NM/35 Bugzilla 39: Base64 decode bug fixes.
418 Patch by Jakob Hirsch.
420 NM/36 Bugzilla 909: Correct connect() call in dcc code.
422 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
424 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
426 NM/39 Bugzilla 911: Fixed MakeLinks build script.
432 TK/01 Add preliminary DKIM support. Currently requires a forked version of
433 ALT-N's libdkim that I have put here:
434 http://duncanthrax.net/exim-experimental/
436 Note to Michael Haardt: I had to rename some vars in sieve.c. They
437 were called 'true' and it seems that C99 defines that as a reserved
438 keyword to be used with 'bool' variable types. That means you could
439 not include C99-style headers which use bools without triggering
440 build errors in sieve.c.
442 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
443 as mailq or other aliases. Changed the --help handling significantly
444 to do whats expected. exim_usage() emits usage/help information.
446 SC/01 Added the -bylocaldomain option to eximstats.
448 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
450 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
452 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
458 PH/01 Another patch from the Sieve maintainer.
460 PH/02 When an IPv6 address is converted to a string for single-key lookup
461 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
462 dots are used instead of colons so that keys in lsearch files need not
463 contain colons. This was done some time before quoting was made available
464 in lsearch files. However, iplsearch files do require colons in IPv6 keys
465 (notated using the quote facility) so as to distinguish them from IPv4
466 keys. This meant that lookups for IP addresses in host lists did not work
467 for iplsearch lookups.
469 This has been fixed by arranging for IPv6 addresses to be expressed with
470 colons if the lookup type is iplsearch. This is not incompatible, because
471 previously such lookups could never work.
473 The situation is now rather anomolous, since one *can* have colons in
474 ordinary lsearch keys. However, making the change in all cases is
475 incompatible and would probably break a number of configurations.
477 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
480 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
481 conversion specification without a maximum field width, thereby enabling
482 a rogue spamd server to cause a buffer overflow. While nobody in their
483 right mind would setup Exim to query an untrusted spamd server, an
484 attacker that gains access to a server running spamd could potentially
485 exploit this vulnerability to run arbitrary code as the Exim user.
487 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
488 $primary_hostname instead of what libspf2 thinks the hosts name is.
490 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
491 a directory entry by the name of the lookup key. Previously, if a
492 symlink pointed to a non-existing file or a file in a directory that
493 Exim lacked permissions to read, a lookup for a key matching that
494 symlink would fail. Now it is enough that a matching directory entry
495 exists, symlink or not. (Bugzilla 503.)
497 PH/03 The body_linecount and body_zerocount variables are now exported in the
500 PH/04 Added the $dnslist_matched variable.
502 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
503 This means they are set thereafter only if the connection becomes
506 PH/06 Added the client_condition to authenticators so that some can be skipped
507 by clients under certain conditions.
509 PH/07 The error message for a badly-placed control=no_multiline_responses left
510 "_responses" off the end of the name.
512 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
514 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
515 (without spaces) instead of just copying the configuration text.
517 PH/10 Added the /noupdate option to the ratelimit ACL condition.
519 PH/11 Added $max_received_linelength.
521 PH/12 Added +ignore_defer and +include_defer to host lists.
523 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
524 way in which PCRE > 7.0 is built.
526 PH/14 Implemented queue_only_load_latch.
528 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
529 MAIL command. The effect was to mangle the value on 64-bit systems.
531 PH/16 Another patch from the Sieve maintainer.
533 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
535 PH/18 If a system quota error occurred while trying to create the file for
536 a maildir delivery, the message "Mailbox is full" was not appended to the
537 bounce if the delivery eventually timed out. Change 4.67/27 below applied
538 only to a quota excession during the actual writing of the file.
540 PH/19 It seems that peer DN values may contain newlines (and other non-printing
541 characters?) which causes problems in log lines. The DN values are now
542 passed through string_printing() before being added to log lines.
544 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
545 and InterBase are left for another time.)
547 PH/21 Added message_body_newlines option.
549 PH/22 Guard against possible overflow in moan_check_errorcopy().
551 PH/23 POSIX allows open() to be a macro; guard against that.
553 PH/24 If the recipient of an error message contained an @ in the local part
554 (suitably quoted, of course), incorrect values were put in $domain and
555 $local_part during the evaluation of errors_copy.
561 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
562 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
565 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
566 issue a MAIL command.
568 PH/02 In an ACL statement such as
570 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
572 if a client was not listed at all, or was listed with a value other than
573 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
574 the condition was not true (as it should be), so access was not denied.
575 The bug was that the ! inversion was incorrectly passed on to the second
576 item. This has been fixed.
578 PH/03 Added additional dnslists conditions == and =& which are different from
579 = and & when the dns lookup returns more than one IP address.
581 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
582 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
584 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
585 FSYNC, which compiles an option called disable_fsync that allows for
586 bypassing fsync(). The documentation is heavily laced with warnings.
588 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
590 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
591 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
592 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
593 including adding "make clean"; (3) Added -fPIC when compiling the test
594 dynamically loaded module, to get rid of a warning.
596 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
597 message fails, move_frozen_messages = true and ignore_bounce_errors_after
598 = 0s. The bug is otherwise harmless.
600 PH/07 There was a bug in the dovecot authenticator such that the value of
601 $auth1 could be overwritten, and so not correctly preserved, after a
602 successful authentication. This usually meant that the value preserved by
603 the server_setid option was incorrect.
605 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
607 PH/09 Installed PCRE release 7.0.
609 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
610 run for batched SMTP input. It is now run at the start of every message
611 in the batch. While fixing this I discovered that the process information
612 (output by running exiwhat) was not always getting set for -bs and -bS
613 input. This is fixed, and it now also says "batched" for BSMTP.
615 PH/11 Added control=no_pipelining.
617 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
618 patch, slightly modified), and move the expansion of helo_data till after
619 the connection is made in the smtp transport (so it can use these
622 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
624 PH/14 Added log_selector = +pid.
626 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
628 PH/16 Add ${if forany and ${if forall.
630 PH/17 Added dsn_from option to vary the From: line in DSNs.
632 PH/18 Flush SMTP output before performing a callout, unless control =
633 no_callout_flush is set.
635 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
636 was true (the default) a successful delivery failed to delete the retry
637 item, thus causing premature timeout of the address. The bug is now
640 PH/20 Added hosts_avoid_pipelining to the smtp transport.
642 PH/21 Long custom messages for fakedefer and fakereject are now split up
643 into multiline reponses in the same way that messages for "deny" and
644 other ACL rejections are.
646 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
647 with slight modification.
649 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
650 draft, changing the syntax and factoring some duplicate code".
652 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
653 for deliveries of the second and subsequent messages over the same SMTP
656 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
657 ${reduce, with only minor "tidies".
659 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
661 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
662 expansion side effects.
664 PH/27 When a message times out after an over-quota error from an Exim-imposed
665 quota, the bounce message says "mailbox is full". This message was not
666 being given when it was a system quota that was exceeded. It now should
669 MH/03 Made $recipients available in local_scan(). local_scan() already has
670 better access to the recipient list through recipients_list[], but
671 $recipients can be useful in postmaster-provided expansion strings.
673 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
674 in the case of a MAIL command with additional options following the
675 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
676 were accidentally chopped off.
678 PH/29 SMTP synchronization checks are implemented when a command is read -
679 there is a check that no more input is waiting when there shouldn't be
680 any. However, for some commands, a delay in an ACL can mean that it is
681 some time before the response is written. In this time, more input might
682 arrive, invalidly. So now there are extra checks after an ACL has run for
683 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
684 pipelining has not been advertised.
686 PH/30 MH's patch to allow iscntrl() characters to be list separators.
688 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
689 returned in the SMTP response when smtp_return_error_details was false.
692 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
693 instead of the C I/O that was originally supplied, because problems were
696 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
697 Exim which did not show up earlier: it was assuming that a call to
698 SSL_CTX_set_info_callback() might give an error value. In fact, there is
699 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
700 was a macro that became an assignment, so it seemed to work. This has
701 changed to a proper function call with a void return, hence the compile
702 error. Exim's code has been fixed.
704 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
707 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
709 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
711 PH/37 If a message is not accepted after it has had an id assigned (e.g.
712 because it turns out to be too big or there is a timeout) there is no
713 "Completed" line in the log. When some messages of this type were
714 selected by exigrep, they were listed as "not completed". Others were
715 picked up by some special patterns. I have improved the selection
716 criteria to be more general.
718 PH/38 The host_find_failed option in the manualroute router can now be set
719 to "ignore", to completely ignore a host whose IP address cannot be
720 found. If all hosts are ignored, the behaviour is controlled by the new
721 host_all_ignored option.
723 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
724 homing or because of multiple MX records with /mx) generated more than
725 one IP address, and the following item turned out to be the local host,
726 all the secondary addresses of the first item were incorrectly removed
727 from the list, along with the local host and any following hosts (which
728 is what is supposed to happen).
730 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
731 whoever called Exim into the -H file. In the case of the daemon it was
732 behaving confusingly. When first started, it used values for whoever
733 started the daemon, but after a SIGHUP it used the Exim user (because it
734 calls itself on a restart). I have changed the code so that it now always
737 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
738 message are rejected with the same error (e.g. no authentication or bad
739 sender address), and a DATA command is nevertheless sent (as can happen
740 with PIPELINING or a stupid MUA), the error message that was given to the
741 RCPT commands is included in the rejection of the DATA command. This is
742 intended to be helpful for MUAs that show only the final error to their
745 PH/42 Another patch from the Sieve maintainer.
747 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
748 Eximstats - Fixed some broken HTML links and added missing column headers
750 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
751 columns for Rejects, Temp Rejects, Ham, and Spam rows.
753 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
755 PH/43 Yet another patch from the Sieve maintainer.
757 PH/44 I found a way to check for a TCP/IP connection going away before sending
758 the response to the final '.' that terminates a message, but only in the
759 case where the client has not sent further data following the '.'
760 (unfortunately, this is allowed). However, in many cases there won't be
761 any further data because there won't be any more messages to send. A call
762 to select() can be used: if it shows that the input is "ready", there is
763 either input waiting, or the socket has been closed. An attempt to read
764 the next input character can distinguish the two cases. Previously, Exim
765 would have sent an OK response which the client would never have see.
766 This could lead to message repetition. This fix should cure that, at
767 least in a lot of common cases.
769 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
770 advertised in response to EHLO.
776 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
777 fixed by 4.65/MH/01 (is this a record?) are fixed:
779 (i) An empty string was always treated as zero by the numeric comparison
780 operators. This behaviour has been restored.
782 (ii) It is documented that the numeric comparison operators always treat
783 their arguments as decimal numbers. This was broken in that numbers
784 starting with 0 were being interpreted as octal.
786 While fixing these problems I realized that there was another issue that
787 hadn't been noticed. Values of message_size_limit (both the global option
788 and the transport option) were treated as octal if they started with 0.
789 The documentation was vague. These values are now always treated as
790 decimal, and I will make that clear in the documentation.
796 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
797 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
800 MH/01 Don't check that the operands of numeric comparison operators are
801 integers when their expansion is in "skipping" mode (fixes bug
802 introduced by 4.64-PH/07).
804 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
805 child addresses, Exim now panics and dies. Previously, because the count
806 is held in a short int, deliveries were likely to be lost. As such a
807 large number of recipients for a single message is ridiculous
808 (performance will be very, very poor), I have chosen to impose a limit
809 rather than extend the field.
815 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
816 leftover -K file (the existence of which was triggered by #402).
817 While we were at it, introduced process PID as part of the -K
818 filename. This should rule out race conditions when creating
821 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
822 processing considerably. Previous code took too long for large mails,
823 triggering a timeout which in turn triggers #401.
825 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
826 in the DK code in transports.c. sendfile() is not really portable,
827 hence the _LINUX specificness.
829 TF/01 In the add_headers option to the mail command in an Exim filter,
830 there was a bug that Exim would claim a syntax error in any
831 header after the first one which had an odd number of characters
834 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
835 callout verification, Exim cached a "reject" for the entire domain. This
836 is correct for most verifications, but it is not correct for a recipient
837 verification with use_sender or use_postmaster set, because in that case
838 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
839 case of MAIL FROM:<> rejection from other early rejections (e.g.
840 rejection of HELO). When verifying a recipient using a non-null MAIL
841 address, the cache is ignored if it shows MAIL FROM:<> rejection.
842 Whatever the result of the callout, the value of the domain cache is
843 left unchanged (for any other kind of callout, getting as far as trying
844 RCPT means that the domain itself is ok).
846 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
849 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
850 manifest itself as EPIPE rather than ECONNECT. When tidying away a
851 session, the daemon ignores ECONNECT errors and logs others; it now
852 ignores EPIPE as well.
854 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
855 (quoted-printable decoding).
857 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
858 later the small subsequent patch to fix an introduced bug.
860 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
862 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
864 PH/08 An error is now given if message_size_limit is specified negative.
866 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
867 to be given (somewhat) arbitrary names.
869 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
872 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
873 miscellaneous code fixes
875 PH/10 Added the log_reject_target ACL modifier to specify where to log
878 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
879 hostname. This is wrong, because it relates to the incoming message (and
880 probably the interface on which it is arriving) and not to the outgoing
881 callout (which could be using a different interface). This has been
882 changed to use the value of the helo_data option from the smtp transport
883 instead - this is what is used when a message is actually being sent. If
884 there is no remote transport (possible with a router that sets up host
885 addresses), $smtp_active_hostname is used.
887 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
888 tweaks were necessary in order to get it to work (see also 21 below):
889 (a) The code assumed that strncpy() returns a negative number on buffer
890 overflow, which isn't the case. Replaced with Exim's string_format()
892 (b) There were several signed/unsigned issues. I just did the minimum
893 hacking in of casts. There is scope for a larger refactoring.
894 (c) The code used strcasecmp() which is not a standard C function.
895 Replaced with Exim's strcmpic() function.
896 (d) The code set only $1; it now sets $auth1 as well.
897 (e) A simple test gave the error "authentication client didn't specify
898 service in request". It would seem that Dovecot has changed its
899 interface. Fortunately there's a specification; I followed it and
900 changed what the client sends and it appears to be working now.
902 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
905 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
906 address A is aliased to B and C, where B exists and C does not. Without
907 -v the output is "A verified" because verification stops after a
908 successful redirection if more than one address is generated. However,
909 with -v the child addresses are also verified. Exim was outputting "A
910 failed to verify" and then showing the successful verification for C,
911 with its parentage. It now outputs "B failed to verify", showing B's
912 parentage before showing the successful verification of C.
914 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
915 look up a TXT record in a specific list after matching in a combined
918 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
919 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
920 they consult the DNS. I had assumed they would set it the way they
921 wanted; and indeed my experiments on Linux seem to show that in some
922 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
923 To be on the safe side, however, I have now made the interface to
924 host_find_byname() similar to host_find_bydns(), with an argument
925 containing the DNS resolver options. The host_find_byname() function now
926 sets these options at its start, just as host_find_bydns() does. The smtp
927 transport options dns_qualify_single and dns_search_parents are passed to
928 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
929 of host_find_byname() use the default settings of RES_DEFNAMES
930 (qualify_single) but not RES_DNSRCH (search_parents).
932 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
933 spool_read_header() do less string testing, by means of a preliminary
934 switch on the second character of optional "-foo" lines. (This is
935 overdue, caused by the large number of possibilities that now exist.
936 Originally there were few.) While I was there, I also converted the
937 str(n)cmp tests so they don't re-test the leading "-" and the first
938 character, in the hope this might squeeze out yet more improvement.
940 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
941 flag allowing group syntax was set by the header_syntax check but not
942 turned off, possible causing trouble later; (2) The flag was not being
943 set at all for the header_verify test, causing "group"-style headers to
944 be rejected. I have now set it in this case, and also caused header_
945 verify to ignore an empty address taken from a group. While doing this, I
946 came across some other cases where the code for allowing group syntax
947 while scanning a header line wasn't quite right (mostly, not resetting
948 the flag correctly in the right place). These bugs could have caused
949 trouble for malformed header lines. I hope it is now all correct.
951 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
952 with the "reply" argument non-NULL. The code, however (which originally
953 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
954 but it didn't always do it. This confused somebody who was copying the
955 code for some other use. I have removed all the tests.
957 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
958 feature that was used to support insecure browsers during the U.S. crypto
959 embargo. It requires special client support, and Exim is probably the
960 only MTA that supported it -- and would never use it because real RSA is
961 always available. This code has been removed, because it had the bad
962 effect of slowing Exim down by computing (never used) parameters for the
963 RSA_EXPORT functionality.
965 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
966 authenticator to fail if there's a tab character in the incoming data
967 (there should never be unless someone is messing about, as it's supposed
968 to be base64-encoded). Also added, on Timo's advice, the "secured" option
969 if the connection is using TLS or if the remote IP is the same as the
970 local IP, and the "valid-client-cert option" if a client certificate has
973 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
974 authenticators. This can be used for authorization after authentication
975 succeeds. (In the case of plaintext, it servers for both authentication
978 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
979 if any retry times were supplied.
981 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
982 connection, where there is no client IP address to check. In this
983 situation, the verify now always succeeds.
985 PH/25 Applied John Jetmore's -Mset patch.
987 PH/26 Added -bem to be like -Mset, but loading a message from a file.
989 PH/27 In a string expansion for a processed (not raw) header when multiple
990 headers of the same name were present, leading whitespace was being
991 removed from all of them, but trailing whitespace was being removed only
992 from the last one. Now trailing whitespace is removed from each header
993 before concatenation. Completely empty headers in a concatenation (as
996 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
997 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
999 PH/29 [Removed. This was a change that I later backed out, and forgot to
1000 correct the ChangeLog entry (that I had efficiently created) before
1001 committing the later change.]
1003 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1004 address errors (4xx response to RCPT) over the same connection as other
1005 messages routed to the same hosts. Such deliveries are always "forced",
1006 so retry times are not inspected. This resulted in far too many retries
1007 for the affected addresses. The effect occurred only when there were more
1008 hosts than the hosts_max_try setting in the smtp transport when it had
1009 the 4xx errors. Those hosts that it had tried were not added to the list
1010 of hosts for which the message was waiting, so if all were tried, there
1011 was no problem. Two fixes have been applied:
1013 (i) If there are any address or message errors in an SMTP delivery, none
1014 of the hosts (tried or untried) are now added to the list of hosts
1015 for which the message is waiting, so the message should not be a
1016 candidate for sending over the same connection that was used for a
1017 successful delivery of some other message. This seems entirely
1018 reasonable: after all the message is NOT "waiting for some host".
1019 This is so "obvious" that I'm not sure why it wasn't done
1020 previously. Hope I haven't missed anything, but it can't do any
1021 harm, as the worst effect is to miss an optimization.
1023 (ii) If, despite (i), such a delivery is accidentally attempted, the
1024 routing retry time is respected, so at least it doesn't keep
1025 hammering the server.
1027 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1028 in ${readsocket because some servers need this prod.
1030 PH/32 Added some extra debug output when updating a wait-xxx database.
1032 PH/33 The hint "could be header name not terminated by colon", which has been
1033 given for certain expansion errors for a long time, was not being given
1034 for the ${if def:h_colon_omitted{... case.
1036 PH/34 The spec says: "With one important exception, whenever a domain list is
1037 being scanned, $domain contains the subject domain." There was at least
1038 one case where this was not true.
1040 PH/35 The error "getsockname() failed: connection reset by peer" was being
1041 written to the panic log as well as the main log, but it isn't really
1042 panic-worthy as it just means the connection died rather early on. I have
1043 removed the panic log writing for the ECONNRESET error when getsockname()
1046 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1047 runs only) independently of the message's sender address. This meant
1048 that, if the 4xx error was in fact related to the sender, a different
1049 message to the same recipient with a different sender could confuse
1050 things. In particualar, this can happen when sending to a greylisting
1051 server, but other circumstances could also provoke similar problems.
1052 I have changed the default so that the retry time for these errors is now
1053 based a combination of the sender and recipient addresses. This change
1054 can be overridden by setting address_retry_include_sender=false in the
1057 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1058 remote server are returned as part of bounce messages. This was not
1059 happening for LMTP over a pipe (the lmtp transport), but now it is the
1060 same for both kinds of LMTP.
1062 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1063 in header lines that were in fact CNAMEs. This is no longer the case.
1065 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1066 and queue runs started by the daemon processed all messages. This has
1067 been fixed so that -R and -S can now usefully be given with -q<time>.
1069 PH/40 Import PCRE release 6.7 (fixes some bugs).
1071 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1073 PH/42 Give an error if -q is specified more than once.
1075 PH/43 Renamed the variables $interface_address and $interface_port as
1076 $received_ip_address and $received_port, to make it clear that these
1077 values apply to message reception, and not to the outgoing interface when
1078 a message is delivered. (The old names remain recognized, of course.)
1080 PH/44 There was no timeout on the connect() call when using a Unix domain
1081 socket in the ${readsocket expansion. There now is.
1083 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1084 be meaningful with "accept".
1086 SC/01 Eximstats V1.43
1087 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1089 SC/02 Eximstats V1.44
1090 Use a glob alias rather than an array ref in the generated
1091 parser. This improves both readability and performance.
1093 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1094 Collect SpamAssassin and rejection statistics.
1095 Don't display local sender or destination tables unless
1096 there is data to show.
1097 Added average volumes into the top table text output.
1099 SC/04 Eximstats V1.46
1100 Collect data on the number of addresses (recipients)
1101 as well as the number of messages.
1103 SC/05 Eximstats V1.47
1104 Added 'Message too big' to the list of mail rejection
1105 reasons (thanks to Marco Gaiarin).
1107 SC/06 Eximstats V1.48
1108 Mainlog lines which have GMT offsets and are too short to
1109 have a flag are now skipped.
1111 SC/07 Eximstats V1.49 (Alain Williams)
1112 Added the -emptyok flag.
1114 SC/08 Eximstats V1.50
1115 Fixes for obtaining the IP address from reject messages.
1117 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1118 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1119 whitesspace changes from 4.64-PH/27
1121 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1124 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1125 are found, allow negative numbers in numeric criteria)
1127 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1129 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1130 to match changes made in 4.64-PH/43
1132 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1134 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1135 "domain" is actually an IP address, and force a failure. This locks out
1136 those revolvers/nameservers that support "A-for-A" lookups, in
1137 contravention of the specifications.
1139 PH/48 When a host name was looked up from an IP address, and the subsequent
1140 forward lookup of the name timed out, the host name was left in
1141 $sender_host_name, contrary to the specification.
1143 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1144 restricted to single-key lookups, Exim was not diagnosing an error if
1145 * or *@ was used with a query-style lookup.
1147 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1149 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1150 long ago, but noone interested enough thought of it. Let's just say that
1151 the "1.1" means that there are some new functions that weren't there at
1152 some point in the past.
1154 PH/51 Error processing for expansion failure of helo_data from an smtp
1155 transport during callout processing was broken.
1157 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1158 tested/used via the -bh/-bhc/-bs options.
1160 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1161 bug, fixed in subsequent PCRE releases).
1163 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1164 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1166 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1172 SC/01 Use a glob alias rather than an array ref in eximstats generated
1173 parser. This improves both readability and performance.
1175 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1176 Don't display local sender or destination tables in eximstats unless
1177 there is data to show.
1178 Added average volumes into the eximstats top table text output.
1180 SC/03 Collect data on the number of addresses (recipients) as well
1181 as the number of messages in eximstats.
1183 TF/01 Correct an error in the documentation for the redirect router. Exim
1184 does (usually) call initgroups() when daemonizing.
1186 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1187 with consistent privilege compared to when running as a daemon.
1189 TF/03 Note in the spec that $authenticated_id is not set for local
1190 submissions from trusted users.
1192 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1193 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1195 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1196 by adding some example configuration directives to the default
1197 configuration file. A little bit of work is required to uncomment the
1198 directives and define how usernames and passwords are checked, but
1199 there is now a framework to start from.
1201 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1202 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1203 without this. I don't know how relevant this is to other LDAP libraries.
1205 PH/02 Add the verb name to the "unknown ACL verb" error.
1207 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1209 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1211 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1212 directory not expanded when it should be if an expanded home directory
1213 was set for the address (which is overridden by the transport).
1215 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1218 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1219 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1220 because it is too late at that time, and has no effect.
1222 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1223 security issue with \' (bugzilla #107). I could not use the
1224 PQescapeStringConn() function, because it needs a PGconn value as one of
1227 PH/08 When testing addresses using -bt, indicate those final addresses that
1228 are duplicates that would not cause an additional delivery. At least one
1229 person was confused, thinking that -bt output corresponded to deliveries.
1230 (Suppressing duplicates isn't a good idea as you lose the information
1231 about possibly different redirections that led to the duplicates.)
1233 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1234 systems where poll() doesn't work, in particular OS X.
1236 PH/10 Added more information to debugging output for retry time not reached.
1238 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1239 operations in malware.c.
1241 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1244 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1245 syslog_duplication set false, log lines that would normally be written
1246 both the the main log and to the reject log were not written to syslog at
1249 PH/14 In the default configuration, change the use of "message" in ACL warn
1250 statements to "add_header".
1252 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1253 not followed by a command (e.g. "seen endif").
1255 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1256 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1259 PH/17 Added extra conditions to the default value of delay_warning_condition
1263 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1264 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1265 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1268 The Auto-Submitted: and various List- headers are standardised, whereas I
1269 don't think Precedence: ever was.
1271 PH/18 Refactored debugging code in route_finduser() to show more information,
1272 in particular, the error code if getpwnam() issues one.
1274 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1275 This is apparently needed in addition to the PH/07 change above to avoid
1276 any possible encoding problems.
1278 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1279 but not after initializing Perl.
1281 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1282 output them only if debugging. By default they are written stderr,
1283 apparently, which is not desirable.
1285 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1288 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1291 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1293 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1294 authenticated or an ident call has been made. Suppress the default
1295 values for $authenticated_id and $authenticated_sender (but permit -oMai
1296 and -oMas) when testing with -bh.
1298 PH/24 Re-jigged the order of the tests in the default configuration so that the
1299 tests for valid domains and recipients precede the DNS black list and CSA
1300 tests, on the grounds that those ones are more expensive.
1302 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1303 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1304 This bug exists in every version of Exim that I still have, right back to
1307 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1308 However, an attempt to turn on case-sensitivity in a regex key by
1309 including (?-i) didn't work because the subject string was already
1310 lowercased, and the effects were non-intuitive. It turns out that a
1311 one-line patch can be used to allow (?-i) to work as expected.
1317 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1318 other effects) broke the use of negated acl sub-conditions.
1320 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1323 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1324 "Deny" causes Exim to reject the incoming connection with a 554 error.
1325 Unfortunately, if there is a major crisis, such as a disk failure,
1326 tcp-wrappers gives "deny", whereas what one would like would be some
1327 kind of temporary error. A kludge has been added to help with this.
1328 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1329 554 error is used if errno is still zero or contains ENOENT (which occurs
1330 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1333 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1335 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1336 errors. Otherwise a message that provokes a temporary error (when other
1337 messages do not) can cause a whole host to time out.
1339 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1340 addresses were routed directly to files or pipes from a redirect router.
1341 File deliveries just didn't batch; pipe deliveries might have suffered
1344 PH/06 A failure to get a lock for a hints database would erroneously always say
1345 "Failed to get write lock", even when it was really a read lock.
1347 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1348 of 0600. This has been changed to use the value of the lockfile_mode
1349 option (which defaults to 0600).
1351 PH/08 Applied small patch from the Sieve maintainer.
1353 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1354 folder from quota calculations, a direct delivery into this folder messed
1355 up the contents of the maildirsize file. This was because the regex was
1356 used only to exclude .Trash (or whatever) when the size of the mailbox
1357 was calculated. There was no check that a delivery was happening into an
1358 excluded directory. This bug has been fixed by ignoring all quota
1359 processing for deliveries into excluded directories.
1361 PH/10 Added the maildirfolder_create_regex option to appendfile.
1367 PH/01 The code for finding all the local interface addresses on a FreeBSD
1368 system running IPv6 was broken. This may well have applied to all BSD
1369 systems, as well as to others that have similar system calls. The broken
1370 code found IPv4 interfaces correctly, but gave incorrect values for the
1371 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1372 that it would not match correctly against @[] and not recognize the IPv6
1375 PH/02 The ipliteral router was not recognizing addresses of the form user@
1376 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1378 PH/03 Added disable_ipv6.
1380 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1381 decoded form, because it is most often used to construct To: headers
1382 lines in autoreplies, and the decoded form may well be syntactically
1383 invalid. However, $reply_address has leading white space removed, and all
1384 newlines turned into spaces so that the autoreply transport does not
1387 PH/05 If group was specified without a user on a router, and no group or user
1388 was specified on a transport, the group from the router was ignored.
1390 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1391 for visible compile-time settings that can be used to change these
1392 numbers, for those that want even more. Backwards compatibility with old
1393 spool files has been maintained. However, going back to a previous Exim
1394 release will lost any variables that are in spool files.
1396 PH/07 Two small changes when running in the test harness: increase delay when
1397 passing a TCP/IP connection to a new process, in case the original
1398 process has to generate a bounce, and remove special handling of
1399 127.0.0.2 (sic), which is no longer necessary.
1401 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1402 be the same on different OS.
1404 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1407 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1408 whether --show-vars was specified or not
1410 JJ/02 exipick: Added support for new ACL variable spool format introduced
1413 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1414 syntactically invalid From: or Reply-to: line, and a filter used this to
1415 generate an autoreply, and therefore failed to obtain an address for the
1416 autoreply, Exim could try to deliver to a non-existent relative file
1417 name, causing unrelated and misleading errors. What now happens is that
1418 it logs this as a hard delivery error, but does not attempt to create a
1421 PH/11 The exinext utility has a -C option for testing purposes, but although
1422 the given file was scanned by exinext itself; it wasn't being passed on
1423 when Exim was called.
1425 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1426 an end-of-file indication when reading a command response.
1428 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1429 compiled. In many other places in Exim, IPv6 addresses are always
1430 recognized, so I have changed this. It also means that IPv4 domain
1431 literals of the form [IPV4:n.n.n.n] are now always recognized.
1433 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1434 used if the router is not running as root, for example, when verifying at
1435 ACL time, or when using -bh. The debugging output from this situation was
1436 non-existent - all you got was a failure to exec. I have made two
1439 (a) Failures to set uid/gid, the current directory, or a process leader
1440 in a subprocess such as that created by queryprogram now generate
1441 suitable debugging ouput when -d is set.
1443 (b) The queryprogram router detects when it is not running as root,
1444 outputs suitable debugging information if -d is set, and then runs
1445 the subprocess without attempting to change uid/gid.
1447 PH/15 Minor change to Makefile for building test_host (undocumented testing
1450 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1451 additional section of a DNS packet that returns MX or SRV records.
1452 Instead, it always explicitly searches for A/AAAA records. This avoids
1453 major problems that occur when a DNS server includes only records of one
1454 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1455 fixed another bug: if SRV records were looked up and the corresponding
1456 address records were *not* found in the additional section, the port
1457 values from the SRV records were lost.
1459 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1460 using the correct key (the original address) when searching the retry
1461 rules in order to find which one to use for generating the retry hint.
1463 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1464 adding the default one. Similarly, if it contains a Reply-To: header, the
1465 errors_reply_to option, if set, is not used.
1467 PH/19 When calculating a retry time, Exim used to measure the "time since
1468 failure" by looking at the "first failed" field in the retry record. Now
1469 it does not use this if it is later than than the arrival time of the
1470 message. Instead it uses the arrival time. This makes for better
1471 behaviour in cases where some deliveries succeed, thus re-setting the
1472 "first failed" field. An example is a quota failure for a huge message
1473 when small messages continue to be delivered. Without this change, the
1474 "time since failure" will always be short, possible causing more frequent
1475 delivery attempts for the huge message than are intended.
1476 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1478 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1479 $1, $2, $3) because the numerical variables can be reset during some
1480 expansion items (e.g. "match"), thereby losing the authentication data.
1482 PH/21 Make -bV show the size of off_t variables so that the test suite can
1483 decide whether to run tests for quotas > 2G.
1485 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1486 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1487 filecount value is greater than 2G or if a quota value is greater than 2G
1488 on a system where the size of off_t is not greater than 4, a panic error
1491 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1492 never match. The debug and -bh output now contains an explicit error
1493 message indicating a malformed IPv4 address or mask.
1495 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1496 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1497 PH/23 above applies.
1499 PH/25 Do not write to syslog when running in the test harness. The only
1500 occasion when this arises is a failure to open the main or panic logs
1501 (for which there is an explicit test).
1503 PH/26 Added the /no_tell option to "control=freeze".
1505 PH/27 If a host name lookup failed very early in a connection, for example, if
1506 the IP address matched host_lookup and the reverse lookup yielded a name
1507 that did not have a forward lookup, an error message of the form "no IP
1508 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1509 could be logged. Now it outputs the IP address instead of "NULL".
1511 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1512 allows the sender and the authenticated sender to be set when
1513 submitting a message from within Exim. Since child_open_exim() is
1514 documented for local_scan(), the new function should be too.
1516 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1517 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1518 results in an empty string is now treated as unset.
1520 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1522 PH/31 Added sender_verify_fail logging option.
1524 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1525 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1526 tidied the source and removed it altogether.
1528 PH/33 When a queue run was abandoned because the load average was too high, a
1529 log line was always written; now it is written only if the queue_run log
1530 selector is set. In addition, the log line for abandonment now contains
1531 information about the queue run such as the pid. This is always present
1532 in "start" and "stop" lines but was omitted from the "abandon" line.
1534 PH/34 Omit spaces between a header name and the colon in the error message that
1535 is given when verify = headers_syntax fails (if there are lots of them,
1536 the message gets confusing).
1538 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1539 names, as there are now some PTR records that contain slashes. This check
1540 is only to protect against broken name servers that fall over on strange
1541 characters, so the fact that it applies to all lookups doesn't matter.
1543 PH/36 Now that the new test suite is complete, we can remove some of the
1544 special code in Exim that was needed for the old test suite. For example,
1545 sorting DNS records because real resolvers return them in an arbitrary
1546 order. The new test suite's fake resolver always returns records in the
1549 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1550 bounces) except when queue_only is set, to avoid logging races between
1551 the different processes.
1553 PH/38 Panic-die if .include specifies a non-absolute path.
1555 PH/39 A tweak to the "H" retry rule from its user.
1557 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1558 a label. They prevented compilation on older perls.
1560 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1561 a warning to be raised on newish perls.
1563 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1564 on queue. Changes to match documented behaviour of showing count of
1565 messages matching specified criteria.
1567 PH/40 Changed the default ident timeout from 30s to 5s.
1569 PH/41 Added support for the use of login_cap features, on those BSD systems
1570 that have them, for controlling the resources used by pipe deliveries.
1572 PH/42 The content-scanning code uses fopen() to create files in which to put
1573 message data. Previously it was not paying any attention to the mode of
1574 the files. Exim runs with umask(0) because the rest of the code creates
1575 files with open(), and sets the required mode explicitly. Thus, these
1576 files were ending up world-writeable. This was not a big issue, because,
1577 being within the spool directory, they were not world-accessible. I have
1578 created a function called modefopen, which takes an additional mode
1579 argument. It sets umask(777), creates the file, chmods it to the required
1580 mode, then resets the umask. All the relevant calls to fopen() in the
1581 content scanning code have been changed to use this function.
1583 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1584 to 24 hours. This avoids potential overflow problems when processing G
1585 and H retry rules. I suspect nobody ever tinkers with this value.
1587 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1589 PH/45 When the plaintext authenticator is running as a client, the server's
1590 challenges are checked to ensure they are valid base64 strings. By
1591 default, the authentication attempt is cancelled if an invalid string is
1592 received. Setting client_ignore_invalid_base64 true ignores these errors.
1593 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1594 they are received. Thus, the responses can be made to depend on the
1595 challenges. If an invalid string is ignored, an empty string is placed in
1598 PH/46 Messages that are created by the autoreply transport now contains a
1599 References: header, in accordance with RFCs 2822 and 3834.
1601 PH/47 Added authenticated_sender_force to the smtp transport.
1603 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1605 PH/49 Installed latest patch from the Sieve maintainer.
1607 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1608 was also set, the appendfile transport was unnecessarily scanning a
1609 directory of message files (e.g. for maildir delivery) to find the count
1610 of files (along with the size), even though it did not need this
1611 information. It now does the scan only if it needs to find either the
1612 size of the count of files.
1614 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1616 PH/52 Two bugs concerned with error handling when the smtp transport is
1619 (i) Exim was not creating retry information for temporary errors given
1620 for individual recipients after the DATA command when the smtp transport
1621 was used in LMTP mode. This meant that they could be retried too
1622 frequently, and not timed out correctly.
1624 (ii) Exim was setting the flag that allows error details to be returned
1625 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1626 recipients that were returned after the DATA command.
1628 PH/53 This is related to PH/52, but is more general: for any failing address,
1629 when detailed error information was permitted to be returned to the
1630 sender, but the error was temporary, then after the final timeout, only
1631 "retry timeout exceeded" was returned. Now it returns the full error as
1632 well as "retry timeout exceeded".
1634 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1635 do this, and (what is worse) MTAs that accept it.
1637 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1638 will now be deprecated.
1640 PH/56 New os.c-cygwin from the Cygwin maintainer.
1642 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1643 formats (previously only available in exim formats via -bpr, -bpru,
1644 and -bpra. Now also available in native and exiqgrep formats)
1646 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1647 with very large, slow to parse queues
1649 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1651 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1653 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1654 responses to authentication challenges, though it was showing the
1655 challenges; (ii) I've removed the CR characters from the debug output for
1658 PH/58 Allow for the insertion of a newline as well as a space when a string
1659 is turned into more than one encoded-word during RFC 2047 encoding. The
1660 Sieve code now uses this.
1662 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1663 data_4xx, lost_connection, tls_required.
1665 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1666 message was being sent as an SMTP response.
1668 PH/61 Add -l and -k options to exicyclog.
1670 PH/62 When verifying, if an address was redirected to one new address, so that
1671 verification continued, and the new address failed or deferred after
1672 having set something in $address_data, the value of $address_data was not
1673 passed back to the ACL. This was different to the case when no
1674 redirection occurred. The value is now passed back in both cases.
1676 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1677 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1678 use login_cap.h, so on its own it isn't the distinguishing feature. The
1679 new name refers directly to the setclassresources() function.
1681 PH/65 Added configuration files for NetBSD3.
1683 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1685 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1686 is preferred over IPv4.
1688 PH/68 The bounce_return_message and bounce_return_body options were not being
1689 honoured for bounces generated during the reception of non-SMTP messages.
1690 In particular, this applied to messages rejected by the ACL. This bug has
1691 been fixed. However, if bounce_return_message is true and bounce_return_
1692 body is false, the headers that are returned for a non-SMTP message
1693 include only those that have been read before the error was detected.
1694 (In the case of an ACL rejection, they have all been read.)
1696 PH/69 The HTML version of the specification is now built in a directory called
1697 spec_html instead of spec.html, because the latter looks like a path with
1698 a MIME-type, and this confuses some software.
1700 PH/70 Catch two compiler warnings in sieve.c.
1702 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1703 function verify_get_ident() calls ip_connect() to connect a socket, but
1704 if the "connect()" function timed out, ip_connect() used to close the
1705 socket. However, verify_get_ident() also closes the socket later, and in
1706 between Exim writes to the log, which may get opened at this point. When
1707 the socket was closed in ip_connect(), the log could get the same file
1708 descriptor number as the socket. This naturally causes chaos. The fix is
1709 not to close the socket in ip_connect(); the socket should be closed by
1710 the function that creates it. There was only one place in the code where
1711 this was missing, in the iplookup router, which I don't think anybody now
1712 uses, but I've fixed it anyway.
1714 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1715 well as to direct DNS lookups. Otherwise the handling of names in host
1716 lists is inconsistent and therefore confusing.
1722 PH/01 Two changes to the default runtime configuration:
1724 (1) Move the checks for relay_from_hosts and authenticated clients from
1725 after to before the (commented out) DNS black list checks.
1727 (2) Add control=submission to the relay_from_hosts and authenticated
1728 clients checks, on the grounds that messages accepted by these
1729 statements are most likely to be submissions.
1731 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1733 (1) Generate an error if the third argument for the ${prvs expansion is
1736 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1739 (3) Reset the variables that are obtained from the first argument of
1740 ${prvscheck and used in the second argument before leaving the code,
1741 because their memory is reclaimed, so using them afterwards may do
1744 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1745 one (it's much easier than Tom thought :-).
1747 (5) Because of (4), we can now allow for the use of $prvscheck_result
1748 inside the third argument.
1750 PH/03 For some reason, the default setting of PATH when running a command from
1751 a pipe transport was just "/usr/bin". I have changed it to
1754 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1755 anything to be listed in the output from -bV.
1757 PH/05 When a filter generated an autoreply, the entire To: header line was
1758 quoted in the delivery log line, like this:
1760 => >A.N.Other <ano@some.domain> <original@ddress> ...
1762 This has been changed so that it extracts the operative address. There
1763 may be more than one such address. If so, they are comma-separated, like
1766 => >ano@some.domain,ona@other.domain <original@ddress> ...
1768 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1769 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1770 not being looked up in the rDNS to get a host name, Exim was showing the
1771 IP address twice in Received: lines, even though the IP addresses were
1772 identical. For example:
1774 Received: from [1.2.3.4] (helo=[1.2.3.4])
1776 However, if the real host name was known, it was omitting the HELO data
1777 if it matched the actual IP address. This has been tidied up so that it
1778 doesn't show the same IP address twice.
1780 PH/07 When both +timestamp and +memory debugging was on, the value given by
1781 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1782 called by the debug printing, thereby overwriting the timestamp buffer.
1783 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1785 PH/08 When the original message was included in an autoreply transport, it
1786 always said "this is a copy of the message, including all the headers",
1787 even if body_only or headers_only was set. It now gives an appropriate
1790 PH/09 Applied a patch from the Sieve maintainer which:
1792 o fixes some comments
1793 o adds the (disabled) notify extension core
1794 o adds some debug output for the result of if/elsif tests
1795 o points to the current vacation draft in the documentation
1796 and documents the missing references header update
1800 o fixes a bug in processing the envelope test (when testing
1801 multiple envelope elements, the last element determinted the
1804 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1805 Electronic Mail") by including:
1807 Auto-submitted: auto-generated
1809 in the messages that it generates (bounce messages and others, such as
1810 warnings). In the case of bounce messages for non-SMTP mesages, there was
1811 also a typo: it was using "Auto_submitted" (underscore instead of
1812 hyphen). Since every message generated by Exim is necessarily in response
1813 to another message, thes have all been changed to:
1815 Auto-Submitted: auto-replied
1817 in accordance with these statements in the RFC:
1819 The auto-replied keyword:
1821 - SHOULD be used on messages sent in direct response to another
1822 message by an automatic process,
1824 - MUST NOT be used on manually-generated messages,
1826 - MAY be used on Delivery Status Notifications (DSNs) and Message
1827 Disposition Notifications (MDNs),
1829 - MUST NOT be used on messages generated by automatic or periodic
1830 processes, except for messages which are automatic responses to
1833 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1834 to the default Received: header definition.
1836 PH/12 Added log selector acl_warn_skipped (default on).
1838 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1839 variables because (a) they are in the wrong storage pool and (b) even if
1840 they were copied, it wouldn't work properly because of the caching.
1842 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1843 checking when decoding. Apparently there are clients that generate
1844 overlong encoded strings. Why am I not surprised?
1846 PH/15 If the first argument of "${if match_address" was not empty, but did not
1847 contain an "@" character, Exim crashed. Now it writes a panic log message
1848 and treats the condition as false.
1850 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1852 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1853 extension "envelope-auth". The code is finished and in agreement with
1854 other implementations, but there is no documentation so far and in fact,
1855 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1856 not changing the active code.
1858 Print executed "if" and "elsif" statements when debugging is used. This
1859 helps a great deal to understand what a filter does.
1861 Document more things not specified clearly in RFC3028. I had all this
1862 sorted out, when out of a sudden new issues came to my mind. Oops."
1864 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1867 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1868 canonical form (as documented). However, after a host name lookup from
1869 the IP address, check_host() was doing a simple string comparison with
1870 addresses acquired from the DNS when checking that the found name did
1871 have the original IP as one of its addresses. Since any found IPv6
1872 addresses are likely to be in abbreviated form, the comparison could
1873 fail. Luckily, there already exists a function for doing the comparison
1874 by converting both addresses to binary, so now that is used instead of
1875 the text comparison.
1877 PH/20 There was another similar case to PH/19, when a complete host name was
1878 given in a host list; looking up its IP address could give an abbreviated
1879 form, whereas the current host's name might or might not be abbreviated.
1880 The same fix has been applied.
1886 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1887 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1890 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1891 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1893 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1895 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1896 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1897 Also, for "Auto-Submitted", treat anything other than "no" as
1898 non-personal, in accordance with RFC 3834. (Previously it treated
1899 anything starting "auto-" as non-personal.)
1901 TF/01 The control=submission/name=... option had a problem with syntax
1902 errors if the name included a slash character. The /name= option
1903 now slurps the rest of the string, so it can include any characters
1904 but it must come last in the list of options (after /sender_retain
1907 PH/05 Some modifications to the interface to the fake nameserver for the new
1915 TK/01 Added the "success_on_redirect" address verification option. See
1916 NewStuff for rationale and an example.
1918 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1920 PH/02 Patch to exigrep to allow it to work on syslog lines.
1922 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1923 fread() to skip over the body file's header line, because in Cygwin the
1924 header line is locked and is inaccessible.
1926 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1927 co-exist for some time) to make it clear that it is the Exim ID that is
1928 referenced, not the Message-ID: header line.
1930 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1931 string_format() function, because snprintf() does not exist on all
1934 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1935 ${stat: expansion item.
1937 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1938 protocol synchronization error", to keep the pedants happy.
1940 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1941 well as for IRIX systems, when gcc is being used. See the host.c source
1944 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1946 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1949 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1950 transport and to the smtp transport in LMTP mode.
1952 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1954 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1955 run a filter in a subprocess. This could lead to confusion in subsequent
1956 lookups in the parent process. There should also be a search_tidyup() at
1957 the end of the subprocess.
1959 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1960 only if the host matched helo_try_verify_hosts, which caused the
1961 verification to occur when the EHLO/HELO command was issued. The ACL just
1962 tested the remembered result. Now, if a previous verification attempt has
1963 not happened, "verify = helo" does it there and then.
1965 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1967 TK/03 Fix log output including CR from clamd.
1969 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1970 exist provoked a memory error which could cause a segfault.
1972 PH/15 Installed PCRE 6.2
1974 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1976 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1977 of the problem. Specifically, suggested +O2 rather than +O1 for the
1980 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1982 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1983 record, the port was not correctly logged when the outgoing_port log
1984 selector was set (it logged the transort's default port).
1986 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1987 fallback_hosts, and "hosts" in the smtp transport.
1989 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1990 host errors such as "Connection refused".
1992 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1993 authentication with radiusclient 0.4.9:
1995 - Error returned from rc_read_config was caught wrongly
1996 - Username/password not passed on to radius server due to wrong length.
1998 The presumption is that some radiusclient API changes for 4.51/PH/17
1999 were not taken care of correctly. The code is still untested by me (my
2000 Linux distribution still has 0.3.2 of radiusclient), but it was
2001 contributed by a Radius user.
2003 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2004 expanding the "port" option of the smtp transport.
2006 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2007 while reading a MIME header. Thanks to Tom Hughes for a patch.
2009 PH/24 Include config.h inside local_scan.h so that configuration settings are
2012 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2013 that in an ACL for RCPT (for example), you can examine exactly what was
2016 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2017 commands, but it was not correctly comparing the address with the actual
2018 client host address. Thus, it would show the EHLO address in Received:
2019 header lines when this was not necessary.
2021 PH/27 Added the % operator to ${eval:}.
2023 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2024 it should be ignoring failures (because with -C, for example, it has lost
2025 privilege). It wasn't ignoring creation failures other than "already
2028 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2031 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2032 because an input line was too long, either on its own, or by virtue of
2033 too many continuations, the temporary file was not being removed, and the
2034 return code was incorrect.
2036 PH/31 Missing "BOOL" in function definition in filtertest.c.
2038 PH/32 Applied Sieve patches from the maintainer.
2040 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2042 PH/33 Added "verify = not_blind".
2044 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2045 Local/Makefile (with some defaults set). These are used in built scripts
2046 such as exicyclog, but they have never been used in the exim_install
2047 script (though there are many overriding facilities there). I have
2048 arranged that the exim_install script now takes note of these two
2051 PH/35 Installed configuration files for Dragonfly.
2053 PH/36 When a locally submitted message by a trusted user did not contain a
2054 From: header, and the sender address was obtained from -f or from an SMTP
2055 MAIL command, and the trusted user did not use -F to supply a sender
2056 name, $originator_name was incorrectly used when constructing a From:
2057 header. Furthermore, $originator_name was used for submission mode
2058 messages from external hosts without From: headers in a similar way,
2059 which is clearly wrong.
2061 PH/37 Added control=suppress_local_fixups.
2063 PH/38 When log_selector = +received_sender was set, and the addition of the
2064 sender made the log line's construction buffer exactly full, or one byte
2065 less than full, an overflow happened when the terminating "\n" was
2068 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2069 when the result of a list match is failure because a DNS lookup failed.
2071 PH/40 RM_COMMAND is now used in the building process.
2073 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2074 the "build-* directories that it finds.
2076 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2077 address in a domain literal was a prefix of an interface address.
2079 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2080 when verifying a sender address, unless rewrite_headers is false.
2082 PH/44 Wrote a long comment about why errors_to addresses are verified as
2083 recipients, not senders.
2085 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2086 the ratelimit ACL was added.
2088 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2090 PH/46 Added extra information about PostgreSQL errors to the error string.
2092 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2093 avoiding the need to install special zones in a real server. This is
2094 backwards compatible; if it can't find the fake resolver, it drops back.
2095 Thus, both old and new test suites can be run.
2097 TF/02 Added util/ratelimit.pl
2099 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2100 clock is set back in time.
2102 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2103 Candler <B.Candler@pobox.com>.
2105 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2106 OK for addresses that are the result of redirections.
2108 PH/48 A number of further additions for the benefit of the new test suite,
2109 including a fake gethostbyname() that interfaces to the fake DNS resolver
2112 TF/06 The fix for widen_domains has also been applied to qualify_single and
2113 search_parents which are the other dnslookup options that can cause
2116 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2119 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2121 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2122 service" if it encountered an MX record with an empty target hostname.
2123 The message is now "an MX or SRV record indicated no SMTP service".
2125 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2126 if the DNS of the sending site is misconfigured. This is quite a
2127 common situation. This change restores the behaviour of treating a
2128 helo verification defer as a failure.
2130 PH/51 If self=fail was set on a router, the bounce message did not include the
2131 actual error message.
2137 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2139 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2140 command itself ended in error, the underlying message about the transport
2141 filter timeout was being overwritten with the pipe command error. Now the
2142 underlying error message should be appended to the second error message.
2144 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2146 PH/02 Reduce the amount of output that "make" produces by default. Full output
2147 can still be requested.
2149 PH/03 The warning log line about a condition test deferring for a "warn" verb
2150 was being output only once per connection, rather than after each
2151 occurrence (because it was using the same function as for successful
2152 "warn" verbs). This seems wrong, so I have changed it.
2154 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2155 it should not have, which might have caused a crash in the right
2156 circumstances, but probably never did.
2158 PH/04 Installed a modified version of Tony Finch's patch to make submission
2159 mode fix the return path as well as the Sender: header line, and to
2160 add a /name= option so that you can make the user's friendly name appear
2163 TF/03 Added the control = fakedefer ACL modifier.
2165 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2166 Mark Lowes for thorough testing.
2168 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2170 TK/03 Merged latest SRS patch from Miles Wilton.
2172 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2173 with the definition in sysexits.h (which is #included earlier).
2174 Fortunately, Exim does not actually use EX_OK. The code used to try to
2175 preserve the sysexits.h value, by assumimg that macro definitions were
2176 scanned for macro replacements. I have been disabused of this notion,
2177 so now the code just undefines EX_OK before #including unistd.h.
2179 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2180 in the smtp transport. When a block could not be written in a single
2181 write() function, the timeout was being re-applied to each part-write.
2182 This seems wrong - if the receiver was accepting one byte at a time it
2183 would take for ever. The timeout is now adjusted when this happens. It
2184 doesn't have to be particularly precise.
2186 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2187 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2189 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2190 without a domain if the check to <postmaster@domain> fails.
2192 SC/01 Eximstats: added -xls and the ability to specify output files
2193 (patch written by Frank Heydlauf).
2195 SC/02 Eximstats: use FileHandles for outputing results.
2197 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2199 SC/04 Eximstats: fixed display of large numbers with -nvr option
2201 SC/05 Eximstats: fixed merging of reports with empty tables.
2203 SC/06 Eximstats: added the -include_original_destination flag
2205 SC/07 Eximstats: removed tabs and trailing whitespace.
2207 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2209 TK/06 MBOX spool code: Add real "From " MBOX separator line
2210 so the .eml file is really in mbox format (even though
2211 most programs do not really care). Patch from Alex Miller.
2213 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2214 The latter is generated from $received_to and is only set if the
2215 message has one envelope recipient. SA can use these headers,
2216 obviously out-of-the-box. Patch from Alex Miller.
2218 PH/08 The ${def test on a variable was returning false if the variable's
2219 value was "0", contrary to what the specification has always said!
2220 The result should be true unless the variable is empty.
2222 PH/09 The syntax error of a character other than { following "${if
2223 def:variable_name" (after optional whitespace) was not being diagnosed.
2224 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2225 accidental colon was present, for example, could give incorrect results.
2227 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2228 result is used (not including appendfile, where other changes are about
2231 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2232 This involved changing a lot of size variables from int to off_t. It
2233 should work with maildirs and everything.
2235 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2236 spamd dying while we are connected to it.
2238 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2241 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2242 function for BDB 4.3.
2244 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2246 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2247 the sources for PCRE have been re-arranged and more files are now
2250 PH/15 The code I had for printing potentially long long variables in PH/11
2251 above was not the best (it lost precision). The length of off_t variables
2252 is now inspected at build time, and an appropriate printing format (%ld
2253 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2254 to be "long long int" or "long int". This is needed for the internal
2255 formatting function string_vformat().
2257 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2258 the configuration file to be ":syslog", then the script "guesses" where
2259 the logs files are, rather than using the compiled in default. In our
2260 case the guess is not the same as the compiled default, so the script
2261 suddenly stopped working when I started to use syslog. The patch checks
2262 to see if log_file_path is "". If so, it attempts to read it from exim
2263 with no configuration file to get the compiled in version, before it
2264 falls back to the previous guessing code."
2266 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2267 implementing BATV in an Exim configuration. See NewStuff for the gory
2270 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2271 Makefile that are specific to HP-UX.
2273 PH/18 If the "use_postmaster" option was set for a recipient callout together
2274 with the "random" option, the postmaster address was used as the MAIL
2275 FROM address for the random test, but not for the subsequent recipient
2276 test. It is now used for both.
2278 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2279 patch removes a few documentation additions to RFC 3028, because the
2280 latest draft now contains them. It adds the new en;ascii-case comparator
2281 and a new error check for 8bit text in MIME parts. Comparator and
2282 require names are now matched exactly. I enabled the subaddress
2283 extension, but it is not well tested yet (read: it works for me)."
2285 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2286 rework some of the code of TK/09 above to avoid the hardwired use of
2287 "%lld" and "long long". Replaced the call to snprintf() with a call to
2290 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2291 records point to non-existent hosts", "retry timeout exceeded", and
2292 "retry time not reached for any host after a long failure period".
2294 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2295 experimental DomainKeys support:
2297 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2298 (2) On an error such as an illegally used "control", the wrong name for
2299 the control was given.
2301 These problems did NOT occur unless DomainKeys support was compiled.
2303 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2305 PH/24 Added ${if match_ip condition.
2307 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2308 databases so that it will be absolutely obvious if a crash occurs in the
2309 DB library. This is a regular occurrence (often caused by mis-matched
2312 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2313 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2314 picked up on a user's system that detects such things. There doesn't seem
2315 to be a gcc warning option for this - only an attribute that has to be
2316 put on the function's prototype. It seems that in Fedora Core 4 they have
2317 set this on a number of new functions. No doubt there will be more in due
2320 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2321 specify a transport. However, if an address that was verified by such a
2322 router was the subject of a callout, Exim crashed because it tried to
2323 read the rcpt_include_affixes from the non-existent transport. Now it
2324 just assumes that the setting of that option is false. This bug was
2325 introduced by 4.51/PH/31.
2327 PH/28 Changed -d+all to exclude +memory, because that information is very
2328 rarely of interest, but it makes the output a lot bigger. People tend to
2329 do -d+all out of habit.
2331 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2332 code in os-type was giving problems when libc.so lives in lib64, like on
2335 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2336 aren't the modern standard, and it seems that some systems' include files
2337 don't always have them. Exim was already checking for some of the newer
2338 ones like T_AAAA, and defining it itself. I've added checks for all the
2339 record types that Exim uses.
2341 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2342 not automatically generating a new one, as it is supposed to. This
2343 prevented TLS from working. If the file did exist, but contained invalid
2344 data, a new version was generated, as expected. It was only the case of a
2345 non-existent file that was broken.
2347 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2348 with a change in libdomainkeys > 0.64.
2350 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2351 from DNS. If the selector record carries the flag, it now has
2352 precedence over the domain-wide flag.
2354 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2356 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2357 the use of a port name that isn't defined in /etc/services) occurred, the
2358 message was deferred as in a normal delivery, and thus remained on the
2359 spool, instead of being failed because of the mua_wrapper setting. This
2360 is now fixed, and I tidied up some of the mua_wrapper messages at the
2363 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2364 the messages in a hash of arrays rather than using individual hashes.
2365 This is a bit cleaner and results in dramatic memory savings, albeit
2366 at a slight CPU cost.
2368 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2369 as requested by Marc Sherman.
2371 SC/10 Eximstats: added histograms for user specified patterns as requested
2374 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2376 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2377 fopen() in the content-scanning modules that did not already have it.
2383 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2384 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2386 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2388 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2390 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2391 to test Sieve filters that use "vacation".
2393 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2394 that changes the way the GnuTLS parameters are stored in the cache file.
2395 The new format can be generated externally. For backward compatibility,
2396 if the data in the cache doesn't make sense, Exim assumes it has read an
2397 old-format file, and it generates new data and writes a new file. This
2398 means that you can't go back to an older release without removing the
2401 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2402 work if there are any delivery delays because "one_time" forces the
2403 parent to be marked "delivered", so its unseen clone is never tried
2404 again. For this reason, Exim now forbids the simultaneous setting of
2407 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2408 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2409 entry if you want to know the details. The fix, however, neglected to
2410 consider the case when local delivery batching is involved. The test for
2411 "previously delivered" was not happening when checking to see if an
2412 address could be batched with a previous (undelivered) one; under
2413 certain circumstances this could lead to multiple deliveries to the same
2416 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2417 in its include files, and this causes problems building Exim.
2419 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2420 header_syntax) but Exim was just ignoring anything given after a slash.
2421 In particular, this caused confusion with an attempt to use "verify =
2422 reverse_host_lookup/defer_ok". An error is now given when options are
2423 supplied for verify items that do not have them. (Maybe reverse_host_
2424 lookup should have a defer_ok option, but that's a different point.)
2426 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2427 defined by RFC 821) to 2048, because there were problems with some AUTH
2428 commands, and RFC 1869 says the size should be increased for extended
2429 SMTP commands that take arguments.
2431 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2434 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2435 "unknown" error; now it says that the functionality isn't in the binary.
2437 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2438 an address' error message when a string expansion fails (syntax or
2439 whatever). Otherwise the password may appear in the log. Following change
2440 PH/42 below, there is no longer a chance of it appearing in a bounce
2443 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2445 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2446 of its addresses was ever tried. (Bugzilla bug #2.)
2448 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2449 the result incorrectly in the debug output. (It correctly added a newline
2450 to what was transported.)
2452 TF/01 Added $received_time.
2454 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2455 commented out examples of how to interface to a virus scanner and to
2456 SpamAssassin. Also added commented examples of av_scanner and
2457 spamd_address settings.
2459 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2460 and controls are allowed in which ACLs. There were a couple of minor
2461 errors. Some of the entries in the conditions table (which is a table of
2462 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2463 negation of where the condition IS allowed.
2465 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2467 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2468 header file does not have a version number, so I've had to invent a new
2469 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2470 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2471 radiusclient), but it was contributed by a Radius user.
2473 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2474 files or directories, for OpenSSL.
2476 PH/19 When an Exim process that is running as root has to create an Exim log
2477 file, it does so in a subprocess that runs as exim:exim so as to get the
2478 ownership right at creation (otherwise, other Exim processes might see
2479 the file with the wrong ownership). There was no test for failure of this
2480 fork() call, which would lead to the process getting stuck as it waited
2481 for a non-existent subprocess. Forks do occasionally fail when resources
2482 run out. I reviewed all the other calls to fork(); they all seem to check
2485 PH/20 When checking for unexpected SMTP input at connect time (before writing
2486 the banner), Exim was not dealing correctly with a non-positive return
2487 from the read() function. If the client had disconnected by this time,
2488 the result was a log entry for a synchronization error with an empty
2489 string after "input=" when read() returned zero. If read() returned -1
2490 (an event I could not check), uninitialized data bytes were printed.
2491 There were reports of junk text (parts of files, etc) appearing after
2494 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2496 PH/22 Added support for macro redefinition, and (re)definition in between
2497 driver and ACL definitions.
2499 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2500 forgetting to use the resulting value; it was using the unexpanded value.
2502 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2503 hadn't been configured. The fix is from Juergen Kreileder, who
2504 understands it better than I do:
2506 "Here's what I see happening with three configured cyrus_sasl
2507 authenticators configured (plain, login, cram-md5):
2509 On startup auth_cyrus_sasl_init() gets called for each of these.
2510 This means three calls to sasl_listmech() without a specified mech_list.
2511 => SASL tests which mechs of all available mechs actually work
2512 => three warnings about OTP not working
2513 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2515 With the patch, sasl_listmech() also gets called three times. But now
2516 SASL's mech_list option is set to the server_mech specified in the the
2517 authenticator. Or in other words, the answer from sasl_listmech()
2518 gets limited to just the mech you're testing for (which is different
2520 => the return list contains just 'plain' or 'login', 'cram-md5' or
2521 nothing depending on the value of ob->server_mech.
2523 I've just tested the patch: Authentication still works fine,
2524 unavailable mechs specified in the exim configuration are still
2525 caught, and the auth.log warnings about OTP are gone."
2527 PH/25 When debugging is enabled, the contents of the command line are added
2528 to the debugging output, even when log_selector=+arguments is not
2531 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2532 answer is "GNU", and only if the return is "GNU/something" is the answer
2535 PH/27 $acl_verify_message is now set immediately after the failure of a
2536 verification in an ACL, and so is available in subsequent modifiers. In
2537 particular, the message can be preserved by coding like this:
2539 warn !verify = sender
2540 set acl_m0 = $acl_verify_message
2542 Previously, $acl_verify_message was set only while expanding "message"
2543 and "log_message" when a very denied access.
2545 PH/28 Modified OS/os.c-Linux with
2547 -#ifndef OS_LOAD_AVERAGE
2548 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2550 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2551 nomenclature these days.)
2553 PH/29 Installed patch from the Sieve maintainer that adds the options
2554 sieve_useraddress and sieve_subaddress to the redirect router.
2556 PH/30 In these circumstances:
2557 . Two addresses routed to the same list of hosts;
2558 . First host does not offer TLS;
2559 . First host accepts first address;
2560 . First host gives temporary error to second address;
2561 . Second host offers TLS and a TLS session is established;
2562 . Second host accepts second address.
2563 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2564 and peerdn, if requested) that were in fact used only for the second
2567 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2568 attention to any local part prefix or suffix that was matched by the
2569 router that accepted the address. It now behaves in the same way as it
2570 does for delivery: the affixes are removed from the local part unless
2571 rcpt_include_affixes is set on the transport.
2573 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2574 timeout during the DATA phase of an incoming message.
2576 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2577 applied a patch sanctioned by the Sieve maintainer.
2579 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2580 the uid or gid is negative. A case of a negative gid caused this to be
2581 noticed. The fix allows for either to be negative.
2583 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2584 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2587 PH/36 The change PH/12 above was broken. Fixed it.
2589 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2590 the grounds that routing the same address twice would always produce the
2591 same answer. This might have been true once, but it is certainly no
2592 longer true now. Routing a child address may depend on the previous
2593 routing that produced that child. Some complicated redirection strategies
2594 went wrong when messages had multiple recipients, and made Exim's
2595 behaviour dependent on the order in which the addresses were given.
2597 I have moved the duplicate checking until after the routing is complete.
2598 Exim scans the addresses that are assigned to local and remote
2599 transports, and removes any duplicates. This means that more work will be
2600 done, as duplicates will always all be routed, but duplicates are
2601 presumably rare, so I don't expect this is of any significance.
2603 For deliveries to pipes, files, and autoreplies, the duplicate checking
2604 still happens during the routing process, since they are not going to be
2607 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2608 It corrects a timeout issue with spamd. This is Ian's comment: "The
2609 background is that sometimes spamd either never reads data from a
2610 connection it has accepted, or it never writes response data. The exiscan
2611 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2612 blindly assumes that writes won't block so it may never time out."
2614 PH/39 Allow G after quota size as well as K and M.
2616 PH/40 The value set for $authenticated_id in an authenticator may not contain
2617 binary zeroes or newlines because the value is written to log lines and
2618 to spool files. There was no check on this. Now the value is run through
2619 the string_printing() function so that such characters are converted to
2620 printable escape sequences.
2622 PH/41 $message_linecount is a new variable that contains the total number of
2623 lines in the message. Compare $body_linecount, which is the count for the
2626 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2627 bounce and delay warning messages, except in certain special cases, which
2630 (a) An SMTP error message from a remote host;
2631 (b) A message specified in a :fail: redirection;
2632 (c) A message specified in a "fail" command in a system filter;
2633 (d) A message specified in a FAIL return from the queryprogram router;
2634 (e) A message specified by the cannot_route_message router option.
2636 In these cases only, Exim does include the error details in bounce and
2637 warning messages. There are also a few cases where bland messages such
2638 as "unrouteable address" or "local delivery error" are given.
2640 PH/43 $value is now also set for the "else" part of a ${run expansion.
2642 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2643 being worked on, but at least Exim now implements the latest version to
2646 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2647 process to complete was treated as a delivery failure, a timeout while
2648 writing the message to the pipe was logged, but erroneously treated as a
2649 successful delivery. Such timeouts include transport filter timeouts. For
2650 consistency with the overall process timeout, these timeouts are now
2651 treated as errors, giving rise to delivery failures by default. However,
2652 there is now a new Boolean option for the pipe transport called
2653 timeout_defer, which, if set TRUE, converts the failures into defers for
2654 both kinds of timeout. A transport filter timeout is now identified in
2657 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2658 systems where "make" and "gmake" are different, calling "gmake" at top
2659 level broke things. I've arranged for the value of $(MAKE) to be passed
2660 from the Makefile to this script so that it can call the same version of
2664 A note about Exim versions 4.44 and 4.50
2665 ----------------------------------------
2667 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2668 changes of various kinds. As a consequence, a big documentation update was
2669 needed. This delayed the release for rather longer than seemed good, especially
2670 in the light of a couple of (minor) security issues. Therefore, the changes
2671 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2672 release. So 4.44 and 4.50 are in effect two different branches that both start
2675 I have left the 4.50 change log unchanged; it contains all the changes since
2676 4.43. The change log for 4.44 is below; many of its items are identical to
2677 those for 4.50. This seems to be the most sensible way to preserve the
2678 historical information.
2684 1. Minor wording change to the doc/README.SIEVE file.
2686 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2687 computation of the current number of files was incorrect.
2689 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2690 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2693 4. Give more explanation in the error message when the command for a transport
2694 filter fails to execute.
2696 5. There are several places where Exim runs a non-Exim command in a
2697 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2698 was being done only for the command run by the queryprogram router. It is
2699 now done for all such subprocesses. The other cases are: ${run, transport
2700 filters, and the commands run by the lmtp and pipe transports.
2702 6. Added CONFIGURE_GROUP build-time option.
2704 7. Some older OS have a limit of 256 on the maximum number of file
2705 descriptors. Exim was using setrlimit() to set 1000 as a large value
2706 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2707 systems. I've change it so that if it can't get 1000, it tries for 256.
2709 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2710 was an oversight, and furthermore, ever since the addition of extra
2711 controls (e.g. 4.43/32), the checks on when to allow different forms of
2712 "control" were broken. There should now be diagnostics for all cases when a
2713 control that does not make sense is encountered.
2715 9. Added the /retain_sender option to "control=submission".
2717 10. $recipients is now available in the predata ACL (oversight).
2719 11. Tidy the search cache before the fork to do a delivery from a message
2720 received from the command line. Otherwise the child will trigger a lookup
2721 failure and thereby defer the delivery if it tries to use (for example) a
2722 cached ldap connection that the parent has called unbind on.
2724 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2725 of $address_data from the recipient verification was clobbered by the
2726 sender verification.
2728 13. The value of address_data from a sender verification is now available in
2729 $sender_address_data in subsequent conditions in the ACL statement.
2731 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2733 15. Added a new option "connect=<time>" to callout options, to set a different
2736 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2737 was its contents. (It was OK if the option was not defined at all.)
2739 17. A "Completed" log line is now written for messages that are removed from
2740 the spool by the -Mrm option.
2742 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2743 information about exactly what failed.
2745 19. Added -dd to debug only the daemon process.
2747 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2748 handles timeouts, both on the server side and network timeouts. Renamed the
2749 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2751 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2752 It is now set to "smtps".
2754 22. $host_address is now set to the target address during the checking of
2755 ignore_target_hosts.
2757 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2758 being passed; this would have caused $sender_host_name to have been used if
2759 matching the list had actually called for a host name (not very likely,
2760 since this list is usually IP addresses). A host name is now passed as
2763 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2764 code that specifies a non-restarting handler (typically sigaction() in
2765 modern systems) in an attempt to fix a rare and obscure crash bug.
2767 25. Narrowed the window for a race in the daemon that could cause it to ignore
2768 SIGCHLD signals. This is not a major problem, because they are used only to
2769 wake it up if nothing else does.
2771 26. A malformed maildirsize file could cause Exim to calculate negative values
2772 for the mailbox size or file count. Odd effects could occur as a result.
2773 The maildirsize information is now recalculated if the size or filecount
2776 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2777 support for a long time. Removed HAVE_SYS_VFS_H.
2779 28. Installed the latest version of exipick from John Jetmore.
2781 29. In an address list, if the pattern was not a regular expression, an empty
2782 subject address (from a bounce message) matched only if the pattern was an
2783 empty string. Non-empty patterns were not even tested. This was the wrong
2784 because it is perfectly reasonable to use an empty address as part of a
2785 database query. An empty address is now tested by patterns that are
2786 lookups. However, all the other forms of pattern expect the subject to
2787 contain a local part and a domain, and therefore, for them, an empty
2788 address still always fails if the pattern is not itself empty.
2790 30. Exim went into a mad DNS loop when attempting to do a callout where the
2791 host was specified on an smtp transport, and looking it up yielded more
2792 than one IP address.
2794 31. Re-factored the code for checking spool and log partition space into a
2795 function that finds that data and another that does the check. The former
2796 is then used to implement four new variables: $spool_space, $log_space,
2797 $spool_inodes, and $log_inodes.
2799 32. The RFC2047 encoding function was originally intended for short strings
2800 such as real names; it was not keeping to the 75-character limit for
2801 encoded words that the RFC imposes. It now respects the limit, and
2802 generates multiple encoded words if necessary. To be on the safe side, I
2803 have increased the buffer size for the ${rfc2047: expansion operator from
2806 33. It is now permitted to omit both strings after an "if" condition; if the
2807 condition is true, the result is "true". As before, when the second string
2808 is omitted, a false condition yields an empty string. This makes it less
2809 cumbersome to write custom ACL and router conditions.
2811 34. Failure to deliver a bounce message always caused it to be frozen, even if
2812 there was an errors_to setting on the router. The errors_to setting is now
2815 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2816 canonical form (fully expanded) before being placed in
2817 $sender_host_address.
2819 36. The table in the code that translates DNS record types into text (T_A to
2820 "A" for instance) was missing entries for NS and CNAME. It is just possible
2821 that this could have caused confusion if both these types were looked up
2822 for the same domain, because the text type is used as part of Exim's
2823 per-process caching. But the chance of anyone hitting this buglet seems
2826 37. The dnsdb lookup has been extended in a number of ways.
2828 (1) There is a new type, "zns", which walks up the domain tree until it
2829 finds some nameserver records. It should be used with care.
2831 (2) There is a new type, "mxh", which is like "mx" except that it returns
2832 just the host names, not the priorities.
2834 (3) It is now possible to give a list of domains (or IP addresses) to be
2835 looked up. The behaviour when one of the lookups defers can be
2836 controlled by a keyword.
2838 (4) It is now possible to specify the separator character for use when
2839 multiple records are returned.
2841 38. The dnslists ACL condition has been extended: it is now possible to supply
2842 a list of IP addresses and/or domains to be looked up in a particular DNS
2845 39. Added log_selector=+queue_time_overall.
2847 40. When running the queue in the test harness, wait just a tad after forking a
2848 delivery process, to get repeatability of debugging output.
2850 41. Include certificate and key file names in error message when GnuTLS fails
2851 to set them up, because the GnuTLS error message doesn't include the name
2852 of the failing file when there is a problem reading it.
2854 42. Allow both -bf and -bF in the same test run.
2856 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2858 44. The "Exiscan patch" is now merged into the mainline Exim source.
2860 45. Sometimes the final signoff response after QUIT could fail to get
2861 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2862 < 0 before doing a fflush(). This bug looks as though it goes back to the
2863 introduction of TLS in release 3.20, but "sometimes" must have been rare
2864 because the tests only now provoked it.
2866 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2867 (this can affect the format of dates).
2869 47. exim_tidydb, when checking for the continued existence of a message for
2870 which it has found a message-specific retry record, was not finding
2871 messages that were in split spool directories. Consequently, it was
2872 deleting retry records that should have stayed in existence.
2874 48. Steve fixed some bugs in eximstats.
2876 49. The SPA authentication driver was not abandoning authentication and moving
2877 on to the next authenticator when an expansion was forced to fail,
2878 contradicting the general specification for all authenticators. Instead it
2879 was generating a temporary error. It now behaves as specified.
2881 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2882 (the order specifies the preference for clients). The order is now AES256,
2883 AES128, 3DES, ARCFOUR128.
2885 51. Small patch to Sieve code - explicitly set From: when generating an
2888 52. Exim crashed if a remote delivery caused a very long error message to be
2889 recorded - for instance if somebody sent an entire SpamAssassin report back
2890 as a large number of 550 error lines. This bug was coincidentally fixed by
2891 increasing the size of one of Exim's internal buffers (big_buffer) that
2892 happened as part of the Exiscan merge. However, to be on the safe side, I
2893 have made the code more robust (and fixed the comments that describe what
2896 53. Now that there can be additional text after "Completed" in log lines (if
2897 the queue_time_overall log selector is set), a one-byte patch to exigrep
2898 was needed to allow it to recognize "Completed" as not the last thing in
2901 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2902 patch that reportedly fixes this has been added. I am not expert enough to
2903 create a test for it. This is what the patch creator wrote:
2905 "I found a little strange behaviour of ldap code when working with
2906 Windows 2003 AD Domain, where users was placed in more than one
2907 Organization Units. When I tried to give exim partial DN, the exit code
2908 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2909 But simultaneously result of request was absolutely normal ldap result,
2910 so I produce this patch..."
2912 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2913 REFERENCE, so I have modified the code to exclude the patch when that macro
2916 55. Some experimental protocols are using DNS PTR records for new purposes. The
2917 keys for these records are domain names, not reversed IP addresses. The
2918 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2919 leaves it alone. Component reversal etc. now happens only for IP addresses.
2922 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2924 57. Double the size of the debug message buffer (to 2048) so that more of very
2925 long debug lines gets shown.
2927 58. The exicyclog utility now does better if the number of log files to keep
2928 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2930 59. Two changes related to the smtp_active_hostname option:
2932 (1) $smtp_active_hostname is now available as a variable.
2933 (2) The default for smtp_banner uses $smtp_active_hostname instead
2934 of $primary_hostname.
2936 60. The host_aton() function is supposed to be passed a string that is known
2937 to be a valid IP address. However, in the case of IPv6 addresses, it was
2938 not checking this. This is a hostage to fortune. Exim now panics and dies
2939 if the condition is not met. A case was found where this could be provoked
2940 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2941 components; fortuitously, this particular loophole had already been fixed
2942 by change 4.50/55 above.
2944 If there are any other similar loopholes, the new check in host_aton()
2945 itself should stop them being exploited. The report I received stated that
2946 data on the command line could provoke the exploit when Exim was running as
2947 exim, but did not say which command line option was involved. All I could
2948 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2949 running as the user.
2952 61. There was a buffer overflow vulnerability in the SPA authentication code
2953 (which came originally from the Samba project). I have added a test to the
2954 spa_base64_to_bits() function which I hope fixes it.
2957 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2958 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2960 63. The daemon start-up calls getloadavg() while still root for those OS that
2961 need the first call to be done as root, but it missed one case: when
2962 deliver_queue_load_max is set with deliver_drop_privilege. This is
2963 necessary for the benefit of the queue runner, because there is no re-exec
2964 when deliver_drop_privilege is set.
2966 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2967 This has been fixed.
2969 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2970 was in use, was not putting the data itself into the right store pool;
2971 consequently, it could be overwritten for a subsequent message in the same
2972 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2975 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2977 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2978 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2979 the function were treating the return as a boolean value, which happened to
2980 work because 0=false and not-0=true, but is not correct code.
2982 68. The host_aton() function was not handling scoped IPv6 addresses (those
2983 with, for example, "%eth0" on the end) correctly.
2985 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2986 negated items (that is, ~something) in unsigned ints. Some compilers
2987 apparently mutter when there is no cast.
2989 70. If an address verification called from an ACL failed, and did not produce a
2990 user-specific message (i.e. there was only a "system" message), nothing was
2991 put in $acl_verify_message. In this situation, it now puts the system
2994 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2995 session; change 4.31/43 added the unwanted input to the log line - except
2996 that it did not do this in the start of session case. It now does.
2998 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2999 This is wrong and can cause the other end to generate a synchronization
3000 error if it is another Exim or anything else that does the synchronization
3001 check. A QUIT command is no longer sent after a timeout.
3003 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3004 during host lookups.
3006 74. The defer_ok option of callout verification was not working if it was used
3007 when verifying addresses in header lines, that is, for this case:
3009 verify = header_sender/callout=defer_ok
3011 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3012 those file descriptors could be used for SMTP connections. If anything
3013 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3014 could be sent to the SMTP client, causing chaos. The daemon now opens
3015 stdin, stdout, and stderr to /dev/null when it puts itself into the
3018 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3019 log by default. The user can override this with suitable Perl magic.
3021 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3022 the log message when discard triggers, was not working for the DATA ACL or
3023 for the non-SMTP ACL.
3025 78. Error message wording change in sieve.c.
3027 79. If smtp_accept_max_per_host was set, the number of connections could be
3028 restricted to fewer than expected, because the daemon was trying to set up
3029 a new connection before checking whether the processes handling previous
3030 connections had finished. The check for completed processes is now done
3031 earlier. On busy systems, this bug wouldn't be noticed because something
3032 else would have woken the daemon, and it would have reaped the completed
3035 80. If a message was submitted locally by a user whose login name contained one
3036 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3037 It caused a spool format error. I have fixed the spool reading code. A
3038 related problem was that the "from" clause in the Received: line became
3039 illegal because of the space(s). It is now covered by ${quote_local_part.
3041 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3044 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3046 83. Patch from Sieve maintainer to fix unterminated string problem in
3047 "vacation" handling.
3049 84. Some minor changes to the Linux configuration files to help with other
3050 OS variants using glibc.
3052 85. One more patch for Sieve to update vacation handling to latest spec.
3055 ----------------------------------------------------
3056 See the note above about the 4.44 and 4.50 releases.
3057 ----------------------------------------------------
3063 1. Change 4.43/35 introduced a bug that caused file counts to be
3064 incorrectly computed when quota_filecount was set in an appendfile
3067 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3068 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3071 3. Give more explanation in the error message when the command for a transport
3072 filter fails to execute.
3074 4. There are several places where Exim runs a non-Exim command in a
3075 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3076 was being done only for the command run by the queryprogram router. It is
3077 now done for all such subprocesses. The other cases are: ${run, transport
3078 filters, and the commands run by the lmtp and pipe transports.
3080 5. Some older OS have a limit of 256 on the maximum number of file
3081 descriptors. Exim was using setrlimit() to set 1000 as a large value
3082 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3083 systems. I've change it so that if it can't get 1000, it tries for 256.
3085 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3086 was an oversight, and furthermore, ever since the addition of extra
3087 controls (e.g. 4.43/32), the checks on when to allow different forms of
3088 "control" were broken. There should now be diagnostics for all cases when a
3089 control that does not make sense is encountered.
3091 7. $recipients is now available in the predata ACL (oversight).
3093 8. Tidy the search cache before the fork to do a delivery from a message
3094 received from the command line. Otherwise the child will trigger a lookup
3095 failure and thereby defer the delivery if it tries to use (for example) a
3096 cached ldap connection that the parent has called unbind on.
3098 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3099 of $address_data from the recipient verification was clobbered by the
3100 sender verification.
3102 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3103 was its contents. (It was OK if the option was not defined at all.)
3105 11. A "Completed" log line is now written for messages that are removed from
3106 the spool by the -Mrm option.
3108 12. $host_address is now set to the target address during the checking of
3109 ignore_target_hosts.
3111 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3112 being passed; this would have caused $sender_host_name to have been used if
3113 matching the list had actually called for a host name (not very likely,
3114 since this list is usually IP addresses). A host name is now passed as
3117 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3118 code that specifies a non-restarting handler (typically sigaction() in
3119 modern systems) in an attempt to fix a rare and obscure crash bug.
3121 15. Narrowed the window for a race in the daemon that could cause it to ignore
3122 SIGCHLD signals. This is not a major problem, because they are used only to
3123 wake it up if nothing else does.
3125 16. A malformed maildirsize file could cause Exim to calculate negative values
3126 for the mailbox size or file count. Odd effects could occur as a result.
3127 The maildirsize information is now recalculated if the size or filecount
3130 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3131 support for a long time. Removed HAVE_SYS_VFS_H.
3133 18. Updated exipick to current release from John Jetmore.
3135 19. Allow an empty sender to be matched against a lookup in an address list.
3136 Previously the only cases considered were a regular expression, or an
3139 20. Exim went into a mad DNS lookup loop when doing a callout where the
3140 host was specified on the transport, if the DNS lookup yielded more than
3143 21. The RFC2047 encoding function was originally intended for short strings
3144 such as real names; it was not keeping to the 75-character limit for
3145 encoded words that the RFC imposes. It now respects the limit, and
3146 generates multiple encoded words if necessary. To be on the safe side, I
3147 have increased the buffer size for the ${rfc2047: expansion operator from
3150 22. Failure to deliver a bounce message always caused it to be frozen, even if
3151 there was an errors_to setting on the router. The errors_to setting is now
3154 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3155 canonical form (fully expanded) before being placed in
3156 $sender_host_address.
3158 24. Updated eximstats to version 1.33
3160 25. Include certificate and key file names in error message when GnuTLS fails
3161 to set them up, because the GnuTLS error message doesn't include the name
3162 of the failing file when there is a problem reading it.
3164 26. Expand error message when OpenSSL has problems setting up cert/key files.
3167 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3168 (this can affect the format of dates).
3170 28. exim_tidydb, when checking for the continued existence of a message for
3171 which it has found a message-specific retry record, was not finding
3172 messages that were in split spool directories. Consequently, it was
3173 deleting retry records that should have stayed in existence.
3175 29. eximstats updated to version 1.35
3176 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3177 1.35 - bugfix such that pie charts by volume are generated correctly
3179 30. The SPA authentication driver was not abandoning authentication and moving
3180 on to the next authenticator when an expansion was forced to fail,
3181 contradicting the general specification for all authenticators. Instead it
3182 was generating a temporary error. It now behaves as specified.
3184 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3185 (the order specifies the preference for clients). The order is now AES256,
3186 AES128, 3DES, ARCFOUR128.
3188 31. Small patch to Sieve code - explicitly set From: when generating an
3191 32. Exim crashed if a remote delivery caused a very long error message to be
3192 recorded - for instance if somebody sent an entire SpamAssassin report back
3193 as a large number of 550 error lines. This bug was coincidentally fixed by
3194 increasing the size of one of Exim's internal buffers (big_buffer) that
3195 happened as part of the Exiscan merge. However, to be on the safe side, I
3196 have made the code more robust (and fixed the comments that describe what
3199 33. Some experimental protocols are using DNS PTR records for new purposes. The
3200 keys for these records are domain names, not reversed IP addresses. The
3201 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3202 leaves it alone. Component reversal etc. now happens only for IP addresses.
3205 34. The host_aton() function is supposed to be passed a string that is known
3206 to be a valid IP address. However, in the case of IPv6 addresses, it was
3207 not checking this. This is a hostage to fortune. Exim now panics and dies
3208 if the condition is not met. A case was found where this could be provoked
3209 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3210 components; fortuitously, this particular loophole had already been fixed
3211 by change 4.50/55 or 4.44/33 above.
3213 If there are any other similar loopholes, the new check in host_aton()
3214 itself should stop them being exploited. The report I received stated that
3215 data on the command line could provoke the exploit when Exim was running as
3216 exim, but did not say which command line option was involved. All I could
3217 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3218 running as the user.
3221 35. There was a buffer overflow vulnerability in the SPA authentication code
3222 (which came originally from the Samba project). I have added a test to the
3223 spa_base64_to_bits() function which I hope fixes it.
3226 36. The daemon start-up calls getloadavg() while still root for those OS that
3227 need the first call to be done as root, but it missed one case: when
3228 deliver_queue_load_max is set with deliver_drop_privilege. This is
3229 necessary for the benefit of the queue runner, because there is no re-exec
3230 when deliver_drop_privilege is set.
3232 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3233 was in use, was not putting the data itself into the right store pool;
3234 consequently, it could be overwritten for a subsequent message in the same
3235 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3238 38. Sometimes the final signoff response after QUIT could fail to get
3239 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3240 < 0 before doing a fflush(). This bug looks as though it goes back to the
3241 introduction of TLS in release 3.20, but "sometimes" must have been rare
3242 because the tests only now provoked it.
3248 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3249 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3250 Now it returns an integer. A number of calls were still expecting a T/F
3251 return. Fortuitously, in all cases, the tests worked in OK situations,
3252 which is the norm. However, things would have gone wrong on any write
3253 failures on the smtp file descriptor. This function is used when sending
3254 messages over SMTP and also when doing verify callouts.
3256 2. When Exim is called to do synchronous delivery of a locally submitted
3257 message (the -odf or -odi options), it no longer closes stderr before doing
3260 3. Implemented the mua_wrapper option.
3262 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3264 5. Implemented the functions header_remove(), header_testname(),
3265 header_add_at_position(), and receive_remove_recipient(), and exported them
3268 6. If an ACL "warn" statement specified the addition of headers, Exim already
3269 inserted X-ACL-Warn: at the start if there was no header name. However, it
3270 was not making this test for the second and subsequent header lines if
3271 there were newlines in the string. This meant that an invalid header could
3272 be inserted if Exim was badly configured.
3274 7. Allow an ACL "warn" statement to add header lines at the start or after all
3275 the Received: headers, as well as at the end.
3277 8. Added the rcpt_4xx retry error code.
3279 9. Added postmaster_mailfrom=xxx to callout verification option.
3281 10. Added mailfrom=xxxx to the callout verification option, for verify=
3284 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3285 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3287 12. Inserted some casts to stop certain compilers warning when using pointer
3288 differences as field lengths or precisions in printf-type calls (mostly
3289 affecting debugging statements).
3291 13. Added optional readline() support for -be (dynamically loaded).
3293 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3294 same clock tick as a message's arrival, so that its received time was the
3295 same as the "first fail" time on the retry record, and that message
3296 remained on the queue past the ultimate address timeout, every queue runner
3297 would try a delivery (because it was past the ultimate address timeout) but
3298 after another failure, the ultimate address timeout, which should have then
3299 bounced the address, did not kick in. This was a "< instead of <=" error;
3300 in most cases the first failure would have been in the next clock tick
3301 after the received time, and all would be well.
3303 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3304 being recognized when the domain list was tested by the match_domain
3305 condition in an expansion string.
3307 16. Added the ${str2b64: operator.
3309 17. Exim was always calling setrlimit() to set a large limit for the number of
3310 processes, without checking whether the existing limit was already
3311 adequate. (It did check for the limit on file descriptors.) Furthermore,
3312 errors from getrlimit() and setrlimit() were being ignored. Now they are
3313 logged to the main and panic logs, but Exim does carry on, to try to do its
3314 job under whatever limits there are.
3316 18. Imported PCRE 5.0.
3318 19. Trivial typo in log message " temporarily refused connection" (the leading
3321 20. If the log selector return_path_on_delivery was set and an address was
3322 redirected to /dev/null, the delivery process crashed because it assumed
3323 that a return path would always be set for a "successful" delivery. In this
3324 case, the whole delivery is bypassed as an optimization, and therefore no
3327 21. Internal re-arrangement: the function for sending a challenge and reading
3328 a response while authentication was assuming a zero-terminated challenge
3329 string. It's now changed to take a pointer and a length, to allow for
3330 binary data in such strings.
3332 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3334 23. Exim was not respecting finduser_retries when seeking the login of the
3335 uid under which it was called; it was always trying 10 times. (The default
3336 setting of finduser_retries is zero.) Also, it was sleeping after the final
3337 failure, which is pointless.
3339 24. Implemented tls_on_connect_ports.
3341 25. Implemented acl_smtp_predata.
3343 26. If the domain in control=submission is set empty, Exim assumes that the
3344 authenticated id is a complete email address when it generates From: or
3345 Sender: header lines.
3347 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3348 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3349 chown and chgrp in /bin and hostname in /usr/bin.
3351 28. Exim was keeping the "process log" file open after each use, just as it
3352 does for the main log. This opens the possibility of it remaining open for
3353 long periods when the USR1 signal hits a daemon. Occasional processlog
3354 errors were reported, that could have been caused by this. Anyway, it seems
3355 much more sensible not to leave this file open at all, so that is what now
3358 29. The long-running daemon process does not normally write to the log once it
3359 has entered its main loop, and it closes the log before doing so. This is
3360 so that log files can straightforwardly be renamed and moved. However,
3361 there are a couple of unusual error situations where the daemon does write
3362 log entries, and I had neglected to close the log afterwards.
3364 30. The text of an SMTP error response that was received during a remote
3365 delivery was being truncated at 512 bytes. This is too short for some of
3366 the long messages that one sometimes sees. I've increased the limit to
3369 31. It is now possible to make retry rules that apply only when a message has a
3370 specific sender, in particular, an empty sender.
3372 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3373 it possible to be selective about when SMTP synchronization is enforced.
3375 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3377 32. Implemented hosts_connection_nolog.
3379 33. Added an ACL for QUIT.
3381 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3384 35. Added mailbox_size and mailbox_filecount to appendfile.
3386 36. Added control = no_multiline_responses to ACLs.
3388 37. There was a bug in the logic of the code that waits for the clock to tick
3389 in the case where the clock went backwards by a substantial amount such
3390 that the microsecond fraction of "now" was more than the microsecond
3391 fraction of "then" (but the whole seconds number was less).
3393 38. Added support for the libradius Radius client library this is found on
3394 FreeBSD (previously only the radiusclient library was supported).
3400 1. When certain lookups returned multiple values in the form name=value, the
3401 quoting of the values was not always being done properly. Specifically:
3402 (a) If the value started with a double quote, but contained no whitespace,
3404 (b) If the value contained whitespace other than a space character (i.e.
3405 tabs or newlines or carriage returns) it was not quoted.
3406 This fix has been applied to the mysql and pgsql lookups by writing a
3407 separate quoting function and calling it from the lookup code. The fix
3408 should probably also be applied to nisplus, ibase and oracle lookups, but
3409 since I cannot test any of those, I have not disturbed their existing code.
3411 2. A hit in the callout cache for a specific address caused a log line with no
3412 reason for rejecting RCPT. Now it says "Previous (cached) callout
3413 verification failure".
3415 3. There was an off-by-one bug in the queryprogram router. An over-long
3416 return line was truncated at 256 instead of 255 characters, thereby
3417 overflowing its buffer with the terminating zero. As well as fixing this, I
3418 have increased the buffer size to 1024 (and made a note to document this).
3420 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3421 when Exim is waiting for an SMTP response from a remote server, Exim
3422 restarts its select() call on the socket, thereby resetting its timeout.
3423 This is not a problem when such interrupts are rare. Somebody set up a cron
3424 job to run exiwhat every 2 minutes, which is less than the normal select()
3425 timeout (5 or 10 minutes). This meant that the select() timeout never
3426 kicked in because it was always reset. I have fixed this by comparing the
3427 time when an interrupt arrives with the time at the start of the first call
3428 to select(). If more time than the timeout has elapsed, the interrupt is
3429 treated as a timeout.
3431 5. Some internal re-factoring in preparation for the addition of Sieve
3432 extensions (by MH). In particular, the "personal" test is moved to a
3433 separate function, and given an option for scanning Cc: and Bcc: (which is
3434 not set for Exim filters).
3436 6. When Exim created an email address using the login of the caller as the
3437 local part (e.g. when creating a From: or Sender: header line), it was not
3438 quoting the local part when it contained special characters such as @.
3440 7. Installed new OpenBSD configuration files.
3442 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3443 try to make them clearer.
3445 9. Callout options, other than the timeout value, were being ignored when
3446 verifying sender addresses in header lines. For example, when using
3448 verify = header_sender/callout=no_cache
3450 the cache was (incorrectly) being used.
3452 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3453 only the Cygwin environment.
3455 11. When return_path_on_delivery was set as a log selector, if different remote
3456 addresses in the same message used different return paths and parallel
3457 remote delivery occurred, the wrong values would sometimes be logged.
3458 (Whenever a remote delivery process finished, the return path value from
3459 the most recently started remote delivery process was logged.)
3461 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3462 lines when AUTH and/or TLS are in use. This is the "received protocol"
3463 field. Exim used to use "asmtp" for authenticated SMTP, without any
3464 indication (in the protocol name) for TLS use. Now it follows the RFC and
3465 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3466 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3467 names appear in log lines as well as in Received: header lines.
3469 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3470 extensions, and comparison tests, and to fix some bugs.
3472 14. Changes to the "personal" filter test:
3474 (1) The test was buggy in that it was just doing the equivalent of
3475 "contains" tests on header lines. For example, if a user's address was
3476 anne@some.where, the "personal" test would incorrectly be true for
3478 To: susanne@some.where
3480 This test is now done by extracting each address from the header in turn,
3481 and checking the entire address. Other tests that are part of "personal"
3482 are now done using regular expressions (for example, to check local parts
3483 of addresses in From: header lines).
3485 (2) The list of non-personal local parts in From: addresses has been
3486 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3487 taken from the Sieve specification recommendations.
3489 (3) If the message contains any header line starting with "List-" it is
3490 treated as non-personal.
3492 (4) The test for "circular" in the Subject: header line has been removed
3493 because it now seems ill-conceived.
3495 15. Minor typos in src/EDITME comments corrected.
3497 16. Installed latest exipick from John Jetmore.
3499 17. If headers_add on a router specified a text string that was too long for
3500 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3501 of string_sprintf() is now avoided.
3503 18. $message_body_size was not set (it was always zero) when running the DATA
3504 ACL and the local_scan() function.
3506 19. For the "mail" command in an Exim filter, no default was being set for
3507 the once_repeat time, causing a random time value to be used if "once" was
3508 specified. (If the value happened to be <= 0, no repeat happened.) The
3509 default is now 0s, meaning "never repeat". The "vacation" command was OK
3510 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3511 (I found it when inspecting the code).
3513 20. There is now an overall timeout for performing a callout verification. It
3514 defaults to 4 times the callout timeout, which applies to individual SMTP
3515 commands during the callout. The overall timeout applies when there is more
3516 than one host that can be tried. The timeout is checked before trying the
3517 next host. This prevents very long delays if there are a large number of
3518 hosts and all are timing out (e.g. when the network connections are timing
3519 out). The value of the overall timeout can be changed by specifying an
3520 additional sub-option for "callout", called "maxwait". For example:
3522 verify = sender/callout=5s,maxwait=20s
3524 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3525 to the end before writing, but this should make it even safer).
3527 22. Exim was forgetting that it had advertised PIPELINING for the second and
3528 subsequent messages on an SMTP connection. It was also not resetting its
3529 memory on STARTTLS and an internal HELO.
3531 23. When Exim logs an SMTP synchronization error within a session, it now
3532 records whether PIPELINING has been advertised or not.
3534 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3535 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3536 rather than long int.
3538 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3540 26. Added the never_mail option to autoreply.
3546 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3547 crash if the getsockname() call failed; this can happen if a connection is
3548 closed very soon after it is established. The problem was simply in the
3549 order in which certain operations were done, causing Exim to try to write
3550 to the SMTP stream before it had set up the file descriptor. The bug has
3551 been fixed by making things happen in the correct order.
3557 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3558 before the connection was closed, thus losing the rejection response.
3560 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3561 some early Solaris releases, but causes trouble in current releases where
3562 socklen_t is defined.
3564 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3567 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3570 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3571 that releasing the top of it at the end releases what was used for sub-
3572 expansions (unless the block got too big). However, discard this block if
3573 the first thing is a variable or header, so that we can use its block when
3574 it is dynamic (useful for very large $message_headers, for example).
3576 6. Lookups now cache *every* query, not just the most recent. A new, separate
3577 store pool is used for this. It can be recovered when all lookup caches are
3578 flushed. Lookups now release memory at the end of their result strings.
3579 This has involved some general refactoring of the lookup sources.
3581 7. Some code has been added to the store_xxx() functions to reduce the amount
3582 of flapping under certain conditions.
3584 8. log_incoming_interface used to affect only the <= reception log lines. Now
3585 it causes the local interface and port to be added to several more SMTP log
3586 lines, for example "SMTP connection from", and rejection lines.
3588 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3590 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3592 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3593 could be overwritten at the end of the current message (or the start of a
3594 new message if it was set in a HELO ACL). The value is now preserved for
3595 the duration of the SMTP connection.
3597 12. If a transport had a headers_rewrite setting, and a matching header line
3598 contained an unqualified address, that address was qualified, even if it
3599 did not match any rewriting rules. The underlying bug was that the values
3600 of the flags that permit the existence of unqualified sender and recipient
3601 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3602 non-local messages, and by -bnq for local messages) were not being
3603 preserved with the message after it was received.
3605 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3606 "next input=" as part of the text comprising the host identity instead of
3607 the correct text. The code was using the same buffer for two different
3608 strings. However, depending on which order the printing function evaluated
3609 its arguments, the bug did not always show up. Under Linux, for example, my
3610 test suite worked just fine.
3612 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3613 doesn't work with some older versions of Perl. It has been changed to "my",
3614 which in any case is probably the better facility to use.
3616 15. A really picky compiler found some instances of statements for creating
3617 error messages that either had too many or two few arguments for the format
3620 16. The size of the buffer for calls to the DNS resolver has been increased
3621 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3622 for addresses that have a lot of PTR records. This alleviates a problem; it
3623 does not fully solve it.
3625 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3626 buffer now truncates the list and logs the incident, which is the same
3627 action as happens when Exim is looking up a host name and its aliases.
3628 Previously in this situation something unpredictable would happen;
3629 sometimes it was "internal error: store_reset failed".
3631 18. If a server dropped the connection unexpectedly when an Exim client was
3632 using GnuTLS and trying to read a response, the client delivery process
3633 crashed while trying to generate an error log message.
3635 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3636 string, for example:
3638 warn message = H1: something\nH2: something
3640 the text was added as a single header line from Exim's point of view
3641 though it ended up OK in the delivered message. However, searching for the
3642 second and subsequent header lines using $h_h2: did not work. This has been
3643 fixed. Similarly, if a system filter added multiple headers in this way,
3644 the routers could not see them.
3646 20. Expanded the error message when iplsearch is called with an invalid key to
3647 suggest using net-iplsearch in a host list.
3649 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3650 ACLs are no longer actually imposed (and a message to that effect is
3653 22. If a "gecos" field in a passwd entry contained escaped characters, in
3654 particular, if it contained a \" sequence, Exim got it wrong when building
3655 a From: or a Sender: header from that name. A second bug also caused
3656 incorrect handling when an unquoted " was present following a character
3657 that needed quoting.
3659 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3660 was not being matched caselessly.
3662 24. Arranged for all hyphens in the exim.8 source to be escaped with
3665 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3666 properly. Recipient callout cache records were still being keyed to include
3667 the sender, even when use_sender was set false. This led to far more
3668 callouts that were necessary. The sender is no longer included in the key
3669 when use_sender is false.
3671 26. Added "control = submission" modifier to ACLs.
3673 27. Added the ${base62d: operator to decode base 62 numbers.
3675 28. dnsdb lookups can now access SRV records.
3677 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3678 the configuration file.
3680 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3681 (-v) mode. This makes the output for a verbose queue run more intelligible.
3683 31. Added a use_postmaster feature to recipient callouts.
3685 32. Added the $body_zerocount variable, containing the number of binary zero
3686 bytes in the message body.
3688 33. The time of last modification of the "new" subdirectory is now used as the
3689 "mailbox time last read" when there is a quota error for a maildir
3692 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3694 35. Added +ignore_unknown as a special item in host lists.
3696 36. Code for decoding IPv6 addresses in host lists is now included, even if
3697 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3698 address was recognized as an IP address, but was then not correctly decoded
3699 into binary, causing unexpected and incorrect effects when compared with
3706 1. Very minor rewording of debugging text in manualroute to say "list of
3707 hosts" instead of "hostlist".
3709 2. If verify=header_syntax was set, and a header line with an unqualified
3710 address (no domain) and a large number of spaces between the end of the
3711 name and the colon was received, the reception process suffered a buffer
3712 overflow, and (when I tested it) crashed. This was caused by some obsolete
3713 code that should have been removed. The fix is to remove it!
3715 3. When running in the test harness, delay a bit after writing a bounce
3716 message to get a bit more predictability in the log output.
3718 4. Added a call to search_tidyup() just before forking a reception process. In
3719 theory, someone could use a lookup in the expansion of smtp_accept_max_
3720 per_host which, without the tidyup, could leave open a database connection.
3722 5. Added the variables $recipient_data and $sender_data which get set from a
3723 lookup success in an ACL "recipients" or "senders" condition, or a router
3724 "senders" option, similar to $domain_data and $local_part_data.
3726 6. Moved the writing of debug_print from before to after the "senders" test
3729 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3730 problems for message scanning, either using a data ACL, or using
3731 local_scan() because the Received: header was not generated till after they
3732 were called (in order to set the time as the time of reception completion).
3733 I have revised the way this works. The header is now generated after the
3734 body is received, but before the ACL or local_scan() are called. After they
3735 are run, the timestamp in the header is updated.
3741 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3742 before starting a queue runner without re-exec. This happened only when
3743 deliver_drop_privilege was set or when the Exim user was set to root. The
3744 effect of the bug was that timeouts during subsequent deliveries caused
3745 crashes instead of being properly handled. The handler is now left at its
3746 default (and expected) setting.
3748 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3749 message, again when deliver_drop_privilege is set or Exim is run as root.
3750 The bug described in (1) was not present in this case, but the tidying up
3751 of the other signals was missing. I have made the two cases consistent.
3753 3. The ignore_target_hosts setting on a manualroute router was being ignored
3754 for hosts that were looked up using the /MX notation.
3756 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3759 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3760 operated on the sender address. After changing the $sender_address to <>
3761 for the sender address verify, Exim was re-instated it as the original
3762 (before rewriting) address, but remembering that it had rewritten it, so it
3763 wasn't rewriting it again. This bug also had the effect of breaking the
3764 sender address verification caching when the sender address was rewritten.
3766 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3767 This has been changed so that if the ip literal address matches
3768 ignore_target_hosts, the router declines.
3770 7. Added expansion conditions match_domain, match_address, and match_local_
3771 part (NOT match_host).
3773 8. The placeholder for the Received: header didn't have a length field set.
3775 9. Added code to Exim itself and to exim_lock to test for a specific race
3776 condition that could lead to file corruption when using MBX delivery. The
3777 issue is with the lockfile that is created in /tmp. If this file is removed
3778 after a process has opened it but before that process has acquired a lock,
3779 there is the potential for a second process to recreate the file and also
3780 acquire a lock. This could lead to two Exim processes writing to the file
3781 at the same time. The added code performs the same test as UW imapd; it
3782 checks after acquiring the lock that its file descriptor still refers to
3783 the same named file.
3785 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3786 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3789 11. Added the smtp_active_hostname option. If used, this will typically be made
3790 to depend on the incoming interface address. Because $interface_address is
3791 not set up until the daemon has forked a reception process, error responses
3792 that can happen earlier (such as "too many connections") no longer contain
3795 12. If an expansion in a condition on a "warn" statement fails because a lookup
3796 defers, the "warn" statement is abandoned, and the next ACL statement is
3797 processed. Previously this caused the whole ACL to be aborted.
3799 13. Added the iplsearch lookup type.
3801 14. Added ident_timeout as a log selector.
3803 15. Added tls_certificate_verified as a log selector.
3805 16. Added a global option tls_require_ciphers (compare the smtp transport
3806 option of the same name). This controls incoming TLS connections.
3808 17. I finally figured out how to make tls_require_ciphers do a similar thing
3809 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3810 before starting the TLS session.
3812 18. Tabs are now shown as \t in -bP output.
3814 19. If the log selector return_path_on_delivery was set, Exim crashed when
3815 bouncing a message because it had too many Received: header lines.
3817 20. If two routers both had headers_remove settings, and the first one included
3818 a superfluous trailing colon, the final name in the first list and the
3819 first name in the second list were incorrectly joined into one item (with a
3820 colon in the middle).
3826 1. Added -C and -D options to the exinext utility, mainly to make it easier
3827 to include in the automated testing, but these could be helpful when
3828 multiple configurations are in use.
3830 2. The exinext utility was not formatting the output nicely when there was
3831 an alternate port involved in the retry record key, nor when there was a
3832 message id as well (for retries that were specific to a specific message
3833 and a specific host). It was also confused by IPv6 addresses, because of
3834 the additional colons they contain. I have fixed the IPv4 problem, and
3835 patched it up to do a reasonable job for IPv6.
3837 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3838 delivery, the log line now contains "pipelined" if PIPELINING was used.
3840 4. An SMTP transport process used to panic and die if the bind() call to set
3841 an explicit outgoing interface failed. This has been changed; it is now
3842 treated in the same way as a connect() failure.
3844 5. A reference to $sender_host_name in the part of a conditional expansion
3845 that was being skipped was still causing a DNS lookup. This no longer
3848 6. The def: expansion condition was not recognizing references to header lines
3849 that used bh_ and bheader_.
3851 7. Added the _cache feature to named lists.
3853 8. The code for checking quota_filecount in the appendfile transport was
3854 allowing one more file than it should have been.
3856 9. For compatibility with Sendmail, the command line option
3864 and sets the incoming protocol and host name (for trusted callers). The
3865 host name and its colon can be omitted when only the protocol is to be set.
3866 Note the Exim already has two private options, -pd and -ps, that refer to
3867 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3868 "s", but I don't think that's a major issue.
3870 10. A number of refactoring changes to the code, none of which should affect
3873 (a) The number of logging options was getting close to filling up the
3874 32-bit word that was used as a bit map. I have split them into two classes:
3875 those that are passed in the argument to log_write(), and those that are
3876 only ever tested independently outside of that function. These are now in
3877 separate 32-bit words, so there is plenty of room for expansion again.
3878 There is no change in the user interface or the logging behaviour.
3880 (b) When building, for example, log lines, the code previously used a
3881 macro that called string_cat() twice, in order to add two strings. This is
3882 not really sufficiently general. Furthermore, there was one instance where
3883 it was actually wrong because one of the argument was used twice, and in
3884 one call a function was used. (As it happened, calling the function twice
3885 did not affect the overall behaviour.) The macro has been replaced by a
3886 function that can join an arbitrary number of extra strings onto a growing
3889 (c) The code for expansion conditions now uses a table and a binary chop
3890 instead of a serial search (which was left over from when there were very
3891 few conditions). Also, it now recognizes conditions like "pam" even when
3892 the relevant support is not compiled in: a suitably worded error message is
3893 given if an attempt is made to use such a condition.
3895 11. Added ${time_interval:xxxxx}.
3897 12. A bug was causing one of the ddress fields not to be passed back correctly
3898 from remote delivery subprocesses. The field in question was not being
3899 subsequently used, so this caused to problems in practice.
3901 13. Added new log selectors queue_time and deliver_time.
3903 14. Might have fixed a bug in maildirsizefile handling that threw up
3904 "unexpected character" debug warnings, and recalculated the data
3905 unnecessarily. In any case, I expanded the warning message to give more
3908 15. Added the message "Restricted characters in address" to the statements in
3909 the default ACL that block characters like @ and % in local parts.
3911 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3912 Three changes have been made:
3914 (a) There was a serious bug; a negative response to MAIL caused the whole
3915 recipient domain to be cached as invalid, thereby blocking all messages
3916 to all local parts at the same domain, from all senders. This bug has
3917 been fixed. The domain is no longer cached after a negative response to
3918 MAIL if the sender used is not empty.
3920 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3923 (c) A new callout option, "use_sender" has been added for people who want
3924 the modified behaviour.
3930 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3933 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3934 indeed breaks things for older releases.
3936 3. Added additional logging to the case where there is a problem reading data
3937 from a filter that is running in a subprocess using a pipe, in order to
3938 try to track down a specific problem.
3940 4. Testing facility fudge: when running in the test harness and attempting
3941 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3942 getting "No route to host". Convert this to a timeout.
3944 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3947 6. Some OS don't have socklen_t but use size_t instead. This affects the
3948 fifth argument of getsockopt() amongst other things. This is now
3949 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3950 can be set for individual OS. I have set it for SunOS5, OSF1, and
3951 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3952 some earlier ones do not.
3954 7. Change 4.30/15 was not doing the test caselessly.
3956 8. The standard form for an IPv6 address literal was being rejected by address
3957 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3958 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3959 this, as well as the form without the "IPv6" on the front (but only when
3960 address literals are enabled, of course).
3962 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3964 10. Exim crashed if a message with an empty sender address specified by -f
3965 encountered a router with an errors_to setting. This could be provoked only
3966 by a command such as
3970 where an empty string was supplied; "<>" did not hit this bug.
3972 11. Installed PCRE release 4.5.
3974 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3975 remained set. It is now erased.
3977 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3978 times from message ids (which are base 36 rather than the normal 62).
3980 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3981 were being counted as actual protocol errors, and logged if the log
3982 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3983 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3984 and DATA following a set of rejected RCPTs do not count as protocol errors.
3985 In other words, Exim assumes they were pipelined, though this may not
3986 actually be the case. Of course, in all cases the client gets an
3987 appropriate error code.
3989 15. If a lookup fails in an ACL condition, a message about the failure may
3990 be available; it is used if testing the ACL cannot continue, because most
3991 such messages specify what the cause of the deferral is. However, some
3992 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3993 that caused an old message to be retained and used if a later statement
3994 caused a defer, replacing the real cause of the deferral.
3996 16. If an IP address had so many PTR records that the DNS lookup buffer
3997 was not large enough to hold them, Exim could crash while trying to process
3998 the truncated data. It now detects and logs this case.
4000 17. Further to 4.21/58, another change has been made: if (and only if) the
4001 first line of a message (the first header line) ends with CRLF, a bare LF
4002 in a subsequent header line has a space inserted after it, so as not to
4003 terminate the header.
4005 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4006 unnecessarily, used atoi() instead of strtol(), and didn't check the
4007 termination when getting file sizes from file names by regex.
4009 19. Completely re-implemented the support for maildirsize files, in the light
4010 of a number of problems with the previous contributed implementation
4011 (4.30/29). In particular:
4013 . If the quota is zero, the maildirsize file is maintained, but no quota is
4016 . If the maildir directory does not exist, it is created before any attempt
4017 to write a maildirsize file.
4019 . The quota value in the file is just a cache; if the quota is changed in
4020 the transport, the new value overrides.
4022 . A regular expression is available for excluding directories from the
4025 20. The autoreply transport checks the characters in options that define the
4026 message's headers; it allows continued headers, but it was checking with
4027 isspace() after an embedded newline instead of explicitly looking for a
4030 21. If all the "regular" hosts to which an address was routed had passed their
4031 expiry times, and had not reached their retry times, the address was
4032 bounced, even if fallback hosts were defined. Now Exim should go on to try
4035 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4036 equivalent code in the SMTP transport. Some hosts send humungous responses
4037 to HELO/EHLO, more than 1024 it seems.
4039 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4040 gives compiler warnings in some environments. I've now done it "properly",
4043 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4044 (because of problems with the built-in one) was declared to return uschar *
4045 instead of char *, causing compiler failure.
4047 25. Fixed a file descriptor leak when processing alias/forward files.
4049 26. Fixed a minor format string issue in dbfn.c.
4051 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4053 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4054 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4057 29. When a sender address is verified, it is cached, to save repeating the test
4058 when there is more than one recipient in a message. However, when the
4059 verification involves a callout, it is possible for different callout
4060 options to be set for different recipients. It is too complicated to keep
4061 track of this in the cache, so now Exim always runs a verification when a
4062 callout is required, relying on the callout cache for the optimization.
4063 The overhead is duplication of the address routing, but this should not be
4066 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4067 to be verified with callout=postmaster, and the main callout worked but the
4068 postmaster check failed, the verification correctly failed. However, if a
4069 subsequent RCPT command asked for sender verification *without* the
4070 postmaster check, incorrect caching caused this verification also to fail,
4073 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4074 it was not caching the DNS options (qualify_single, search_parents) that
4075 were used when the lookup failed. A subsequent lookup with different
4076 options therefore always gave the same answer, though there were cases
4077 where it should not have. (Example: a "domains = !$mx_any" option on a
4078 dnslookup router: the "domains" option is always processed without any
4079 widening, but the router might have qualify_single set.) Now Exim uses the
4080 cached value only when the same options are set.
4082 32. Added John Jetmore's "exipick" utility to the distribution.
4084 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4085 than a timeout (e.g. a certificate is required, and is not provided), an
4086 Exim server now closes the connection immediately. Previously it waited for
4087 the client to close - but if the client is SSL, it seems that they each
4088 wait for each other, leading to a delay before one of them times out.
4090 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4091 maintained 0.8.x compatibility because I don't think many are using it, and
4092 it is clearly obsolete.
4094 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4097 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4098 client certificate was expired. A simple patch fixes this, though I don't
4099 understand the full logic of why the verify callback is called multiple
4102 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4103 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4104 which causes problems with some clients (such as the Certicom SSL Plus
4105 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4106 disables the coutermeasure allowing Eudora to connect."
4108 38. Exim was not checking that a write() to a log file succeeded. This could
4109 lead to Bad Things if a log got too big, in particular if it hit a file
4110 size limit. Exim now panics and dies if it cannot write to a log file, just
4111 as it does if it cannot open a log file.
4113 39. Modified OS/Makefile-Linux so that it now contains
4115 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4117 The two -D definitions ensure that Exim is compiled with large file
4118 support, which makes it possible to handle log files that are bigger than
4121 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4122 instance) a domain was checked against a named list that involved a lookup,
4123 causing $domain_data to be set, then another domain was checked against the
4124 same list, then the first domain was re-checked, the value of $domain_data
4125 after the final check could be wrong. In particular, if the second check
4126 failed, it could be set empty. This bug probably also applied to
4129 41. The strip_trailing_dot option was not being applied to the address given
4130 with the -f command-line option.
4132 42. The code for reading a message's header from the spool was incrementing
4133 $received_count, but never initializing it. This meant that the value was
4134 incorrect (doubled) while delivering a message in the same process in which
4135 it was received. In the most common configuration of Exim, this never
4136 happens - a fresh exec is done - but it can happen when
4137 deliver_drop_privilege is set.
4139 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4140 it now includes up to 150 characters of the unexpected data in the log
4143 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4144 and building data strings. The size of both of these buffers was 10 000
4145 bytes - far larger than anybody would *ever* want, thought I. Needless to
4146 say, somebody hit the limit. I have increased the maximum line length to
4147 20 000 and the maximum data length of concatenated lines to 100 000. I have
4148 also fixed two bugs, because there was no checking on these buffers. Tsk,
4149 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4150 buffer is too small.
4152 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4153 lsearch lookups. Now it does.
4155 46. When parsing a route_list item in a manualroute router, a fixed-length
4156 buffer was used for the list of hosts. I made this 1024 bytes long,
4157 thinking that nobody would ever have a list of hosts that long. Wrong.
4158 Somebody had a whole pile of complicated expansion conditions, and the
4159 string was silently truncated, leading to an expansion error. It turns out
4160 that it is easier to change to an unlimited length (owing to other changes
4161 that have happened since this code was originally written) than to build
4162 structure for giving a limitation error. The length of the item that
4163 expands into the list of hosts is now unlimited.
4165 47. The lsearch lookup could not handle data where the length of text line was
4166 more than 4095 characters. Such lines were truncated, leading to shortened
4167 data being returned. It should now handle lines of any length.
4169 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4170 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4173 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4174 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4175 no longer happens in comment lines. A list of replacements is now placed
4176 at the head of all of the source files, except those whose only change is
4177 to replace PERL_COMMAND in the very first #! line.
4179 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4180 messages on the queue, with a bottom-up merge sort, using code contributed
4181 by Michael Haardt. This should make operations like -bp somewhat faster on
4182 large queues. It won't affect queue runners, except when queue_run_in_order
4185 51. Installed eximstats 1.31 in the distribution.
4187 52. Added support for SRV lookups to the dnslookup router.
4189 53. If an ACL referred to $message_body or $message_body_end, the value was not
4190 reset for any messages that followed in the same SMTP session.
4192 54. The store-handling optimization for building very long strings was not
4193 differentiating between the different store pools. I don't think this
4194 actually made any difference in practice, but I've tidied it.
4196 55. While running the routers to verify a sender address, $sender_address
4197 was still set to the sender address. This is wrong, because when routing to
4198 send a bounce to the sender, it would be empty. Therefore, I have changed
4199 it so that, while verifying a sender address, $sender_address is set to <>.
4200 (There is no change to what happens when verifying a recipient address.)
4202 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4203 target A or AAAA records (if not already returned) without resetting the
4204 qualify_single or search_parents options of the DNS resolver. These are
4205 inappropriate in this case because the targets of MX and SRV records must
4206 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4207 target that, when qualified, matched something in the local domain. These
4208 two options are now turned off when doing these lookups.
4210 57. It seems that at least some releases of Reiserfs (which does not have the
4211 concept of a fixed number of inodes) returns zero and not -1 for the
4212 number of available inodes. This interacted badly with check_spool_inodes,
4213 which assumed that -1 was the "no such thing" setting. What I have done is
4214 to check that the total number of inodes is greater than zero before doing
4215 the test of how many are available.
4217 58. When a "warn" ACL statement has a log_message modifier, the message is
4218 remembered, and not repeated. This is to avoid a lot of repetition when a
4219 message has many recipients that cause the same warning to be written.
4220 Howewer, Exim was preserving the list of already written lines for an
4221 entire SMTP session, which doesn't seem right. The memory is now reset if a
4222 new message is started.
4224 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4225 debugging output unless log_rewrite was also set.
4227 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4228 of (int)(handle) when we know that handle contains (void *)(-1).
4230 61. The Exim daemon panic-logs an error return when it closes the incoming
4231 connection. However "connection reset by peer" seems to be common, and
4232 isn't really an error worthy of noting specially, so that particular error
4235 62. When Exim is trying to find all the local interfaces, it used to panic and
4236 die if the ioctl to get the interface flags failed. However, it seems that
4237 on at least one OS (Solaris 9) it is possible to have an interface that is
4238 included in the list of interfaces, but for which you get a failure error
4239 for this call. This happens when the interface is not "plumbed" into a
4240 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4241 failure of the "get flags" call assumes that the interface is down.
4243 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4244 makes life easier for people who are doing arithmetic on fields extracted
4245 from dates, where you often get leading zeros that should not be
4246 interpreted as octal.
4248 64. Added qualify_domain to the redirect router, to override the global
4251 65. If a pathologically long header line contained very many addresses (the
4252 report of this problem mentioned 10 000) and each of them was rewritten,
4253 Exim could use up a very large amount of memory. (It kept on making new
4254 copies of the header line as it rewrote, and never released the old ones.)
4255 At the expense of a bit more processing, the header rewriting function has
4256 been changed so that it no longer eats memory in this way.
4258 66. The generation of the Received: header has been moved from the time that a
4259 message starts to be received, to the time that it finishes. The timestamp
4260 in the Received: header should now be very close to that of the <= log
4261 line. There are two side-effects of this change:
4263 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4264 logged header lines no longer include the local Received: line, because
4265 it has not yet been created. The same applies to a copy of the message
4266 that is returned to a non-SMTP sender when a message is rejected.
4268 (b) When a filter file is tested using -bf, no additional Received: header
4269 is added to the test message. After some thought, I decided that this
4272 This change does not affect the value of $received_for. It is still set
4273 after address rewriting, but before local_scan() is called.
4275 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4277 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4278 gave an unhelpful panic error message, and a defer error. I have managed to
4279 change this behaviour so that it now rejects any supplied certificate,
4280 which seems right, as the list of acceptable certificates is empty.
4282 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4283 gave an unhelpful defer error. I have not managed to make this reject any
4284 supplied certificates, but the error message it gives is "no certificate
4285 supplied", which is not helpful.
4287 70. exigrep's output now also includes lines that are not associated with any
4288 message, but which match the given pattern. Implemented by a patch from
4289 Martin Sluka, which also tidied up the Perl a bit.
4291 71. Recipient callout verification, like sender verification, was using <> in
4292 the MAIL FROM command. This isn't really the right thing, since the actual
4293 sender may affect whether the remote host accepts the recipient or not. I
4294 have changed it to use the actual sender in the callout; this means that
4295 the cache record is now keyed on a recipient/sender pair, not just the
4296 recipient address. There doesn't seem to be a real danger of callout loops,
4297 since a callout by the remote host to check the sender would use <>.
4298 [SEE ABOVE: changed after hitting problems.]
4300 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4301 temporary errors. However, in the case of such a code being given after
4302 the end of a data transmission (i.e. after ".") Exim was failing to write
4303 a retry record for the message. (Yes, there was some broken host that was
4304 actually sending 8xx at this point.)
4306 73. An unknown lookup type in a host list could cause Exim to panic-die when
4307 the list was checked. (An example that provoked this was putting <; in the
4308 middle of a list instead of at the start.) If this happened during a DATA
4309 ACL check, a -D file could be left lying around. This kind of configuration
4310 error no longer causes Exim to die; instead it causes a defer errror. The
4311 incident is still logged to the main and panic logs.
4313 74. Buglet left over from Exim 3 conversion. The message "too many messages
4314 in one connection" was written to the rejectlog but not the mainlog, except
4315 when address rewriting (yes!) was being logged.
4317 75. Added write_rejectlog option.
4319 76. When a system filter was run not as root (that is, when system_filter_user
4320 was set), the values of the $n variables were not being returned to the
4321 main process; thus, they were not subsequently available in the $sn
4324 77. Added +return_path_on_delivery log selector.
4326 78. A connection timeout was being treated differently from recipients deferred
4327 when testing hosts_max_try with a message that was older than the host's
4328 retry timeout. (The host should not be counted, thus allowing all hosts to
4329 be tried at least once before bouncing.) This may have been the cause of an
4330 occasionally reported bug whereby a message would remain on the queue
4331 longer than the retry timeout, but would be bounced if a delivery was
4332 forced. I say "may" because I never totally pinned down the problem;
4333 setting up timeout/retry tests is difficult. See also the next item.
4335 79. The ultimate address timeout was not being applied to errors that involved
4336 a combination of host plus message (for example, a timeout on a MAIL
4337 command). When an address resolved to a number of possible hosts, and they
4338 were not all tried for each delivery (e.g. because of hosts_max_try), a
4339 message could remain on the queue longer than the retry timeout.
4341 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4344 81. Fixed an obscure SMTP outgoing bug which required at least the following
4345 conditions: (a) there was another message waiting for the same server;
4346 (b) the server returned 5xx to all RCPT commands in the first message so
4347 that the message was not completed; (c) the server dropped the connection
4348 or gave a negative response to the RSET that Exim sends to abort the
4349 transaction. The observed case was a dropped connection after DATA that had
4350 been sent in pipelining mode. That is, the server had advertised PIPELINING
4351 but was not implementing it correctly. The effect of the bug was incorrect
4352 behaviour, such as trying another host, and this could lead to a crash.
4358 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4359 and daemon.c were passed as pointers to ints; they should have been
4360 pointers to socklen_t variables (which are typically unsigned ints).
4362 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4365 3. Fixed a really odd bug that affected only the testing scheme; patching a
4366 certain fixed string in the binary changed the value of another string that
4367 happened to be identical to the end of the original first string.
4369 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4370 name", it returns that address as the IP address. On some operating
4371 systems (e.g. Solaris), it also passes back the IP address string as the
4372 "host name". However, on others (e.g. Linux), it passes back an empty
4373 string. Exim wasn't checking for this, and was changing the host name to an
4374 empty string, assuming it had been canonicized.
4376 5. Although rare, it is permitted to have more than one PTR record for a given
4377 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4378 all the names associated with an address, because they do in Solaris.
4379 However, it seems that they do not in Linux for data that comes from the
4380 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4381 I found this out when I moved to a new Linux workstation and tried to run
4382 the Exim test suite.
4384 To get round this problem I have changed the code so that it now does its
4385 own call to the DNS to look up PTR records when searching for a host name.
4386 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4387 addresses that are only in /etc/hosts are still found.
4389 This behaviour is, however, controlled by an option called host_lookup_
4390 order, which defaults to "bydns:byaddr". If people want to use the other
4391 order, or indeed, just use one or the other means of lookup, they can
4392 specify it in this variable.
4394 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4395 some operating systems, this comes back from gethostbyaddr() as an empty
4396 string, and this is what Exim used to test for. However, it seems that in
4397 other systems, "." is yielded. Exim now tests for this case too.
4399 7. The values of check_spool_space and check_log_space are now held internally
4400 as a number of kilobytes instead of an absolute number of bytes. If a
4401 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4402 kilobyte. This means that much larger values can be stored.
4404 8. Exim monitor: an attempt to get the action menu when not actually pointing
4405 at a message produces an empty menu entitled "No message selected". This
4406 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4407 no entries in it ("Shell widget menu has zero width and/or height"). So I
4408 have added a single, blank menu entry in this case.
4410 9. Added ${quote_local_part.
4412 10. MIME decoding is now applied to the contents of Subject: header lines when
4415 11. Now that a reference to $sender_host_address automatically causes a reverse
4416 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4417 host lookup before query-style lookups in lists that might use this
4418 variable. This has therefore been abolished, and the "net-" prefix is no
4419 longer necessary for query-style lookups.
4421 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4422 have been a typo for LFLAGS, so it has been changed.
4424 13. The install script calls Exim with "-C /dev/null" in order to find the
4425 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4426 to be output. Howeve, since Exim outputs its version number before the
4427 error, it didn't break the script. It just looked ugly. I fixed this by
4428 always allowing "-C /dev/null" if the caller is root.
4430 14. Ignore overlarge ACL variable number when reading spool file - insurance
4431 against a later release with more variables having written the file.
4433 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4434 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4435 this, as well as the form without the "IPv6" on the front.
4437 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4438 OS/Makefile-Darwin file.
4440 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4441 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4443 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4444 variable. However, some OS use ints for inodes. I've added cast to long int
4445 to get rid of the compiler warning.
4447 19. I had forgotten to lock out "/../" in configuration file names when
4448 ALT_CONFIG_PREFIX was set.
4450 20. Routers used for verification do not need to specify transports. However,
4451 if such a router generated a host list, and callout was configured, Exim
4452 crashed, because it could not find a port number from the (non-existent)
4453 transport. It now assumes port 25 in this circumstance.
4455 21. Added the -t option to exigrep.
4457 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4458 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4459 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4460 them is set, LOOKUP_LSEARCH is forced.
4462 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4464 24. Added sender and host information to the "rejected by local_scan()" log
4465 line; previously there was no indication of these.
4467 25. Added .include_if_exists.
4469 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4470 Linux. It turns out that not all file systems support this. Apparently some
4471 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4472 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4473 not supported on the file descriptor, is now ignored when Exim is trying to
4474 sync a directory. This applies only to Linux.
4476 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4478 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4479 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4482 29. Added support for maildirsize files from supplied patch (modified a bit).
4484 30. The use of :fail: followed by an empty string could lead Exim to respond to
4485 sender verification failures with (e.g.):
4487 550 Verification failed for <xxx>
4488 550 Sender verify failed
4490 where the first response line was missing the '-' that indicates it is not
4491 the final line of the response.
4493 31. The loop for finding the name of the user that called Exim had a hardwired
4494 limit of 10; it now uses the value of finduser_retries, which is used for
4495 all other user lookups.
4497 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4500 33. Exim was neglecting to zero errno before one call of strtol() when
4501 expanding a string and expecting an integer value. On some systems this
4502 resulted in spurious "integer overflow" errors. Also, it was casting the
4503 result into an int without checking.
4505 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4506 did not work. The code looks as if it has *never* worked, though it appears
4507 to have been documented since at least releast 1.62. I have made it work.
4509 35. The "timeout_DNS" error in retry rules, also documented since at least
4510 1.62, also never worked. As it isn't clear exactly what this means, and
4511 clearly it isn't a major issue, I have abolished the feature by treating it
4512 as "timeout", and writing a warning to the main and panic logs.
4514 36. The display of retry rules for -brt wasn't always showing the error code
4517 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4518 timeout_connect_A, timeout_connect_MX.
4520 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4521 to the empty sender.
4523 39. The daemon was not analysing the content of -oX till after it had closed
4524 stderr and disconnected from the controlling terminal. This meant that any
4525 syntax errors were only noted on the panic log, and the return code from
4526 the command was 0. By re-arranging the code a little, I've made the
4527 decoding happen first, so such errors now appear on stderr, and the return
4528 code is 1. However, the actual setting up of the sockets still happens in
4529 the disconnected process, so errors there are still only recorded on the
4532 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4533 connections (as happens on some IP stacks) was logged at start up time as
4534 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4535 it from "IPv6 and IPv4", which means that two separate sockets are being
4538 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4539 says whether AF_INET or AF_INET6 was passed as an argument.
4541 42. Exiwhat output was messed up when time zones were included in log
4544 43. Exiwhat now gives more information about the daemon's listening ports,
4545 and whether -tls-on-connect was used.
4547 44. The "port" option of the smtp transport is now expanded.
4549 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4550 silently ignored. Now an error message is written to the main and panic
4553 46. There's a new ACL modifier called "logwrite" which writes to a log file
4554 as soon as it is encountered.
4556 47. Added $local_user_uid and $local_user_gid at routing time.
4558 48. Exim crashed when trying to verify a sender address that was being
4561 49. Exim was recognizing only a space character after ".include". It now also
4562 recognizes a tab character.
4564 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4565 extracting the relevant information from the specification. The man page no
4566 longer contains scrambled data for the -d option, and I've added a section
4567 at the front about calling Exim under different names.
4569 51. Added "extra_headers" argument to the "mail" command in filter files.
4571 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4574 53. Installed eximstats 1.29.
4576 54. Added transport_filter_timeout as a generic transport option.
4578 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4579 Cc: header lines. This was required by RFC 822, but it not required by RFC
4582 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4583 incoming messages that did not have them. Now it does so only if the
4584 message originates locally, that is, if there is no associated remote host
4585 address. When Resent- header lines are present, this applies to the Resent-
4586 lines rather than the non-Resent- lines.
4588 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4589 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4591 58. Messages for configuration errors now include the name of the main
4592 configuration file - useful now that there may be more than one file in a
4593 list (.included file names were always shown).
4595 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4596 for those rare installations that do not start the daemon as root or run it
4597 setuid root. I've cut out the call to initgroups() if the daemon is not
4600 60. The Exim user and group can now be bound into the binary as text strings
4601 that are looked up at the start of Exim's processing.
4603 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4605 62. Added $mailstore_basename variable.
4607 63. Installed patch to sieve.c from Michael Haardt.
4609 64. When Exim failed to open the panic log after failing to open the main log,
4610 the original message it was trying to log was written to stderr and debug
4611 output, but if they were not available (the usual case in production), it
4612 was lost. Now it is written to syslog before the two lines that record the
4613 failures to open the logs.
4615 65. Users' Exim filters run in subprocesses under the user's uid. It is
4616 possible for a "deliver" command or an alias in a "personal" command to
4617 provoke an address rewrite. If logging of address rewriting is configured,
4618 this fails because the process is not running as root or exim. There may be
4619 a better way of dealing with this, but for the moment (because 4.30 needs
4620 to be released), I have disabled address rewrite logging when running a
4621 filter in a non-root, non-exim process.
4627 1. The buildconfig auxiliary program wasn't quoting the value set for
4628 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4629 not defined. This bug was masked in 4.22 by the effect that was fixed in
4632 2. Some messages that were rejected after a message id was allocated were
4633 shown as "incomplete" by exigrep. It no longer does this for messages that
4634 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4636 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4637 have allow_domain_literals set, the ID did not get logged in the <= line.
4638 Domain literals are now always recognized in Message-ID: header lines.
4640 4. The first argument for a ${extract expansion item is the key name or field
4641 number. Leading and trailing spaces in this item were not being ignored,
4642 causing some misleading effects.
4644 5. When deliver_drop_privilege was set, single queue runner processes started
4645 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4646 same command in the process it spins off) were not dropping privilege.
4648 6. When the daemon running as "exim" started a queue runner, it always
4649 re-executed Exim in the spun-off process. This is a waste of effort when
4650 deliver_drop_privilege is set. The new process now just calls the
4651 queue-runner function directly.
4657 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4660 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4661 ignored. Though the use of -oP was forcing the writing of a pid file, it
4662 was always written to the default place.
4664 3. If the message "no IP address found for host xxxx" is generated during
4665 incoming verification, it is now followed by identification of the incoming
4666 connection (so you can more easily find what provoked it).
4668 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4670 5. Added some features to "harden" Exim a bit more against certain attacks:
4672 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4673 be put in Local/Makefile. This is like the never_users runtime option,
4674 but it cannot be overridden. The default setting is "root".
4676 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4677 prefix string with which any file named in a -C command line option
4680 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4681 is retained for -C and -D only if the caller of Exim is root. Without
4682 it, the exim user may also use -C and -D and retain privilege.
4684 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4685 command line option is disabled.
4687 6. Macro names set by the -D option must start with an upper case letter, just
4688 like macro names defined in the configuration file.
4690 7. Added "dereference=" facility to LDAP.
4692 8. Two instances of the typo "uknown" in the source files are fixed.
4694 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4695 the Configure-Makefile script screwed up while processing it.
4697 10. Incorporated PCRE 4.4.
4699 11. The SMTP synchronization check was not operating right at the start of an
4700 SMTP session. For example, it could not catch a HELO sent before the client
4701 waited for the greeting. There is now a check for outstanding input at the
4702 point when the greeting is written. Because of the duplex, asynchronous
4703 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4704 way, but not yet received, when the check is performed.
4706 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4707 on TCP/IP sockets, because this apparently causes some broken clients to
4710 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4711 unchanged) from the Cygwin maintainer.
4713 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4714 was supported instead of testing for mbx. Effectively a typo.
4716 15. The spa authenticator server code was not checking that the input it
4717 received was valid base64.
4719 16. The debug output line for the "set" modifier in ACLs was not showing the
4720 name of the variable that was being set.
4722 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4724 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4725 lookup on its own. Something else was needed to trigger the lookup. For
4726 example, a match in host_lookup or the need for a host name in a host list.
4727 Now, if $sender_host_name is referenced and the host name has not yet been
4728 looked up, a lookup is performed. If the lookup fails, the variable remains
4729 empty, and $host_lookup_failed is set to "1".
4731 19. Added "eqi" as a case-independent comparison operator.
4733 20. The saslauthd authentication condition could segfault if neither service
4734 nor realm was specified.
4736 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4737 error message that was logged was misleading, and incoming SMTP
4738 connections were dropped. The message is now more accurate, and temporary
4739 errors are given to SMTP connections.
4741 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4742 (except RSET) with a 421 error, until QUIT is received. However, it was
4743 failing to send a response to QUIT.
4745 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4746 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4747 ACL is now run after the helo_try_verify_hosts code.
4749 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4750 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4751 case-independent, so other case variants are also recognized). Apparently
4752 some systems use these upper case variants.
4754 25. If more than two messages were waiting for the same host, and a transport
4755 filter was specified for the transport, Exim sent two messages over the
4756 same TCP/IP connection, and then failed with "socket operation on non-
4757 socket" when it tried to send the third.
4759 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4761 27. The extern definition of crypt16() in expand.c was not being excluded when
4762 the OS had its own crypt16() function.
4764 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4765 as a preferred synonym for return_size_limit, both as an option and as an
4768 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4770 30. Changed the default configuration ACL to relax the local part checking rule
4771 for addresses that are not in any local domains. For these addresses,
4772 slashes and pipe symbols are allowed within local parts, but the sequence
4773 /../ is explicitly forbidden.
4775 31. SPA server authentication was not clearing the challenge buffer before
4778 32. log_message in a "warn" ACL statement was writing to the reject log as
4779 well as to the main log, which contradicts the documentation and doesn't
4780 seem right (because no rejection is happening). So I have stopped it.
4782 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4783 However, I am unable to do any testing of this.
4785 34. Fixed an infelicity in the appendfile transport. When checking directories
4786 for a mailbox, to see if any needed to be created, it was accidentally
4787 using path names with one or more superfluous leading slashes; tracing
4788 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4790 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4791 contents are added to the log line that is written for every discarded
4792 recipient. (Previously a log_message setting was ignored.)
4794 36. The ${quote: operator now quotes the string if it is empty.
4796 37. The install script runs exim in order to find its version number. If for
4797 some reason other than non-existence or emptiness, which it checks, it
4798 could not run './exim', it was installing it with an empty version number,
4799 i.e. as "exim-". This error state is now caught, and the installation is
4802 38. An argument was missing from the function that creates an error message
4803 when Exim fails to connect to the socket for saslauthd authentication.
4804 This could cause Exim to crash, or give a corrupted message.
4806 39. Added isip, isip4, and isip6 to ${if conditions.
4808 40. The ACL variables $acl_xx are now saved with the message, and can be
4809 accessed later in routers, transports, and filters.
4811 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4812 strings in the file are not string-expanded.
4814 42. If a MAIL command specified a SIZE value that was too large to fit into an
4815 int variable, the check against message_size_limit failed. Such values are
4816 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4817 day this will have to be increased, but I don't think I want to be around
4818 when emails are that large.
4825 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4826 iconv() is not standard in FreeBSD.
4828 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4829 IPv6 enabled. The observed symptom was a segmentation fault on return from
4830 the function os_common_find_running_interfaces() in src/os.c.
4832 3. In the check_special_case() function in daemon.c I had used "errno" as an
4833 argument name, which causes warnings on some systems. This was basically a
4834 typo, since it was named "eno" in the comments!
4836 4. The code that waits for the clock to tick (at a resolution of some fraction
4837 of a second) so as to ensure message-id uniqueness was always waiting for
4838 at least one whole tick, when it could have waited for less. [This is
4839 almost certainly not relevant at current processor speeds, where it is
4840 unlikely to ever wait at all. But we try to future-proof.]
4842 5. The function that sleeps for a time interval that includes fractions of a
4843 second contained a race. It did not block SIGALRM between setting the
4844 timer, and suspending (a couple of lines later). If the interval was short
4845 and the sigsuspend() was delayed until after it had expired, the suspension
4846 never ended. On busy systems this could lead to processes getting stuck for
4849 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4850 process, before it forks any delivery processes. The open lookup caching
4851 mechanism meant that the open file or database connection was passed into
4852 the delivery process. The problem was that delivery processes always tidy
4853 up cached lookup data. This could cause a problem for the next delivery
4854 process started by the queue runner, because the external queue runner
4855 process does not know about the closure. So the next delivery process
4856 still has data in the lookup cache. In the case of a file lookup, there was
4857 no problem because closing a file descriptor in a subprocess doesn't affect
4858 the parent. However, if the lookup was caching a connection to a database,
4859 the connection was closed, and the second delivery process was likely to
4860 see errors such as "PGSQL: query failed: server closed the connection
4861 unexpectedly". The problem has been fixed by closing all cached lookups
4862 in a queue runner before running a delivery process.
4864 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4865 seem to have the "const" qualifier which it has on other OS. I've
4868 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4869 *of the same type* (client or server) with the same public name that an
4870 error should be diagnosed.
4872 9. When Exim looked up a host name for an IP address, but failed to find the
4873 original IP address when looking up the host name (a safety check), it
4874 output the message "<ip address> does not match any IP for NULL", which was
4875 confusing, to say the least. The bug was that the host name should have
4876 appeared instead of "NULL".
4878 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4879 user that is built into the binary, and the -C or -D options is used, root
4880 privilege is dropped before the configuration file is read. In addition,
4881 logging is switched to stderr instead of the normal log files. If the
4882 configuration then re-defines the Exim user, the unprivileged environment
4883 is probably not what is expected, so Exim logs a panic warning message (but
4886 However, if deliver_drop_privilege is set, the unprivileged state may well
4887 be exactly what is intended, so the warning has been cut out in that case,
4888 and Exim is allowed to try to write to its normal log files.
4894 1. smtp_return_error_details was not giving details for temporary sender
4895 or receiver verification errors.
4897 2. Diagnose a configuration error if two authenticators have the same public
4900 3. Exim used not to create the message log file for a message until the first
4901 delivery attempt. This could be confusing when incoming messages were held
4902 for policy or load reasons. The message log file is now created at the time
4903 the message is received, and an initial "Received" line is written to it.
4905 4. The automatically generated man page for command line options had a minor
4906 bug that caused no ill effects; however, a more serious problem was that
4907 the procedure for building the man page automatically didn't always
4908 operate. Consequently, release 4.20 contains an out-of-date version. This
4909 shouldn't happen again.
4911 5. When building Exim with embedded Perl support, the script that builds the
4912 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4913 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4915 6. The freeze_tell option was not being used for messages that were frozen on
4916 arrival, either by an ACL or by local_scan().
4918 7. Added the smtp_incomplete_transaction log selector.
4920 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4921 was accepting AUTH without a new EHLO.
4923 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4924 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4927 10. Logging of TCP/IP connections (when configured) now happens in the main
4928 daemon process instead of the child process, so that the TCP/IP connection
4929 count is more accurate (but it can never be perfect).
4931 11. The use of "drop" in a nested ACL was not being handled correctly in the
4932 outer ACL. Now, if condition failure induced by the nested "drop" causes
4933 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4934 or "require"), the connection is dropped.
4936 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4937 that yield "discard" can now be used with an "accept" or a "discard" verb,
4938 but an error is generated for any others (because I can't see a useful way
4939 to define what should happen).
4941 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4942 are now processed in the same way as lines in the Exim configuration file.
4943 In particular, continuation lines are supported.
4945 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4947 15. Added -ti meaning -t -i.
4949 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4950 domains, and warn by logging if others are found.
4952 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4953 returned from GIFCONF when Exim is trying to find the list of interfaces on
4954 a host. The code in os.c has been modified to copy each ifreq to an aligned
4955 structure in all cases.
4957 Also, in some cases, the returned ifreq's were being copied to a 'struct
4958 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4959 means the last couple of bytes of an IPv6 address could be chopped if the
4960 ifreq contained only a normal sockaddr (14 bytes storage).
4962 18. Named domain lists were not supported in the hosts_treat_as_local option.
4963 An entry such as +xxxx was not recognized, and was treated as a literal
4966 19. Ensure that header lines added by a DATA ACL are included in the reject log
4967 if the ACL subsequently rejects the message.
4969 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4970 MD5 (which is deprecated).
4972 21. When testing a filter file using -bf, Exim was writing a message when it
4973 took the sender from a "From " line in the message, but it was not doing so
4974 when it took $return_path from a Return-Path: header line. It now does.
4976 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4977 with a valid header line field name (a series of printing characters
4978 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4980 23. Changed "disc" in the source to "disk" to conform to the documentation and
4981 the book and for uniformity.
4983 24. Ignore Sendmail's -Ooption=value command line item.
4985 25. When execve() failed while trying to run a command in a pipe transport,
4986 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4987 could be confused with a return value of 69 from the command itself. This
4988 has been changed to 127, the value the shell returns if it is asked to run
4989 a non-existent command. The wording for the related log line suggests a
4990 non-existent command as the problem.
4992 26. If received_header_text expands to an empty string, do not add a Received:
4993 header line to the message. (Well, it adds a token one on the spool, but
4994 marks it "old" so that it doesn't get used or transmitted.)
4996 27. Installed eximstats 1.28 (addition of -nt option).
4998 28. There was no check for failure on the call to getsockname() in the daemon
4999 code. This can fail if there is a shortage of resources on the system, with
5000 ENOMEM, for example. A temporary error is now given on failure.
5002 29. Contrary to the C standard, it seems that in some environments, the
5003 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5004 program. Exim now does this explicitly; it affects the formatting of
5005 timestamps using strftime().
5007 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5008 complaints. I've now initialized all the variables, to avoid this.
5010 32. Header lines added by a system filter were not being "seen" during
5011 transport-time rewrites.
5013 33. The info_callback() function passed to OpenSSL is set up with type void
5014 (*)(SSL *, int, int), as described somewhere. However, when calling the
5015 function (actually a macro) that sets it up, the type void(*)() is
5016 expected. I've put in a cast to prevent warnings from picky compilers.
5018 34. If a DNS black list lookup found a CNAME record, but there were no A
5019 records associated with the domain it pointed at, Exim crashed.
5021 35. If a DNS black list lookup returned more than one A record, Exim ignored
5022 all but the first. It now scans all returned addresses if a particular IP
5023 value is being sought. In this situation, the contents of the
5024 $dnslist_value variable are a list of all the addresses, separated by a
5027 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5028 to accept a host name and all its aliases if the forward lookup for any of
5029 them yielded the IP address of the incoming connection. Now it accepts only
5030 those names whose forward lookup yields the correct IP address. Any other
5031 names are discarded. This closes a loophole whereby a rogue DNS
5032 administrator could create reverse DNS records to break through a
5033 wildcarded host restriction in an ACL.
5035 37. If a user filter or a system filter that ran in a subprocess used any of
5036 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5037 the wrong values were passed to the pipe command ($thisaddress had the
5038 value of $0, $0 had the value of $1, etc). This bug was introduced by
5039 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5041 38. Improved the line breaking for long SMTP error messages from ACLs.
5042 Previously, if there was no break point between 40 and 75 characters, Exim
5043 left the rest of the message alone. Two changes have been made: (a) I've
5044 reduced the minimum length to 35 characters; (b) if it can't find a break
5045 point between 35 and 75 characters, it looks ahead and uses the first one
5046 that it finds. This may give the occasional overlong line, but at least the
5047 remaining text gets split now.
5049 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5050 file descriptors might be low, and that setting 1000 would always raise it.
5051 It turns out that in some environments, the limit is already over 1000 and
5052 that lowering it causes trouble. So now Exim takes care not to decrease it.
5054 40. When delivering a message, the value of $return_path is set to $sender_
5055 address at the start of routing (routers may change the value). By an
5056 oversight, this default was not being set up when an address was tested by
5057 -bt or -bv, which affected the outcome if any router or filter referred to
5060 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5061 log only when "message" or "log_message" are set. However, if one of the
5062 conditions was an address verification, or a call to a nested ACL, the
5063 messages generated by the underlying test were being passed through. This
5064 no longer happens. The underlying message is available in $acl_verify_
5065 message for both "message" and "log_message" expansions, so it can be
5066 passed through if needed.
5068 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5069 new expansion $bh_ to give the encoded byte string without charset
5070 translation. Translation happens only if iconv() is available; HAVE_ICONV
5071 indicates this at build time. HEADERS_CHARSET gives the charset to
5072 translate to; headers_charset can change it in the configuration, and
5073 "headers charset" can change it in an individual filter file.
5075 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5076 that creates RFC 2047 encoded "words" labels them as that charset instead
5077 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5078 expansion operator; (ii) when Exim creates a From: line for a local
5079 message; (iii) when a header line is rewritten to include a "phrase" part.
5081 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5082 buggy, causing it to skip the first lines of messages whose message ID
5083 ended in 'D'. This would not have bitten before Exim release 4.14, because
5084 message IDs were unlikely to end in 'D' before then. The effect was to have
5085 incorrect size information for certain domains.
5087 45. #include "config.h" was missing at the start of the crypt16.c module. This
5088 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5091 46. If there was a timeout during a "random" callout check, Exim treated it as
5092 a failure of the random address, and carried on sending RSET and the real
5093 address. If the delay was just some slowness somewhere, the response to the
5094 original RCPT would be taken as a response to RSET and so on, causing
5095 mayhem of various kinds.
5097 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5098 when I implemented it. It didn't allow for the fact that some option values
5099 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5100 the right test for positive values.
5102 48. Domain names in DNS records are case-independent. Exim always looks them up
5103 in lower case. Some resolvers return domain names in exactly the case they
5104 appear in the zone file, that is, they may contain uppercase letters. Not
5105 all resolvers do this - some return always lower case. Exim was treating a
5106 change of case by a resolver as a change of domain, similar to a widening
5107 of a domain abbreviation. This triggered its re-routing code and so it was
5108 trying to route what was effectively the same domain again. This normally
5109 caused routing to fail (because the router wouldn't handle the domain
5110 twice). Now Exim checks for this case specially, and just changes the
5111 casing of the domain that it ultimately uses when it transmits the message
5114 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5117 50. If a filter generated a file delivery with a non-absolute name (possible if
5118 no home directory exists for the router), the forbid_file option was not
5121 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5122 the existing equality matching.
5124 52. Exim was using ints instead of ino_t variables in some places where it was
5125 dealing with inode numbers.
5127 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5128 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5129 called TMPDIR, and if it finds it is different, it changes its value.
5131 54. The smtp_printf() function is now made available to local_scan() so
5132 additional output lines can be written before returning. There is also an
5133 smtp_fflush() function to enable the detection of a dropped connection.
5134 The variables smtp_input and smtp_batched_input are exported to
5137 55. Changed the default runtime configuration: the message "Unknown user"
5138 has been removed from the ACL, and instead placed on the localuser router,
5139 using the cannot_route_message feature. This means that any verification
5140 failures that generate their own messages won't get overridden. Similarly,
5141 the "Unrouteable address" message that was in the ACL for unverifiable
5142 relay addresses has also been removed.
5144 56. Added hosts_avoid_esmtp to the smtp transport.
5146 57. The exicyclog script was not checking for the esoteric option
5147 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5148 will work only if exicyclog is run under the appropriate euid.
5150 58. Following a discussion on the list, the rules by which Exim recognises line
5151 endings on incoming messages have been changed. The -dropcr and drop_cr
5152 options are now no-ops, retained only for backwards compatibility. The
5153 following line terminators are recognized: LF CRLF CR. However, special
5154 processing applies to CR:
5156 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5157 nor a local message in the state where . is a terminator.
5159 (ii) If a bare CR is encountered in a header line, an extra space is added
5160 after the line terminator so as not to end the header. The reasoning
5161 behind this is that bare CRs in header lines are most likely either
5162 to be mistakes, or people trying to play silly games.
5164 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5165 was being incorrectly given as 18 bytes larger than it should have been.
5166 This is a VOB (very old bug).
5168 60. This may never have affected anything current, but just in case it has:
5169 When the local host is found other than at the start of a list of hosts,
5170 the local host, those with the same MX, and any that follow, are discarded.
5171 When the list in question was part of a longer list of hosts, the following
5172 hosts (not currently being processed) were also being discarded. This no
5173 longer happens. I'm not sure if this situation could ever has previously
5176 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5179 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5180 header. This is something that is recommended in a new Internet Draft, and
5181 is something that is documented as being done by Sendmail. There are two
5182 possible values. For messages generated by the autoreply transport, Exim
5185 Auto-Submitted: auto-replied
5187 whereas for all other generated messages (e.g. bounces) it adds
5189 Auto-Submitted: auto-generated
5191 63. The "personal" condition in filters now includes a test for the
5192 Auto-Submitted: header. If it contains the string "auto-" the message it
5193 not considered personal.
5195 64. Added rcpt_include_affixes as a generic transport option.
5197 65. Added queue_only_override (default true).
5199 66. Added the syslog_duplication option.
5201 67. If what should have been the first header line of a message consisted of
5202 a space followed by a colon, Exim was mis-interpreting it as a header line.
5203 It isn't of course - it is syntactically invalid and should therefore be
5204 treated as the start of the message body. The misbehaviour could have
5205 caused a number of strange effects, including loss of data in subsequent
5206 header lines, and spool format errors.
5208 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5209 client host had authenticated. This control can now be exercised by an ACL
5210 for more flexibility.
5212 69. By default, callouts do not happen when testing with -bh. There is now a
5213 variant, -bhc, which does actually run the callout code, including
5214 consulting and updating the callout cache.
5216 70. Added support for saslauthd authentication, courtesy of Alexander
5219 71. If statvfs() failed on the spool or log directories while checking their
5220 size for availability, Exim confusingly gave the error "space shortage".
5221 Furthermore, in debugging mode it crashed with a floating point exception.
5222 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5223 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5224 serious problem, Exim now writes to the main and panic logs when this
5225 happens, with details of the failure. It then refuses to accept the
5226 incoming message, giving the message "spool directory problem" or "log
5227 directory problem" with a 421 code for SMTP messages.
5229 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5230 0, 1, and 2 exist, because some OS complain for execs without them (see
5231 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5232 descriptors. However, the code omitted to check that the open succeeded,
5233 causing mysterious errors if for some reason the permissions on /dev/null
5234 got screwed. Now Exim writes a message to the main and panic logs, and
5235 bombs out if it can't open /dev/null.
5237 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5238 interact so that it is all more flexible. It is supposed to remain
5239 backwards compatible. Also added extra_local_interfaces.
5241 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5242 to bomb out with an assertion failure - to the client this appears as a
5243 connection drop. This problem occurs in the part of the code that was taken
5244 from the Samba project. Fortunately, the assertion is in a very simple
5245 function, so I have fixed this by reproducing the function inline in the
5246 one place where it is called, and arranging for authentication to fail
5247 instead of killing the process with assert().
5249 75. The SPA client code was not working when the server requested OEM rather
5250 than Unicode encoding.
5252 76. Added code to make require_files with a specific uid setting more usable in
5253 the case where statting the file as root fails - usually a non-root-mounted
5254 NFS file system. When this happens and the failure is EACCES, Exim now
5255 forks a subprocess and does the per-uid checking as the relevant uid.
5257 77. Added process_log_path.
5259 78. If log_file_path was not explicitly set, a setting of check_log_space or
5260 check_log_inodes was ignored.
5262 79. If a space check for the spool or log partitions fails, the incident is now
5263 logged. Of course, in the latter case the data may get lost...
5265 80. Added the %p formatting code to string_format() so that it can be used to
5266 print addresses in debug_print(). Adjusted all the address printing in the
5267 debugging in store.c to use %p rather than %d.
5269 81. There was a concern that a line of code in smtp_in.c could overflow a
5270 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5271 initially expressed, the concern was not well-founded, because trailing
5272 spaces are removed early. However, if the trailing spaces were followed by
5273 a NULL, they did not get removed, so the overflow was possible. Two fixes
5276 (a) I re-wrote the offending code in a cleaner fashion.
5277 (b) If an incoming SMTP command contains a NULL character, it is rejected
5280 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5281 runs initgroups(), so that if the Exim user is in any additional groups,
5282 they will be used during message reception.
5288 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob