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 dependency in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
44 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
45 Patch from Jakob Hirsch.
47 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
48 SQL string expansion failure details.
49 Patch from Andrey Oktyabrski.
51 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
52 Patch from Simon Arlott.
54 PP/09 New lookups functionality failed to compile on old gcc which rejects
55 extern declarations in function scope.
56 Patch from Oliver Fleischmann
58 PP/10 Use sig_atomic_t for flags set from signal handlers.
59 Check getgroups() return and improve debugging.
60 Fixed developed for diagnosis in bug 927 (which turned out to be
63 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
64 Patch from Mark Zealey.
66 PP/12 Bugzilla 1056: Improved spamd server selection.
67 Patch from Mark Zealey.
69 PP/13 Bugzilla 1086: Deal with maildir quota file races.
70 Based on patch from Heiko Schlittermann.
72 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
73 Patch from Uwe Doering, sign-off by Michael Haardt.
75 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
76 variable declaration deep within a block. Bug and patch from
79 PP/15 lookups-Makefile IRIX compatibilty coercion.
85 TF/01 Failure to get a lock on a hints database can have serious
86 consequences so log it to the panic log.
88 TF/02 Log LMTP confirmation messages in the same way as SMTP,
89 controlled using the smtp_confirmation log selector.
91 TF/03 Include the error message when we fail to unlink a spool file.
93 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
94 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
95 for maintaining out-of-tree patches for some time.
97 PP/01 Bugzilla 139: Documentation and portability issues.
98 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
99 Handle per-OS dynamic-module compilation flags.
101 PP/02 Let /dev/null have normal permissions.
102 The 4.73 fixes were a little too stringent and complained about the
103 permissions on /dev/null. Exempt it from some checks.
104 Reported by Andreas M. Kirchwitz.
106 PP/03 Report version information for many libraries, including
107 Exim version information for dynamically loaded libraries. Created
108 version.h, now support a version extension string for distributors
109 who patch heavily. Dynamic module ABI change.
111 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
112 privilege escalation vulnerability whereby the Exim run-time user
113 can cause root to append content of the attacker's choosing to
116 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
119 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
120 If dropping privileges for untrusted macros, we disabled normal logging
121 on the basis that it would fail; for the Exim run-time user, this is not
122 the case, and it resulted in successful deliveries going unlogged.
123 Fixed. Reported by Andreas Metzler.
129 PP/01 Date: & Message-Id: revert to normally being appended to a message,
130 only prepend for the Resent-* case. Fixes regression introduced in
131 Exim 4.70 by NM/22 for Bugzilla 607.
133 PP/02 Include check_rfc2047_length in configure.default because we're seeing
134 increasing numbers of administrators be bitten by this.
136 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
138 PP/03 Bugzilla 994: added openssl_options main configuration option.
140 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
142 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
144 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
146 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
147 without a peer certificate, leading to a segfault because of an
148 assumption that peers always have certificates. Be a little more
149 paranoid. Problem reported by Martin Tscholak.
151 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
152 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
153 NB: ClamAV planning to remove STREAM in "middle of 2010".
154 CL also introduces -bmalware, various -d+acl logging additions and
155 more caution in buffer sizes.
157 PP/09 Implemented reverse_ip expansion operator.
159 PP/10 Bugzilla 937: provide a "debug" ACL control.
161 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
163 PP/12 Bugzilla 973: Implement --version.
165 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
167 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
169 PP/15 Bugzilla 816: support multiple condition rules on Routers.
171 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
172 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
173 ignore trailing whitespace.
175 JJ/02 prevent non-panic DKIM error from being sent to paniclog
177 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
180 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
181 Notification from Dr Andrew Aitchison.
183 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
184 ExtendedDetectionInfo response format.
185 Notification from John Horne.
187 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
190 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
191 XSL and documented dependency on system catalogs, with examples of how
194 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
197 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
198 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
199 configuration file which is writeable by the Exim user or group.
201 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
202 of configuration files to cover files specified with the -C option if
203 they are going to be used with root privileges, not just the default
206 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
207 option (effectively making it always true).
209 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
210 files to be used while preserving root privileges.
212 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
213 that rogue child processes cannot use them.
215 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
216 run-time user, instead of root.
218 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
219 Exim run-time user without dropping privileges.
221 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
222 result string, instead of calling string_vformat() twice with the same
225 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
226 for other users. Others should always drop root privileges if they use
227 -C on the command line, even for a whitelisted configure file.
229 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
231 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
237 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
238 $data_path, and $header_path variables; fixed documentation bugs and
241 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
242 exipick to access non-standard spools, including the "frozen" queue
245 NM/01 Bugzilla 965: Support mysql stored procedures.
246 Patch from Alain Williams
248 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
250 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
251 Patch from Andreas Metzler
253 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
254 Patch from Kirill Miazine
256 NM/05 Bugzilla 671: Added umask to procmail example.
258 JJ/03 installed exipick 20100323.0, fixing doc bug
260 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
261 directory. Notification and patch from Dan Rosenberg.
263 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
265 TK/02 Improve log output when DKIM signing operation fails.
267 MH/01 Treat the transport option dkim_domain as a colon separated
268 list, not as a single string, and sign the message with each element,
269 omitting multiple occurences of the same signer.
271 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
272 Bugzilla 985, 986. Patch by Simon Arlott
274 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
275 Patch by Simon Arlott
277 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
278 MBX locking. Notification from Dan Rosenberg.
284 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
286 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
288 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
290 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
292 NM/04 Bugzilla 915: Flag broken perl installation during build.
298 TK/01 Added patch by Johannes Berg that expands the main option
299 "spamd_address" if it starts with a dollar sign.
301 TK/02 Write list of recipients to X-Envelope-Sender header when building
302 the mbox-format spool file for content scanning (suggested by Jakob
305 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
306 (http://www.dcc-servers.net/) support via dccifd. Activated by
307 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
309 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
310 by Mark Daniel Reidel <mr@df.eu>.
312 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
313 When building exim an external PCRE library is now needed -
314 PCRE is a system library on the majority of modern systems.
315 See entry on PCRE_LIBS in EDITME file.
317 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
318 conversation. Added nologin parameter to request.
319 Patch contributed by Kirill Miazine.
321 TF/01 Do not log submission mode rewrites if they do not change the address.
323 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
325 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
326 log files in place. Contributed by Roberto Lima.
328 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
330 TF/03 Bugzilla 615: When checking the local_parts router precondition
331 after a local_part_suffix or local_part_prefix option, Exim now
332 does not use the address's named list lookup cache, since this
333 contains cached lookups for the whole local part.
335 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
336 Robert Millan. Documentation is in experimental-spec.txt.
338 TF/04 Bugzilla 668: Fix parallel build (make -j).
340 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
342 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
343 Patch provided by Jan Srzednicki.
345 TF/05 Leading white space used to be stripped from $spam_report which
346 wrecked the formatting. Now it is preserved.
348 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
349 that they are available at delivery time.
351 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
353 TF/08 TLS error reporting now respects the incoming_interface and
354 incoming_port log selectors.
356 TF/09 Produce a more useful error message if an SMTP transport's hosts
357 setting expands to an empty string.
359 NM/06 Bugzilla 744: EXPN did not work under TLS.
360 Patch provided by Phil Pennock.
362 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
363 Patch provided by Richard Godbee.
365 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
366 acl_smtp_notquit, added index entry.
368 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
369 Patch provided by Eugene Bujak.
371 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
372 accept(). Patch provided by Maxim Dounin.
374 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
375 Patch provided by Phil Pennock.
377 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
379 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
380 Patch provided by Brad "anomie" Jorsch.
382 NM/14 Bugzilla 622: Added timeout setting to transport filter.
383 Patch provided by Dean Brooks.
385 TK/05 Add native DKIM support (does not depend on external libraries).
387 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
388 Patch provided by Graeme Fowler.
390 NM/16 Bugzilla 851: Documentation example syntax fix.
392 NM/17 Changed NOTICE file to remove references to embedded PCRE.
394 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
397 NM/19 Bugzilla 745: TLS version reporting.
398 Patch provided by Phil Pennock.
400 NM/20 Bugzilla 167: bool: condition support.
401 Patch provided by Phil Pennock.
403 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
404 clients. Patch provided by Phil Pennock.
406 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
407 Patch provided by Brad "anomie" Jorsch.
409 NM/23 Bugzilla 687: Fix misparses in eximstats.
410 Patch provided by Heiko Schlittermann.
412 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
413 Patch provided by Heiko Schlittermann.
415 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
416 plus update to original patch.
418 NM/26 Bugzilla 799: Documentation correction for ratelimit.
420 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
421 Patch provided by David Brownlee.
423 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
425 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
427 NM/30 Bugzilla 888: TLS documentation bugfixes.
429 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
431 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
432 Unlike the original bugzilla I have changed all shell scripts in src tree.
434 NM/33 Bugzilla 898: Transport filter timeout fix.
435 Patch by Todd Rinaldo.
437 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
438 Patch by Serge Demonchaux.
440 NM/35 Bugzilla 39: Base64 decode bug fixes.
441 Patch by Jakob Hirsch.
443 NM/36 Bugzilla 909: Correct connect() call in dcc code.
445 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
447 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
449 NM/39 Bugzilla 911: Fixed MakeLinks build script.
455 TK/01 Add preliminary DKIM support. Currently requires a forked version of
456 ALT-N's libdkim that I have put here:
457 http://duncanthrax.net/exim-experimental/
459 Note to Michael Haardt: I had to rename some vars in sieve.c. They
460 were called 'true' and it seems that C99 defines that as a reserved
461 keyword to be used with 'bool' variable types. That means you could
462 not include C99-style headers which use bools without triggering
463 build errors in sieve.c.
465 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
466 as mailq or other aliases. Changed the --help handling significantly
467 to do whats expected. exim_usage() emits usage/help information.
469 SC/01 Added the -bylocaldomain option to eximstats.
471 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
473 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
475 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
481 PH/01 Another patch from the Sieve maintainer.
483 PH/02 When an IPv6 address is converted to a string for single-key lookup
484 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
485 dots are used instead of colons so that keys in lsearch files need not
486 contain colons. This was done some time before quoting was made available
487 in lsearch files. However, iplsearch files do require colons in IPv6 keys
488 (notated using the quote facility) so as to distinguish them from IPv4
489 keys. This meant that lookups for IP addresses in host lists did not work
490 for iplsearch lookups.
492 This has been fixed by arranging for IPv6 addresses to be expressed with
493 colons if the lookup type is iplsearch. This is not incompatible, because
494 previously such lookups could never work.
496 The situation is now rather anomolous, since one *can* have colons in
497 ordinary lsearch keys. However, making the change in all cases is
498 incompatible and would probably break a number of configurations.
500 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
503 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
504 conversion specification without a maximum field width, thereby enabling
505 a rogue spamd server to cause a buffer overflow. While nobody in their
506 right mind would setup Exim to query an untrusted spamd server, an
507 attacker that gains access to a server running spamd could potentially
508 exploit this vulnerability to run arbitrary code as the Exim user.
510 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
511 $primary_hostname instead of what libspf2 thinks the hosts name is.
513 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
514 a directory entry by the name of the lookup key. Previously, if a
515 symlink pointed to a non-existing file or a file in a directory that
516 Exim lacked permissions to read, a lookup for a key matching that
517 symlink would fail. Now it is enough that a matching directory entry
518 exists, symlink or not. (Bugzilla 503.)
520 PH/03 The body_linecount and body_zerocount variables are now exported in the
523 PH/04 Added the $dnslist_matched variable.
525 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
526 This means they are set thereafter only if the connection becomes
529 PH/06 Added the client_condition to authenticators so that some can be skipped
530 by clients under certain conditions.
532 PH/07 The error message for a badly-placed control=no_multiline_responses left
533 "_responses" off the end of the name.
535 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
537 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
538 (without spaces) instead of just copying the configuration text.
540 PH/10 Added the /noupdate option to the ratelimit ACL condition.
542 PH/11 Added $max_received_linelength.
544 PH/12 Added +ignore_defer and +include_defer to host lists.
546 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
547 way in which PCRE > 7.0 is built.
549 PH/14 Implemented queue_only_load_latch.
551 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
552 MAIL command. The effect was to mangle the value on 64-bit systems.
554 PH/16 Another patch from the Sieve maintainer.
556 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
558 PH/18 If a system quota error occurred while trying to create the file for
559 a maildir delivery, the message "Mailbox is full" was not appended to the
560 bounce if the delivery eventually timed out. Change 4.67/27 below applied
561 only to a quota excession during the actual writing of the file.
563 PH/19 It seems that peer DN values may contain newlines (and other non-printing
564 characters?) which causes problems in log lines. The DN values are now
565 passed through string_printing() before being added to log lines.
567 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
568 and InterBase are left for another time.)
570 PH/21 Added message_body_newlines option.
572 PH/22 Guard against possible overflow in moan_check_errorcopy().
574 PH/23 POSIX allows open() to be a macro; guard against that.
576 PH/24 If the recipient of an error message contained an @ in the local part
577 (suitably quoted, of course), incorrect values were put in $domain and
578 $local_part during the evaluation of errors_copy.
584 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
585 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
588 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
589 issue a MAIL command.
591 PH/02 In an ACL statement such as
593 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
595 if a client was not listed at all, or was listed with a value other than
596 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
597 the condition was not true (as it should be), so access was not denied.
598 The bug was that the ! inversion was incorrectly passed on to the second
599 item. This has been fixed.
601 PH/03 Added additional dnslists conditions == and =& which are different from
602 = and & when the dns lookup returns more than one IP address.
604 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
605 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
607 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
608 FSYNC, which compiles an option called disable_fsync that allows for
609 bypassing fsync(). The documentation is heavily laced with warnings.
611 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
613 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
614 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
615 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
616 including adding "make clean"; (3) Added -fPIC when compiling the test
617 dynamically loaded module, to get rid of a warning.
619 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
620 message fails, move_frozen_messages = true and ignore_bounce_errors_after
621 = 0s. The bug is otherwise harmless.
623 PH/07 There was a bug in the dovecot authenticator such that the value of
624 $auth1 could be overwritten, and so not correctly preserved, after a
625 successful authentication. This usually meant that the value preserved by
626 the server_setid option was incorrect.
628 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
630 PH/09 Installed PCRE release 7.0.
632 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
633 run for batched SMTP input. It is now run at the start of every message
634 in the batch. While fixing this I discovered that the process information
635 (output by running exiwhat) was not always getting set for -bs and -bS
636 input. This is fixed, and it now also says "batched" for BSMTP.
638 PH/11 Added control=no_pipelining.
640 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
641 patch, slightly modified), and move the expansion of helo_data till after
642 the connection is made in the smtp transport (so it can use these
645 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
647 PH/14 Added log_selector = +pid.
649 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
651 PH/16 Add ${if forany and ${if forall.
653 PH/17 Added dsn_from option to vary the From: line in DSNs.
655 PH/18 Flush SMTP output before performing a callout, unless control =
656 no_callout_flush is set.
658 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
659 was true (the default) a successful delivery failed to delete the retry
660 item, thus causing premature timeout of the address. The bug is now
663 PH/20 Added hosts_avoid_pipelining to the smtp transport.
665 PH/21 Long custom messages for fakedefer and fakereject are now split up
666 into multiline reponses in the same way that messages for "deny" and
667 other ACL rejections are.
669 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
670 with slight modification.
672 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
673 draft, changing the syntax and factoring some duplicate code".
675 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
676 for deliveries of the second and subsequent messages over the same SMTP
679 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
680 ${reduce, with only minor "tidies".
682 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
684 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
685 expansion side effects.
687 PH/27 When a message times out after an over-quota error from an Exim-imposed
688 quota, the bounce message says "mailbox is full". This message was not
689 being given when it was a system quota that was exceeded. It now should
692 MH/03 Made $recipients available in local_scan(). local_scan() already has
693 better access to the recipient list through recipients_list[], but
694 $recipients can be useful in postmaster-provided expansion strings.
696 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
697 in the case of a MAIL command with additional options following the
698 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
699 were accidentally chopped off.
701 PH/29 SMTP synchronization checks are implemented when a command is read -
702 there is a check that no more input is waiting when there shouldn't be
703 any. However, for some commands, a delay in an ACL can mean that it is
704 some time before the response is written. In this time, more input might
705 arrive, invalidly. So now there are extra checks after an ACL has run for
706 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
707 pipelining has not been advertised.
709 PH/30 MH's patch to allow iscntrl() characters to be list separators.
711 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
712 returned in the SMTP response when smtp_return_error_details was false.
715 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
716 instead of the C I/O that was originally supplied, because problems were
719 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
720 Exim which did not show up earlier: it was assuming that a call to
721 SSL_CTX_set_info_callback() might give an error value. In fact, there is
722 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
723 was a macro that became an assignment, so it seemed to work. This has
724 changed to a proper function call with a void return, hence the compile
725 error. Exim's code has been fixed.
727 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
730 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
732 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
734 PH/37 If a message is not accepted after it has had an id assigned (e.g.
735 because it turns out to be too big or there is a timeout) there is no
736 "Completed" line in the log. When some messages of this type were
737 selected by exigrep, they were listed as "not completed". Others were
738 picked up by some special patterns. I have improved the selection
739 criteria to be more general.
741 PH/38 The host_find_failed option in the manualroute router can now be set
742 to "ignore", to completely ignore a host whose IP address cannot be
743 found. If all hosts are ignored, the behaviour is controlled by the new
744 host_all_ignored option.
746 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
747 homing or because of multiple MX records with /mx) generated more than
748 one IP address, and the following item turned out to be the local host,
749 all the secondary addresses of the first item were incorrectly removed
750 from the list, along with the local host and any following hosts (which
751 is what is supposed to happen).
753 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
754 whoever called Exim into the -H file. In the case of the daemon it was
755 behaving confusingly. When first started, it used values for whoever
756 started the daemon, but after a SIGHUP it used the Exim user (because it
757 calls itself on a restart). I have changed the code so that it now always
760 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
761 message are rejected with the same error (e.g. no authentication or bad
762 sender address), and a DATA command is nevertheless sent (as can happen
763 with PIPELINING or a stupid MUA), the error message that was given to the
764 RCPT commands is included in the rejection of the DATA command. This is
765 intended to be helpful for MUAs that show only the final error to their
768 PH/42 Another patch from the Sieve maintainer.
770 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
771 Eximstats - Fixed some broken HTML links and added missing column headers
773 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
774 columns for Rejects, Temp Rejects, Ham, and Spam rows.
776 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
778 PH/43 Yet another patch from the Sieve maintainer.
780 PH/44 I found a way to check for a TCP/IP connection going away before sending
781 the response to the final '.' that terminates a message, but only in the
782 case where the client has not sent further data following the '.'
783 (unfortunately, this is allowed). However, in many cases there won't be
784 any further data because there won't be any more messages to send. A call
785 to select() can be used: if it shows that the input is "ready", there is
786 either input waiting, or the socket has been closed. An attempt to read
787 the next input character can distinguish the two cases. Previously, Exim
788 would have sent an OK response which the client would never have see.
789 This could lead to message repetition. This fix should cure that, at
790 least in a lot of common cases.
792 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
793 advertised in response to EHLO.
799 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
800 fixed by 4.65/MH/01 (is this a record?) are fixed:
802 (i) An empty string was always treated as zero by the numeric comparison
803 operators. This behaviour has been restored.
805 (ii) It is documented that the numeric comparison operators always treat
806 their arguments as decimal numbers. This was broken in that numbers
807 starting with 0 were being interpreted as octal.
809 While fixing these problems I realized that there was another issue that
810 hadn't been noticed. Values of message_size_limit (both the global option
811 and the transport option) were treated as octal if they started with 0.
812 The documentation was vague. These values are now always treated as
813 decimal, and I will make that clear in the documentation.
819 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
820 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
823 MH/01 Don't check that the operands of numeric comparison operators are
824 integers when their expansion is in "skipping" mode (fixes bug
825 introduced by 4.64-PH/07).
827 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
828 child addresses, Exim now panics and dies. Previously, because the count
829 is held in a short int, deliveries were likely to be lost. As such a
830 large number of recipients for a single message is ridiculous
831 (performance will be very, very poor), I have chosen to impose a limit
832 rather than extend the field.
838 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
839 leftover -K file (the existence of which was triggered by #402).
840 While we were at it, introduced process PID as part of the -K
841 filename. This should rule out race conditions when creating
844 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
845 processing considerably. Previous code took too long for large mails,
846 triggering a timeout which in turn triggers #401.
848 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
849 in the DK code in transports.c. sendfile() is not really portable,
850 hence the _LINUX specificness.
852 TF/01 In the add_headers option to the mail command in an Exim filter,
853 there was a bug that Exim would claim a syntax error in any
854 header after the first one which had an odd number of characters
857 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
858 callout verification, Exim cached a "reject" for the entire domain. This
859 is correct for most verifications, but it is not correct for a recipient
860 verification with use_sender or use_postmaster set, because in that case
861 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
862 case of MAIL FROM:<> rejection from other early rejections (e.g.
863 rejection of HELO). When verifying a recipient using a non-null MAIL
864 address, the cache is ignored if it shows MAIL FROM:<> rejection.
865 Whatever the result of the callout, the value of the domain cache is
866 left unchanged (for any other kind of callout, getting as far as trying
867 RCPT means that the domain itself is ok).
869 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
872 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
873 manifest itself as EPIPE rather than ECONNECT. When tidying away a
874 session, the daemon ignores ECONNECT errors and logs others; it now
875 ignores EPIPE as well.
877 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
878 (quoted-printable decoding).
880 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
881 later the small subsequent patch to fix an introduced bug.
883 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
885 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
887 PH/08 An error is now given if message_size_limit is specified negative.
889 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
890 to be given (somewhat) arbitrary names.
892 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
895 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
896 miscellaneous code fixes
898 PH/10 Added the log_reject_target ACL modifier to specify where to log
901 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
902 hostname. This is wrong, because it relates to the incoming message (and
903 probably the interface on which it is arriving) and not to the outgoing
904 callout (which could be using a different interface). This has been
905 changed to use the value of the helo_data option from the smtp transport
906 instead - this is what is used when a message is actually being sent. If
907 there is no remote transport (possible with a router that sets up host
908 addresses), $smtp_active_hostname is used.
910 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
911 tweaks were necessary in order to get it to work (see also 21 below):
912 (a) The code assumed that strncpy() returns a negative number on buffer
913 overflow, which isn't the case. Replaced with Exim's string_format()
915 (b) There were several signed/unsigned issues. I just did the minimum
916 hacking in of casts. There is scope for a larger refactoring.
917 (c) The code used strcasecmp() which is not a standard C function.
918 Replaced with Exim's strcmpic() function.
919 (d) The code set only $1; it now sets $auth1 as well.
920 (e) A simple test gave the error "authentication client didn't specify
921 service in request". It would seem that Dovecot has changed its
922 interface. Fortunately there's a specification; I followed it and
923 changed what the client sends and it appears to be working now.
925 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
928 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
929 address A is aliased to B and C, where B exists and C does not. Without
930 -v the output is "A verified" because verification stops after a
931 successful redirection if more than one address is generated. However,
932 with -v the child addresses are also verified. Exim was outputting "A
933 failed to verify" and then showing the successful verification for C,
934 with its parentage. It now outputs "B failed to verify", showing B's
935 parentage before showing the successful verification of C.
937 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
938 look up a TXT record in a specific list after matching in a combined
941 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
942 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
943 they consult the DNS. I had assumed they would set it the way they
944 wanted; and indeed my experiments on Linux seem to show that in some
945 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
946 To be on the safe side, however, I have now made the interface to
947 host_find_byname() similar to host_find_bydns(), with an argument
948 containing the DNS resolver options. The host_find_byname() function now
949 sets these options at its start, just as host_find_bydns() does. The smtp
950 transport options dns_qualify_single and dns_search_parents are passed to
951 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
952 of host_find_byname() use the default settings of RES_DEFNAMES
953 (qualify_single) but not RES_DNSRCH (search_parents).
955 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
956 spool_read_header() do less string testing, by means of a preliminary
957 switch on the second character of optional "-foo" lines. (This is
958 overdue, caused by the large number of possibilities that now exist.
959 Originally there were few.) While I was there, I also converted the
960 str(n)cmp tests so they don't re-test the leading "-" and the first
961 character, in the hope this might squeeze out yet more improvement.
963 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
964 flag allowing group syntax was set by the header_syntax check but not
965 turned off, possible causing trouble later; (2) The flag was not being
966 set at all for the header_verify test, causing "group"-style headers to
967 be rejected. I have now set it in this case, and also caused header_
968 verify to ignore an empty address taken from a group. While doing this, I
969 came across some other cases where the code for allowing group syntax
970 while scanning a header line wasn't quite right (mostly, not resetting
971 the flag correctly in the right place). These bugs could have caused
972 trouble for malformed header lines. I hope it is now all correct.
974 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
975 with the "reply" argument non-NULL. The code, however (which originally
976 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
977 but it didn't always do it. This confused somebody who was copying the
978 code for some other use. I have removed all the tests.
980 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
981 feature that was used to support insecure browsers during the U.S. crypto
982 embargo. It requires special client support, and Exim is probably the
983 only MTA that supported it -- and would never use it because real RSA is
984 always available. This code has been removed, because it had the bad
985 effect of slowing Exim down by computing (never used) parameters for the
986 RSA_EXPORT functionality.
988 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
989 authenticator to fail if there's a tab character in the incoming data
990 (there should never be unless someone is messing about, as it's supposed
991 to be base64-encoded). Also added, on Timo's advice, the "secured" option
992 if the connection is using TLS or if the remote IP is the same as the
993 local IP, and the "valid-client-cert option" if a client certificate has
996 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
997 authenticators. This can be used for authorization after authentication
998 succeeds. (In the case of plaintext, it servers for both authentication
1001 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1002 if any retry times were supplied.
1004 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1005 connection, where there is no client IP address to check. In this
1006 situation, the verify now always succeeds.
1008 PH/25 Applied John Jetmore's -Mset patch.
1010 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1012 PH/27 In a string expansion for a processed (not raw) header when multiple
1013 headers of the same name were present, leading whitespace was being
1014 removed from all of them, but trailing whitespace was being removed only
1015 from the last one. Now trailing whitespace is removed from each header
1016 before concatenation. Completely empty headers in a concatenation (as
1017 before) are ignored.
1019 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1020 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1022 PH/29 [Removed. This was a change that I later backed out, and forgot to
1023 correct the ChangeLog entry (that I had efficiently created) before
1024 committing the later change.]
1026 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1027 address errors (4xx response to RCPT) over the same connection as other
1028 messages routed to the same hosts. Such deliveries are always "forced",
1029 so retry times are not inspected. This resulted in far too many retries
1030 for the affected addresses. The effect occurred only when there were more
1031 hosts than the hosts_max_try setting in the smtp transport when it had
1032 the 4xx errors. Those hosts that it had tried were not added to the list
1033 of hosts for which the message was waiting, so if all were tried, there
1034 was no problem. Two fixes have been applied:
1036 (i) If there are any address or message errors in an SMTP delivery, none
1037 of the hosts (tried or untried) are now added to the list of hosts
1038 for which the message is waiting, so the message should not be a
1039 candidate for sending over the same connection that was used for a
1040 successful delivery of some other message. This seems entirely
1041 reasonable: after all the message is NOT "waiting for some host".
1042 This is so "obvious" that I'm not sure why it wasn't done
1043 previously. Hope I haven't missed anything, but it can't do any
1044 harm, as the worst effect is to miss an optimization.
1046 (ii) If, despite (i), such a delivery is accidentally attempted, the
1047 routing retry time is respected, so at least it doesn't keep
1048 hammering the server.
1050 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1051 in ${readsocket because some servers need this prod.
1053 PH/32 Added some extra debug output when updating a wait-xxx database.
1055 PH/33 The hint "could be header name not terminated by colon", which has been
1056 given for certain expansion errors for a long time, was not being given
1057 for the ${if def:h_colon_omitted{... case.
1059 PH/34 The spec says: "With one important exception, whenever a domain list is
1060 being scanned, $domain contains the subject domain." There was at least
1061 one case where this was not true.
1063 PH/35 The error "getsockname() failed: connection reset by peer" was being
1064 written to the panic log as well as the main log, but it isn't really
1065 panic-worthy as it just means the connection died rather early on. I have
1066 removed the panic log writing for the ECONNRESET error when getsockname()
1069 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1070 runs only) independently of the message's sender address. This meant
1071 that, if the 4xx error was in fact related to the sender, a different
1072 message to the same recipient with a different sender could confuse
1073 things. In particualar, this can happen when sending to a greylisting
1074 server, but other circumstances could also provoke similar problems.
1075 I have changed the default so that the retry time for these errors is now
1076 based a combination of the sender and recipient addresses. This change
1077 can be overridden by setting address_retry_include_sender=false in the
1080 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1081 remote server are returned as part of bounce messages. This was not
1082 happening for LMTP over a pipe (the lmtp transport), but now it is the
1083 same for both kinds of LMTP.
1085 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1086 in header lines that were in fact CNAMEs. This is no longer the case.
1088 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1089 and queue runs started by the daemon processed all messages. This has
1090 been fixed so that -R and -S can now usefully be given with -q<time>.
1092 PH/40 Import PCRE release 6.7 (fixes some bugs).
1094 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1096 PH/42 Give an error if -q is specified more than once.
1098 PH/43 Renamed the variables $interface_address and $interface_port as
1099 $received_ip_address and $received_port, to make it clear that these
1100 values apply to message reception, and not to the outgoing interface when
1101 a message is delivered. (The old names remain recognized, of course.)
1103 PH/44 There was no timeout on the connect() call when using a Unix domain
1104 socket in the ${readsocket expansion. There now is.
1106 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1107 be meaningful with "accept".
1109 SC/01 Eximstats V1.43
1110 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1112 SC/02 Eximstats V1.44
1113 Use a glob alias rather than an array ref in the generated
1114 parser. This improves both readability and performance.
1116 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1117 Collect SpamAssassin and rejection statistics.
1118 Don't display local sender or destination tables unless
1119 there is data to show.
1120 Added average volumes into the top table text output.
1122 SC/04 Eximstats V1.46
1123 Collect data on the number of addresses (recipients)
1124 as well as the number of messages.
1126 SC/05 Eximstats V1.47
1127 Added 'Message too big' to the list of mail rejection
1128 reasons (thanks to Marco Gaiarin).
1130 SC/06 Eximstats V1.48
1131 Mainlog lines which have GMT offsets and are too short to
1132 have a flag are now skipped.
1134 SC/07 Eximstats V1.49 (Alain Williams)
1135 Added the -emptyok flag.
1137 SC/08 Eximstats V1.50
1138 Fixes for obtaining the IP address from reject messages.
1140 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1141 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1142 whitesspace changes from 4.64-PH/27
1144 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1147 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1148 are found, allow negative numbers in numeric criteria)
1150 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1152 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1153 to match changes made in 4.64-PH/43
1155 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1157 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1158 "domain" is actually an IP address, and force a failure. This locks out
1159 those revolvers/nameservers that support "A-for-A" lookups, in
1160 contravention of the specifications.
1162 PH/48 When a host name was looked up from an IP address, and the subsequent
1163 forward lookup of the name timed out, the host name was left in
1164 $sender_host_name, contrary to the specification.
1166 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1167 restricted to single-key lookups, Exim was not diagnosing an error if
1168 * or *@ was used with a query-style lookup.
1170 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1172 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1173 long ago, but noone interested enough thought of it. Let's just say that
1174 the "1.1" means that there are some new functions that weren't there at
1175 some point in the past.
1177 PH/51 Error processing for expansion failure of helo_data from an smtp
1178 transport during callout processing was broken.
1180 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1181 tested/used via the -bh/-bhc/-bs options.
1183 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1184 bug, fixed in subsequent PCRE releases).
1186 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1187 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1189 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1195 SC/01 Use a glob alias rather than an array ref in eximstats generated
1196 parser. This improves both readability and performance.
1198 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1199 Don't display local sender or destination tables in eximstats unless
1200 there is data to show.
1201 Added average volumes into the eximstats top table text output.
1203 SC/03 Collect data on the number of addresses (recipients) as well
1204 as the number of messages in eximstats.
1206 TF/01 Correct an error in the documentation for the redirect router. Exim
1207 does (usually) call initgroups() when daemonizing.
1209 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1210 with consistent privilege compared to when running as a daemon.
1212 TF/03 Note in the spec that $authenticated_id is not set for local
1213 submissions from trusted users.
1215 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1216 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1218 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1219 by adding some example configuration directives to the default
1220 configuration file. A little bit of work is required to uncomment the
1221 directives and define how usernames and passwords are checked, but
1222 there is now a framework to start from.
1224 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1225 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1226 without this. I don't know how relevant this is to other LDAP libraries.
1228 PH/02 Add the verb name to the "unknown ACL verb" error.
1230 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1232 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1234 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1235 directory not expanded when it should be if an expanded home directory
1236 was set for the address (which is overridden by the transport).
1238 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1241 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1242 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1243 because it is too late at that time, and has no effect.
1245 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1246 security issue with \' (bugzilla #107). I could not use the
1247 PQescapeStringConn() function, because it needs a PGconn value as one of
1250 PH/08 When testing addresses using -bt, indicate those final addresses that
1251 are duplicates that would not cause an additional delivery. At least one
1252 person was confused, thinking that -bt output corresponded to deliveries.
1253 (Suppressing duplicates isn't a good idea as you lose the information
1254 about possibly different redirections that led to the duplicates.)
1256 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1257 systems where poll() doesn't work, in particular OS X.
1259 PH/10 Added more information to debugging output for retry time not reached.
1261 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1262 operations in malware.c.
1264 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1267 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1268 syslog_duplication set false, log lines that would normally be written
1269 both the the main log and to the reject log were not written to syslog at
1272 PH/14 In the default configuration, change the use of "message" in ACL warn
1273 statements to "add_header".
1275 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1276 not followed by a command (e.g. "seen endif").
1278 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1279 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1282 PH/17 Added extra conditions to the default value of delay_warning_condition
1286 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1287 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1288 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1291 The Auto-Submitted: and various List- headers are standardised, whereas I
1292 don't think Precedence: ever was.
1294 PH/18 Refactored debugging code in route_finduser() to show more information,
1295 in particular, the error code if getpwnam() issues one.
1297 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1298 This is apparently needed in addition to the PH/07 change above to avoid
1299 any possible encoding problems.
1301 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1302 but not after initializing Perl.
1304 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1305 output them only if debugging. By default they are written stderr,
1306 apparently, which is not desirable.
1308 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1311 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1314 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1316 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1317 authenticated or an ident call has been made. Suppress the default
1318 values for $authenticated_id and $authenticated_sender (but permit -oMai
1319 and -oMas) when testing with -bh.
1321 PH/24 Re-jigged the order of the tests in the default configuration so that the
1322 tests for valid domains and recipients precede the DNS black list and CSA
1323 tests, on the grounds that those ones are more expensive.
1325 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1326 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1327 This bug exists in every version of Exim that I still have, right back to
1330 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1331 However, an attempt to turn on case-sensitivity in a regex key by
1332 including (?-i) didn't work because the subject string was already
1333 lowercased, and the effects were non-intuitive. It turns out that a
1334 one-line patch can be used to allow (?-i) to work as expected.
1340 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1341 other effects) broke the use of negated acl sub-conditions.
1343 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1346 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1347 "Deny" causes Exim to reject the incoming connection with a 554 error.
1348 Unfortunately, if there is a major crisis, such as a disk failure,
1349 tcp-wrappers gives "deny", whereas what one would like would be some
1350 kind of temporary error. A kludge has been added to help with this.
1351 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1352 554 error is used if errno is still zero or contains ENOENT (which occurs
1353 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1356 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1358 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1359 errors. Otherwise a message that provokes a temporary error (when other
1360 messages do not) can cause a whole host to time out.
1362 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1363 addresses were routed directly to files or pipes from a redirect router.
1364 File deliveries just didn't batch; pipe deliveries might have suffered
1367 PH/06 A failure to get a lock for a hints database would erroneously always say
1368 "Failed to get write lock", even when it was really a read lock.
1370 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1371 of 0600. This has been changed to use the value of the lockfile_mode
1372 option (which defaults to 0600).
1374 PH/08 Applied small patch from the Sieve maintainer.
1376 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1377 folder from quota calculations, a direct delivery into this folder messed
1378 up the contents of the maildirsize file. This was because the regex was
1379 used only to exclude .Trash (or whatever) when the size of the mailbox
1380 was calculated. There was no check that a delivery was happening into an
1381 excluded directory. This bug has been fixed by ignoring all quota
1382 processing for deliveries into excluded directories.
1384 PH/10 Added the maildirfolder_create_regex option to appendfile.
1390 PH/01 The code for finding all the local interface addresses on a FreeBSD
1391 system running IPv6 was broken. This may well have applied to all BSD
1392 systems, as well as to others that have similar system calls. The broken
1393 code found IPv4 interfaces correctly, but gave incorrect values for the
1394 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1395 that it would not match correctly against @[] and not recognize the IPv6
1398 PH/02 The ipliteral router was not recognizing addresses of the form user@
1399 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1401 PH/03 Added disable_ipv6.
1403 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1404 decoded form, because it is most often used to construct To: headers
1405 lines in autoreplies, and the decoded form may well be syntactically
1406 invalid. However, $reply_address has leading white space removed, and all
1407 newlines turned into spaces so that the autoreply transport does not
1410 PH/05 If group was specified without a user on a router, and no group or user
1411 was specified on a transport, the group from the router was ignored.
1413 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1414 for visible compile-time settings that can be used to change these
1415 numbers, for those that want even more. Backwards compatibility with old
1416 spool files has been maintained. However, going back to a previous Exim
1417 release will lost any variables that are in spool files.
1419 PH/07 Two small changes when running in the test harness: increase delay when
1420 passing a TCP/IP connection to a new process, in case the original
1421 process has to generate a bounce, and remove special handling of
1422 127.0.0.2 (sic), which is no longer necessary.
1424 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1425 be the same on different OS.
1427 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1430 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1431 whether --show-vars was specified or not
1433 JJ/02 exipick: Added support for new ACL variable spool format introduced
1436 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1437 syntactically invalid From: or Reply-to: line, and a filter used this to
1438 generate an autoreply, and therefore failed to obtain an address for the
1439 autoreply, Exim could try to deliver to a non-existent relative file
1440 name, causing unrelated and misleading errors. What now happens is that
1441 it logs this as a hard delivery error, but does not attempt to create a
1444 PH/11 The exinext utility has a -C option for testing purposes, but although
1445 the given file was scanned by exinext itself; it wasn't being passed on
1446 when Exim was called.
1448 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1449 an end-of-file indication when reading a command response.
1451 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1452 compiled. In many other places in Exim, IPv6 addresses are always
1453 recognized, so I have changed this. It also means that IPv4 domain
1454 literals of the form [IPV4:n.n.n.n] are now always recognized.
1456 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1457 used if the router is not running as root, for example, when verifying at
1458 ACL time, or when using -bh. The debugging output from this situation was
1459 non-existent - all you got was a failure to exec. I have made two
1462 (a) Failures to set uid/gid, the current directory, or a process leader
1463 in a subprocess such as that created by queryprogram now generate
1464 suitable debugging ouput when -d is set.
1466 (b) The queryprogram router detects when it is not running as root,
1467 outputs suitable debugging information if -d is set, and then runs
1468 the subprocess without attempting to change uid/gid.
1470 PH/15 Minor change to Makefile for building test_host (undocumented testing
1473 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1474 additional section of a DNS packet that returns MX or SRV records.
1475 Instead, it always explicitly searches for A/AAAA records. This avoids
1476 major problems that occur when a DNS server includes only records of one
1477 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1478 fixed another bug: if SRV records were looked up and the corresponding
1479 address records were *not* found in the additional section, the port
1480 values from the SRV records were lost.
1482 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1483 using the correct key (the original address) when searching the retry
1484 rules in order to find which one to use for generating the retry hint.
1486 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1487 adding the default one. Similarly, if it contains a Reply-To: header, the
1488 errors_reply_to option, if set, is not used.
1490 PH/19 When calculating a retry time, Exim used to measure the "time since
1491 failure" by looking at the "first failed" field in the retry record. Now
1492 it does not use this if it is later than than the arrival time of the
1493 message. Instead it uses the arrival time. This makes for better
1494 behaviour in cases where some deliveries succeed, thus re-setting the
1495 "first failed" field. An example is a quota failure for a huge message
1496 when small messages continue to be delivered. Without this change, the
1497 "time since failure" will always be short, possible causing more frequent
1498 delivery attempts for the huge message than are intended.
1499 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1501 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1502 $1, $2, $3) because the numerical variables can be reset during some
1503 expansion items (e.g. "match"), thereby losing the authentication data.
1505 PH/21 Make -bV show the size of off_t variables so that the test suite can
1506 decide whether to run tests for quotas > 2G.
1508 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1509 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1510 filecount value is greater than 2G or if a quota value is greater than 2G
1511 on a system where the size of off_t is not greater than 4, a panic error
1514 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1515 never match. The debug and -bh output now contains an explicit error
1516 message indicating a malformed IPv4 address or mask.
1518 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1519 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1520 PH/23 above applies.
1522 PH/25 Do not write to syslog when running in the test harness. The only
1523 occasion when this arises is a failure to open the main or panic logs
1524 (for which there is an explicit test).
1526 PH/26 Added the /no_tell option to "control=freeze".
1528 PH/27 If a host name lookup failed very early in a connection, for example, if
1529 the IP address matched host_lookup and the reverse lookup yielded a name
1530 that did not have a forward lookup, an error message of the form "no IP
1531 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1532 could be logged. Now it outputs the IP address instead of "NULL".
1534 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1535 allows the sender and the authenticated sender to be set when
1536 submitting a message from within Exim. Since child_open_exim() is
1537 documented for local_scan(), the new function should be too.
1539 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1540 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1541 results in an empty string is now treated as unset.
1543 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1545 PH/31 Added sender_verify_fail logging option.
1547 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1548 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1549 tidied the source and removed it altogether.
1551 PH/33 When a queue run was abandoned because the load average was too high, a
1552 log line was always written; now it is written only if the queue_run log
1553 selector is set. In addition, the log line for abandonment now contains
1554 information about the queue run such as the pid. This is always present
1555 in "start" and "stop" lines but was omitted from the "abandon" line.
1557 PH/34 Omit spaces between a header name and the colon in the error message that
1558 is given when verify = headers_syntax fails (if there are lots of them,
1559 the message gets confusing).
1561 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1562 names, as there are now some PTR records that contain slashes. This check
1563 is only to protect against broken name servers that fall over on strange
1564 characters, so the fact that it applies to all lookups doesn't matter.
1566 PH/36 Now that the new test suite is complete, we can remove some of the
1567 special code in Exim that was needed for the old test suite. For example,
1568 sorting DNS records because real resolvers return them in an arbitrary
1569 order. The new test suite's fake resolver always returns records in the
1572 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1573 bounces) except when queue_only is set, to avoid logging races between
1574 the different processes.
1576 PH/38 Panic-die if .include specifies a non-absolute path.
1578 PH/39 A tweak to the "H" retry rule from its user.
1580 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1581 a label. They prevented compilation on older perls.
1583 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1584 a warning to be raised on newish perls.
1586 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1587 on queue. Changes to match documented behaviour of showing count of
1588 messages matching specified criteria.
1590 PH/40 Changed the default ident timeout from 30s to 5s.
1592 PH/41 Added support for the use of login_cap features, on those BSD systems
1593 that have them, for controlling the resources used by pipe deliveries.
1595 PH/42 The content-scanning code uses fopen() to create files in which to put
1596 message data. Previously it was not paying any attention to the mode of
1597 the files. Exim runs with umask(0) because the rest of the code creates
1598 files with open(), and sets the required mode explicitly. Thus, these
1599 files were ending up world-writeable. This was not a big issue, because,
1600 being within the spool directory, they were not world-accessible. I have
1601 created a function called modefopen, which takes an additional mode
1602 argument. It sets umask(777), creates the file, chmods it to the required
1603 mode, then resets the umask. All the relevant calls to fopen() in the
1604 content scanning code have been changed to use this function.
1606 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1607 to 24 hours. This avoids potential overflow problems when processing G
1608 and H retry rules. I suspect nobody ever tinkers with this value.
1610 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1612 PH/45 When the plaintext authenticator is running as a client, the server's
1613 challenges are checked to ensure they are valid base64 strings. By
1614 default, the authentication attempt is cancelled if an invalid string is
1615 received. Setting client_ignore_invalid_base64 true ignores these errors.
1616 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1617 they are received. Thus, the responses can be made to depend on the
1618 challenges. If an invalid string is ignored, an empty string is placed in
1621 PH/46 Messages that are created by the autoreply transport now contains a
1622 References: header, in accordance with RFCs 2822 and 3834.
1624 PH/47 Added authenticated_sender_force to the smtp transport.
1626 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1628 PH/49 Installed latest patch from the Sieve maintainer.
1630 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1631 was also set, the appendfile transport was unnecessarily scanning a
1632 directory of message files (e.g. for maildir delivery) to find the count
1633 of files (along with the size), even though it did not need this
1634 information. It now does the scan only if it needs to find either the
1635 size of the count of files.
1637 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1639 PH/52 Two bugs concerned with error handling when the smtp transport is
1642 (i) Exim was not creating retry information for temporary errors given
1643 for individual recipients after the DATA command when the smtp transport
1644 was used in LMTP mode. This meant that they could be retried too
1645 frequently, and not timed out correctly.
1647 (ii) Exim was setting the flag that allows error details to be returned
1648 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1649 recipients that were returned after the DATA command.
1651 PH/53 This is related to PH/52, but is more general: for any failing address,
1652 when detailed error information was permitted to be returned to the
1653 sender, but the error was temporary, then after the final timeout, only
1654 "retry timeout exceeded" was returned. Now it returns the full error as
1655 well as "retry timeout exceeded".
1657 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1658 do this, and (what is worse) MTAs that accept it.
1660 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1661 will now be deprecated.
1663 PH/56 New os.c-cygwin from the Cygwin maintainer.
1665 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1666 formats (previously only available in exim formats via -bpr, -bpru,
1667 and -bpra. Now also available in native and exiqgrep formats)
1669 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1670 with very large, slow to parse queues
1672 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1674 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1676 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1677 responses to authentication challenges, though it was showing the
1678 challenges; (ii) I've removed the CR characters from the debug output for
1681 PH/58 Allow for the insertion of a newline as well as a space when a string
1682 is turned into more than one encoded-word during RFC 2047 encoding. The
1683 Sieve code now uses this.
1685 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1686 data_4xx, lost_connection, tls_required.
1688 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1689 message was being sent as an SMTP response.
1691 PH/61 Add -l and -k options to exicyclog.
1693 PH/62 When verifying, if an address was redirected to one new address, so that
1694 verification continued, and the new address failed or deferred after
1695 having set something in $address_data, the value of $address_data was not
1696 passed back to the ACL. This was different to the case when no
1697 redirection occurred. The value is now passed back in both cases.
1699 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1700 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1701 use login_cap.h, so on its own it isn't the distinguishing feature. The
1702 new name refers directly to the setclassresources() function.
1704 PH/65 Added configuration files for NetBSD3.
1706 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1708 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1709 is preferred over IPv4.
1711 PH/68 The bounce_return_message and bounce_return_body options were not being
1712 honoured for bounces generated during the reception of non-SMTP messages.
1713 In particular, this applied to messages rejected by the ACL. This bug has
1714 been fixed. However, if bounce_return_message is true and bounce_return_
1715 body is false, the headers that are returned for a non-SMTP message
1716 include only those that have been read before the error was detected.
1717 (In the case of an ACL rejection, they have all been read.)
1719 PH/69 The HTML version of the specification is now built in a directory called
1720 spec_html instead of spec.html, because the latter looks like a path with
1721 a MIME-type, and this confuses some software.
1723 PH/70 Catch two compiler warnings in sieve.c.
1725 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1726 function verify_get_ident() calls ip_connect() to connect a socket, but
1727 if the "connect()" function timed out, ip_connect() used to close the
1728 socket. However, verify_get_ident() also closes the socket later, and in
1729 between Exim writes to the log, which may get opened at this point. When
1730 the socket was closed in ip_connect(), the log could get the same file
1731 descriptor number as the socket. This naturally causes chaos. The fix is
1732 not to close the socket in ip_connect(); the socket should be closed by
1733 the function that creates it. There was only one place in the code where
1734 this was missing, in the iplookup router, which I don't think anybody now
1735 uses, but I've fixed it anyway.
1737 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1738 well as to direct DNS lookups. Otherwise the handling of names in host
1739 lists is inconsistent and therefore confusing.
1745 PH/01 Two changes to the default runtime configuration:
1747 (1) Move the checks for relay_from_hosts and authenticated clients from
1748 after to before the (commented out) DNS black list checks.
1750 (2) Add control=submission to the relay_from_hosts and authenticated
1751 clients checks, on the grounds that messages accepted by these
1752 statements are most likely to be submissions.
1754 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1756 (1) Generate an error if the third argument for the ${prvs expansion is
1759 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1762 (3) Reset the variables that are obtained from the first argument of
1763 ${prvscheck and used in the second argument before leaving the code,
1764 because their memory is reclaimed, so using them afterwards may do
1767 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1768 one (it's much easier than Tom thought :-).
1770 (5) Because of (4), we can now allow for the use of $prvscheck_result
1771 inside the third argument.
1773 PH/03 For some reason, the default setting of PATH when running a command from
1774 a pipe transport was just "/usr/bin". I have changed it to
1777 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1778 anything to be listed in the output from -bV.
1780 PH/05 When a filter generated an autoreply, the entire To: header line was
1781 quoted in the delivery log line, like this:
1783 => >A.N.Other <ano@some.domain> <original@ddress> ...
1785 This has been changed so that it extracts the operative address. There
1786 may be more than one such address. If so, they are comma-separated, like
1789 => >ano@some.domain,ona@other.domain <original@ddress> ...
1791 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1792 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1793 not being looked up in the rDNS to get a host name, Exim was showing the
1794 IP address twice in Received: lines, even though the IP addresses were
1795 identical. For example:
1797 Received: from [1.2.3.4] (helo=[1.2.3.4])
1799 However, if the real host name was known, it was omitting the HELO data
1800 if it matched the actual IP address. This has been tidied up so that it
1801 doesn't show the same IP address twice.
1803 PH/07 When both +timestamp and +memory debugging was on, the value given by
1804 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1805 called by the debug printing, thereby overwriting the timestamp buffer.
1806 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1808 PH/08 When the original message was included in an autoreply transport, it
1809 always said "this is a copy of the message, including all the headers",
1810 even if body_only or headers_only was set. It now gives an appropriate
1813 PH/09 Applied a patch from the Sieve maintainer which:
1815 o fixes some comments
1816 o adds the (disabled) notify extension core
1817 o adds some debug output for the result of if/elsif tests
1818 o points to the current vacation draft in the documentation
1819 and documents the missing references header update
1823 o fixes a bug in processing the envelope test (when testing
1824 multiple envelope elements, the last element determinted the
1827 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1828 Electronic Mail") by including:
1830 Auto-submitted: auto-generated
1832 in the messages that it generates (bounce messages and others, such as
1833 warnings). In the case of bounce messages for non-SMTP mesages, there was
1834 also a typo: it was using "Auto_submitted" (underscore instead of
1835 hyphen). Since every message generated by Exim is necessarily in response
1836 to another message, thes have all been changed to:
1838 Auto-Submitted: auto-replied
1840 in accordance with these statements in the RFC:
1842 The auto-replied keyword:
1844 - SHOULD be used on messages sent in direct response to another
1845 message by an automatic process,
1847 - MUST NOT be used on manually-generated messages,
1849 - MAY be used on Delivery Status Notifications (DSNs) and Message
1850 Disposition Notifications (MDNs),
1852 - MUST NOT be used on messages generated by automatic or periodic
1853 processes, except for messages which are automatic responses to
1856 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1857 to the default Received: header definition.
1859 PH/12 Added log selector acl_warn_skipped (default on).
1861 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1862 variables because (a) they are in the wrong storage pool and (b) even if
1863 they were copied, it wouldn't work properly because of the caching.
1865 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1866 checking when decoding. Apparently there are clients that generate
1867 overlong encoded strings. Why am I not surprised?
1869 PH/15 If the first argument of "${if match_address" was not empty, but did not
1870 contain an "@" character, Exim crashed. Now it writes a panic log message
1871 and treats the condition as false.
1873 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1875 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1876 extension "envelope-auth". The code is finished and in agreement with
1877 other implementations, but there is no documentation so far and in fact,
1878 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1879 not changing the active code.
1881 Print executed "if" and "elsif" statements when debugging is used. This
1882 helps a great deal to understand what a filter does.
1884 Document more things not specified clearly in RFC3028. I had all this
1885 sorted out, when out of a sudden new issues came to my mind. Oops."
1887 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1890 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1891 canonical form (as documented). However, after a host name lookup from
1892 the IP address, check_host() was doing a simple string comparison with
1893 addresses acquired from the DNS when checking that the found name did
1894 have the original IP as one of its addresses. Since any found IPv6
1895 addresses are likely to be in abbreviated form, the comparison could
1896 fail. Luckily, there already exists a function for doing the comparison
1897 by converting both addresses to binary, so now that is used instead of
1898 the text comparison.
1900 PH/20 There was another similar case to PH/19, when a complete host name was
1901 given in a host list; looking up its IP address could give an abbreviated
1902 form, whereas the current host's name might or might not be abbreviated.
1903 The same fix has been applied.
1909 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1910 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1913 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1914 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1916 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1918 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1919 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1920 Also, for "Auto-Submitted", treat anything other than "no" as
1921 non-personal, in accordance with RFC 3834. (Previously it treated
1922 anything starting "auto-" as non-personal.)
1924 TF/01 The control=submission/name=... option had a problem with syntax
1925 errors if the name included a slash character. The /name= option
1926 now slurps the rest of the string, so it can include any characters
1927 but it must come last in the list of options (after /sender_retain
1930 PH/05 Some modifications to the interface to the fake nameserver for the new
1938 TK/01 Added the "success_on_redirect" address verification option. See
1939 NewStuff for rationale and an example.
1941 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1943 PH/02 Patch to exigrep to allow it to work on syslog lines.
1945 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1946 fread() to skip over the body file's header line, because in Cygwin the
1947 header line is locked and is inaccessible.
1949 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1950 co-exist for some time) to make it clear that it is the Exim ID that is
1951 referenced, not the Message-ID: header line.
1953 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1954 string_format() function, because snprintf() does not exist on all
1957 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1958 ${stat: expansion item.
1960 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1961 protocol synchronization error", to keep the pedants happy.
1963 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1964 well as for IRIX systems, when gcc is being used. See the host.c source
1967 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1969 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1972 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1973 transport and to the smtp transport in LMTP mode.
1975 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1977 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1978 run a filter in a subprocess. This could lead to confusion in subsequent
1979 lookups in the parent process. There should also be a search_tidyup() at
1980 the end of the subprocess.
1982 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1983 only if the host matched helo_try_verify_hosts, which caused the
1984 verification to occur when the EHLO/HELO command was issued. The ACL just
1985 tested the remembered result. Now, if a previous verification attempt has
1986 not happened, "verify = helo" does it there and then.
1988 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1990 TK/03 Fix log output including CR from clamd.
1992 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1993 exist provoked a memory error which could cause a segfault.
1995 PH/15 Installed PCRE 6.2
1997 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1999 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2000 of the problem. Specifically, suggested +O2 rather than +O1 for the
2003 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2005 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2006 record, the port was not correctly logged when the outgoing_port log
2007 selector was set (it logged the transort's default port).
2009 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2010 fallback_hosts, and "hosts" in the smtp transport.
2012 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2013 host errors such as "Connection refused".
2015 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2016 authentication with radiusclient 0.4.9:
2018 - Error returned from rc_read_config was caught wrongly
2019 - Username/password not passed on to radius server due to wrong length.
2021 The presumption is that some radiusclient API changes for 4.51/PH/17
2022 were not taken care of correctly. The code is still untested by me (my
2023 Linux distribution still has 0.3.2 of radiusclient), but it was
2024 contributed by a Radius user.
2026 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2027 expanding the "port" option of the smtp transport.
2029 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2030 while reading a MIME header. Thanks to Tom Hughes for a patch.
2032 PH/24 Include config.h inside local_scan.h so that configuration settings are
2035 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2036 that in an ACL for RCPT (for example), you can examine exactly what was
2039 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2040 commands, but it was not correctly comparing the address with the actual
2041 client host address. Thus, it would show the EHLO address in Received:
2042 header lines when this was not necessary.
2044 PH/27 Added the % operator to ${eval:}.
2046 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2047 it should be ignoring failures (because with -C, for example, it has lost
2048 privilege). It wasn't ignoring creation failures other than "already
2051 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2054 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2055 because an input line was too long, either on its own, or by virtue of
2056 too many continuations, the temporary file was not being removed, and the
2057 return code was incorrect.
2059 PH/31 Missing "BOOL" in function definition in filtertest.c.
2061 PH/32 Applied Sieve patches from the maintainer.
2063 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2065 PH/33 Added "verify = not_blind".
2067 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2068 Local/Makefile (with some defaults set). These are used in built scripts
2069 such as exicyclog, but they have never been used in the exim_install
2070 script (though there are many overriding facilities there). I have
2071 arranged that the exim_install script now takes note of these two
2074 PH/35 Installed configuration files for Dragonfly.
2076 PH/36 When a locally submitted message by a trusted user did not contain a
2077 From: header, and the sender address was obtained from -f or from an SMTP
2078 MAIL command, and the trusted user did not use -F to supply a sender
2079 name, $originator_name was incorrectly used when constructing a From:
2080 header. Furthermore, $originator_name was used for submission mode
2081 messages from external hosts without From: headers in a similar way,
2082 which is clearly wrong.
2084 PH/37 Added control=suppress_local_fixups.
2086 PH/38 When log_selector = +received_sender was set, and the addition of the
2087 sender made the log line's construction buffer exactly full, or one byte
2088 less than full, an overflow happened when the terminating "\n" was
2091 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2092 when the result of a list match is failure because a DNS lookup failed.
2094 PH/40 RM_COMMAND is now used in the building process.
2096 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2097 the "build-* directories that it finds.
2099 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2100 address in a domain literal was a prefix of an interface address.
2102 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2103 when verifying a sender address, unless rewrite_headers is false.
2105 PH/44 Wrote a long comment about why errors_to addresses are verified as
2106 recipients, not senders.
2108 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2109 the ratelimit ACL was added.
2111 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2113 PH/46 Added extra information about PostgreSQL errors to the error string.
2115 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2116 avoiding the need to install special zones in a real server. This is
2117 backwards compatible; if it can't find the fake resolver, it drops back.
2118 Thus, both old and new test suites can be run.
2120 TF/02 Added util/ratelimit.pl
2122 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2123 clock is set back in time.
2125 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2126 Candler <B.Candler@pobox.com>.
2128 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2129 OK for addresses that are the result of redirections.
2131 PH/48 A number of further additions for the benefit of the new test suite,
2132 including a fake gethostbyname() that interfaces to the fake DNS resolver
2135 TF/06 The fix for widen_domains has also been applied to qualify_single and
2136 search_parents which are the other dnslookup options that can cause
2139 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2142 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2144 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2145 service" if it encountered an MX record with an empty target hostname.
2146 The message is now "an MX or SRV record indicated no SMTP service".
2148 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2149 if the DNS of the sending site is misconfigured. This is quite a
2150 common situation. This change restores the behaviour of treating a
2151 helo verification defer as a failure.
2153 PH/51 If self=fail was set on a router, the bounce message did not include the
2154 actual error message.
2160 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2162 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2163 command itself ended in error, the underlying message about the transport
2164 filter timeout was being overwritten with the pipe command error. Now the
2165 underlying error message should be appended to the second error message.
2167 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2169 PH/02 Reduce the amount of output that "make" produces by default. Full output
2170 can still be requested.
2172 PH/03 The warning log line about a condition test deferring for a "warn" verb
2173 was being output only once per connection, rather than after each
2174 occurrence (because it was using the same function as for successful
2175 "warn" verbs). This seems wrong, so I have changed it.
2177 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2178 it should not have, which might have caused a crash in the right
2179 circumstances, but probably never did.
2181 PH/04 Installed a modified version of Tony Finch's patch to make submission
2182 mode fix the return path as well as the Sender: header line, and to
2183 add a /name= option so that you can make the user's friendly name appear
2186 TF/03 Added the control = fakedefer ACL modifier.
2188 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2189 Mark Lowes for thorough testing.
2191 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2193 TK/03 Merged latest SRS patch from Miles Wilton.
2195 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2196 with the definition in sysexits.h (which is #included earlier).
2197 Fortunately, Exim does not actually use EX_OK. The code used to try to
2198 preserve the sysexits.h value, by assumimg that macro definitions were
2199 scanned for macro replacements. I have been disabused of this notion,
2200 so now the code just undefines EX_OK before #including unistd.h.
2202 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2203 in the smtp transport. When a block could not be written in a single
2204 write() function, the timeout was being re-applied to each part-write.
2205 This seems wrong - if the receiver was accepting one byte at a time it
2206 would take for ever. The timeout is now adjusted when this happens. It
2207 doesn't have to be particularly precise.
2209 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2210 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2212 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2213 without a domain if the check to <postmaster@domain> fails.
2215 SC/01 Eximstats: added -xls and the ability to specify output files
2216 (patch written by Frank Heydlauf).
2218 SC/02 Eximstats: use FileHandles for outputing results.
2220 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2222 SC/04 Eximstats: fixed display of large numbers with -nvr option
2224 SC/05 Eximstats: fixed merging of reports with empty tables.
2226 SC/06 Eximstats: added the -include_original_destination flag
2228 SC/07 Eximstats: removed tabs and trailing whitespace.
2230 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2232 TK/06 MBOX spool code: Add real "From " MBOX separator line
2233 so the .eml file is really in mbox format (even though
2234 most programs do not really care). Patch from Alex Miller.
2236 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2237 The latter is generated from $received_to and is only set if the
2238 message has one envelope recipient. SA can use these headers,
2239 obviously out-of-the-box. Patch from Alex Miller.
2241 PH/08 The ${def test on a variable was returning false if the variable's
2242 value was "0", contrary to what the specification has always said!
2243 The result should be true unless the variable is empty.
2245 PH/09 The syntax error of a character other than { following "${if
2246 def:variable_name" (after optional whitespace) was not being diagnosed.
2247 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2248 accidental colon was present, for example, could give incorrect results.
2250 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2251 result is used (not including appendfile, where other changes are about
2254 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2255 This involved changing a lot of size variables from int to off_t. It
2256 should work with maildirs and everything.
2258 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2259 spamd dying while we are connected to it.
2261 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2264 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2265 function for BDB 4.3.
2267 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2269 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2270 the sources for PCRE have been re-arranged and more files are now
2273 PH/15 The code I had for printing potentially long long variables in PH/11
2274 above was not the best (it lost precision). The length of off_t variables
2275 is now inspected at build time, and an appropriate printing format (%ld
2276 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2277 to be "long long int" or "long int". This is needed for the internal
2278 formatting function string_vformat().
2280 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2281 the configuration file to be ":syslog", then the script "guesses" where
2282 the logs files are, rather than using the compiled in default. In our
2283 case the guess is not the same as the compiled default, so the script
2284 suddenly stopped working when I started to use syslog. The patch checks
2285 to see if log_file_path is "". If so, it attempts to read it from exim
2286 with no configuration file to get the compiled in version, before it
2287 falls back to the previous guessing code."
2289 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2290 implementing BATV in an Exim configuration. See NewStuff for the gory
2293 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2294 Makefile that are specific to HP-UX.
2296 PH/18 If the "use_postmaster" option was set for a recipient callout together
2297 with the "random" option, the postmaster address was used as the MAIL
2298 FROM address for the random test, but not for the subsequent recipient
2299 test. It is now used for both.
2301 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2302 patch removes a few documentation additions to RFC 3028, because the
2303 latest draft now contains them. It adds the new en;ascii-case comparator
2304 and a new error check for 8bit text in MIME parts. Comparator and
2305 require names are now matched exactly. I enabled the subaddress
2306 extension, but it is not well tested yet (read: it works for me)."
2308 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2309 rework some of the code of TK/09 above to avoid the hardwired use of
2310 "%lld" and "long long". Replaced the call to snprintf() with a call to
2313 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2314 records point to non-existent hosts", "retry timeout exceeded", and
2315 "retry time not reached for any host after a long failure period".
2317 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2318 experimental DomainKeys support:
2320 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2321 (2) On an error such as an illegally used "control", the wrong name for
2322 the control was given.
2324 These problems did NOT occur unless DomainKeys support was compiled.
2326 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2328 PH/24 Added ${if match_ip condition.
2330 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2331 databases so that it will be absolutely obvious if a crash occurs in the
2332 DB library. This is a regular occurrence (often caused by mis-matched
2335 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2336 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2337 picked up on a user's system that detects such things. There doesn't seem
2338 to be a gcc warning option for this - only an attribute that has to be
2339 put on the function's prototype. It seems that in Fedora Core 4 they have
2340 set this on a number of new functions. No doubt there will be more in due
2343 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2344 specify a transport. However, if an address that was verified by such a
2345 router was the subject of a callout, Exim crashed because it tried to
2346 read the rcpt_include_affixes from the non-existent transport. Now it
2347 just assumes that the setting of that option is false. This bug was
2348 introduced by 4.51/PH/31.
2350 PH/28 Changed -d+all to exclude +memory, because that information is very
2351 rarely of interest, but it makes the output a lot bigger. People tend to
2352 do -d+all out of habit.
2354 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2355 code in os-type was giving problems when libc.so lives in lib64, like on
2358 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2359 aren't the modern standard, and it seems that some systems' include files
2360 don't always have them. Exim was already checking for some of the newer
2361 ones like T_AAAA, and defining it itself. I've added checks for all the
2362 record types that Exim uses.
2364 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2365 not automatically generating a new one, as it is supposed to. This
2366 prevented TLS from working. If the file did exist, but contained invalid
2367 data, a new version was generated, as expected. It was only the case of a
2368 non-existent file that was broken.
2370 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2371 with a change in libdomainkeys > 0.64.
2373 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2374 from DNS. If the selector record carries the flag, it now has
2375 precedence over the domain-wide flag.
2377 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2379 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2380 the use of a port name that isn't defined in /etc/services) occurred, the
2381 message was deferred as in a normal delivery, and thus remained on the
2382 spool, instead of being failed because of the mua_wrapper setting. This
2383 is now fixed, and I tidied up some of the mua_wrapper messages at the
2386 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2387 the messages in a hash of arrays rather than using individual hashes.
2388 This is a bit cleaner and results in dramatic memory savings, albeit
2389 at a slight CPU cost.
2391 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2392 as requested by Marc Sherman.
2394 SC/10 Eximstats: added histograms for user specified patterns as requested
2397 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2399 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2400 fopen() in the content-scanning modules that did not already have it.
2406 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2407 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2409 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2411 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2413 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2414 to test Sieve filters that use "vacation".
2416 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2417 that changes the way the GnuTLS parameters are stored in the cache file.
2418 The new format can be generated externally. For backward compatibility,
2419 if the data in the cache doesn't make sense, Exim assumes it has read an
2420 old-format file, and it generates new data and writes a new file. This
2421 means that you can't go back to an older release without removing the
2424 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2425 work if there are any delivery delays because "one_time" forces the
2426 parent to be marked "delivered", so its unseen clone is never tried
2427 again. For this reason, Exim now forbids the simultaneous setting of
2430 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2431 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2432 entry if you want to know the details. The fix, however, neglected to
2433 consider the case when local delivery batching is involved. The test for
2434 "previously delivered" was not happening when checking to see if an
2435 address could be batched with a previous (undelivered) one; under
2436 certain circumstances this could lead to multiple deliveries to the same
2439 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2440 in its include files, and this causes problems building Exim.
2442 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2443 header_syntax) but Exim was just ignoring anything given after a slash.
2444 In particular, this caused confusion with an attempt to use "verify =
2445 reverse_host_lookup/defer_ok". An error is now given when options are
2446 supplied for verify items that do not have them. (Maybe reverse_host_
2447 lookup should have a defer_ok option, but that's a different point.)
2449 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2450 defined by RFC 821) to 2048, because there were problems with some AUTH
2451 commands, and RFC 1869 says the size should be increased for extended
2452 SMTP commands that take arguments.
2454 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2457 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2458 "unknown" error; now it says that the functionality isn't in the binary.
2460 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2461 an address' error message when a string expansion fails (syntax or
2462 whatever). Otherwise the password may appear in the log. Following change
2463 PH/42 below, there is no longer a chance of it appearing in a bounce
2466 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2468 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2469 of its addresses was ever tried. (Bugzilla bug #2.)
2471 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2472 the result incorrectly in the debug output. (It correctly added a newline
2473 to what was transported.)
2475 TF/01 Added $received_time.
2477 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2478 commented out examples of how to interface to a virus scanner and to
2479 SpamAssassin. Also added commented examples of av_scanner and
2480 spamd_address settings.
2482 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2483 and controls are allowed in which ACLs. There were a couple of minor
2484 errors. Some of the entries in the conditions table (which is a table of
2485 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2486 negation of where the condition IS allowed.
2488 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2490 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2491 header file does not have a version number, so I've had to invent a new
2492 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2493 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2494 radiusclient), but it was contributed by a Radius user.
2496 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2497 files or directories, for OpenSSL.
2499 PH/19 When an Exim process that is running as root has to create an Exim log
2500 file, it does so in a subprocess that runs as exim:exim so as to get the
2501 ownership right at creation (otherwise, other Exim processes might see
2502 the file with the wrong ownership). There was no test for failure of this
2503 fork() call, which would lead to the process getting stuck as it waited
2504 for a non-existent subprocess. Forks do occasionally fail when resources
2505 run out. I reviewed all the other calls to fork(); they all seem to check
2508 PH/20 When checking for unexpected SMTP input at connect time (before writing
2509 the banner), Exim was not dealing correctly with a non-positive return
2510 from the read() function. If the client had disconnected by this time,
2511 the result was a log entry for a synchronization error with an empty
2512 string after "input=" when read() returned zero. If read() returned -1
2513 (an event I could not check), uninitialized data bytes were printed.
2514 There were reports of junk text (parts of files, etc) appearing after
2517 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2519 PH/22 Added support for macro redefinition, and (re)definition in between
2520 driver and ACL definitions.
2522 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2523 forgetting to use the resulting value; it was using the unexpanded value.
2525 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2526 hadn't been configured. The fix is from Juergen Kreileder, who
2527 understands it better than I do:
2529 "Here's what I see happening with three configured cyrus_sasl
2530 authenticators configured (plain, login, cram-md5):
2532 On startup auth_cyrus_sasl_init() gets called for each of these.
2533 This means three calls to sasl_listmech() without a specified mech_list.
2534 => SASL tests which mechs of all available mechs actually work
2535 => three warnings about OTP not working
2536 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2538 With the patch, sasl_listmech() also gets called three times. But now
2539 SASL's mech_list option is set to the server_mech specified in the the
2540 authenticator. Or in other words, the answer from sasl_listmech()
2541 gets limited to just the mech you're testing for (which is different
2543 => the return list contains just 'plain' or 'login', 'cram-md5' or
2544 nothing depending on the value of ob->server_mech.
2546 I've just tested the patch: Authentication still works fine,
2547 unavailable mechs specified in the exim configuration are still
2548 caught, and the auth.log warnings about OTP are gone."
2550 PH/25 When debugging is enabled, the contents of the command line are added
2551 to the debugging output, even when log_selector=+arguments is not
2554 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2555 answer is "GNU", and only if the return is "GNU/something" is the answer
2558 PH/27 $acl_verify_message is now set immediately after the failure of a
2559 verification in an ACL, and so is available in subsequent modifiers. In
2560 particular, the message can be preserved by coding like this:
2562 warn !verify = sender
2563 set acl_m0 = $acl_verify_message
2565 Previously, $acl_verify_message was set only while expanding "message"
2566 and "log_message" when a very denied access.
2568 PH/28 Modified OS/os.c-Linux with
2570 -#ifndef OS_LOAD_AVERAGE
2571 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2573 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2574 nomenclature these days.)
2576 PH/29 Installed patch from the Sieve maintainer that adds the options
2577 sieve_useraddress and sieve_subaddress to the redirect router.
2579 PH/30 In these circumstances:
2580 . Two addresses routed to the same list of hosts;
2581 . First host does not offer TLS;
2582 . First host accepts first address;
2583 . First host gives temporary error to second address;
2584 . Second host offers TLS and a TLS session is established;
2585 . Second host accepts second address.
2586 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2587 and peerdn, if requested) that were in fact used only for the second
2590 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2591 attention to any local part prefix or suffix that was matched by the
2592 router that accepted the address. It now behaves in the same way as it
2593 does for delivery: the affixes are removed from the local part unless
2594 rcpt_include_affixes is set on the transport.
2596 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2597 timeout during the DATA phase of an incoming message.
2599 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2600 applied a patch sanctioned by the Sieve maintainer.
2602 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2603 the uid or gid is negative. A case of a negative gid caused this to be
2604 noticed. The fix allows for either to be negative.
2606 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2607 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2610 PH/36 The change PH/12 above was broken. Fixed it.
2612 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2613 the grounds that routing the same address twice would always produce the
2614 same answer. This might have been true once, but it is certainly no
2615 longer true now. Routing a child address may depend on the previous
2616 routing that produced that child. Some complicated redirection strategies
2617 went wrong when messages had multiple recipients, and made Exim's
2618 behaviour dependent on the order in which the addresses were given.
2620 I have moved the duplicate checking until after the routing is complete.
2621 Exim scans the addresses that are assigned to local and remote
2622 transports, and removes any duplicates. This means that more work will be
2623 done, as duplicates will always all be routed, but duplicates are
2624 presumably rare, so I don't expect this is of any significance.
2626 For deliveries to pipes, files, and autoreplies, the duplicate checking
2627 still happens during the routing process, since they are not going to be
2630 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2631 It corrects a timeout issue with spamd. This is Ian's comment: "The
2632 background is that sometimes spamd either never reads data from a
2633 connection it has accepted, or it never writes response data. The exiscan
2634 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2635 blindly assumes that writes won't block so it may never time out."
2637 PH/39 Allow G after quota size as well as K and M.
2639 PH/40 The value set for $authenticated_id in an authenticator may not contain
2640 binary zeroes or newlines because the value is written to log lines and
2641 to spool files. There was no check on this. Now the value is run through
2642 the string_printing() function so that such characters are converted to
2643 printable escape sequences.
2645 PH/41 $message_linecount is a new variable that contains the total number of
2646 lines in the message. Compare $body_linecount, which is the count for the
2649 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2650 bounce and delay warning messages, except in certain special cases, which
2653 (a) An SMTP error message from a remote host;
2654 (b) A message specified in a :fail: redirection;
2655 (c) A message specified in a "fail" command in a system filter;
2656 (d) A message specified in a FAIL return from the queryprogram router;
2657 (e) A message specified by the cannot_route_message router option.
2659 In these cases only, Exim does include the error details in bounce and
2660 warning messages. There are also a few cases where bland messages such
2661 as "unrouteable address" or "local delivery error" are given.
2663 PH/43 $value is now also set for the "else" part of a ${run expansion.
2665 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2666 being worked on, but at least Exim now implements the latest version to
2669 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2670 process to complete was treated as a delivery failure, a timeout while
2671 writing the message to the pipe was logged, but erroneously treated as a
2672 successful delivery. Such timeouts include transport filter timeouts. For
2673 consistency with the overall process timeout, these timeouts are now
2674 treated as errors, giving rise to delivery failures by default. However,
2675 there is now a new Boolean option for the pipe transport called
2676 timeout_defer, which, if set TRUE, converts the failures into defers for
2677 both kinds of timeout. A transport filter timeout is now identified in
2680 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2681 systems where "make" and "gmake" are different, calling "gmake" at top
2682 level broke things. I've arranged for the value of $(MAKE) to be passed
2683 from the Makefile to this script so that it can call the same version of
2687 A note about Exim versions 4.44 and 4.50
2688 ----------------------------------------
2690 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2691 changes of various kinds. As a consequence, a big documentation update was
2692 needed. This delayed the release for rather longer than seemed good, especially
2693 in the light of a couple of (minor) security issues. Therefore, the changes
2694 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2695 release. So 4.44 and 4.50 are in effect two different branches that both start
2698 I have left the 4.50 change log unchanged; it contains all the changes since
2699 4.43. The change log for 4.44 is below; many of its items are identical to
2700 those for 4.50. This seems to be the most sensible way to preserve the
2701 historical information.
2707 1. Minor wording change to the doc/README.SIEVE file.
2709 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2710 computation of the current number of files was incorrect.
2712 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2713 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2716 4. Give more explanation in the error message when the command for a transport
2717 filter fails to execute.
2719 5. There are several places where Exim runs a non-Exim command in a
2720 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2721 was being done only for the command run by the queryprogram router. It is
2722 now done for all such subprocesses. The other cases are: ${run, transport
2723 filters, and the commands run by the lmtp and pipe transports.
2725 6. Added CONFIGURE_GROUP build-time option.
2727 7. Some older OS have a limit of 256 on the maximum number of file
2728 descriptors. Exim was using setrlimit() to set 1000 as a large value
2729 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2730 systems. I've change it so that if it can't get 1000, it tries for 256.
2732 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2733 was an oversight, and furthermore, ever since the addition of extra
2734 controls (e.g. 4.43/32), the checks on when to allow different forms of
2735 "control" were broken. There should now be diagnostics for all cases when a
2736 control that does not make sense is encountered.
2738 9. Added the /retain_sender option to "control=submission".
2740 10. $recipients is now available in the predata ACL (oversight).
2742 11. Tidy the search cache before the fork to do a delivery from a message
2743 received from the command line. Otherwise the child will trigger a lookup
2744 failure and thereby defer the delivery if it tries to use (for example) a
2745 cached ldap connection that the parent has called unbind on.
2747 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2748 of $address_data from the recipient verification was clobbered by the
2749 sender verification.
2751 13. The value of address_data from a sender verification is now available in
2752 $sender_address_data in subsequent conditions in the ACL statement.
2754 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2756 15. Added a new option "connect=<time>" to callout options, to set a different
2759 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2760 was its contents. (It was OK if the option was not defined at all.)
2762 17. A "Completed" log line is now written for messages that are removed from
2763 the spool by the -Mrm option.
2765 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2766 information about exactly what failed.
2768 19. Added -dd to debug only the daemon process.
2770 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2771 handles timeouts, both on the server side and network timeouts. Renamed the
2772 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2774 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2775 It is now set to "smtps".
2777 22. $host_address is now set to the target address during the checking of
2778 ignore_target_hosts.
2780 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2781 being passed; this would have caused $sender_host_name to have been used if
2782 matching the list had actually called for a host name (not very likely,
2783 since this list is usually IP addresses). A host name is now passed as
2786 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2787 code that specifies a non-restarting handler (typically sigaction() in
2788 modern systems) in an attempt to fix a rare and obscure crash bug.
2790 25. Narrowed the window for a race in the daemon that could cause it to ignore
2791 SIGCHLD signals. This is not a major problem, because they are used only to
2792 wake it up if nothing else does.
2794 26. A malformed maildirsize file could cause Exim to calculate negative values
2795 for the mailbox size or file count. Odd effects could occur as a result.
2796 The maildirsize information is now recalculated if the size or filecount
2799 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2800 support for a long time. Removed HAVE_SYS_VFS_H.
2802 28. Installed the latest version of exipick from John Jetmore.
2804 29. In an address list, if the pattern was not a regular expression, an empty
2805 subject address (from a bounce message) matched only if the pattern was an
2806 empty string. Non-empty patterns were not even tested. This was the wrong
2807 because it is perfectly reasonable to use an empty address as part of a
2808 database query. An empty address is now tested by patterns that are
2809 lookups. However, all the other forms of pattern expect the subject to
2810 contain a local part and a domain, and therefore, for them, an empty
2811 address still always fails if the pattern is not itself empty.
2813 30. Exim went into a mad DNS loop when attempting to do a callout where the
2814 host was specified on an smtp transport, and looking it up yielded more
2815 than one IP address.
2817 31. Re-factored the code for checking spool and log partition space into a
2818 function that finds that data and another that does the check. The former
2819 is then used to implement four new variables: $spool_space, $log_space,
2820 $spool_inodes, and $log_inodes.
2822 32. The RFC2047 encoding function was originally intended for short strings
2823 such as real names; it was not keeping to the 75-character limit for
2824 encoded words that the RFC imposes. It now respects the limit, and
2825 generates multiple encoded words if necessary. To be on the safe side, I
2826 have increased the buffer size for the ${rfc2047: expansion operator from
2829 33. It is now permitted to omit both strings after an "if" condition; if the
2830 condition is true, the result is "true". As before, when the second string
2831 is omitted, a false condition yields an empty string. This makes it less
2832 cumbersome to write custom ACL and router conditions.
2834 34. Failure to deliver a bounce message always caused it to be frozen, even if
2835 there was an errors_to setting on the router. The errors_to setting is now
2838 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2839 canonical form (fully expanded) before being placed in
2840 $sender_host_address.
2842 36. The table in the code that translates DNS record types into text (T_A to
2843 "A" for instance) was missing entries for NS and CNAME. It is just possible
2844 that this could have caused confusion if both these types were looked up
2845 for the same domain, because the text type is used as part of Exim's
2846 per-process caching. But the chance of anyone hitting this buglet seems
2849 37. The dnsdb lookup has been extended in a number of ways.
2851 (1) There is a new type, "zns", which walks up the domain tree until it
2852 finds some nameserver records. It should be used with care.
2854 (2) There is a new type, "mxh", which is like "mx" except that it returns
2855 just the host names, not the priorities.
2857 (3) It is now possible to give a list of domains (or IP addresses) to be
2858 looked up. The behaviour when one of the lookups defers can be
2859 controlled by a keyword.
2861 (4) It is now possible to specify the separator character for use when
2862 multiple records are returned.
2864 38. The dnslists ACL condition has been extended: it is now possible to supply
2865 a list of IP addresses and/or domains to be looked up in a particular DNS
2868 39. Added log_selector=+queue_time_overall.
2870 40. When running the queue in the test harness, wait just a tad after forking a
2871 delivery process, to get repeatability of debugging output.
2873 41. Include certificate and key file names in error message when GnuTLS fails
2874 to set them up, because the GnuTLS error message doesn't include the name
2875 of the failing file when there is a problem reading it.
2877 42. Allow both -bf and -bF in the same test run.
2879 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2881 44. The "Exiscan patch" is now merged into the mainline Exim source.
2883 45. Sometimes the final signoff response after QUIT could fail to get
2884 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2885 < 0 before doing a fflush(). This bug looks as though it goes back to the
2886 introduction of TLS in release 3.20, but "sometimes" must have been rare
2887 because the tests only now provoked it.
2889 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2890 (this can affect the format of dates).
2892 47. exim_tidydb, when checking for the continued existence of a message for
2893 which it has found a message-specific retry record, was not finding
2894 messages that were in split spool directories. Consequently, it was
2895 deleting retry records that should have stayed in existence.
2897 48. Steve fixed some bugs in eximstats.
2899 49. The SPA authentication driver was not abandoning authentication and moving
2900 on to the next authenticator when an expansion was forced to fail,
2901 contradicting the general specification for all authenticators. Instead it
2902 was generating a temporary error. It now behaves as specified.
2904 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2905 (the order specifies the preference for clients). The order is now AES256,
2906 AES128, 3DES, ARCFOUR128.
2908 51. Small patch to Sieve code - explicitly set From: when generating an
2911 52. Exim crashed if a remote delivery caused a very long error message to be
2912 recorded - for instance if somebody sent an entire SpamAssassin report back
2913 as a large number of 550 error lines. This bug was coincidentally fixed by
2914 increasing the size of one of Exim's internal buffers (big_buffer) that
2915 happened as part of the Exiscan merge. However, to be on the safe side, I
2916 have made the code more robust (and fixed the comments that describe what
2919 53. Now that there can be additional text after "Completed" in log lines (if
2920 the queue_time_overall log selector is set), a one-byte patch to exigrep
2921 was needed to allow it to recognize "Completed" as not the last thing in
2924 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2925 patch that reportedly fixes this has been added. I am not expert enough to
2926 create a test for it. This is what the patch creator wrote:
2928 "I found a little strange behaviour of ldap code when working with
2929 Windows 2003 AD Domain, where users was placed in more than one
2930 Organization Units. When I tried to give exim partial DN, the exit code
2931 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2932 But simultaneously result of request was absolutely normal ldap result,
2933 so I produce this patch..."
2935 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2936 REFERENCE, so I have modified the code to exclude the patch when that macro
2939 55. Some experimental protocols are using DNS PTR records for new purposes. The
2940 keys for these records are domain names, not reversed IP addresses. The
2941 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2942 leaves it alone. Component reversal etc. now happens only for IP addresses.
2945 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2947 57. Double the size of the debug message buffer (to 2048) so that more of very
2948 long debug lines gets shown.
2950 58. The exicyclog utility now does better if the number of log files to keep
2951 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2953 59. Two changes related to the smtp_active_hostname option:
2955 (1) $smtp_active_hostname is now available as a variable.
2956 (2) The default for smtp_banner uses $smtp_active_hostname instead
2957 of $primary_hostname.
2959 60. The host_aton() function is supposed to be passed a string that is known
2960 to be a valid IP address. However, in the case of IPv6 addresses, it was
2961 not checking this. This is a hostage to fortune. Exim now panics and dies
2962 if the condition is not met. A case was found where this could be provoked
2963 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2964 components; fortuitously, this particular loophole had already been fixed
2965 by change 4.50/55 above.
2967 If there are any other similar loopholes, the new check in host_aton()
2968 itself should stop them being exploited. The report I received stated that
2969 data on the command line could provoke the exploit when Exim was running as
2970 exim, but did not say which command line option was involved. All I could
2971 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2972 running as the user.
2975 61. There was a buffer overflow vulnerability in the SPA authentication code
2976 (which came originally from the Samba project). I have added a test to the
2977 spa_base64_to_bits() function which I hope fixes it.
2980 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2981 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2983 63. The daemon start-up calls getloadavg() while still root for those OS that
2984 need the first call to be done as root, but it missed one case: when
2985 deliver_queue_load_max is set with deliver_drop_privilege. This is
2986 necessary for the benefit of the queue runner, because there is no re-exec
2987 when deliver_drop_privilege is set.
2989 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2990 This has been fixed.
2992 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2993 was in use, was not putting the data itself into the right store pool;
2994 consequently, it could be overwritten for a subsequent message in the same
2995 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2998 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3000 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3001 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3002 the function were treating the return as a boolean value, which happened to
3003 work because 0=false and not-0=true, but is not correct code.
3005 68. The host_aton() function was not handling scoped IPv6 addresses (those
3006 with, for example, "%eth0" on the end) correctly.
3008 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3009 negated items (that is, ~something) in unsigned ints. Some compilers
3010 apparently mutter when there is no cast.
3012 70. If an address verification called from an ACL failed, and did not produce a
3013 user-specific message (i.e. there was only a "system" message), nothing was
3014 put in $acl_verify_message. In this situation, it now puts the system
3017 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3018 session; change 4.31/43 added the unwanted input to the log line - except
3019 that it did not do this in the start of session case. It now does.
3021 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3022 This is wrong and can cause the other end to generate a synchronization
3023 error if it is another Exim or anything else that does the synchronization
3024 check. A QUIT command is no longer sent after a timeout.
3026 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3027 during host lookups.
3029 74. The defer_ok option of callout verification was not working if it was used
3030 when verifying addresses in header lines, that is, for this case:
3032 verify = header_sender/callout=defer_ok
3034 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3035 those file descriptors could be used for SMTP connections. If anything
3036 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3037 could be sent to the SMTP client, causing chaos. The daemon now opens
3038 stdin, stdout, and stderr to /dev/null when it puts itself into the
3041 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3042 log by default. The user can override this with suitable Perl magic.
3044 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3045 the log message when discard triggers, was not working for the DATA ACL or
3046 for the non-SMTP ACL.
3048 78. Error message wording change in sieve.c.
3050 79. If smtp_accept_max_per_host was set, the number of connections could be
3051 restricted to fewer than expected, because the daemon was trying to set up
3052 a new connection before checking whether the processes handling previous
3053 connections had finished. The check for completed processes is now done
3054 earlier. On busy systems, this bug wouldn't be noticed because something
3055 else would have woken the daemon, and it would have reaped the completed
3058 80. If a message was submitted locally by a user whose login name contained one
3059 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3060 It caused a spool format error. I have fixed the spool reading code. A
3061 related problem was that the "from" clause in the Received: line became
3062 illegal because of the space(s). It is now covered by ${quote_local_part.
3064 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3067 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3069 83. Patch from Sieve maintainer to fix unterminated string problem in
3070 "vacation" handling.
3072 84. Some minor changes to the Linux configuration files to help with other
3073 OS variants using glibc.
3075 85. One more patch for Sieve to update vacation handling to latest spec.
3078 ----------------------------------------------------
3079 See the note above about the 4.44 and 4.50 releases.
3080 ----------------------------------------------------
3086 1. Change 4.43/35 introduced a bug that caused file counts to be
3087 incorrectly computed when quota_filecount was set in an appendfile
3090 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3091 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3094 3. Give more explanation in the error message when the command for a transport
3095 filter fails to execute.
3097 4. There are several places where Exim runs a non-Exim command in a
3098 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3099 was being done only for the command run by the queryprogram router. It is
3100 now done for all such subprocesses. The other cases are: ${run, transport
3101 filters, and the commands run by the lmtp and pipe transports.
3103 5. Some older OS have a limit of 256 on the maximum number of file
3104 descriptors. Exim was using setrlimit() to set 1000 as a large value
3105 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3106 systems. I've change it so that if it can't get 1000, it tries for 256.
3108 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3109 was an oversight, and furthermore, ever since the addition of extra
3110 controls (e.g. 4.43/32), the checks on when to allow different forms of
3111 "control" were broken. There should now be diagnostics for all cases when a
3112 control that does not make sense is encountered.
3114 7. $recipients is now available in the predata ACL (oversight).
3116 8. Tidy the search cache before the fork to do a delivery from a message
3117 received from the command line. Otherwise the child will trigger a lookup
3118 failure and thereby defer the delivery if it tries to use (for example) a
3119 cached ldap connection that the parent has called unbind on.
3121 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3122 of $address_data from the recipient verification was clobbered by the
3123 sender verification.
3125 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3126 was its contents. (It was OK if the option was not defined at all.)
3128 11. A "Completed" log line is now written for messages that are removed from
3129 the spool by the -Mrm option.
3131 12. $host_address is now set to the target address during the checking of
3132 ignore_target_hosts.
3134 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3135 being passed; this would have caused $sender_host_name to have been used if
3136 matching the list had actually called for a host name (not very likely,
3137 since this list is usually IP addresses). A host name is now passed as
3140 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3141 code that specifies a non-restarting handler (typically sigaction() in
3142 modern systems) in an attempt to fix a rare and obscure crash bug.
3144 15. Narrowed the window for a race in the daemon that could cause it to ignore
3145 SIGCHLD signals. This is not a major problem, because they are used only to
3146 wake it up if nothing else does.
3148 16. A malformed maildirsize file could cause Exim to calculate negative values
3149 for the mailbox size or file count. Odd effects could occur as a result.
3150 The maildirsize information is now recalculated if the size or filecount
3153 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3154 support for a long time. Removed HAVE_SYS_VFS_H.
3156 18. Updated exipick to current release from John Jetmore.
3158 19. Allow an empty sender to be matched against a lookup in an address list.
3159 Previously the only cases considered were a regular expression, or an
3162 20. Exim went into a mad DNS lookup loop when doing a callout where the
3163 host was specified on the transport, if the DNS lookup yielded more than
3166 21. The RFC2047 encoding function was originally intended for short strings
3167 such as real names; it was not keeping to the 75-character limit for
3168 encoded words that the RFC imposes. It now respects the limit, and
3169 generates multiple encoded words if necessary. To be on the safe side, I
3170 have increased the buffer size for the ${rfc2047: expansion operator from
3173 22. Failure to deliver a bounce message always caused it to be frozen, even if
3174 there was an errors_to setting on the router. The errors_to setting is now
3177 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3178 canonical form (fully expanded) before being placed in
3179 $sender_host_address.
3181 24. Updated eximstats to version 1.33
3183 25. Include certificate and key file names in error message when GnuTLS fails
3184 to set them up, because the GnuTLS error message doesn't include the name
3185 of the failing file when there is a problem reading it.
3187 26. Expand error message when OpenSSL has problems setting up cert/key files.
3190 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3191 (this can affect the format of dates).
3193 28. exim_tidydb, when checking for the continued existence of a message for
3194 which it has found a message-specific retry record, was not finding
3195 messages that were in split spool directories. Consequently, it was
3196 deleting retry records that should have stayed in existence.
3198 29. eximstats updated to version 1.35
3199 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3200 1.35 - bugfix such that pie charts by volume are generated correctly
3202 30. The SPA authentication driver was not abandoning authentication and moving
3203 on to the next authenticator when an expansion was forced to fail,
3204 contradicting the general specification for all authenticators. Instead it
3205 was generating a temporary error. It now behaves as specified.
3207 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3208 (the order specifies the preference for clients). The order is now AES256,
3209 AES128, 3DES, ARCFOUR128.
3211 31. Small patch to Sieve code - explicitly set From: when generating an
3214 32. Exim crashed if a remote delivery caused a very long error message to be
3215 recorded - for instance if somebody sent an entire SpamAssassin report back
3216 as a large number of 550 error lines. This bug was coincidentally fixed by
3217 increasing the size of one of Exim's internal buffers (big_buffer) that
3218 happened as part of the Exiscan merge. However, to be on the safe side, I
3219 have made the code more robust (and fixed the comments that describe what
3222 33. Some experimental protocols are using DNS PTR records for new purposes. The
3223 keys for these records are domain names, not reversed IP addresses. The
3224 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3225 leaves it alone. Component reversal etc. now happens only for IP addresses.
3228 34. The host_aton() function is supposed to be passed a string that is known
3229 to be a valid IP address. However, in the case of IPv6 addresses, it was
3230 not checking this. This is a hostage to fortune. Exim now panics and dies
3231 if the condition is not met. A case was found where this could be provoked
3232 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3233 components; fortuitously, this particular loophole had already been fixed
3234 by change 4.50/55 or 4.44/33 above.
3236 If there are any other similar loopholes, the new check in host_aton()
3237 itself should stop them being exploited. The report I received stated that
3238 data on the command line could provoke the exploit when Exim was running as
3239 exim, but did not say which command line option was involved. All I could
3240 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3241 running as the user.
3244 35. There was a buffer overflow vulnerability in the SPA authentication code
3245 (which came originally from the Samba project). I have added a test to the
3246 spa_base64_to_bits() function which I hope fixes it.
3249 36. The daemon start-up calls getloadavg() while still root for those OS that
3250 need the first call to be done as root, but it missed one case: when
3251 deliver_queue_load_max is set with deliver_drop_privilege. This is
3252 necessary for the benefit of the queue runner, because there is no re-exec
3253 when deliver_drop_privilege is set.
3255 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3256 was in use, was not putting the data itself into the right store pool;
3257 consequently, it could be overwritten for a subsequent message in the same
3258 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3261 38. Sometimes the final signoff response after QUIT could fail to get
3262 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3263 < 0 before doing a fflush(). This bug looks as though it goes back to the
3264 introduction of TLS in release 3.20, but "sometimes" must have been rare
3265 because the tests only now provoked it.
3271 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3272 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3273 Now it returns an integer. A number of calls were still expecting a T/F
3274 return. Fortuitously, in all cases, the tests worked in OK situations,
3275 which is the norm. However, things would have gone wrong on any write
3276 failures on the smtp file descriptor. This function is used when sending
3277 messages over SMTP and also when doing verify callouts.
3279 2. When Exim is called to do synchronous delivery of a locally submitted
3280 message (the -odf or -odi options), it no longer closes stderr before doing
3283 3. Implemented the mua_wrapper option.
3285 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3287 5. Implemented the functions header_remove(), header_testname(),
3288 header_add_at_position(), and receive_remove_recipient(), and exported them
3291 6. If an ACL "warn" statement specified the addition of headers, Exim already
3292 inserted X-ACL-Warn: at the start if there was no header name. However, it
3293 was not making this test for the second and subsequent header lines if
3294 there were newlines in the string. This meant that an invalid header could
3295 be inserted if Exim was badly configured.
3297 7. Allow an ACL "warn" statement to add header lines at the start or after all
3298 the Received: headers, as well as at the end.
3300 8. Added the rcpt_4xx retry error code.
3302 9. Added postmaster_mailfrom=xxx to callout verification option.
3304 10. Added mailfrom=xxxx to the callout verification option, for verify=
3307 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3308 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3310 12. Inserted some casts to stop certain compilers warning when using pointer
3311 differences as field lengths or precisions in printf-type calls (mostly
3312 affecting debugging statements).
3314 13. Added optional readline() support for -be (dynamically loaded).
3316 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3317 same clock tick as a message's arrival, so that its received time was the
3318 same as the "first fail" time on the retry record, and that message
3319 remained on the queue past the ultimate address timeout, every queue runner
3320 would try a delivery (because it was past the ultimate address timeout) but
3321 after another failure, the ultimate address timeout, which should have then
3322 bounced the address, did not kick in. This was a "< instead of <=" error;
3323 in most cases the first failure would have been in the next clock tick
3324 after the received time, and all would be well.
3326 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3327 being recognized when the domain list was tested by the match_domain
3328 condition in an expansion string.
3330 16. Added the ${str2b64: operator.
3332 17. Exim was always calling setrlimit() to set a large limit for the number of
3333 processes, without checking whether the existing limit was already
3334 adequate. (It did check for the limit on file descriptors.) Furthermore,
3335 errors from getrlimit() and setrlimit() were being ignored. Now they are
3336 logged to the main and panic logs, but Exim does carry on, to try to do its
3337 job under whatever limits there are.
3339 18. Imported PCRE 5.0.
3341 19. Trivial typo in log message " temporarily refused connection" (the leading
3344 20. If the log selector return_path_on_delivery was set and an address was
3345 redirected to /dev/null, the delivery process crashed because it assumed
3346 that a return path would always be set for a "successful" delivery. In this
3347 case, the whole delivery is bypassed as an optimization, and therefore no
3350 21. Internal re-arrangement: the function for sending a challenge and reading
3351 a response while authentication was assuming a zero-terminated challenge
3352 string. It's now changed to take a pointer and a length, to allow for
3353 binary data in such strings.
3355 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3357 23. Exim was not respecting finduser_retries when seeking the login of the
3358 uid under which it was called; it was always trying 10 times. (The default
3359 setting of finduser_retries is zero.) Also, it was sleeping after the final
3360 failure, which is pointless.
3362 24. Implemented tls_on_connect_ports.
3364 25. Implemented acl_smtp_predata.
3366 26. If the domain in control=submission is set empty, Exim assumes that the
3367 authenticated id is a complete email address when it generates From: or
3368 Sender: header lines.
3370 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3371 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3372 chown and chgrp in /bin and hostname in /usr/bin.
3374 28. Exim was keeping the "process log" file open after each use, just as it
3375 does for the main log. This opens the possibility of it remaining open for
3376 long periods when the USR1 signal hits a daemon. Occasional processlog
3377 errors were reported, that could have been caused by this. Anyway, it seems
3378 much more sensible not to leave this file open at all, so that is what now
3381 29. The long-running daemon process does not normally write to the log once it
3382 has entered its main loop, and it closes the log before doing so. This is
3383 so that log files can straightforwardly be renamed and moved. However,
3384 there are a couple of unusual error situations where the daemon does write
3385 log entries, and I had neglected to close the log afterwards.
3387 30. The text of an SMTP error response that was received during a remote
3388 delivery was being truncated at 512 bytes. This is too short for some of
3389 the long messages that one sometimes sees. I've increased the limit to
3392 31. It is now possible to make retry rules that apply only when a message has a
3393 specific sender, in particular, an empty sender.
3395 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3396 it possible to be selective about when SMTP synchronization is enforced.
3398 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3400 32. Implemented hosts_connection_nolog.
3402 33. Added an ACL for QUIT.
3404 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3407 35. Added mailbox_size and mailbox_filecount to appendfile.
3409 36. Added control = no_multiline_responses to ACLs.
3411 37. There was a bug in the logic of the code that waits for the clock to tick
3412 in the case where the clock went backwards by a substantial amount such
3413 that the microsecond fraction of "now" was more than the microsecond
3414 fraction of "then" (but the whole seconds number was less).
3416 38. Added support for the libradius Radius client library this is found on
3417 FreeBSD (previously only the radiusclient library was supported).
3423 1. When certain lookups returned multiple values in the form name=value, the
3424 quoting of the values was not always being done properly. Specifically:
3425 (a) If the value started with a double quote, but contained no whitespace,
3427 (b) If the value contained whitespace other than a space character (i.e.
3428 tabs or newlines or carriage returns) it was not quoted.
3429 This fix has been applied to the mysql and pgsql lookups by writing a
3430 separate quoting function and calling it from the lookup code. The fix
3431 should probably also be applied to nisplus, ibase and oracle lookups, but
3432 since I cannot test any of those, I have not disturbed their existing code.
3434 2. A hit in the callout cache for a specific address caused a log line with no
3435 reason for rejecting RCPT. Now it says "Previous (cached) callout
3436 verification failure".
3438 3. There was an off-by-one bug in the queryprogram router. An over-long
3439 return line was truncated at 256 instead of 255 characters, thereby
3440 overflowing its buffer with the terminating zero. As well as fixing this, I
3441 have increased the buffer size to 1024 (and made a note to document this).
3443 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3444 when Exim is waiting for an SMTP response from a remote server, Exim
3445 restarts its select() call on the socket, thereby resetting its timeout.
3446 This is not a problem when such interrupts are rare. Somebody set up a cron
3447 job to run exiwhat every 2 minutes, which is less than the normal select()
3448 timeout (5 or 10 minutes). This meant that the select() timeout never
3449 kicked in because it was always reset. I have fixed this by comparing the
3450 time when an interrupt arrives with the time at the start of the first call
3451 to select(). If more time than the timeout has elapsed, the interrupt is
3452 treated as a timeout.
3454 5. Some internal re-factoring in preparation for the addition of Sieve
3455 extensions (by MH). In particular, the "personal" test is moved to a
3456 separate function, and given an option for scanning Cc: and Bcc: (which is
3457 not set for Exim filters).
3459 6. When Exim created an email address using the login of the caller as the
3460 local part (e.g. when creating a From: or Sender: header line), it was not
3461 quoting the local part when it contained special characters such as @.
3463 7. Installed new OpenBSD configuration files.
3465 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3466 try to make them clearer.
3468 9. Callout options, other than the timeout value, were being ignored when
3469 verifying sender addresses in header lines. For example, when using
3471 verify = header_sender/callout=no_cache
3473 the cache was (incorrectly) being used.
3475 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3476 only the Cygwin environment.
3478 11. When return_path_on_delivery was set as a log selector, if different remote
3479 addresses in the same message used different return paths and parallel
3480 remote delivery occurred, the wrong values would sometimes be logged.
3481 (Whenever a remote delivery process finished, the return path value from
3482 the most recently started remote delivery process was logged.)
3484 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3485 lines when AUTH and/or TLS are in use. This is the "received protocol"
3486 field. Exim used to use "asmtp" for authenticated SMTP, without any
3487 indication (in the protocol name) for TLS use. Now it follows the RFC and
3488 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3489 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3490 names appear in log lines as well as in Received: header lines.
3492 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3493 extensions, and comparison tests, and to fix some bugs.
3495 14. Changes to the "personal" filter test:
3497 (1) The test was buggy in that it was just doing the equivalent of
3498 "contains" tests on header lines. For example, if a user's address was
3499 anne@some.where, the "personal" test would incorrectly be true for
3501 To: susanne@some.where
3503 This test is now done by extracting each address from the header in turn,
3504 and checking the entire address. Other tests that are part of "personal"
3505 are now done using regular expressions (for example, to check local parts
3506 of addresses in From: header lines).
3508 (2) The list of non-personal local parts in From: addresses has been
3509 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3510 taken from the Sieve specification recommendations.
3512 (3) If the message contains any header line starting with "List-" it is
3513 treated as non-personal.
3515 (4) The test for "circular" in the Subject: header line has been removed
3516 because it now seems ill-conceived.
3518 15. Minor typos in src/EDITME comments corrected.
3520 16. Installed latest exipick from John Jetmore.
3522 17. If headers_add on a router specified a text string that was too long for
3523 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3524 of string_sprintf() is now avoided.
3526 18. $message_body_size was not set (it was always zero) when running the DATA
3527 ACL and the local_scan() function.
3529 19. For the "mail" command in an Exim filter, no default was being set for
3530 the once_repeat time, causing a random time value to be used if "once" was
3531 specified. (If the value happened to be <= 0, no repeat happened.) The
3532 default is now 0s, meaning "never repeat". The "vacation" command was OK
3533 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3534 (I found it when inspecting the code).
3536 20. There is now an overall timeout for performing a callout verification. It
3537 defaults to 4 times the callout timeout, which applies to individual SMTP
3538 commands during the callout. The overall timeout applies when there is more
3539 than one host that can be tried. The timeout is checked before trying the
3540 next host. This prevents very long delays if there are a large number of
3541 hosts and all are timing out (e.g. when the network connections are timing
3542 out). The value of the overall timeout can be changed by specifying an
3543 additional sub-option for "callout", called "maxwait". For example:
3545 verify = sender/callout=5s,maxwait=20s
3547 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3548 to the end before writing, but this should make it even safer).
3550 22. Exim was forgetting that it had advertised PIPELINING for the second and
3551 subsequent messages on an SMTP connection. It was also not resetting its
3552 memory on STARTTLS and an internal HELO.
3554 23. When Exim logs an SMTP synchronization error within a session, it now
3555 records whether PIPELINING has been advertised or not.
3557 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3558 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3559 rather than long int.
3561 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3563 26. Added the never_mail option to autoreply.
3569 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3570 crash if the getsockname() call failed; this can happen if a connection is
3571 closed very soon after it is established. The problem was simply in the
3572 order in which certain operations were done, causing Exim to try to write
3573 to the SMTP stream before it had set up the file descriptor. The bug has
3574 been fixed by making things happen in the correct order.
3580 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3581 before the connection was closed, thus losing the rejection response.
3583 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3584 some early Solaris releases, but causes trouble in current releases where
3585 socklen_t is defined.
3587 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3590 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3593 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3594 that releasing the top of it at the end releases what was used for sub-
3595 expansions (unless the block got too big). However, discard this block if
3596 the first thing is a variable or header, so that we can use its block when
3597 it is dynamic (useful for very large $message_headers, for example).
3599 6. Lookups now cache *every* query, not just the most recent. A new, separate
3600 store pool is used for this. It can be recovered when all lookup caches are
3601 flushed. Lookups now release memory at the end of their result strings.
3602 This has involved some general refactoring of the lookup sources.
3604 7. Some code has been added to the store_xxx() functions to reduce the amount
3605 of flapping under certain conditions.
3607 8. log_incoming_interface used to affect only the <= reception log lines. Now
3608 it causes the local interface and port to be added to several more SMTP log
3609 lines, for example "SMTP connection from", and rejection lines.
3611 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3613 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3615 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3616 could be overwritten at the end of the current message (or the start of a
3617 new message if it was set in a HELO ACL). The value is now preserved for
3618 the duration of the SMTP connection.
3620 12. If a transport had a headers_rewrite setting, and a matching header line
3621 contained an unqualified address, that address was qualified, even if it
3622 did not match any rewriting rules. The underlying bug was that the values
3623 of the flags that permit the existence of unqualified sender and recipient
3624 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3625 non-local messages, and by -bnq for local messages) were not being
3626 preserved with the message after it was received.
3628 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3629 "next input=" as part of the text comprising the host identity instead of
3630 the correct text. The code was using the same buffer for two different
3631 strings. However, depending on which order the printing function evaluated
3632 its arguments, the bug did not always show up. Under Linux, for example, my
3633 test suite worked just fine.
3635 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3636 doesn't work with some older versions of Perl. It has been changed to "my",
3637 which in any case is probably the better facility to use.
3639 15. A really picky compiler found some instances of statements for creating
3640 error messages that either had too many or two few arguments for the format
3643 16. The size of the buffer for calls to the DNS resolver has been increased
3644 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3645 for addresses that have a lot of PTR records. This alleviates a problem; it
3646 does not fully solve it.
3648 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3649 buffer now truncates the list and logs the incident, which is the same
3650 action as happens when Exim is looking up a host name and its aliases.
3651 Previously in this situation something unpredictable would happen;
3652 sometimes it was "internal error: store_reset failed".
3654 18. If a server dropped the connection unexpectedly when an Exim client was
3655 using GnuTLS and trying to read a response, the client delivery process
3656 crashed while trying to generate an error log message.
3658 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3659 string, for example:
3661 warn message = H1: something\nH2: something
3663 the text was added as a single header line from Exim's point of view
3664 though it ended up OK in the delivered message. However, searching for the
3665 second and subsequent header lines using $h_h2: did not work. This has been
3666 fixed. Similarly, if a system filter added multiple headers in this way,
3667 the routers could not see them.
3669 20. Expanded the error message when iplsearch is called with an invalid key to
3670 suggest using net-iplsearch in a host list.
3672 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3673 ACLs are no longer actually imposed (and a message to that effect is
3676 22. If a "gecos" field in a passwd entry contained escaped characters, in
3677 particular, if it contained a \" sequence, Exim got it wrong when building
3678 a From: or a Sender: header from that name. A second bug also caused
3679 incorrect handling when an unquoted " was present following a character
3680 that needed quoting.
3682 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3683 was not being matched caselessly.
3685 24. Arranged for all hyphens in the exim.8 source to be escaped with
3688 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3689 properly. Recipient callout cache records were still being keyed to include
3690 the sender, even when use_sender was set false. This led to far more
3691 callouts that were necessary. The sender is no longer included in the key
3692 when use_sender is false.
3694 26. Added "control = submission" modifier to ACLs.
3696 27. Added the ${base62d: operator to decode base 62 numbers.
3698 28. dnsdb lookups can now access SRV records.
3700 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3701 the configuration file.
3703 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3704 (-v) mode. This makes the output for a verbose queue run more intelligible.
3706 31. Added a use_postmaster feature to recipient callouts.
3708 32. Added the $body_zerocount variable, containing the number of binary zero
3709 bytes in the message body.
3711 33. The time of last modification of the "new" subdirectory is now used as the
3712 "mailbox time last read" when there is a quota error for a maildir
3715 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3717 35. Added +ignore_unknown as a special item in host lists.
3719 36. Code for decoding IPv6 addresses in host lists is now included, even if
3720 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3721 address was recognized as an IP address, but was then not correctly decoded
3722 into binary, causing unexpected and incorrect effects when compared with
3729 1. Very minor rewording of debugging text in manualroute to say "list of
3730 hosts" instead of "hostlist".
3732 2. If verify=header_syntax was set, and a header line with an unqualified
3733 address (no domain) and a large number of spaces between the end of the
3734 name and the colon was received, the reception process suffered a buffer
3735 overflow, and (when I tested it) crashed. This was caused by some obsolete
3736 code that should have been removed. The fix is to remove it!
3738 3. When running in the test harness, delay a bit after writing a bounce
3739 message to get a bit more predictability in the log output.
3741 4. Added a call to search_tidyup() just before forking a reception process. In
3742 theory, someone could use a lookup in the expansion of smtp_accept_max_
3743 per_host which, without the tidyup, could leave open a database connection.
3745 5. Added the variables $recipient_data and $sender_data which get set from a
3746 lookup success in an ACL "recipients" or "senders" condition, or a router
3747 "senders" option, similar to $domain_data and $local_part_data.
3749 6. Moved the writing of debug_print from before to after the "senders" test
3752 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3753 problems for message scanning, either using a data ACL, or using
3754 local_scan() because the Received: header was not generated till after they
3755 were called (in order to set the time as the time of reception completion).
3756 I have revised the way this works. The header is now generated after the
3757 body is received, but before the ACL or local_scan() are called. After they
3758 are run, the timestamp in the header is updated.
3764 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3765 before starting a queue runner without re-exec. This happened only when
3766 deliver_drop_privilege was set or when the Exim user was set to root. The
3767 effect of the bug was that timeouts during subsequent deliveries caused
3768 crashes instead of being properly handled. The handler is now left at its
3769 default (and expected) setting.
3771 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3772 message, again when deliver_drop_privilege is set or Exim is run as root.
3773 The bug described in (1) was not present in this case, but the tidying up
3774 of the other signals was missing. I have made the two cases consistent.
3776 3. The ignore_target_hosts setting on a manualroute router was being ignored
3777 for hosts that were looked up using the /MX notation.
3779 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3782 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3783 operated on the sender address. After changing the $sender_address to <>
3784 for the sender address verify, Exim was re-instated it as the original
3785 (before rewriting) address, but remembering that it had rewritten it, so it
3786 wasn't rewriting it again. This bug also had the effect of breaking the
3787 sender address verification caching when the sender address was rewritten.
3789 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3790 This has been changed so that if the ip literal address matches
3791 ignore_target_hosts, the router declines.
3793 7. Added expansion conditions match_domain, match_address, and match_local_
3794 part (NOT match_host).
3796 8. The placeholder for the Received: header didn't have a length field set.
3798 9. Added code to Exim itself and to exim_lock to test for a specific race
3799 condition that could lead to file corruption when using MBX delivery. The
3800 issue is with the lockfile that is created in /tmp. If this file is removed
3801 after a process has opened it but before that process has acquired a lock,
3802 there is the potential for a second process to recreate the file and also
3803 acquire a lock. This could lead to two Exim processes writing to the file
3804 at the same time. The added code performs the same test as UW imapd; it
3805 checks after acquiring the lock that its file descriptor still refers to
3806 the same named file.
3808 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3809 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3812 11. Added the smtp_active_hostname option. If used, this will typically be made
3813 to depend on the incoming interface address. Because $interface_address is
3814 not set up until the daemon has forked a reception process, error responses
3815 that can happen earlier (such as "too many connections") no longer contain
3818 12. If an expansion in a condition on a "warn" statement fails because a lookup
3819 defers, the "warn" statement is abandoned, and the next ACL statement is
3820 processed. Previously this caused the whole ACL to be aborted.
3822 13. Added the iplsearch lookup type.
3824 14. Added ident_timeout as a log selector.
3826 15. Added tls_certificate_verified as a log selector.
3828 16. Added a global option tls_require_ciphers (compare the smtp transport
3829 option of the same name). This controls incoming TLS connections.
3831 17. I finally figured out how to make tls_require_ciphers do a similar thing
3832 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3833 before starting the TLS session.
3835 18. Tabs are now shown as \t in -bP output.
3837 19. If the log selector return_path_on_delivery was set, Exim crashed when
3838 bouncing a message because it had too many Received: header lines.
3840 20. If two routers both had headers_remove settings, and the first one included
3841 a superfluous trailing colon, the final name in the first list and the
3842 first name in the second list were incorrectly joined into one item (with a
3843 colon in the middle).
3849 1. Added -C and -D options to the exinext utility, mainly to make it easier
3850 to include in the automated testing, but these could be helpful when
3851 multiple configurations are in use.
3853 2. The exinext utility was not formatting the output nicely when there was
3854 an alternate port involved in the retry record key, nor when there was a
3855 message id as well (for retries that were specific to a specific message
3856 and a specific host). It was also confused by IPv6 addresses, because of
3857 the additional colons they contain. I have fixed the IPv4 problem, and
3858 patched it up to do a reasonable job for IPv6.
3860 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3861 delivery, the log line now contains "pipelined" if PIPELINING was used.
3863 4. An SMTP transport process used to panic and die if the bind() call to set
3864 an explicit outgoing interface failed. This has been changed; it is now
3865 treated in the same way as a connect() failure.
3867 5. A reference to $sender_host_name in the part of a conditional expansion
3868 that was being skipped was still causing a DNS lookup. This no longer
3871 6. The def: expansion condition was not recognizing references to header lines
3872 that used bh_ and bheader_.
3874 7. Added the _cache feature to named lists.
3876 8. The code for checking quota_filecount in the appendfile transport was
3877 allowing one more file than it should have been.
3879 9. For compatibility with Sendmail, the command line option
3887 and sets the incoming protocol and host name (for trusted callers). The
3888 host name and its colon can be omitted when only the protocol is to be set.
3889 Note the Exim already has two private options, -pd and -ps, that refer to
3890 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3891 "s", but I don't think that's a major issue.
3893 10. A number of refactoring changes to the code, none of which should affect
3896 (a) The number of logging options was getting close to filling up the
3897 32-bit word that was used as a bit map. I have split them into two classes:
3898 those that are passed in the argument to log_write(), and those that are
3899 only ever tested independently outside of that function. These are now in
3900 separate 32-bit words, so there is plenty of room for expansion again.
3901 There is no change in the user interface or the logging behaviour.
3903 (b) When building, for example, log lines, the code previously used a
3904 macro that called string_cat() twice, in order to add two strings. This is
3905 not really sufficiently general. Furthermore, there was one instance where
3906 it was actually wrong because one of the argument was used twice, and in
3907 one call a function was used. (As it happened, calling the function twice
3908 did not affect the overall behaviour.) The macro has been replaced by a
3909 function that can join an arbitrary number of extra strings onto a growing
3912 (c) The code for expansion conditions now uses a table and a binary chop
3913 instead of a serial search (which was left over from when there were very
3914 few conditions). Also, it now recognizes conditions like "pam" even when
3915 the relevant support is not compiled in: a suitably worded error message is
3916 given if an attempt is made to use such a condition.
3918 11. Added ${time_interval:xxxxx}.
3920 12. A bug was causing one of the ddress fields not to be passed back correctly
3921 from remote delivery subprocesses. The field in question was not being
3922 subsequently used, so this caused to problems in practice.
3924 13. Added new log selectors queue_time and deliver_time.
3926 14. Might have fixed a bug in maildirsizefile handling that threw up
3927 "unexpected character" debug warnings, and recalculated the data
3928 unnecessarily. In any case, I expanded the warning message to give more
3931 15. Added the message "Restricted characters in address" to the statements in
3932 the default ACL that block characters like @ and % in local parts.
3934 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3935 Three changes have been made:
3937 (a) There was a serious bug; a negative response to MAIL caused the whole
3938 recipient domain to be cached as invalid, thereby blocking all messages
3939 to all local parts at the same domain, from all senders. This bug has
3940 been fixed. The domain is no longer cached after a negative response to
3941 MAIL if the sender used is not empty.
3943 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3946 (c) A new callout option, "use_sender" has been added for people who want
3947 the modified behaviour.
3953 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3956 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3957 indeed breaks things for older releases.
3959 3. Added additional logging to the case where there is a problem reading data
3960 from a filter that is running in a subprocess using a pipe, in order to
3961 try to track down a specific problem.
3963 4. Testing facility fudge: when running in the test harness and attempting
3964 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3965 getting "No route to host". Convert this to a timeout.
3967 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3970 6. Some OS don't have socklen_t but use size_t instead. This affects the
3971 fifth argument of getsockopt() amongst other things. This is now
3972 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3973 can be set for individual OS. I have set it for SunOS5, OSF1, and
3974 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3975 some earlier ones do not.
3977 7. Change 4.30/15 was not doing the test caselessly.
3979 8. The standard form for an IPv6 address literal was being rejected by address
3980 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3981 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3982 this, as well as the form without the "IPv6" on the front (but only when
3983 address literals are enabled, of course).
3985 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3987 10. Exim crashed if a message with an empty sender address specified by -f
3988 encountered a router with an errors_to setting. This could be provoked only
3989 by a command such as
3993 where an empty string was supplied; "<>" did not hit this bug.
3995 11. Installed PCRE release 4.5.
3997 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3998 remained set. It is now erased.
4000 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4001 times from message ids (which are base 36 rather than the normal 62).
4003 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4004 were being counted as actual protocol errors, and logged if the log
4005 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4006 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4007 and DATA following a set of rejected RCPTs do not count as protocol errors.
4008 In other words, Exim assumes they were pipelined, though this may not
4009 actually be the case. Of course, in all cases the client gets an
4010 appropriate error code.
4012 15. If a lookup fails in an ACL condition, a message about the failure may
4013 be available; it is used if testing the ACL cannot continue, because most
4014 such messages specify what the cause of the deferral is. However, some
4015 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4016 that caused an old message to be retained and used if a later statement
4017 caused a defer, replacing the real cause of the deferral.
4019 16. If an IP address had so many PTR records that the DNS lookup buffer
4020 was not large enough to hold them, Exim could crash while trying to process
4021 the truncated data. It now detects and logs this case.
4023 17. Further to 4.21/58, another change has been made: if (and only if) the
4024 first line of a message (the first header line) ends with CRLF, a bare LF
4025 in a subsequent header line has a space inserted after it, so as not to
4026 terminate the header.
4028 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4029 unnecessarily, used atoi() instead of strtol(), and didn't check the
4030 termination when getting file sizes from file names by regex.
4032 19. Completely re-implemented the support for maildirsize files, in the light
4033 of a number of problems with the previous contributed implementation
4034 (4.30/29). In particular:
4036 . If the quota is zero, the maildirsize file is maintained, but no quota is
4039 . If the maildir directory does not exist, it is created before any attempt
4040 to write a maildirsize file.
4042 . The quota value in the file is just a cache; if the quota is changed in
4043 the transport, the new value overrides.
4045 . A regular expression is available for excluding directories from the
4048 20. The autoreply transport checks the characters in options that define the
4049 message's headers; it allows continued headers, but it was checking with
4050 isspace() after an embedded newline instead of explicitly looking for a
4053 21. If all the "regular" hosts to which an address was routed had passed their
4054 expiry times, and had not reached their retry times, the address was
4055 bounced, even if fallback hosts were defined. Now Exim should go on to try
4058 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4059 equivalent code in the SMTP transport. Some hosts send humungous responses
4060 to HELO/EHLO, more than 1024 it seems.
4062 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4063 gives compiler warnings in some environments. I've now done it "properly",
4066 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4067 (because of problems with the built-in one) was declared to return uschar *
4068 instead of char *, causing compiler failure.
4070 25. Fixed a file descriptor leak when processing alias/forward files.
4072 26. Fixed a minor format string issue in dbfn.c.
4074 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4076 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4077 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4080 29. When a sender address is verified, it is cached, to save repeating the test
4081 when there is more than one recipient in a message. However, when the
4082 verification involves a callout, it is possible for different callout
4083 options to be set for different recipients. It is too complicated to keep
4084 track of this in the cache, so now Exim always runs a verification when a
4085 callout is required, relying on the callout cache for the optimization.
4086 The overhead is duplication of the address routing, but this should not be
4089 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4090 to be verified with callout=postmaster, and the main callout worked but the
4091 postmaster check failed, the verification correctly failed. However, if a
4092 subsequent RCPT command asked for sender verification *without* the
4093 postmaster check, incorrect caching caused this verification also to fail,
4096 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4097 it was not caching the DNS options (qualify_single, search_parents) that
4098 were used when the lookup failed. A subsequent lookup with different
4099 options therefore always gave the same answer, though there were cases
4100 where it should not have. (Example: a "domains = !$mx_any" option on a
4101 dnslookup router: the "domains" option is always processed without any
4102 widening, but the router might have qualify_single set.) Now Exim uses the
4103 cached value only when the same options are set.
4105 32. Added John Jetmore's "exipick" utility to the distribution.
4107 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4108 than a timeout (e.g. a certificate is required, and is not provided), an
4109 Exim server now closes the connection immediately. Previously it waited for
4110 the client to close - but if the client is SSL, it seems that they each
4111 wait for each other, leading to a delay before one of them times out.
4113 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4114 maintained 0.8.x compatibility because I don't think many are using it, and
4115 it is clearly obsolete.
4117 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4120 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4121 client certificate was expired. A simple patch fixes this, though I don't
4122 understand the full logic of why the verify callback is called multiple
4125 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4126 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4127 which causes problems with some clients (such as the Certicom SSL Plus
4128 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4129 disables the coutermeasure allowing Eudora to connect."
4131 38. Exim was not checking that a write() to a log file succeeded. This could
4132 lead to Bad Things if a log got too big, in particular if it hit a file
4133 size limit. Exim now panics and dies if it cannot write to a log file, just
4134 as it does if it cannot open a log file.
4136 39. Modified OS/Makefile-Linux so that it now contains
4138 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4140 The two -D definitions ensure that Exim is compiled with large file
4141 support, which makes it possible to handle log files that are bigger than
4144 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4145 instance) a domain was checked against a named list that involved a lookup,
4146 causing $domain_data to be set, then another domain was checked against the
4147 same list, then the first domain was re-checked, the value of $domain_data
4148 after the final check could be wrong. In particular, if the second check
4149 failed, it could be set empty. This bug probably also applied to
4152 41. The strip_trailing_dot option was not being applied to the address given
4153 with the -f command-line option.
4155 42. The code for reading a message's header from the spool was incrementing
4156 $received_count, but never initializing it. This meant that the value was
4157 incorrect (doubled) while delivering a message in the same process in which
4158 it was received. In the most common configuration of Exim, this never
4159 happens - a fresh exec is done - but it can happen when
4160 deliver_drop_privilege is set.
4162 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4163 it now includes up to 150 characters of the unexpected data in the log
4166 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4167 and building data strings. The size of both of these buffers was 10 000
4168 bytes - far larger than anybody would *ever* want, thought I. Needless to
4169 say, somebody hit the limit. I have increased the maximum line length to
4170 20 000 and the maximum data length of concatenated lines to 100 000. I have
4171 also fixed two bugs, because there was no checking on these buffers. Tsk,
4172 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4173 buffer is too small.
4175 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4176 lsearch lookups. Now it does.
4178 46. When parsing a route_list item in a manualroute router, a fixed-length
4179 buffer was used for the list of hosts. I made this 1024 bytes long,
4180 thinking that nobody would ever have a list of hosts that long. Wrong.
4181 Somebody had a whole pile of complicated expansion conditions, and the
4182 string was silently truncated, leading to an expansion error. It turns out
4183 that it is easier to change to an unlimited length (owing to other changes
4184 that have happened since this code was originally written) than to build
4185 structure for giving a limitation error. The length of the item that
4186 expands into the list of hosts is now unlimited.
4188 47. The lsearch lookup could not handle data where the length of text line was
4189 more than 4095 characters. Such lines were truncated, leading to shortened
4190 data being returned. It should now handle lines of any length.
4192 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4193 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4196 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4197 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4198 no longer happens in comment lines. A list of replacements is now placed
4199 at the head of all of the source files, except those whose only change is
4200 to replace PERL_COMMAND in the very first #! line.
4202 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4203 messages on the queue, with a bottom-up merge sort, using code contributed
4204 by Michael Haardt. This should make operations like -bp somewhat faster on
4205 large queues. It won't affect queue runners, except when queue_run_in_order
4208 51. Installed eximstats 1.31 in the distribution.
4210 52. Added support for SRV lookups to the dnslookup router.
4212 53. If an ACL referred to $message_body or $message_body_end, the value was not
4213 reset for any messages that followed in the same SMTP session.
4215 54. The store-handling optimization for building very long strings was not
4216 differentiating between the different store pools. I don't think this
4217 actually made any difference in practice, but I've tidied it.
4219 55. While running the routers to verify a sender address, $sender_address
4220 was still set to the sender address. This is wrong, because when routing to
4221 send a bounce to the sender, it would be empty. Therefore, I have changed
4222 it so that, while verifying a sender address, $sender_address is set to <>.
4223 (There is no change to what happens when verifying a recipient address.)
4225 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4226 target A or AAAA records (if not already returned) without resetting the
4227 qualify_single or search_parents options of the DNS resolver. These are
4228 inappropriate in this case because the targets of MX and SRV records must
4229 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4230 target that, when qualified, matched something in the local domain. These
4231 two options are now turned off when doing these lookups.
4233 57. It seems that at least some releases of Reiserfs (which does not have the
4234 concept of a fixed number of inodes) returns zero and not -1 for the
4235 number of available inodes. This interacted badly with check_spool_inodes,
4236 which assumed that -1 was the "no such thing" setting. What I have done is
4237 to check that the total number of inodes is greater than zero before doing
4238 the test of how many are available.
4240 58. When a "warn" ACL statement has a log_message modifier, the message is
4241 remembered, and not repeated. This is to avoid a lot of repetition when a
4242 message has many recipients that cause the same warning to be written.
4243 Howewer, Exim was preserving the list of already written lines for an
4244 entire SMTP session, which doesn't seem right. The memory is now reset if a
4245 new message is started.
4247 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4248 debugging output unless log_rewrite was also set.
4250 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4251 of (int)(handle) when we know that handle contains (void *)(-1).
4253 61. The Exim daemon panic-logs an error return when it closes the incoming
4254 connection. However "connection reset by peer" seems to be common, and
4255 isn't really an error worthy of noting specially, so that particular error
4258 62. When Exim is trying to find all the local interfaces, it used to panic and
4259 die if the ioctl to get the interface flags failed. However, it seems that
4260 on at least one OS (Solaris 9) it is possible to have an interface that is
4261 included in the list of interfaces, but for which you get a failure error
4262 for this call. This happens when the interface is not "plumbed" into a
4263 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4264 failure of the "get flags" call assumes that the interface is down.
4266 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4267 makes life easier for people who are doing arithmetic on fields extracted
4268 from dates, where you often get leading zeros that should not be
4269 interpreted as octal.
4271 64. Added qualify_domain to the redirect router, to override the global
4274 65. If a pathologically long header line contained very many addresses (the
4275 report of this problem mentioned 10 000) and each of them was rewritten,
4276 Exim could use up a very large amount of memory. (It kept on making new
4277 copies of the header line as it rewrote, and never released the old ones.)
4278 At the expense of a bit more processing, the header rewriting function has
4279 been changed so that it no longer eats memory in this way.
4281 66. The generation of the Received: header has been moved from the time that a
4282 message starts to be received, to the time that it finishes. The timestamp
4283 in the Received: header should now be very close to that of the <= log
4284 line. There are two side-effects of this change:
4286 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4287 logged header lines no longer include the local Received: line, because
4288 it has not yet been created. The same applies to a copy of the message
4289 that is returned to a non-SMTP sender when a message is rejected.
4291 (b) When a filter file is tested using -bf, no additional Received: header
4292 is added to the test message. After some thought, I decided that this
4295 This change does not affect the value of $received_for. It is still set
4296 after address rewriting, but before local_scan() is called.
4298 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4300 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4301 gave an unhelpful panic error message, and a defer error. I have managed to
4302 change this behaviour so that it now rejects any supplied certificate,
4303 which seems right, as the list of acceptable certificates is empty.
4305 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4306 gave an unhelpful defer error. I have not managed to make this reject any
4307 supplied certificates, but the error message it gives is "no certificate
4308 supplied", which is not helpful.
4310 70. exigrep's output now also includes lines that are not associated with any
4311 message, but which match the given pattern. Implemented by a patch from
4312 Martin Sluka, which also tidied up the Perl a bit.
4314 71. Recipient callout verification, like sender verification, was using <> in
4315 the MAIL FROM command. This isn't really the right thing, since the actual
4316 sender may affect whether the remote host accepts the recipient or not. I
4317 have changed it to use the actual sender in the callout; this means that
4318 the cache record is now keyed on a recipient/sender pair, not just the
4319 recipient address. There doesn't seem to be a real danger of callout loops,
4320 since a callout by the remote host to check the sender would use <>.
4321 [SEE ABOVE: changed after hitting problems.]
4323 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4324 temporary errors. However, in the case of such a code being given after
4325 the end of a data transmission (i.e. after ".") Exim was failing to write
4326 a retry record for the message. (Yes, there was some broken host that was
4327 actually sending 8xx at this point.)
4329 73. An unknown lookup type in a host list could cause Exim to panic-die when
4330 the list was checked. (An example that provoked this was putting <; in the
4331 middle of a list instead of at the start.) If this happened during a DATA
4332 ACL check, a -D file could be left lying around. This kind of configuration
4333 error no longer causes Exim to die; instead it causes a defer errror. The
4334 incident is still logged to the main and panic logs.
4336 74. Buglet left over from Exim 3 conversion. The message "too many messages
4337 in one connection" was written to the rejectlog but not the mainlog, except
4338 when address rewriting (yes!) was being logged.
4340 75. Added write_rejectlog option.
4342 76. When a system filter was run not as root (that is, when system_filter_user
4343 was set), the values of the $n variables were not being returned to the
4344 main process; thus, they were not subsequently available in the $sn
4347 77. Added +return_path_on_delivery log selector.
4349 78. A connection timeout was being treated differently from recipients deferred
4350 when testing hosts_max_try with a message that was older than the host's
4351 retry timeout. (The host should not be counted, thus allowing all hosts to
4352 be tried at least once before bouncing.) This may have been the cause of an
4353 occasionally reported bug whereby a message would remain on the queue
4354 longer than the retry timeout, but would be bounced if a delivery was
4355 forced. I say "may" because I never totally pinned down the problem;
4356 setting up timeout/retry tests is difficult. See also the next item.
4358 79. The ultimate address timeout was not being applied to errors that involved
4359 a combination of host plus message (for example, a timeout on a MAIL
4360 command). When an address resolved to a number of possible hosts, and they
4361 were not all tried for each delivery (e.g. because of hosts_max_try), a
4362 message could remain on the queue longer than the retry timeout.
4364 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4367 81. Fixed an obscure SMTP outgoing bug which required at least the following
4368 conditions: (a) there was another message waiting for the same server;
4369 (b) the server returned 5xx to all RCPT commands in the first message so
4370 that the message was not completed; (c) the server dropped the connection
4371 or gave a negative response to the RSET that Exim sends to abort the
4372 transaction. The observed case was a dropped connection after DATA that had
4373 been sent in pipelining mode. That is, the server had advertised PIPELINING
4374 but was not implementing it correctly. The effect of the bug was incorrect
4375 behaviour, such as trying another host, and this could lead to a crash.
4381 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4382 and daemon.c were passed as pointers to ints; they should have been
4383 pointers to socklen_t variables (which are typically unsigned ints).
4385 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4388 3. Fixed a really odd bug that affected only the testing scheme; patching a
4389 certain fixed string in the binary changed the value of another string that
4390 happened to be identical to the end of the original first string.
4392 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4393 name", it returns that address as the IP address. On some operating
4394 systems (e.g. Solaris), it also passes back the IP address string as the
4395 "host name". However, on others (e.g. Linux), it passes back an empty
4396 string. Exim wasn't checking for this, and was changing the host name to an
4397 empty string, assuming it had been canonicized.
4399 5. Although rare, it is permitted to have more than one PTR record for a given
4400 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4401 all the names associated with an address, because they do in Solaris.
4402 However, it seems that they do not in Linux for data that comes from the
4403 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4404 I found this out when I moved to a new Linux workstation and tried to run
4405 the Exim test suite.
4407 To get round this problem I have changed the code so that it now does its
4408 own call to the DNS to look up PTR records when searching for a host name.
4409 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4410 addresses that are only in /etc/hosts are still found.
4412 This behaviour is, however, controlled by an option called host_lookup_
4413 order, which defaults to "bydns:byaddr". If people want to use the other
4414 order, or indeed, just use one or the other means of lookup, they can
4415 specify it in this variable.
4417 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4418 some operating systems, this comes back from gethostbyaddr() as an empty
4419 string, and this is what Exim used to test for. However, it seems that in
4420 other systems, "." is yielded. Exim now tests for this case too.
4422 7. The values of check_spool_space and check_log_space are now held internally
4423 as a number of kilobytes instead of an absolute number of bytes. If a
4424 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4425 kilobyte. This means that much larger values can be stored.
4427 8. Exim monitor: an attempt to get the action menu when not actually pointing
4428 at a message produces an empty menu entitled "No message selected". This
4429 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4430 no entries in it ("Shell widget menu has zero width and/or height"). So I
4431 have added a single, blank menu entry in this case.
4433 9. Added ${quote_local_part.
4435 10. MIME decoding is now applied to the contents of Subject: header lines when
4438 11. Now that a reference to $sender_host_address automatically causes a reverse
4439 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4440 host lookup before query-style lookups in lists that might use this
4441 variable. This has therefore been abolished, and the "net-" prefix is no
4442 longer necessary for query-style lookups.
4444 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4445 have been a typo for LFLAGS, so it has been changed.
4447 13. The install script calls Exim with "-C /dev/null" in order to find the
4448 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4449 to be output. Howeve, since Exim outputs its version number before the
4450 error, it didn't break the script. It just looked ugly. I fixed this by
4451 always allowing "-C /dev/null" if the caller is root.
4453 14. Ignore overlarge ACL variable number when reading spool file - insurance
4454 against a later release with more variables having written the file.
4456 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4457 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4458 this, as well as the form without the "IPv6" on the front.
4460 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4461 OS/Makefile-Darwin file.
4463 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4464 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4466 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4467 variable. However, some OS use ints for inodes. I've added cast to long int
4468 to get rid of the compiler warning.
4470 19. I had forgotten to lock out "/../" in configuration file names when
4471 ALT_CONFIG_PREFIX was set.
4473 20. Routers used for verification do not need to specify transports. However,
4474 if such a router generated a host list, and callout was configured, Exim
4475 crashed, because it could not find a port number from the (non-existent)
4476 transport. It now assumes port 25 in this circumstance.
4478 21. Added the -t option to exigrep.
4480 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4481 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4482 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4483 them is set, LOOKUP_LSEARCH is forced.
4485 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4487 24. Added sender and host information to the "rejected by local_scan()" log
4488 line; previously there was no indication of these.
4490 25. Added .include_if_exists.
4492 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4493 Linux. It turns out that not all file systems support this. Apparently some
4494 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4495 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4496 not supported on the file descriptor, is now ignored when Exim is trying to
4497 sync a directory. This applies only to Linux.
4499 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4501 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4502 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4505 29. Added support for maildirsize files from supplied patch (modified a bit).
4507 30. The use of :fail: followed by an empty string could lead Exim to respond to
4508 sender verification failures with (e.g.):
4510 550 Verification failed for <xxx>
4511 550 Sender verify failed
4513 where the first response line was missing the '-' that indicates it is not
4514 the final line of the response.
4516 31. The loop for finding the name of the user that called Exim had a hardwired
4517 limit of 10; it now uses the value of finduser_retries, which is used for
4518 all other user lookups.
4520 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4523 33. Exim was neglecting to zero errno before one call of strtol() when
4524 expanding a string and expecting an integer value. On some systems this
4525 resulted in spurious "integer overflow" errors. Also, it was casting the
4526 result into an int without checking.
4528 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4529 did not work. The code looks as if it has *never* worked, though it appears
4530 to have been documented since at least releast 1.62. I have made it work.
4532 35. The "timeout_DNS" error in retry rules, also documented since at least
4533 1.62, also never worked. As it isn't clear exactly what this means, and
4534 clearly it isn't a major issue, I have abolished the feature by treating it
4535 as "timeout", and writing a warning to the main and panic logs.
4537 36. The display of retry rules for -brt wasn't always showing the error code
4540 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4541 timeout_connect_A, timeout_connect_MX.
4543 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4544 to the empty sender.
4546 39. The daemon was not analysing the content of -oX till after it had closed
4547 stderr and disconnected from the controlling terminal. This meant that any
4548 syntax errors were only noted on the panic log, and the return code from
4549 the command was 0. By re-arranging the code a little, I've made the
4550 decoding happen first, so such errors now appear on stderr, and the return
4551 code is 1. However, the actual setting up of the sockets still happens in
4552 the disconnected process, so errors there are still only recorded on the
4555 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4556 connections (as happens on some IP stacks) was logged at start up time as
4557 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4558 it from "IPv6 and IPv4", which means that two separate sockets are being
4561 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4562 says whether AF_INET or AF_INET6 was passed as an argument.
4564 42. Exiwhat output was messed up when time zones were included in log
4567 43. Exiwhat now gives more information about the daemon's listening ports,
4568 and whether -tls-on-connect was used.
4570 44. The "port" option of the smtp transport is now expanded.
4572 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4573 silently ignored. Now an error message is written to the main and panic
4576 46. There's a new ACL modifier called "logwrite" which writes to a log file
4577 as soon as it is encountered.
4579 47. Added $local_user_uid and $local_user_gid at routing time.
4581 48. Exim crashed when trying to verify a sender address that was being
4584 49. Exim was recognizing only a space character after ".include". It now also
4585 recognizes a tab character.
4587 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4588 extracting the relevant information from the specification. The man page no
4589 longer contains scrambled data for the -d option, and I've added a section
4590 at the front about calling Exim under different names.
4592 51. Added "extra_headers" argument to the "mail" command in filter files.
4594 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4597 53. Installed eximstats 1.29.
4599 54. Added transport_filter_timeout as a generic transport option.
4601 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4602 Cc: header lines. This was required by RFC 822, but it not required by RFC
4605 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4606 incoming messages that did not have them. Now it does so only if the
4607 message originates locally, that is, if there is no associated remote host
4608 address. When Resent- header lines are present, this applies to the Resent-
4609 lines rather than the non-Resent- lines.
4611 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4612 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4614 58. Messages for configuration errors now include the name of the main
4615 configuration file - useful now that there may be more than one file in a
4616 list (.included file names were always shown).
4618 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4619 for those rare installations that do not start the daemon as root or run it
4620 setuid root. I've cut out the call to initgroups() if the daemon is not
4623 60. The Exim user and group can now be bound into the binary as text strings
4624 that are looked up at the start of Exim's processing.
4626 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4628 62. Added $mailstore_basename variable.
4630 63. Installed patch to sieve.c from Michael Haardt.
4632 64. When Exim failed to open the panic log after failing to open the main log,
4633 the original message it was trying to log was written to stderr and debug
4634 output, but if they were not available (the usual case in production), it
4635 was lost. Now it is written to syslog before the two lines that record the
4636 failures to open the logs.
4638 65. Users' Exim filters run in subprocesses under the user's uid. It is
4639 possible for a "deliver" command or an alias in a "personal" command to
4640 provoke an address rewrite. If logging of address rewriting is configured,
4641 this fails because the process is not running as root or exim. There may be
4642 a better way of dealing with this, but for the moment (because 4.30 needs
4643 to be released), I have disabled address rewrite logging when running a
4644 filter in a non-root, non-exim process.
4650 1. The buildconfig auxiliary program wasn't quoting the value set for
4651 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4652 not defined. This bug was masked in 4.22 by the effect that was fixed in
4655 2. Some messages that were rejected after a message id was allocated were
4656 shown as "incomplete" by exigrep. It no longer does this for messages that
4657 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4659 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4660 have allow_domain_literals set, the ID did not get logged in the <= line.
4661 Domain literals are now always recognized in Message-ID: header lines.
4663 4. The first argument for a ${extract expansion item is the key name or field
4664 number. Leading and trailing spaces in this item were not being ignored,
4665 causing some misleading effects.
4667 5. When deliver_drop_privilege was set, single queue runner processes started
4668 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4669 same command in the process it spins off) were not dropping privilege.
4671 6. When the daemon running as "exim" started a queue runner, it always
4672 re-executed Exim in the spun-off process. This is a waste of effort when
4673 deliver_drop_privilege is set. The new process now just calls the
4674 queue-runner function directly.
4680 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4683 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4684 ignored. Though the use of -oP was forcing the writing of a pid file, it
4685 was always written to the default place.
4687 3. If the message "no IP address found for host xxxx" is generated during
4688 incoming verification, it is now followed by identification of the incoming
4689 connection (so you can more easily find what provoked it).
4691 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4693 5. Added some features to "harden" Exim a bit more against certain attacks:
4695 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4696 be put in Local/Makefile. This is like the never_users runtime option,
4697 but it cannot be overridden. The default setting is "root".
4699 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4700 prefix string with which any file named in a -C command line option
4703 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4704 is retained for -C and -D only if the caller of Exim is root. Without
4705 it, the exim user may also use -C and -D and retain privilege.
4707 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4708 command line option is disabled.
4710 6. Macro names set by the -D option must start with an upper case letter, just
4711 like macro names defined in the configuration file.
4713 7. Added "dereference=" facility to LDAP.
4715 8. Two instances of the typo "uknown" in the source files are fixed.
4717 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4718 the Configure-Makefile script screwed up while processing it.
4720 10. Incorporated PCRE 4.4.
4722 11. The SMTP synchronization check was not operating right at the start of an
4723 SMTP session. For example, it could not catch a HELO sent before the client
4724 waited for the greeting. There is now a check for outstanding input at the
4725 point when the greeting is written. Because of the duplex, asynchronous
4726 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4727 way, but not yet received, when the check is performed.
4729 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4730 on TCP/IP sockets, because this apparently causes some broken clients to
4733 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4734 unchanged) from the Cygwin maintainer.
4736 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4737 was supported instead of testing for mbx. Effectively a typo.
4739 15. The spa authenticator server code was not checking that the input it
4740 received was valid base64.
4742 16. The debug output line for the "set" modifier in ACLs was not showing the
4743 name of the variable that was being set.
4745 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4747 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4748 lookup on its own. Something else was needed to trigger the lookup. For
4749 example, a match in host_lookup or the need for a host name in a host list.
4750 Now, if $sender_host_name is referenced and the host name has not yet been
4751 looked up, a lookup is performed. If the lookup fails, the variable remains
4752 empty, and $host_lookup_failed is set to "1".
4754 19. Added "eqi" as a case-independent comparison operator.
4756 20. The saslauthd authentication condition could segfault if neither service
4757 nor realm was specified.
4759 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4760 error message that was logged was misleading, and incoming SMTP
4761 connections were dropped. The message is now more accurate, and temporary
4762 errors are given to SMTP connections.
4764 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4765 (except RSET) with a 421 error, until QUIT is received. However, it was
4766 failing to send a response to QUIT.
4768 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4769 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4770 ACL is now run after the helo_try_verify_hosts code.
4772 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4773 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4774 case-independent, so other case variants are also recognized). Apparently
4775 some systems use these upper case variants.
4777 25. If more than two messages were waiting for the same host, and a transport
4778 filter was specified for the transport, Exim sent two messages over the
4779 same TCP/IP connection, and then failed with "socket operation on non-
4780 socket" when it tried to send the third.
4782 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4784 27. The extern definition of crypt16() in expand.c was not being excluded when
4785 the OS had its own crypt16() function.
4787 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4788 as a preferred synonym for return_size_limit, both as an option and as an
4791 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4793 30. Changed the default configuration ACL to relax the local part checking rule
4794 for addresses that are not in any local domains. For these addresses,
4795 slashes and pipe symbols are allowed within local parts, but the sequence
4796 /../ is explicitly forbidden.
4798 31. SPA server authentication was not clearing the challenge buffer before
4801 32. log_message in a "warn" ACL statement was writing to the reject log as
4802 well as to the main log, which contradicts the documentation and doesn't
4803 seem right (because no rejection is happening). So I have stopped it.
4805 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4806 However, I am unable to do any testing of this.
4808 34. Fixed an infelicity in the appendfile transport. When checking directories
4809 for a mailbox, to see if any needed to be created, it was accidentally
4810 using path names with one or more superfluous leading slashes; tracing
4811 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4813 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4814 contents are added to the log line that is written for every discarded
4815 recipient. (Previously a log_message setting was ignored.)
4817 36. The ${quote: operator now quotes the string if it is empty.
4819 37. The install script runs exim in order to find its version number. If for
4820 some reason other than non-existence or emptiness, which it checks, it
4821 could not run './exim', it was installing it with an empty version number,
4822 i.e. as "exim-". This error state is now caught, and the installation is
4825 38. An argument was missing from the function that creates an error message
4826 when Exim fails to connect to the socket for saslauthd authentication.
4827 This could cause Exim to crash, or give a corrupted message.
4829 39. Added isip, isip4, and isip6 to ${if conditions.
4831 40. The ACL variables $acl_xx are now saved with the message, and can be
4832 accessed later in routers, transports, and filters.
4834 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4835 strings in the file are not string-expanded.
4837 42. If a MAIL command specified a SIZE value that was too large to fit into an
4838 int variable, the check against message_size_limit failed. Such values are
4839 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4840 day this will have to be increased, but I don't think I want to be around
4841 when emails are that large.
4848 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4849 iconv() is not standard in FreeBSD.
4851 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4852 IPv6 enabled. The observed symptom was a segmentation fault on return from
4853 the function os_common_find_running_interfaces() in src/os.c.
4855 3. In the check_special_case() function in daemon.c I had used "errno" as an
4856 argument name, which causes warnings on some systems. This was basically a
4857 typo, since it was named "eno" in the comments!
4859 4. The code that waits for the clock to tick (at a resolution of some fraction
4860 of a second) so as to ensure message-id uniqueness was always waiting for
4861 at least one whole tick, when it could have waited for less. [This is
4862 almost certainly not relevant at current processor speeds, where it is
4863 unlikely to ever wait at all. But we try to future-proof.]
4865 5. The function that sleeps for a time interval that includes fractions of a
4866 second contained a race. It did not block SIGALRM between setting the
4867 timer, and suspending (a couple of lines later). If the interval was short
4868 and the sigsuspend() was delayed until after it had expired, the suspension
4869 never ended. On busy systems this could lead to processes getting stuck for
4872 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4873 process, before it forks any delivery processes. The open lookup caching
4874 mechanism meant that the open file or database connection was passed into
4875 the delivery process. The problem was that delivery processes always tidy
4876 up cached lookup data. This could cause a problem for the next delivery
4877 process started by the queue runner, because the external queue runner
4878 process does not know about the closure. So the next delivery process
4879 still has data in the lookup cache. In the case of a file lookup, there was
4880 no problem because closing a file descriptor in a subprocess doesn't affect
4881 the parent. However, if the lookup was caching a connection to a database,
4882 the connection was closed, and the second delivery process was likely to
4883 see errors such as "PGSQL: query failed: server closed the connection
4884 unexpectedly". The problem has been fixed by closing all cached lookups
4885 in a queue runner before running a delivery process.
4887 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4888 seem to have the "const" qualifier which it has on other OS. I've
4891 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4892 *of the same type* (client or server) with the same public name that an
4893 error should be diagnosed.
4895 9. When Exim looked up a host name for an IP address, but failed to find the
4896 original IP address when looking up the host name (a safety check), it
4897 output the message "<ip address> does not match any IP for NULL", which was
4898 confusing, to say the least. The bug was that the host name should have
4899 appeared instead of "NULL".
4901 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4902 user that is built into the binary, and the -C or -D options is used, root
4903 privilege is dropped before the configuration file is read. In addition,
4904 logging is switched to stderr instead of the normal log files. If the
4905 configuration then re-defines the Exim user, the unprivileged environment
4906 is probably not what is expected, so Exim logs a panic warning message (but
4909 However, if deliver_drop_privilege is set, the unprivileged state may well
4910 be exactly what is intended, so the warning has been cut out in that case,
4911 and Exim is allowed to try to write to its normal log files.
4917 1. smtp_return_error_details was not giving details for temporary sender
4918 or receiver verification errors.
4920 2. Diagnose a configuration error if two authenticators have the same public
4923 3. Exim used not to create the message log file for a message until the first
4924 delivery attempt. This could be confusing when incoming messages were held
4925 for policy or load reasons. The message log file is now created at the time
4926 the message is received, and an initial "Received" line is written to it.
4928 4. The automatically generated man page for command line options had a minor
4929 bug that caused no ill effects; however, a more serious problem was that
4930 the procedure for building the man page automatically didn't always
4931 operate. Consequently, release 4.20 contains an out-of-date version. This
4932 shouldn't happen again.
4934 5. When building Exim with embedded Perl support, the script that builds the
4935 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4936 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4938 6. The freeze_tell option was not being used for messages that were frozen on
4939 arrival, either by an ACL or by local_scan().
4941 7. Added the smtp_incomplete_transaction log selector.
4943 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4944 was accepting AUTH without a new EHLO.
4946 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4947 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4950 10. Logging of TCP/IP connections (when configured) now happens in the main
4951 daemon process instead of the child process, so that the TCP/IP connection
4952 count is more accurate (but it can never be perfect).
4954 11. The use of "drop" in a nested ACL was not being handled correctly in the
4955 outer ACL. Now, if condition failure induced by the nested "drop" causes
4956 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4957 or "require"), the connection is dropped.
4959 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4960 that yield "discard" can now be used with an "accept" or a "discard" verb,
4961 but an error is generated for any others (because I can't see a useful way
4962 to define what should happen).
4964 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4965 are now processed in the same way as lines in the Exim configuration file.
4966 In particular, continuation lines are supported.
4968 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4970 15. Added -ti meaning -t -i.
4972 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4973 domains, and warn by logging if others are found.
4975 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4976 returned from GIFCONF when Exim is trying to find the list of interfaces on
4977 a host. The code in os.c has been modified to copy each ifreq to an aligned
4978 structure in all cases.
4980 Also, in some cases, the returned ifreq's were being copied to a 'struct
4981 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4982 means the last couple of bytes of an IPv6 address could be chopped if the
4983 ifreq contained only a normal sockaddr (14 bytes storage).
4985 18. Named domain lists were not supported in the hosts_treat_as_local option.
4986 An entry such as +xxxx was not recognized, and was treated as a literal
4989 19. Ensure that header lines added by a DATA ACL are included in the reject log
4990 if the ACL subsequently rejects the message.
4992 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4993 MD5 (which is deprecated).
4995 21. When testing a filter file using -bf, Exim was writing a message when it
4996 took the sender from a "From " line in the message, but it was not doing so
4997 when it took $return_path from a Return-Path: header line. It now does.
4999 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5000 with a valid header line field name (a series of printing characters
5001 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5003 23. Changed "disc" in the source to "disk" to conform to the documentation and
5004 the book and for uniformity.
5006 24. Ignore Sendmail's -Ooption=value command line item.
5008 25. When execve() failed while trying to run a command in a pipe transport,
5009 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5010 could be confused with a return value of 69 from the command itself. This
5011 has been changed to 127, the value the shell returns if it is asked to run
5012 a non-existent command. The wording for the related log line suggests a
5013 non-existent command as the problem.
5015 26. If received_header_text expands to an empty string, do not add a Received:
5016 header line to the message. (Well, it adds a token one on the spool, but
5017 marks it "old" so that it doesn't get used or transmitted.)
5019 27. Installed eximstats 1.28 (addition of -nt option).
5021 28. There was no check for failure on the call to getsockname() in the daemon
5022 code. This can fail if there is a shortage of resources on the system, with
5023 ENOMEM, for example. A temporary error is now given on failure.
5025 29. Contrary to the C standard, it seems that in some environments, the
5026 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5027 program. Exim now does this explicitly; it affects the formatting of
5028 timestamps using strftime().
5030 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5031 complaints. I've now initialized all the variables, to avoid this.
5033 32. Header lines added by a system filter were not being "seen" during
5034 transport-time rewrites.
5036 33. The info_callback() function passed to OpenSSL is set up with type void
5037 (*)(SSL *, int, int), as described somewhere. However, when calling the
5038 function (actually a macro) that sets it up, the type void(*)() is
5039 expected. I've put in a cast to prevent warnings from picky compilers.
5041 34. If a DNS black list lookup found a CNAME record, but there were no A
5042 records associated with the domain it pointed at, Exim crashed.
5044 35. If a DNS black list lookup returned more than one A record, Exim ignored
5045 all but the first. It now scans all returned addresses if a particular IP
5046 value is being sought. In this situation, the contents of the
5047 $dnslist_value variable are a list of all the addresses, separated by a
5050 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5051 to accept a host name and all its aliases if the forward lookup for any of
5052 them yielded the IP address of the incoming connection. Now it accepts only
5053 those names whose forward lookup yields the correct IP address. Any other
5054 names are discarded. This closes a loophole whereby a rogue DNS
5055 administrator could create reverse DNS records to break through a
5056 wildcarded host restriction in an ACL.
5058 37. If a user filter or a system filter that ran in a subprocess used any of
5059 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5060 the wrong values were passed to the pipe command ($thisaddress had the
5061 value of $0, $0 had the value of $1, etc). This bug was introduced by
5062 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5064 38. Improved the line breaking for long SMTP error messages from ACLs.
5065 Previously, if there was no break point between 40 and 75 characters, Exim
5066 left the rest of the message alone. Two changes have been made: (a) I've
5067 reduced the minimum length to 35 characters; (b) if it can't find a break
5068 point between 35 and 75 characters, it looks ahead and uses the first one
5069 that it finds. This may give the occasional overlong line, but at least the
5070 remaining text gets split now.
5072 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5073 file descriptors might be low, and that setting 1000 would always raise it.
5074 It turns out that in some environments, the limit is already over 1000 and
5075 that lowering it causes trouble. So now Exim takes care not to decrease it.
5077 40. When delivering a message, the value of $return_path is set to $sender_
5078 address at the start of routing (routers may change the value). By an
5079 oversight, this default was not being set up when an address was tested by
5080 -bt or -bv, which affected the outcome if any router or filter referred to
5083 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5084 log only when "message" or "log_message" are set. However, if one of the
5085 conditions was an address verification, or a call to a nested ACL, the
5086 messages generated by the underlying test were being passed through. This
5087 no longer happens. The underlying message is available in $acl_verify_
5088 message for both "message" and "log_message" expansions, so it can be
5089 passed through if needed.
5091 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5092 new expansion $bh_ to give the encoded byte string without charset
5093 translation. Translation happens only if iconv() is available; HAVE_ICONV
5094 indicates this at build time. HEADERS_CHARSET gives the charset to
5095 translate to; headers_charset can change it in the configuration, and
5096 "headers charset" can change it in an individual filter file.
5098 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5099 that creates RFC 2047 encoded "words" labels them as that charset instead
5100 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5101 expansion operator; (ii) when Exim creates a From: line for a local
5102 message; (iii) when a header line is rewritten to include a "phrase" part.
5104 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5105 buggy, causing it to skip the first lines of messages whose message ID
5106 ended in 'D'. This would not have bitten before Exim release 4.14, because
5107 message IDs were unlikely to end in 'D' before then. The effect was to have
5108 incorrect size information for certain domains.
5110 45. #include "config.h" was missing at the start of the crypt16.c module. This
5111 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5114 46. If there was a timeout during a "random" callout check, Exim treated it as
5115 a failure of the random address, and carried on sending RSET and the real
5116 address. If the delay was just some slowness somewhere, the response to the
5117 original RCPT would be taken as a response to RSET and so on, causing
5118 mayhem of various kinds.
5120 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5121 when I implemented it. It didn't allow for the fact that some option values
5122 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5123 the right test for positive values.
5125 48. Domain names in DNS records are case-independent. Exim always looks them up
5126 in lower case. Some resolvers return domain names in exactly the case they
5127 appear in the zone file, that is, they may contain uppercase letters. Not
5128 all resolvers do this - some return always lower case. Exim was treating a
5129 change of case by a resolver as a change of domain, similar to a widening
5130 of a domain abbreviation. This triggered its re-routing code and so it was
5131 trying to route what was effectively the same domain again. This normally
5132 caused routing to fail (because the router wouldn't handle the domain
5133 twice). Now Exim checks for this case specially, and just changes the
5134 casing of the domain that it ultimately uses when it transmits the message
5137 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5140 50. If a filter generated a file delivery with a non-absolute name (possible if
5141 no home directory exists for the router), the forbid_file option was not
5144 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5145 the existing equality matching.
5147 52. Exim was using ints instead of ino_t variables in some places where it was
5148 dealing with inode numbers.
5150 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5151 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5152 called TMPDIR, and if it finds it is different, it changes its value.
5154 54. The smtp_printf() function is now made available to local_scan() so
5155 additional output lines can be written before returning. There is also an
5156 smtp_fflush() function to enable the detection of a dropped connection.
5157 The variables smtp_input and smtp_batched_input are exported to
5160 55. Changed the default runtime configuration: the message "Unknown user"
5161 has been removed from the ACL, and instead placed on the localuser router,
5162 using the cannot_route_message feature. This means that any verification
5163 failures that generate their own messages won't get overridden. Similarly,
5164 the "Unrouteable address" message that was in the ACL for unverifiable
5165 relay addresses has also been removed.
5167 56. Added hosts_avoid_esmtp to the smtp transport.
5169 57. The exicyclog script was not checking for the esoteric option
5170 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5171 will work only if exicyclog is run under the appropriate euid.
5173 58. Following a discussion on the list, the rules by which Exim recognises line
5174 endings on incoming messages have been changed. The -dropcr and drop_cr
5175 options are now no-ops, retained only for backwards compatibility. The
5176 following line terminators are recognized: LF CRLF CR. However, special
5177 processing applies to CR:
5179 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5180 nor a local message in the state where . is a terminator.
5182 (ii) If a bare CR is encountered in a header line, an extra space is added
5183 after the line terminator so as not to end the header. The reasoning
5184 behind this is that bare CRs in header lines are most likely either
5185 to be mistakes, or people trying to play silly games.
5187 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5188 was being incorrectly given as 18 bytes larger than it should have been.
5189 This is a VOB (very old bug).
5191 60. This may never have affected anything current, but just in case it has:
5192 When the local host is found other than at the start of a list of hosts,
5193 the local host, those with the same MX, and any that follow, are discarded.
5194 When the list in question was part of a longer list of hosts, the following
5195 hosts (not currently being processed) were also being discarded. This no
5196 longer happens. I'm not sure if this situation could ever has previously
5199 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5202 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5203 header. This is something that is recommended in a new Internet Draft, and
5204 is something that is documented as being done by Sendmail. There are two
5205 possible values. For messages generated by the autoreply transport, Exim
5208 Auto-Submitted: auto-replied
5210 whereas for all other generated messages (e.g. bounces) it adds
5212 Auto-Submitted: auto-generated
5214 63. The "personal" condition in filters now includes a test for the
5215 Auto-Submitted: header. If it contains the string "auto-" the message it
5216 not considered personal.
5218 64. Added rcpt_include_affixes as a generic transport option.
5220 65. Added queue_only_override (default true).
5222 66. Added the syslog_duplication option.
5224 67. If what should have been the first header line of a message consisted of
5225 a space followed by a colon, Exim was mis-interpreting it as a header line.
5226 It isn't of course - it is syntactically invalid and should therefore be
5227 treated as the start of the message body. The misbehaviour could have
5228 caused a number of strange effects, including loss of data in subsequent
5229 header lines, and spool format errors.
5231 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5232 client host had authenticated. This control can now be exercised by an ACL
5233 for more flexibility.
5235 69. By default, callouts do not happen when testing with -bh. There is now a
5236 variant, -bhc, which does actually run the callout code, including
5237 consulting and updating the callout cache.
5239 70. Added support for saslauthd authentication, courtesy of Alexander
5242 71. If statvfs() failed on the spool or log directories while checking their
5243 size for availability, Exim confusingly gave the error "space shortage".
5244 Furthermore, in debugging mode it crashed with a floating point exception.
5245 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5246 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5247 serious problem, Exim now writes to the main and panic logs when this
5248 happens, with details of the failure. It then refuses to accept the
5249 incoming message, giving the message "spool directory problem" or "log
5250 directory problem" with a 421 code for SMTP messages.
5252 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5253 0, 1, and 2 exist, because some OS complain for execs without them (see
5254 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5255 descriptors. However, the code omitted to check that the open succeeded,
5256 causing mysterious errors if for some reason the permissions on /dev/null
5257 got screwed. Now Exim writes a message to the main and panic logs, and
5258 bombs out if it can't open /dev/null.
5260 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5261 interact so that it is all more flexible. It is supposed to remain
5262 backwards compatible. Also added extra_local_interfaces.
5264 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5265 to bomb out with an assertion failure - to the client this appears as a
5266 connection drop. This problem occurs in the part of the code that was taken
5267 from the Samba project. Fortunately, the assertion is in a very simple
5268 function, so I have fixed this by reproducing the function inline in the
5269 one place where it is called, and arranging for authentication to fail
5270 instead of killing the process with assert().
5272 75. The SPA client code was not working when the server requested OEM rather
5273 than Unicode encoding.
5275 76. Added code to make require_files with a specific uid setting more usable in
5276 the case where statting the file as root fails - usually a non-root-mounted
5277 NFS file system. When this happens and the failure is EACCES, Exim now
5278 forks a subprocess and does the per-uid checking as the relevant uid.
5280 77. Added process_log_path.
5282 78. If log_file_path was not explicitly set, a setting of check_log_space or
5283 check_log_inodes was ignored.
5285 79. If a space check for the spool or log partitions fails, the incident is now
5286 logged. Of course, in the latter case the data may get lost...
5288 80. Added the %p formatting code to string_format() so that it can be used to
5289 print addresses in debug_print(). Adjusted all the address printing in the
5290 debugging in store.c to use %p rather than %d.
5292 81. There was a concern that a line of code in smtp_in.c could overflow a
5293 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5294 initially expressed, the concern was not well-founded, because trailing
5295 spaces are removed early. However, if the trailing spaces were followed by
5296 a NULL, they did not get removed, so the overflow was possible. Two fixes
5299 (a) I re-wrote the offending code in a cleaner fashion.
5300 (b) If an incoming SMTP command contains a NULL character, it is rejected
5303 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5304 runs initgroups(), so that if the Exim user is in any additional groups,
5305 they will be used during message reception.
5311 The change log for 4.20 and earlier releases has been archived.