1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
44 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
45 Patch from Jakob Hirsch.
47 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
48 SQL string expansion failure details.
49 Patch from Andrey Oktyabrski.
51 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
52 Patch from Simon Arlott.
54 PP/09 New lookups functionality failed to compile on old gcc which rejects
55 extern declarations in function scope.
56 Patch from Oliver Fleischmann
58 PP/10 Use sig_atomic_t for flags set from signal handlers.
59 Check getgroups() return and improve debugging.
60 Fixed developed for diagnosis in bug 927 (which turned out to be
63 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
64 Patch from Mark Zealey.
66 PP/12 Bugzilla 1056: Improved spamd server selection.
67 Patch from Mark Zealey.
73 TF/01 Failure to get a lock on a hints database can have serious
74 consequences so log it to the panic log.
76 TF/02 Log LMTP confirmation messages in the same way as SMTP,
77 controlled using the smtp_confirmation log selector.
79 TF/03 Include the error message when we fail to unlink a spool file.
81 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
82 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
83 for maintaining out-of-tree patches for some time.
85 PP/01 Bugzilla 139: Documentation and portability issues.
86 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
87 Handle per-OS dynamic-module compilation flags.
89 PP/02 Let /dev/null have normal permissions.
90 The 4.73 fixes were a little too stringent and complained about the
91 permissions on /dev/null. Exempt it from some checks.
92 Reported by Andreas M. Kirchwitz.
94 PP/03 Report version information for many libraries, including
95 Exim version information for dynamically loaded libraries. Created
96 version.h, now support a version extension string for distributors
97 who patch heavily. Dynamic module ABI change.
99 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
100 privilege escalation vulnerability whereby the Exim run-time user
101 can cause root to append content of the attacker's choosing to
104 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
107 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
108 If dropping privileges for untrusted macros, we disabled normal logging
109 on the basis that it would fail; for the Exim run-time user, this is not
110 the case, and it resulted in successful deliveries going unlogged.
111 Fixed. Reported by Andreas Metzler.
117 PP/01 Date: & Message-Id: revert to normally being appended to a message,
118 only prepend for the Resent-* case. Fixes regression introduced in
119 Exim 4.70 by NM/22 for Bugzilla 607.
121 PP/02 Include check_rfc2047_length in configure.default because we're seeing
122 increasing numbers of administrators be bitten by this.
124 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
126 PP/03 Bugzilla 994: added openssl_options main configuration option.
128 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
130 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
132 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
134 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
135 without a peer certificate, leading to a segfault because of an
136 assumption that peers always have certificates. Be a little more
137 paranoid. Problem reported by Martin Tscholak.
139 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
140 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
141 NB: ClamAV planning to remove STREAM in "middle of 2010".
142 CL also introduces -bmalware, various -d+acl logging additions and
143 more caution in buffer sizes.
145 PP/09 Implemented reverse_ip expansion operator.
147 PP/10 Bugzilla 937: provide a "debug" ACL control.
149 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
151 PP/12 Bugzilla 973: Implement --version.
153 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
155 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
157 PP/15 Bugzilla 816: support multiple condition rules on Routers.
159 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
160 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
161 ignore trailing whitespace.
163 JJ/02 prevent non-panic DKIM error from being sent to paniclog
165 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
168 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
169 Notification from Dr Andrew Aitchison.
171 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
172 ExtendedDetectionInfo response format.
173 Notification from John Horne.
175 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
178 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
179 XSL and documented dependency on system catalogs, with examples of how
182 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
185 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
186 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
187 configuration file which is writeable by the Exim user or group.
189 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
190 of configuration files to cover files specified with the -C option if
191 they are going to be used with root privileges, not just the default
194 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
195 option (effectively making it always true).
197 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
198 files to be used while preserving root privileges.
200 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
201 that rogue child processes cannot use them.
203 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
204 run-time user, instead of root.
206 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
207 Exim run-time user without dropping privileges.
209 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
210 result string, instead of calling string_vformat() twice with the same
213 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
214 for other users. Others should always drop root privileges if they use
215 -C on the command line, even for a whitelisted configure file.
217 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
219 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
225 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
226 $data_path, and $header_path variables; fixed documentation bugs and
229 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
230 exipick to access non-standard spools, including the "frozen" queue
233 NM/01 Bugzilla 965: Support mysql stored procedures.
234 Patch from Alain Williams
236 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
238 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
239 Patch from Andreas Metzler
241 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
242 Patch from Kirill Miazine
244 NM/05 Bugzilla 671: Added umask to procmail example.
246 JJ/03 installed exipick 20100323.0, fixing doc bug
248 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
249 directory. Notification and patch from Dan Rosenberg.
251 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
253 TK/02 Improve log output when DKIM signing operation fails.
255 MH/01 Treat the transport option dkim_domain as a colon separated
256 list, not as a single string, and sign the message with each element,
257 omitting multiple occurences of the same signer.
259 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
260 Bugzilla 985, 986. Patch by Simon Arlott
262 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
263 Patch by Simon Arlott
265 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
266 MBX locking. Notification from Dan Rosenberg.
272 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
274 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
276 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
278 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
280 NM/04 Bugzilla 915: Flag broken perl installation during build.
286 TK/01 Added patch by Johannes Berg that expands the main option
287 "spamd_address" if it starts with a dollar sign.
289 TK/02 Write list of recipients to X-Envelope-Sender header when building
290 the mbox-format spool file for content scanning (suggested by Jakob
293 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
294 (http://www.dcc-servers.net/) support via dccifd. Activated by
295 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
297 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
298 by Mark Daniel Reidel <mr@df.eu>.
300 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
301 When building exim an external PCRE library is now needed -
302 PCRE is a system library on the majority of modern systems.
303 See entry on PCRE_LIBS in EDITME file.
305 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
306 conversation. Added nologin parameter to request.
307 Patch contributed by Kirill Miazine.
309 TF/01 Do not log submission mode rewrites if they do not change the address.
311 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
313 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
314 log files in place. Contributed by Roberto Lima.
316 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
318 TF/03 Bugzilla 615: When checking the local_parts router precondition
319 after a local_part_suffix or local_part_prefix option, Exim now
320 does not use the address's named list lookup cache, since this
321 contains cached lookups for the whole local part.
323 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
324 Robert Millan. Documentation is in experimental-spec.txt.
326 TF/04 Bugzilla 668: Fix parallel build (make -j).
328 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
330 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
331 Patch provided by Jan Srzednicki.
333 TF/05 Leading white space used to be stripped from $spam_report which
334 wrecked the formatting. Now it is preserved.
336 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
337 that they are available at delivery time.
339 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
341 TF/08 TLS error reporting now respects the incoming_interface and
342 incoming_port log selectors.
344 TF/09 Produce a more useful error message if an SMTP transport's hosts
345 setting expands to an empty string.
347 NM/06 Bugzilla 744: EXPN did not work under TLS.
348 Patch provided by Phil Pennock.
350 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
351 Patch provided by Richard Godbee.
353 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
354 acl_smtp_notquit, added index entry.
356 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
357 Patch provided by Eugene Bujak.
359 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
360 accept(). Patch provided by Maxim Dounin.
362 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
363 Patch provided by Phil Pennock.
365 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
367 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
368 Patch provided by Brad "anomie" Jorsch.
370 NM/14 Bugzilla 622: Added timeout setting to transport filter.
371 Patch provided by Dean Brooks.
373 TK/05 Add native DKIM support (does not depend on external libraries).
375 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
376 Patch provided by Graeme Fowler.
378 NM/16 Bugzilla 851: Documentation example syntax fix.
380 NM/17 Changed NOTICE file to remove references to embedded PCRE.
382 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
385 NM/19 Bugzilla 745: TLS version reporting.
386 Patch provided by Phil Pennock.
388 NM/20 Bugzilla 167: bool: condition support.
389 Patch provided by Phil Pennock.
391 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
392 clients. Patch provided by Phil Pennock.
394 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
395 Patch provided by Brad "anomie" Jorsch.
397 NM/23 Bugzilla 687: Fix misparses in eximstats.
398 Patch provided by Heiko Schlittermann.
400 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
401 Patch provided by Heiko Schlittermann.
403 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
404 plus update to original patch.
406 NM/26 Bugzilla 799: Documentation correction for ratelimit.
408 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
409 Patch provided by David Brownlee.
411 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
413 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
415 NM/30 Bugzilla 888: TLS documentation bugfixes.
417 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
419 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
420 Unlike the original bugzilla I have changed all shell scripts in src tree.
422 NM/33 Bugzilla 898: Transport filter timeout fix.
423 Patch by Todd Rinaldo.
425 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
426 Patch by Serge Demonchaux.
428 NM/35 Bugzilla 39: Base64 decode bug fixes.
429 Patch by Jakob Hirsch.
431 NM/36 Bugzilla 909: Correct connect() call in dcc code.
433 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
435 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
437 NM/39 Bugzilla 911: Fixed MakeLinks build script.
443 TK/01 Add preliminary DKIM support. Currently requires a forked version of
444 ALT-N's libdkim that I have put here:
445 http://duncanthrax.net/exim-experimental/
447 Note to Michael Haardt: I had to rename some vars in sieve.c. They
448 were called 'true' and it seems that C99 defines that as a reserved
449 keyword to be used with 'bool' variable types. That means you could
450 not include C99-style headers which use bools without triggering
451 build errors in sieve.c.
453 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
454 as mailq or other aliases. Changed the --help handling significantly
455 to do whats expected. exim_usage() emits usage/help information.
457 SC/01 Added the -bylocaldomain option to eximstats.
459 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
461 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
463 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
469 PH/01 Another patch from the Sieve maintainer.
471 PH/02 When an IPv6 address is converted to a string for single-key lookup
472 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
473 dots are used instead of colons so that keys in lsearch files need not
474 contain colons. This was done some time before quoting was made available
475 in lsearch files. However, iplsearch files do require colons in IPv6 keys
476 (notated using the quote facility) so as to distinguish them from IPv4
477 keys. This meant that lookups for IP addresses in host lists did not work
478 for iplsearch lookups.
480 This has been fixed by arranging for IPv6 addresses to be expressed with
481 colons if the lookup type is iplsearch. This is not incompatible, because
482 previously such lookups could never work.
484 The situation is now rather anomolous, since one *can* have colons in
485 ordinary lsearch keys. However, making the change in all cases is
486 incompatible and would probably break a number of configurations.
488 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
491 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
492 conversion specification without a maximum field width, thereby enabling
493 a rogue spamd server to cause a buffer overflow. While nobody in their
494 right mind would setup Exim to query an untrusted spamd server, an
495 attacker that gains access to a server running spamd could potentially
496 exploit this vulnerability to run arbitrary code as the Exim user.
498 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
499 $primary_hostname instead of what libspf2 thinks the hosts name is.
501 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
502 a directory entry by the name of the lookup key. Previously, if a
503 symlink pointed to a non-existing file or a file in a directory that
504 Exim lacked permissions to read, a lookup for a key matching that
505 symlink would fail. Now it is enough that a matching directory entry
506 exists, symlink or not. (Bugzilla 503.)
508 PH/03 The body_linecount and body_zerocount variables are now exported in the
511 PH/04 Added the $dnslist_matched variable.
513 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
514 This means they are set thereafter only if the connection becomes
517 PH/06 Added the client_condition to authenticators so that some can be skipped
518 by clients under certain conditions.
520 PH/07 The error message for a badly-placed control=no_multiline_responses left
521 "_responses" off the end of the name.
523 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
525 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
526 (without spaces) instead of just copying the configuration text.
528 PH/10 Added the /noupdate option to the ratelimit ACL condition.
530 PH/11 Added $max_received_linelength.
532 PH/12 Added +ignore_defer and +include_defer to host lists.
534 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
535 way in which PCRE > 7.0 is built.
537 PH/14 Implemented queue_only_load_latch.
539 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
540 MAIL command. The effect was to mangle the value on 64-bit systems.
542 PH/16 Another patch from the Sieve maintainer.
544 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
546 PH/18 If a system quota error occurred while trying to create the file for
547 a maildir delivery, the message "Mailbox is full" was not appended to the
548 bounce if the delivery eventually timed out. Change 4.67/27 below applied
549 only to a quota excession during the actual writing of the file.
551 PH/19 It seems that peer DN values may contain newlines (and other non-printing
552 characters?) which causes problems in log lines. The DN values are now
553 passed through string_printing() before being added to log lines.
555 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
556 and InterBase are left for another time.)
558 PH/21 Added message_body_newlines option.
560 PH/22 Guard against possible overflow in moan_check_errorcopy().
562 PH/23 POSIX allows open() to be a macro; guard against that.
564 PH/24 If the recipient of an error message contained an @ in the local part
565 (suitably quoted, of course), incorrect values were put in $domain and
566 $local_part during the evaluation of errors_copy.
572 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
573 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
576 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
577 issue a MAIL command.
579 PH/02 In an ACL statement such as
581 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
583 if a client was not listed at all, or was listed with a value other than
584 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
585 the condition was not true (as it should be), so access was not denied.
586 The bug was that the ! inversion was incorrectly passed on to the second
587 item. This has been fixed.
589 PH/03 Added additional dnslists conditions == and =& which are different from
590 = and & when the dns lookup returns more than one IP address.
592 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
593 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
595 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
596 FSYNC, which compiles an option called disable_fsync that allows for
597 bypassing fsync(). The documentation is heavily laced with warnings.
599 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
601 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
602 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
603 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
604 including adding "make clean"; (3) Added -fPIC when compiling the test
605 dynamically loaded module, to get rid of a warning.
607 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
608 message fails, move_frozen_messages = true and ignore_bounce_errors_after
609 = 0s. The bug is otherwise harmless.
611 PH/07 There was a bug in the dovecot authenticator such that the value of
612 $auth1 could be overwritten, and so not correctly preserved, after a
613 successful authentication. This usually meant that the value preserved by
614 the server_setid option was incorrect.
616 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
618 PH/09 Installed PCRE release 7.0.
620 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
621 run for batched SMTP input. It is now run at the start of every message
622 in the batch. While fixing this I discovered that the process information
623 (output by running exiwhat) was not always getting set for -bs and -bS
624 input. This is fixed, and it now also says "batched" for BSMTP.
626 PH/11 Added control=no_pipelining.
628 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
629 patch, slightly modified), and move the expansion of helo_data till after
630 the connection is made in the smtp transport (so it can use these
633 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
635 PH/14 Added log_selector = +pid.
637 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
639 PH/16 Add ${if forany and ${if forall.
641 PH/17 Added dsn_from option to vary the From: line in DSNs.
643 PH/18 Flush SMTP output before performing a callout, unless control =
644 no_callout_flush is set.
646 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
647 was true (the default) a successful delivery failed to delete the retry
648 item, thus causing premature timeout of the address. The bug is now
651 PH/20 Added hosts_avoid_pipelining to the smtp transport.
653 PH/21 Long custom messages for fakedefer and fakereject are now split up
654 into multiline reponses in the same way that messages for "deny" and
655 other ACL rejections are.
657 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
658 with slight modification.
660 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
661 draft, changing the syntax and factoring some duplicate code".
663 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
664 for deliveries of the second and subsequent messages over the same SMTP
667 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
668 ${reduce, with only minor "tidies".
670 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
672 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
673 expansion side effects.
675 PH/27 When a message times out after an over-quota error from an Exim-imposed
676 quota, the bounce message says "mailbox is full". This message was not
677 being given when it was a system quota that was exceeded. It now should
680 MH/03 Made $recipients available in local_scan(). local_scan() already has
681 better access to the recipient list through recipients_list[], but
682 $recipients can be useful in postmaster-provided expansion strings.
684 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
685 in the case of a MAIL command with additional options following the
686 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
687 were accidentally chopped off.
689 PH/29 SMTP synchronization checks are implemented when a command is read -
690 there is a check that no more input is waiting when there shouldn't be
691 any. However, for some commands, a delay in an ACL can mean that it is
692 some time before the response is written. In this time, more input might
693 arrive, invalidly. So now there are extra checks after an ACL has run for
694 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
695 pipelining has not been advertised.
697 PH/30 MH's patch to allow iscntrl() characters to be list separators.
699 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
700 returned in the SMTP response when smtp_return_error_details was false.
703 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
704 instead of the C I/O that was originally supplied, because problems were
707 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
708 Exim which did not show up earlier: it was assuming that a call to
709 SSL_CTX_set_info_callback() might give an error value. In fact, there is
710 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
711 was a macro that became an assignment, so it seemed to work. This has
712 changed to a proper function call with a void return, hence the compile
713 error. Exim's code has been fixed.
715 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
718 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
720 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
722 PH/37 If a message is not accepted after it has had an id assigned (e.g.
723 because it turns out to be too big or there is a timeout) there is no
724 "Completed" line in the log. When some messages of this type were
725 selected by exigrep, they were listed as "not completed". Others were
726 picked up by some special patterns. I have improved the selection
727 criteria to be more general.
729 PH/38 The host_find_failed option in the manualroute router can now be set
730 to "ignore", to completely ignore a host whose IP address cannot be
731 found. If all hosts are ignored, the behaviour is controlled by the new
732 host_all_ignored option.
734 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
735 homing or because of multiple MX records with /mx) generated more than
736 one IP address, and the following item turned out to be the local host,
737 all the secondary addresses of the first item were incorrectly removed
738 from the list, along with the local host and any following hosts (which
739 is what is supposed to happen).
741 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
742 whoever called Exim into the -H file. In the case of the daemon it was
743 behaving confusingly. When first started, it used values for whoever
744 started the daemon, but after a SIGHUP it used the Exim user (because it
745 calls itself on a restart). I have changed the code so that it now always
748 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
749 message are rejected with the same error (e.g. no authentication or bad
750 sender address), and a DATA command is nevertheless sent (as can happen
751 with PIPELINING or a stupid MUA), the error message that was given to the
752 RCPT commands is included in the rejection of the DATA command. This is
753 intended to be helpful for MUAs that show only the final error to their
756 PH/42 Another patch from the Sieve maintainer.
758 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
759 Eximstats - Fixed some broken HTML links and added missing column headers
761 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
762 columns for Rejects, Temp Rejects, Ham, and Spam rows.
764 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
766 PH/43 Yet another patch from the Sieve maintainer.
768 PH/44 I found a way to check for a TCP/IP connection going away before sending
769 the response to the final '.' that terminates a message, but only in the
770 case where the client has not sent further data following the '.'
771 (unfortunately, this is allowed). However, in many cases there won't be
772 any further data because there won't be any more messages to send. A call
773 to select() can be used: if it shows that the input is "ready", there is
774 either input waiting, or the socket has been closed. An attempt to read
775 the next input character can distinguish the two cases. Previously, Exim
776 would have sent an OK response which the client would never have see.
777 This could lead to message repetition. This fix should cure that, at
778 least in a lot of common cases.
780 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
781 advertised in response to EHLO.
787 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
788 fixed by 4.65/MH/01 (is this a record?) are fixed:
790 (i) An empty string was always treated as zero by the numeric comparison
791 operators. This behaviour has been restored.
793 (ii) It is documented that the numeric comparison operators always treat
794 their arguments as decimal numbers. This was broken in that numbers
795 starting with 0 were being interpreted as octal.
797 While fixing these problems I realized that there was another issue that
798 hadn't been noticed. Values of message_size_limit (both the global option
799 and the transport option) were treated as octal if they started with 0.
800 The documentation was vague. These values are now always treated as
801 decimal, and I will make that clear in the documentation.
807 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
808 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
811 MH/01 Don't check that the operands of numeric comparison operators are
812 integers when their expansion is in "skipping" mode (fixes bug
813 introduced by 4.64-PH/07).
815 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
816 child addresses, Exim now panics and dies. Previously, because the count
817 is held in a short int, deliveries were likely to be lost. As such a
818 large number of recipients for a single message is ridiculous
819 (performance will be very, very poor), I have chosen to impose a limit
820 rather than extend the field.
826 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
827 leftover -K file (the existence of which was triggered by #402).
828 While we were at it, introduced process PID as part of the -K
829 filename. This should rule out race conditions when creating
832 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
833 processing considerably. Previous code took too long for large mails,
834 triggering a timeout which in turn triggers #401.
836 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
837 in the DK code in transports.c. sendfile() is not really portable,
838 hence the _LINUX specificness.
840 TF/01 In the add_headers option to the mail command in an Exim filter,
841 there was a bug that Exim would claim a syntax error in any
842 header after the first one which had an odd number of characters
845 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
846 callout verification, Exim cached a "reject" for the entire domain. This
847 is correct for most verifications, but it is not correct for a recipient
848 verification with use_sender or use_postmaster set, because in that case
849 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
850 case of MAIL FROM:<> rejection from other early rejections (e.g.
851 rejection of HELO). When verifying a recipient using a non-null MAIL
852 address, the cache is ignored if it shows MAIL FROM:<> rejection.
853 Whatever the result of the callout, the value of the domain cache is
854 left unchanged (for any other kind of callout, getting as far as trying
855 RCPT means that the domain itself is ok).
857 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
860 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
861 manifest itself as EPIPE rather than ECONNECT. When tidying away a
862 session, the daemon ignores ECONNECT errors and logs others; it now
863 ignores EPIPE as well.
865 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
866 (quoted-printable decoding).
868 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
869 later the small subsequent patch to fix an introduced bug.
871 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
873 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
875 PH/08 An error is now given if message_size_limit is specified negative.
877 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
878 to be given (somewhat) arbitrary names.
880 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
883 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
884 miscellaneous code fixes
886 PH/10 Added the log_reject_target ACL modifier to specify where to log
889 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
890 hostname. This is wrong, because it relates to the incoming message (and
891 probably the interface on which it is arriving) and not to the outgoing
892 callout (which could be using a different interface). This has been
893 changed to use the value of the helo_data option from the smtp transport
894 instead - this is what is used when a message is actually being sent. If
895 there is no remote transport (possible with a router that sets up host
896 addresses), $smtp_active_hostname is used.
898 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
899 tweaks were necessary in order to get it to work (see also 21 below):
900 (a) The code assumed that strncpy() returns a negative number on buffer
901 overflow, which isn't the case. Replaced with Exim's string_format()
903 (b) There were several signed/unsigned issues. I just did the minimum
904 hacking in of casts. There is scope for a larger refactoring.
905 (c) The code used strcasecmp() which is not a standard C function.
906 Replaced with Exim's strcmpic() function.
907 (d) The code set only $1; it now sets $auth1 as well.
908 (e) A simple test gave the error "authentication client didn't specify
909 service in request". It would seem that Dovecot has changed its
910 interface. Fortunately there's a specification; I followed it and
911 changed what the client sends and it appears to be working now.
913 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
916 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
917 address A is aliased to B and C, where B exists and C does not. Without
918 -v the output is "A verified" because verification stops after a
919 successful redirection if more than one address is generated. However,
920 with -v the child addresses are also verified. Exim was outputting "A
921 failed to verify" and then showing the successful verification for C,
922 with its parentage. It now outputs "B failed to verify", showing B's
923 parentage before showing the successful verification of C.
925 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
926 look up a TXT record in a specific list after matching in a combined
929 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
930 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
931 they consult the DNS. I had assumed they would set it the way they
932 wanted; and indeed my experiments on Linux seem to show that in some
933 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
934 To be on the safe side, however, I have now made the interface to
935 host_find_byname() similar to host_find_bydns(), with an argument
936 containing the DNS resolver options. The host_find_byname() function now
937 sets these options at its start, just as host_find_bydns() does. The smtp
938 transport options dns_qualify_single and dns_search_parents are passed to
939 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
940 of host_find_byname() use the default settings of RES_DEFNAMES
941 (qualify_single) but not RES_DNSRCH (search_parents).
943 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
944 spool_read_header() do less string testing, by means of a preliminary
945 switch on the second character of optional "-foo" lines. (This is
946 overdue, caused by the large number of possibilities that now exist.
947 Originally there were few.) While I was there, I also converted the
948 str(n)cmp tests so they don't re-test the leading "-" and the first
949 character, in the hope this might squeeze out yet more improvement.
951 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
952 flag allowing group syntax was set by the header_syntax check but not
953 turned off, possible causing trouble later; (2) The flag was not being
954 set at all for the header_verify test, causing "group"-style headers to
955 be rejected. I have now set it in this case, and also caused header_
956 verify to ignore an empty address taken from a group. While doing this, I
957 came across some other cases where the code for allowing group syntax
958 while scanning a header line wasn't quite right (mostly, not resetting
959 the flag correctly in the right place). These bugs could have caused
960 trouble for malformed header lines. I hope it is now all correct.
962 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
963 with the "reply" argument non-NULL. The code, however (which originally
964 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
965 but it didn't always do it. This confused somebody who was copying the
966 code for some other use. I have removed all the tests.
968 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
969 feature that was used to support insecure browsers during the U.S. crypto
970 embargo. It requires special client support, and Exim is probably the
971 only MTA that supported it -- and would never use it because real RSA is
972 always available. This code has been removed, because it had the bad
973 effect of slowing Exim down by computing (never used) parameters for the
974 RSA_EXPORT functionality.
976 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
977 authenticator to fail if there's a tab character in the incoming data
978 (there should never be unless someone is messing about, as it's supposed
979 to be base64-encoded). Also added, on Timo's advice, the "secured" option
980 if the connection is using TLS or if the remote IP is the same as the
981 local IP, and the "valid-client-cert option" if a client certificate has
984 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
985 authenticators. This can be used for authorization after authentication
986 succeeds. (In the case of plaintext, it servers for both authentication
989 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
990 if any retry times were supplied.
992 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
993 connection, where there is no client IP address to check. In this
994 situation, the verify now always succeeds.
996 PH/25 Applied John Jetmore's -Mset patch.
998 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1000 PH/27 In a string expansion for a processed (not raw) header when multiple
1001 headers of the same name were present, leading whitespace was being
1002 removed from all of them, but trailing whitespace was being removed only
1003 from the last one. Now trailing whitespace is removed from each header
1004 before concatenation. Completely empty headers in a concatenation (as
1005 before) are ignored.
1007 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1008 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1010 PH/29 [Removed. This was a change that I later backed out, and forgot to
1011 correct the ChangeLog entry (that I had efficiently created) before
1012 committing the later change.]
1014 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1015 address errors (4xx response to RCPT) over the same connection as other
1016 messages routed to the same hosts. Such deliveries are always "forced",
1017 so retry times are not inspected. This resulted in far too many retries
1018 for the affected addresses. The effect occurred only when there were more
1019 hosts than the hosts_max_try setting in the smtp transport when it had
1020 the 4xx errors. Those hosts that it had tried were not added to the list
1021 of hosts for which the message was waiting, so if all were tried, there
1022 was no problem. Two fixes have been applied:
1024 (i) If there are any address or message errors in an SMTP delivery, none
1025 of the hosts (tried or untried) are now added to the list of hosts
1026 for which the message is waiting, so the message should not be a
1027 candidate for sending over the same connection that was used for a
1028 successful delivery of some other message. This seems entirely
1029 reasonable: after all the message is NOT "waiting for some host".
1030 This is so "obvious" that I'm not sure why it wasn't done
1031 previously. Hope I haven't missed anything, but it can't do any
1032 harm, as the worst effect is to miss an optimization.
1034 (ii) If, despite (i), such a delivery is accidentally attempted, the
1035 routing retry time is respected, so at least it doesn't keep
1036 hammering the server.
1038 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1039 in ${readsocket because some servers need this prod.
1041 PH/32 Added some extra debug output when updating a wait-xxx database.
1043 PH/33 The hint "could be header name not terminated by colon", which has been
1044 given for certain expansion errors for a long time, was not being given
1045 for the ${if def:h_colon_omitted{... case.
1047 PH/34 The spec says: "With one important exception, whenever a domain list is
1048 being scanned, $domain contains the subject domain." There was at least
1049 one case where this was not true.
1051 PH/35 The error "getsockname() failed: connection reset by peer" was being
1052 written to the panic log as well as the main log, but it isn't really
1053 panic-worthy as it just means the connection died rather early on. I have
1054 removed the panic log writing for the ECONNRESET error when getsockname()
1057 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1058 runs only) independently of the message's sender address. This meant
1059 that, if the 4xx error was in fact related to the sender, a different
1060 message to the same recipient with a different sender could confuse
1061 things. In particualar, this can happen when sending to a greylisting
1062 server, but other circumstances could also provoke similar problems.
1063 I have changed the default so that the retry time for these errors is now
1064 based a combination of the sender and recipient addresses. This change
1065 can be overridden by setting address_retry_include_sender=false in the
1068 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1069 remote server are returned as part of bounce messages. This was not
1070 happening for LMTP over a pipe (the lmtp transport), but now it is the
1071 same for both kinds of LMTP.
1073 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1074 in header lines that were in fact CNAMEs. This is no longer the case.
1076 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1077 and queue runs started by the daemon processed all messages. This has
1078 been fixed so that -R and -S can now usefully be given with -q<time>.
1080 PH/40 Import PCRE release 6.7 (fixes some bugs).
1082 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1084 PH/42 Give an error if -q is specified more than once.
1086 PH/43 Renamed the variables $interface_address and $interface_port as
1087 $received_ip_address and $received_port, to make it clear that these
1088 values apply to message reception, and not to the outgoing interface when
1089 a message is delivered. (The old names remain recognized, of course.)
1091 PH/44 There was no timeout on the connect() call when using a Unix domain
1092 socket in the ${readsocket expansion. There now is.
1094 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1095 be meaningful with "accept".
1097 SC/01 Eximstats V1.43
1098 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1100 SC/02 Eximstats V1.44
1101 Use a glob alias rather than an array ref in the generated
1102 parser. This improves both readability and performance.
1104 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1105 Collect SpamAssassin and rejection statistics.
1106 Don't display local sender or destination tables unless
1107 there is data to show.
1108 Added average volumes into the top table text output.
1110 SC/04 Eximstats V1.46
1111 Collect data on the number of addresses (recipients)
1112 as well as the number of messages.
1114 SC/05 Eximstats V1.47
1115 Added 'Message too big' to the list of mail rejection
1116 reasons (thanks to Marco Gaiarin).
1118 SC/06 Eximstats V1.48
1119 Mainlog lines which have GMT offsets and are too short to
1120 have a flag are now skipped.
1122 SC/07 Eximstats V1.49 (Alain Williams)
1123 Added the -emptyok flag.
1125 SC/08 Eximstats V1.50
1126 Fixes for obtaining the IP address from reject messages.
1128 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1129 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1130 whitesspace changes from 4.64-PH/27
1132 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1135 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1136 are found, allow negative numbers in numeric criteria)
1138 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1140 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1141 to match changes made in 4.64-PH/43
1143 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1145 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1146 "domain" is actually an IP address, and force a failure. This locks out
1147 those revolvers/nameservers that support "A-for-A" lookups, in
1148 contravention of the specifications.
1150 PH/48 When a host name was looked up from an IP address, and the subsequent
1151 forward lookup of the name timed out, the host name was left in
1152 $sender_host_name, contrary to the specification.
1154 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1155 restricted to single-key lookups, Exim was not diagnosing an error if
1156 * or *@ was used with a query-style lookup.
1158 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1160 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1161 long ago, but noone interested enough thought of it. Let's just say that
1162 the "1.1" means that there are some new functions that weren't there at
1163 some point in the past.
1165 PH/51 Error processing for expansion failure of helo_data from an smtp
1166 transport during callout processing was broken.
1168 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1169 tested/used via the -bh/-bhc/-bs options.
1171 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1172 bug, fixed in subsequent PCRE releases).
1174 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1175 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1177 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1183 SC/01 Use a glob alias rather than an array ref in eximstats generated
1184 parser. This improves both readability and performance.
1186 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1187 Don't display local sender or destination tables in eximstats unless
1188 there is data to show.
1189 Added average volumes into the eximstats top table text output.
1191 SC/03 Collect data on the number of addresses (recipients) as well
1192 as the number of messages in eximstats.
1194 TF/01 Correct an error in the documentation for the redirect router. Exim
1195 does (usually) call initgroups() when daemonizing.
1197 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1198 with consistent privilege compared to when running as a daemon.
1200 TF/03 Note in the spec that $authenticated_id is not set for local
1201 submissions from trusted users.
1203 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1204 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1206 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1207 by adding some example configuration directives to the default
1208 configuration file. A little bit of work is required to uncomment the
1209 directives and define how usernames and passwords are checked, but
1210 there is now a framework to start from.
1212 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1213 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1214 without this. I don't know how relevant this is to other LDAP libraries.
1216 PH/02 Add the verb name to the "unknown ACL verb" error.
1218 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1220 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1222 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1223 directory not expanded when it should be if an expanded home directory
1224 was set for the address (which is overridden by the transport).
1226 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1229 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1230 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1231 because it is too late at that time, and has no effect.
1233 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1234 security issue with \' (bugzilla #107). I could not use the
1235 PQescapeStringConn() function, because it needs a PGconn value as one of
1238 PH/08 When testing addresses using -bt, indicate those final addresses that
1239 are duplicates that would not cause an additional delivery. At least one
1240 person was confused, thinking that -bt output corresponded to deliveries.
1241 (Suppressing duplicates isn't a good idea as you lose the information
1242 about possibly different redirections that led to the duplicates.)
1244 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1245 systems where poll() doesn't work, in particular OS X.
1247 PH/10 Added more information to debugging output for retry time not reached.
1249 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1250 operations in malware.c.
1252 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1255 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1256 syslog_duplication set false, log lines that would normally be written
1257 both the the main log and to the reject log were not written to syslog at
1260 PH/14 In the default configuration, change the use of "message" in ACL warn
1261 statements to "add_header".
1263 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1264 not followed by a command (e.g. "seen endif").
1266 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1267 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1270 PH/17 Added extra conditions to the default value of delay_warning_condition
1274 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1275 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1276 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1279 The Auto-Submitted: and various List- headers are standardised, whereas I
1280 don't think Precedence: ever was.
1282 PH/18 Refactored debugging code in route_finduser() to show more information,
1283 in particular, the error code if getpwnam() issues one.
1285 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1286 This is apparently needed in addition to the PH/07 change above to avoid
1287 any possible encoding problems.
1289 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1290 but not after initializing Perl.
1292 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1293 output them only if debugging. By default they are written stderr,
1294 apparently, which is not desirable.
1296 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1299 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1302 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1304 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1305 authenticated or an ident call has been made. Suppress the default
1306 values for $authenticated_id and $authenticated_sender (but permit -oMai
1307 and -oMas) when testing with -bh.
1309 PH/24 Re-jigged the order of the tests in the default configuration so that the
1310 tests for valid domains and recipients precede the DNS black list and CSA
1311 tests, on the grounds that those ones are more expensive.
1313 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1314 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1315 This bug exists in every version of Exim that I still have, right back to
1318 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1319 However, an attempt to turn on case-sensitivity in a regex key by
1320 including (?-i) didn't work because the subject string was already
1321 lowercased, and the effects were non-intuitive. It turns out that a
1322 one-line patch can be used to allow (?-i) to work as expected.
1328 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1329 other effects) broke the use of negated acl sub-conditions.
1331 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1334 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1335 "Deny" causes Exim to reject the incoming connection with a 554 error.
1336 Unfortunately, if there is a major crisis, such as a disk failure,
1337 tcp-wrappers gives "deny", whereas what one would like would be some
1338 kind of temporary error. A kludge has been added to help with this.
1339 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1340 554 error is used if errno is still zero or contains ENOENT (which occurs
1341 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1344 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1346 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1347 errors. Otherwise a message that provokes a temporary error (when other
1348 messages do not) can cause a whole host to time out.
1350 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1351 addresses were routed directly to files or pipes from a redirect router.
1352 File deliveries just didn't batch; pipe deliveries might have suffered
1355 PH/06 A failure to get a lock for a hints database would erroneously always say
1356 "Failed to get write lock", even when it was really a read lock.
1358 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1359 of 0600. This has been changed to use the value of the lockfile_mode
1360 option (which defaults to 0600).
1362 PH/08 Applied small patch from the Sieve maintainer.
1364 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1365 folder from quota calculations, a direct delivery into this folder messed
1366 up the contents of the maildirsize file. This was because the regex was
1367 used only to exclude .Trash (or whatever) when the size of the mailbox
1368 was calculated. There was no check that a delivery was happening into an
1369 excluded directory. This bug has been fixed by ignoring all quota
1370 processing for deliveries into excluded directories.
1372 PH/10 Added the maildirfolder_create_regex option to appendfile.
1378 PH/01 The code for finding all the local interface addresses on a FreeBSD
1379 system running IPv6 was broken. This may well have applied to all BSD
1380 systems, as well as to others that have similar system calls. The broken
1381 code found IPv4 interfaces correctly, but gave incorrect values for the
1382 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1383 that it would not match correctly against @[] and not recognize the IPv6
1386 PH/02 The ipliteral router was not recognizing addresses of the form user@
1387 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1389 PH/03 Added disable_ipv6.
1391 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1392 decoded form, because it is most often used to construct To: headers
1393 lines in autoreplies, and the decoded form may well be syntactically
1394 invalid. However, $reply_address has leading white space removed, and all
1395 newlines turned into spaces so that the autoreply transport does not
1398 PH/05 If group was specified without a user on a router, and no group or user
1399 was specified on a transport, the group from the router was ignored.
1401 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1402 for visible compile-time settings that can be used to change these
1403 numbers, for those that want even more. Backwards compatibility with old
1404 spool files has been maintained. However, going back to a previous Exim
1405 release will lost any variables that are in spool files.
1407 PH/07 Two small changes when running in the test harness: increase delay when
1408 passing a TCP/IP connection to a new process, in case the original
1409 process has to generate a bounce, and remove special handling of
1410 127.0.0.2 (sic), which is no longer necessary.
1412 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1413 be the same on different OS.
1415 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1418 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1419 whether --show-vars was specified or not
1421 JJ/02 exipick: Added support for new ACL variable spool format introduced
1424 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1425 syntactically invalid From: or Reply-to: line, and a filter used this to
1426 generate an autoreply, and therefore failed to obtain an address for the
1427 autoreply, Exim could try to deliver to a non-existent relative file
1428 name, causing unrelated and misleading errors. What now happens is that
1429 it logs this as a hard delivery error, but does not attempt to create a
1432 PH/11 The exinext utility has a -C option for testing purposes, but although
1433 the given file was scanned by exinext itself; it wasn't being passed on
1434 when Exim was called.
1436 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1437 an end-of-file indication when reading a command response.
1439 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1440 compiled. In many other places in Exim, IPv6 addresses are always
1441 recognized, so I have changed this. It also means that IPv4 domain
1442 literals of the form [IPV4:n.n.n.n] are now always recognized.
1444 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1445 used if the router is not running as root, for example, when verifying at
1446 ACL time, or when using -bh. The debugging output from this situation was
1447 non-existent - all you got was a failure to exec. I have made two
1450 (a) Failures to set uid/gid, the current directory, or a process leader
1451 in a subprocess such as that created by queryprogram now generate
1452 suitable debugging ouput when -d is set.
1454 (b) The queryprogram router detects when it is not running as root,
1455 outputs suitable debugging information if -d is set, and then runs
1456 the subprocess without attempting to change uid/gid.
1458 PH/15 Minor change to Makefile for building test_host (undocumented testing
1461 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1462 additional section of a DNS packet that returns MX or SRV records.
1463 Instead, it always explicitly searches for A/AAAA records. This avoids
1464 major problems that occur when a DNS server includes only records of one
1465 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1466 fixed another bug: if SRV records were looked up and the corresponding
1467 address records were *not* found in the additional section, the port
1468 values from the SRV records were lost.
1470 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1471 using the correct key (the original address) when searching the retry
1472 rules in order to find which one to use for generating the retry hint.
1474 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1475 adding the default one. Similarly, if it contains a Reply-To: header, the
1476 errors_reply_to option, if set, is not used.
1478 PH/19 When calculating a retry time, Exim used to measure the "time since
1479 failure" by looking at the "first failed" field in the retry record. Now
1480 it does not use this if it is later than than the arrival time of the
1481 message. Instead it uses the arrival time. This makes for better
1482 behaviour in cases where some deliveries succeed, thus re-setting the
1483 "first failed" field. An example is a quota failure for a huge message
1484 when small messages continue to be delivered. Without this change, the
1485 "time since failure" will always be short, possible causing more frequent
1486 delivery attempts for the huge message than are intended.
1487 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1489 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1490 $1, $2, $3) because the numerical variables can be reset during some
1491 expansion items (e.g. "match"), thereby losing the authentication data.
1493 PH/21 Make -bV show the size of off_t variables so that the test suite can
1494 decide whether to run tests for quotas > 2G.
1496 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1497 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1498 filecount value is greater than 2G or if a quota value is greater than 2G
1499 on a system where the size of off_t is not greater than 4, a panic error
1502 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1503 never match. The debug and -bh output now contains an explicit error
1504 message indicating a malformed IPv4 address or mask.
1506 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1507 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1508 PH/23 above applies.
1510 PH/25 Do not write to syslog when running in the test harness. The only
1511 occasion when this arises is a failure to open the main or panic logs
1512 (for which there is an explicit test).
1514 PH/26 Added the /no_tell option to "control=freeze".
1516 PH/27 If a host name lookup failed very early in a connection, for example, if
1517 the IP address matched host_lookup and the reverse lookup yielded a name
1518 that did not have a forward lookup, an error message of the form "no IP
1519 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1520 could be logged. Now it outputs the IP address instead of "NULL".
1522 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1523 allows the sender and the authenticated sender to be set when
1524 submitting a message from within Exim. Since child_open_exim() is
1525 documented for local_scan(), the new function should be too.
1527 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1528 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1529 results in an empty string is now treated as unset.
1531 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1533 PH/31 Added sender_verify_fail logging option.
1535 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1536 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1537 tidied the source and removed it altogether.
1539 PH/33 When a queue run was abandoned because the load average was too high, a
1540 log line was always written; now it is written only if the queue_run log
1541 selector is set. In addition, the log line for abandonment now contains
1542 information about the queue run such as the pid. This is always present
1543 in "start" and "stop" lines but was omitted from the "abandon" line.
1545 PH/34 Omit spaces between a header name and the colon in the error message that
1546 is given when verify = headers_syntax fails (if there are lots of them,
1547 the message gets confusing).
1549 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1550 names, as there are now some PTR records that contain slashes. This check
1551 is only to protect against broken name servers that fall over on strange
1552 characters, so the fact that it applies to all lookups doesn't matter.
1554 PH/36 Now that the new test suite is complete, we can remove some of the
1555 special code in Exim that was needed for the old test suite. For example,
1556 sorting DNS records because real resolvers return them in an arbitrary
1557 order. The new test suite's fake resolver always returns records in the
1560 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1561 bounces) except when queue_only is set, to avoid logging races between
1562 the different processes.
1564 PH/38 Panic-die if .include specifies a non-absolute path.
1566 PH/39 A tweak to the "H" retry rule from its user.
1568 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1569 a label. They prevented compilation on older perls.
1571 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1572 a warning to be raised on newish perls.
1574 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1575 on queue. Changes to match documented behaviour of showing count of
1576 messages matching specified criteria.
1578 PH/40 Changed the default ident timeout from 30s to 5s.
1580 PH/41 Added support for the use of login_cap features, on those BSD systems
1581 that have them, for controlling the resources used by pipe deliveries.
1583 PH/42 The content-scanning code uses fopen() to create files in which to put
1584 message data. Previously it was not paying any attention to the mode of
1585 the files. Exim runs with umask(0) because the rest of the code creates
1586 files with open(), and sets the required mode explicitly. Thus, these
1587 files were ending up world-writeable. This was not a big issue, because,
1588 being within the spool directory, they were not world-accessible. I have
1589 created a function called modefopen, which takes an additional mode
1590 argument. It sets umask(777), creates the file, chmods it to the required
1591 mode, then resets the umask. All the relevant calls to fopen() in the
1592 content scanning code have been changed to use this function.
1594 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1595 to 24 hours. This avoids potential overflow problems when processing G
1596 and H retry rules. I suspect nobody ever tinkers with this value.
1598 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1600 PH/45 When the plaintext authenticator is running as a client, the server's
1601 challenges are checked to ensure they are valid base64 strings. By
1602 default, the authentication attempt is cancelled if an invalid string is
1603 received. Setting client_ignore_invalid_base64 true ignores these errors.
1604 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1605 they are received. Thus, the responses can be made to depend on the
1606 challenges. If an invalid string is ignored, an empty string is placed in
1609 PH/46 Messages that are created by the autoreply transport now contains a
1610 References: header, in accordance with RFCs 2822 and 3834.
1612 PH/47 Added authenticated_sender_force to the smtp transport.
1614 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1616 PH/49 Installed latest patch from the Sieve maintainer.
1618 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1619 was also set, the appendfile transport was unnecessarily scanning a
1620 directory of message files (e.g. for maildir delivery) to find the count
1621 of files (along with the size), even though it did not need this
1622 information. It now does the scan only if it needs to find either the
1623 size of the count of files.
1625 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1627 PH/52 Two bugs concerned with error handling when the smtp transport is
1630 (i) Exim was not creating retry information for temporary errors given
1631 for individual recipients after the DATA command when the smtp transport
1632 was used in LMTP mode. This meant that they could be retried too
1633 frequently, and not timed out correctly.
1635 (ii) Exim was setting the flag that allows error details to be returned
1636 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1637 recipients that were returned after the DATA command.
1639 PH/53 This is related to PH/52, but is more general: for any failing address,
1640 when detailed error information was permitted to be returned to the
1641 sender, but the error was temporary, then after the final timeout, only
1642 "retry timeout exceeded" was returned. Now it returns the full error as
1643 well as "retry timeout exceeded".
1645 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1646 do this, and (what is worse) MTAs that accept it.
1648 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1649 will now be deprecated.
1651 PH/56 New os.c-cygwin from the Cygwin maintainer.
1653 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1654 formats (previously only available in exim formats via -bpr, -bpru,
1655 and -bpra. Now also available in native and exiqgrep formats)
1657 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1658 with very large, slow to parse queues
1660 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1662 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1664 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1665 responses to authentication challenges, though it was showing the
1666 challenges; (ii) I've removed the CR characters from the debug output for
1669 PH/58 Allow for the insertion of a newline as well as a space when a string
1670 is turned into more than one encoded-word during RFC 2047 encoding. The
1671 Sieve code now uses this.
1673 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1674 data_4xx, lost_connection, tls_required.
1676 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1677 message was being sent as an SMTP response.
1679 PH/61 Add -l and -k options to exicyclog.
1681 PH/62 When verifying, if an address was redirected to one new address, so that
1682 verification continued, and the new address failed or deferred after
1683 having set something in $address_data, the value of $address_data was not
1684 passed back to the ACL. This was different to the case when no
1685 redirection occurred. The value is now passed back in both cases.
1687 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1688 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1689 use login_cap.h, so on its own it isn't the distinguishing feature. The
1690 new name refers directly to the setclassresources() function.
1692 PH/65 Added configuration files for NetBSD3.
1694 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1696 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1697 is preferred over IPv4.
1699 PH/68 The bounce_return_message and bounce_return_body options were not being
1700 honoured for bounces generated during the reception of non-SMTP messages.
1701 In particular, this applied to messages rejected by the ACL. This bug has
1702 been fixed. However, if bounce_return_message is true and bounce_return_
1703 body is false, the headers that are returned for a non-SMTP message
1704 include only those that have been read before the error was detected.
1705 (In the case of an ACL rejection, they have all been read.)
1707 PH/69 The HTML version of the specification is now built in a directory called
1708 spec_html instead of spec.html, because the latter looks like a path with
1709 a MIME-type, and this confuses some software.
1711 PH/70 Catch two compiler warnings in sieve.c.
1713 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1714 function verify_get_ident() calls ip_connect() to connect a socket, but
1715 if the "connect()" function timed out, ip_connect() used to close the
1716 socket. However, verify_get_ident() also closes the socket later, and in
1717 between Exim writes to the log, which may get opened at this point. When
1718 the socket was closed in ip_connect(), the log could get the same file
1719 descriptor number as the socket. This naturally causes chaos. The fix is
1720 not to close the socket in ip_connect(); the socket should be closed by
1721 the function that creates it. There was only one place in the code where
1722 this was missing, in the iplookup router, which I don't think anybody now
1723 uses, but I've fixed it anyway.
1725 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1726 well as to direct DNS lookups. Otherwise the handling of names in host
1727 lists is inconsistent and therefore confusing.
1733 PH/01 Two changes to the default runtime configuration:
1735 (1) Move the checks for relay_from_hosts and authenticated clients from
1736 after to before the (commented out) DNS black list checks.
1738 (2) Add control=submission to the relay_from_hosts and authenticated
1739 clients checks, on the grounds that messages accepted by these
1740 statements are most likely to be submissions.
1742 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1744 (1) Generate an error if the third argument for the ${prvs expansion is
1747 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1750 (3) Reset the variables that are obtained from the first argument of
1751 ${prvscheck and used in the second argument before leaving the code,
1752 because their memory is reclaimed, so using them afterwards may do
1755 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1756 one (it's much easier than Tom thought :-).
1758 (5) Because of (4), we can now allow for the use of $prvscheck_result
1759 inside the third argument.
1761 PH/03 For some reason, the default setting of PATH when running a command from
1762 a pipe transport was just "/usr/bin". I have changed it to
1765 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1766 anything to be listed in the output from -bV.
1768 PH/05 When a filter generated an autoreply, the entire To: header line was
1769 quoted in the delivery log line, like this:
1771 => >A.N.Other <ano@some.domain> <original@ddress> ...
1773 This has been changed so that it extracts the operative address. There
1774 may be more than one such address. If so, they are comma-separated, like
1777 => >ano@some.domain,ona@other.domain <original@ddress> ...
1779 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1780 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1781 not being looked up in the rDNS to get a host name, Exim was showing the
1782 IP address twice in Received: lines, even though the IP addresses were
1783 identical. For example:
1785 Received: from [1.2.3.4] (helo=[1.2.3.4])
1787 However, if the real host name was known, it was omitting the HELO data
1788 if it matched the actual IP address. This has been tidied up so that it
1789 doesn't show the same IP address twice.
1791 PH/07 When both +timestamp and +memory debugging was on, the value given by
1792 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1793 called by the debug printing, thereby overwriting the timestamp buffer.
1794 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1796 PH/08 When the original message was included in an autoreply transport, it
1797 always said "this is a copy of the message, including all the headers",
1798 even if body_only or headers_only was set. It now gives an appropriate
1801 PH/09 Applied a patch from the Sieve maintainer which:
1803 o fixes some comments
1804 o adds the (disabled) notify extension core
1805 o adds some debug output for the result of if/elsif tests
1806 o points to the current vacation draft in the documentation
1807 and documents the missing references header update
1811 o fixes a bug in processing the envelope test (when testing
1812 multiple envelope elements, the last element determinted the
1815 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1816 Electronic Mail") by including:
1818 Auto-submitted: auto-generated
1820 in the messages that it generates (bounce messages and others, such as
1821 warnings). In the case of bounce messages for non-SMTP mesages, there was
1822 also a typo: it was using "Auto_submitted" (underscore instead of
1823 hyphen). Since every message generated by Exim is necessarily in response
1824 to another message, thes have all been changed to:
1826 Auto-Submitted: auto-replied
1828 in accordance with these statements in the RFC:
1830 The auto-replied keyword:
1832 - SHOULD be used on messages sent in direct response to another
1833 message by an automatic process,
1835 - MUST NOT be used on manually-generated messages,
1837 - MAY be used on Delivery Status Notifications (DSNs) and Message
1838 Disposition Notifications (MDNs),
1840 - MUST NOT be used on messages generated by automatic or periodic
1841 processes, except for messages which are automatic responses to
1844 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1845 to the default Received: header definition.
1847 PH/12 Added log selector acl_warn_skipped (default on).
1849 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1850 variables because (a) they are in the wrong storage pool and (b) even if
1851 they were copied, it wouldn't work properly because of the caching.
1853 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1854 checking when decoding. Apparently there are clients that generate
1855 overlong encoded strings. Why am I not surprised?
1857 PH/15 If the first argument of "${if match_address" was not empty, but did not
1858 contain an "@" character, Exim crashed. Now it writes a panic log message
1859 and treats the condition as false.
1861 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1863 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1864 extension "envelope-auth". The code is finished and in agreement with
1865 other implementations, but there is no documentation so far and in fact,
1866 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1867 not changing the active code.
1869 Print executed "if" and "elsif" statements when debugging is used. This
1870 helps a great deal to understand what a filter does.
1872 Document more things not specified clearly in RFC3028. I had all this
1873 sorted out, when out of a sudden new issues came to my mind. Oops."
1875 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1878 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1879 canonical form (as documented). However, after a host name lookup from
1880 the IP address, check_host() was doing a simple string comparison with
1881 addresses acquired from the DNS when checking that the found name did
1882 have the original IP as one of its addresses. Since any found IPv6
1883 addresses are likely to be in abbreviated form, the comparison could
1884 fail. Luckily, there already exists a function for doing the comparison
1885 by converting both addresses to binary, so now that is used instead of
1886 the text comparison.
1888 PH/20 There was another similar case to PH/19, when a complete host name was
1889 given in a host list; looking up its IP address could give an abbreviated
1890 form, whereas the current host's name might or might not be abbreviated.
1891 The same fix has been applied.
1897 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1898 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1901 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1902 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1904 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1906 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1907 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1908 Also, for "Auto-Submitted", treat anything other than "no" as
1909 non-personal, in accordance with RFC 3834. (Previously it treated
1910 anything starting "auto-" as non-personal.)
1912 TF/01 The control=submission/name=... option had a problem with syntax
1913 errors if the name included a slash character. The /name= option
1914 now slurps the rest of the string, so it can include any characters
1915 but it must come last in the list of options (after /sender_retain
1918 PH/05 Some modifications to the interface to the fake nameserver for the new
1926 TK/01 Added the "success_on_redirect" address verification option. See
1927 NewStuff for rationale and an example.
1929 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1931 PH/02 Patch to exigrep to allow it to work on syslog lines.
1933 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1934 fread() to skip over the body file's header line, because in Cygwin the
1935 header line is locked and is inaccessible.
1937 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1938 co-exist for some time) to make it clear that it is the Exim ID that is
1939 referenced, not the Message-ID: header line.
1941 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1942 string_format() function, because snprintf() does not exist on all
1945 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1946 ${stat: expansion item.
1948 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1949 protocol synchronization error", to keep the pedants happy.
1951 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1952 well as for IRIX systems, when gcc is being used. See the host.c source
1955 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1957 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1960 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1961 transport and to the smtp transport in LMTP mode.
1963 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1965 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1966 run a filter in a subprocess. This could lead to confusion in subsequent
1967 lookups in the parent process. There should also be a search_tidyup() at
1968 the end of the subprocess.
1970 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1971 only if the host matched helo_try_verify_hosts, which caused the
1972 verification to occur when the EHLO/HELO command was issued. The ACL just
1973 tested the remembered result. Now, if a previous verification attempt has
1974 not happened, "verify = helo" does it there and then.
1976 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1978 TK/03 Fix log output including CR from clamd.
1980 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1981 exist provoked a memory error which could cause a segfault.
1983 PH/15 Installed PCRE 6.2
1985 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1987 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1988 of the problem. Specifically, suggested +O2 rather than +O1 for the
1991 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1993 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1994 record, the port was not correctly logged when the outgoing_port log
1995 selector was set (it logged the transort's default port).
1997 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1998 fallback_hosts, and "hosts" in the smtp transport.
2000 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2001 host errors such as "Connection refused".
2003 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2004 authentication with radiusclient 0.4.9:
2006 - Error returned from rc_read_config was caught wrongly
2007 - Username/password not passed on to radius server due to wrong length.
2009 The presumption is that some radiusclient API changes for 4.51/PH/17
2010 were not taken care of correctly. The code is still untested by me (my
2011 Linux distribution still has 0.3.2 of radiusclient), but it was
2012 contributed by a Radius user.
2014 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2015 expanding the "port" option of the smtp transport.
2017 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2018 while reading a MIME header. Thanks to Tom Hughes for a patch.
2020 PH/24 Include config.h inside local_scan.h so that configuration settings are
2023 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2024 that in an ACL for RCPT (for example), you can examine exactly what was
2027 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2028 commands, but it was not correctly comparing the address with the actual
2029 client host address. Thus, it would show the EHLO address in Received:
2030 header lines when this was not necessary.
2032 PH/27 Added the % operator to ${eval:}.
2034 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2035 it should be ignoring failures (because with -C, for example, it has lost
2036 privilege). It wasn't ignoring creation failures other than "already
2039 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2042 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2043 because an input line was too long, either on its own, or by virtue of
2044 too many continuations, the temporary file was not being removed, and the
2045 return code was incorrect.
2047 PH/31 Missing "BOOL" in function definition in filtertest.c.
2049 PH/32 Applied Sieve patches from the maintainer.
2051 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2053 PH/33 Added "verify = not_blind".
2055 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2056 Local/Makefile (with some defaults set). These are used in built scripts
2057 such as exicyclog, but they have never been used in the exim_install
2058 script (though there are many overriding facilities there). I have
2059 arranged that the exim_install script now takes note of these two
2062 PH/35 Installed configuration files for Dragonfly.
2064 PH/36 When a locally submitted message by a trusted user did not contain a
2065 From: header, and the sender address was obtained from -f or from an SMTP
2066 MAIL command, and the trusted user did not use -F to supply a sender
2067 name, $originator_name was incorrectly used when constructing a From:
2068 header. Furthermore, $originator_name was used for submission mode
2069 messages from external hosts without From: headers in a similar way,
2070 which is clearly wrong.
2072 PH/37 Added control=suppress_local_fixups.
2074 PH/38 When log_selector = +received_sender was set, and the addition of the
2075 sender made the log line's construction buffer exactly full, or one byte
2076 less than full, an overflow happened when the terminating "\n" was
2079 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2080 when the result of a list match is failure because a DNS lookup failed.
2082 PH/40 RM_COMMAND is now used in the building process.
2084 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2085 the "build-* directories that it finds.
2087 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2088 address in a domain literal was a prefix of an interface address.
2090 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2091 when verifying a sender address, unless rewrite_headers is false.
2093 PH/44 Wrote a long comment about why errors_to addresses are verified as
2094 recipients, not senders.
2096 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2097 the ratelimit ACL was added.
2099 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2101 PH/46 Added extra information about PostgreSQL errors to the error string.
2103 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2104 avoiding the need to install special zones in a real server. This is
2105 backwards compatible; if it can't find the fake resolver, it drops back.
2106 Thus, both old and new test suites can be run.
2108 TF/02 Added util/ratelimit.pl
2110 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2111 clock is set back in time.
2113 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2114 Candler <B.Candler@pobox.com>.
2116 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2117 OK for addresses that are the result of redirections.
2119 PH/48 A number of further additions for the benefit of the new test suite,
2120 including a fake gethostbyname() that interfaces to the fake DNS resolver
2123 TF/06 The fix for widen_domains has also been applied to qualify_single and
2124 search_parents which are the other dnslookup options that can cause
2127 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2130 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2132 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2133 service" if it encountered an MX record with an empty target hostname.
2134 The message is now "an MX or SRV record indicated no SMTP service".
2136 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2137 if the DNS of the sending site is misconfigured. This is quite a
2138 common situation. This change restores the behaviour of treating a
2139 helo verification defer as a failure.
2141 PH/51 If self=fail was set on a router, the bounce message did not include the
2142 actual error message.
2148 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2150 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2151 command itself ended in error, the underlying message about the transport
2152 filter timeout was being overwritten with the pipe command error. Now the
2153 underlying error message should be appended to the second error message.
2155 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2157 PH/02 Reduce the amount of output that "make" produces by default. Full output
2158 can still be requested.
2160 PH/03 The warning log line about a condition test deferring for a "warn" verb
2161 was being output only once per connection, rather than after each
2162 occurrence (because it was using the same function as for successful
2163 "warn" verbs). This seems wrong, so I have changed it.
2165 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2166 it should not have, which might have caused a crash in the right
2167 circumstances, but probably never did.
2169 PH/04 Installed a modified version of Tony Finch's patch to make submission
2170 mode fix the return path as well as the Sender: header line, and to
2171 add a /name= option so that you can make the user's friendly name appear
2174 TF/03 Added the control = fakedefer ACL modifier.
2176 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2177 Mark Lowes for thorough testing.
2179 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2181 TK/03 Merged latest SRS patch from Miles Wilton.
2183 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2184 with the definition in sysexits.h (which is #included earlier).
2185 Fortunately, Exim does not actually use EX_OK. The code used to try to
2186 preserve the sysexits.h value, by assumimg that macro definitions were
2187 scanned for macro replacements. I have been disabused of this notion,
2188 so now the code just undefines EX_OK before #including unistd.h.
2190 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2191 in the smtp transport. When a block could not be written in a single
2192 write() function, the timeout was being re-applied to each part-write.
2193 This seems wrong - if the receiver was accepting one byte at a time it
2194 would take for ever. The timeout is now adjusted when this happens. It
2195 doesn't have to be particularly precise.
2197 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2198 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2200 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2201 without a domain if the check to <postmaster@domain> fails.
2203 SC/01 Eximstats: added -xls and the ability to specify output files
2204 (patch written by Frank Heydlauf).
2206 SC/02 Eximstats: use FileHandles for outputing results.
2208 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2210 SC/04 Eximstats: fixed display of large numbers with -nvr option
2212 SC/05 Eximstats: fixed merging of reports with empty tables.
2214 SC/06 Eximstats: added the -include_original_destination flag
2216 SC/07 Eximstats: removed tabs and trailing whitespace.
2218 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2220 TK/06 MBOX spool code: Add real "From " MBOX separator line
2221 so the .eml file is really in mbox format (even though
2222 most programs do not really care). Patch from Alex Miller.
2224 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2225 The latter is generated from $received_to and is only set if the
2226 message has one envelope recipient. SA can use these headers,
2227 obviously out-of-the-box. Patch from Alex Miller.
2229 PH/08 The ${def test on a variable was returning false if the variable's
2230 value was "0", contrary to what the specification has always said!
2231 The result should be true unless the variable is empty.
2233 PH/09 The syntax error of a character other than { following "${if
2234 def:variable_name" (after optional whitespace) was not being diagnosed.
2235 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2236 accidental colon was present, for example, could give incorrect results.
2238 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2239 result is used (not including appendfile, where other changes are about
2242 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2243 This involved changing a lot of size variables from int to off_t. It
2244 should work with maildirs and everything.
2246 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2247 spamd dying while we are connected to it.
2249 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2252 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2253 function for BDB 4.3.
2255 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2257 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2258 the sources for PCRE have been re-arranged and more files are now
2261 PH/15 The code I had for printing potentially long long variables in PH/11
2262 above was not the best (it lost precision). The length of off_t variables
2263 is now inspected at build time, and an appropriate printing format (%ld
2264 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2265 to be "long long int" or "long int". This is needed for the internal
2266 formatting function string_vformat().
2268 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2269 the configuration file to be ":syslog", then the script "guesses" where
2270 the logs files are, rather than using the compiled in default. In our
2271 case the guess is not the same as the compiled default, so the script
2272 suddenly stopped working when I started to use syslog. The patch checks
2273 to see if log_file_path is "". If so, it attempts to read it from exim
2274 with no configuration file to get the compiled in version, before it
2275 falls back to the previous guessing code."
2277 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2278 implementing BATV in an Exim configuration. See NewStuff for the gory
2281 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2282 Makefile that are specific to HP-UX.
2284 PH/18 If the "use_postmaster" option was set for a recipient callout together
2285 with the "random" option, the postmaster address was used as the MAIL
2286 FROM address for the random test, but not for the subsequent recipient
2287 test. It is now used for both.
2289 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2290 patch removes a few documentation additions to RFC 3028, because the
2291 latest draft now contains them. It adds the new en;ascii-case comparator
2292 and a new error check for 8bit text in MIME parts. Comparator and
2293 require names are now matched exactly. I enabled the subaddress
2294 extension, but it is not well tested yet (read: it works for me)."
2296 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2297 rework some of the code of TK/09 above to avoid the hardwired use of
2298 "%lld" and "long long". Replaced the call to snprintf() with a call to
2301 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2302 records point to non-existent hosts", "retry timeout exceeded", and
2303 "retry time not reached for any host after a long failure period".
2305 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2306 experimental DomainKeys support:
2308 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2309 (2) On an error such as an illegally used "control", the wrong name for
2310 the control was given.
2312 These problems did NOT occur unless DomainKeys support was compiled.
2314 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2316 PH/24 Added ${if match_ip condition.
2318 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2319 databases so that it will be absolutely obvious if a crash occurs in the
2320 DB library. This is a regular occurrence (often caused by mis-matched
2323 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2324 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2325 picked up on a user's system that detects such things. There doesn't seem
2326 to be a gcc warning option for this - only an attribute that has to be
2327 put on the function's prototype. It seems that in Fedora Core 4 they have
2328 set this on a number of new functions. No doubt there will be more in due
2331 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2332 specify a transport. However, if an address that was verified by such a
2333 router was the subject of a callout, Exim crashed because it tried to
2334 read the rcpt_include_affixes from the non-existent transport. Now it
2335 just assumes that the setting of that option is false. This bug was
2336 introduced by 4.51/PH/31.
2338 PH/28 Changed -d+all to exclude +memory, because that information is very
2339 rarely of interest, but it makes the output a lot bigger. People tend to
2340 do -d+all out of habit.
2342 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2343 code in os-type was giving problems when libc.so lives in lib64, like on
2346 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2347 aren't the modern standard, and it seems that some systems' include files
2348 don't always have them. Exim was already checking for some of the newer
2349 ones like T_AAAA, and defining it itself. I've added checks for all the
2350 record types that Exim uses.
2352 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2353 not automatically generating a new one, as it is supposed to. This
2354 prevented TLS from working. If the file did exist, but contained invalid
2355 data, a new version was generated, as expected. It was only the case of a
2356 non-existent file that was broken.
2358 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2359 with a change in libdomainkeys > 0.64.
2361 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2362 from DNS. If the selector record carries the flag, it now has
2363 precedence over the domain-wide flag.
2365 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2367 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2368 the use of a port name that isn't defined in /etc/services) occurred, the
2369 message was deferred as in a normal delivery, and thus remained on the
2370 spool, instead of being failed because of the mua_wrapper setting. This
2371 is now fixed, and I tidied up some of the mua_wrapper messages at the
2374 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2375 the messages in a hash of arrays rather than using individual hashes.
2376 This is a bit cleaner and results in dramatic memory savings, albeit
2377 at a slight CPU cost.
2379 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2380 as requested by Marc Sherman.
2382 SC/10 Eximstats: added histograms for user specified patterns as requested
2385 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2387 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2388 fopen() in the content-scanning modules that did not already have it.
2394 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2395 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2397 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2399 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2401 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2402 to test Sieve filters that use "vacation".
2404 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2405 that changes the way the GnuTLS parameters are stored in the cache file.
2406 The new format can be generated externally. For backward compatibility,
2407 if the data in the cache doesn't make sense, Exim assumes it has read an
2408 old-format file, and it generates new data and writes a new file. This
2409 means that you can't go back to an older release without removing the
2412 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2413 work if there are any delivery delays because "one_time" forces the
2414 parent to be marked "delivered", so its unseen clone is never tried
2415 again. For this reason, Exim now forbids the simultaneous setting of
2418 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2419 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2420 entry if you want to know the details. The fix, however, neglected to
2421 consider the case when local delivery batching is involved. The test for
2422 "previously delivered" was not happening when checking to see if an
2423 address could be batched with a previous (undelivered) one; under
2424 certain circumstances this could lead to multiple deliveries to the same
2427 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2428 in its include files, and this causes problems building Exim.
2430 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2431 header_syntax) but Exim was just ignoring anything given after a slash.
2432 In particular, this caused confusion with an attempt to use "verify =
2433 reverse_host_lookup/defer_ok". An error is now given when options are
2434 supplied for verify items that do not have them. (Maybe reverse_host_
2435 lookup should have a defer_ok option, but that's a different point.)
2437 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2438 defined by RFC 821) to 2048, because there were problems with some AUTH
2439 commands, and RFC 1869 says the size should be increased for extended
2440 SMTP commands that take arguments.
2442 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2445 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2446 "unknown" error; now it says that the functionality isn't in the binary.
2448 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2449 an address' error message when a string expansion fails (syntax or
2450 whatever). Otherwise the password may appear in the log. Following change
2451 PH/42 below, there is no longer a chance of it appearing in a bounce
2454 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2456 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2457 of its addresses was ever tried. (Bugzilla bug #2.)
2459 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2460 the result incorrectly in the debug output. (It correctly added a newline
2461 to what was transported.)
2463 TF/01 Added $received_time.
2465 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2466 commented out examples of how to interface to a virus scanner and to
2467 SpamAssassin. Also added commented examples of av_scanner and
2468 spamd_address settings.
2470 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2471 and controls are allowed in which ACLs. There were a couple of minor
2472 errors. Some of the entries in the conditions table (which is a table of
2473 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2474 negation of where the condition IS allowed.
2476 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2478 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2479 header file does not have a version number, so I've had to invent a new
2480 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2481 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2482 radiusclient), but it was contributed by a Radius user.
2484 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2485 files or directories, for OpenSSL.
2487 PH/19 When an Exim process that is running as root has to create an Exim log
2488 file, it does so in a subprocess that runs as exim:exim so as to get the
2489 ownership right at creation (otherwise, other Exim processes might see
2490 the file with the wrong ownership). There was no test for failure of this
2491 fork() call, which would lead to the process getting stuck as it waited
2492 for a non-existent subprocess. Forks do occasionally fail when resources
2493 run out. I reviewed all the other calls to fork(); they all seem to check
2496 PH/20 When checking for unexpected SMTP input at connect time (before writing
2497 the banner), Exim was not dealing correctly with a non-positive return
2498 from the read() function. If the client had disconnected by this time,
2499 the result was a log entry for a synchronization error with an empty
2500 string after "input=" when read() returned zero. If read() returned -1
2501 (an event I could not check), uninitialized data bytes were printed.
2502 There were reports of junk text (parts of files, etc) appearing after
2505 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2507 PH/22 Added support for macro redefinition, and (re)definition in between
2508 driver and ACL definitions.
2510 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2511 forgetting to use the resulting value; it was using the unexpanded value.
2513 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2514 hadn't been configured. The fix is from Juergen Kreileder, who
2515 understands it better than I do:
2517 "Here's what I see happening with three configured cyrus_sasl
2518 authenticators configured (plain, login, cram-md5):
2520 On startup auth_cyrus_sasl_init() gets called for each of these.
2521 This means three calls to sasl_listmech() without a specified mech_list.
2522 => SASL tests which mechs of all available mechs actually work
2523 => three warnings about OTP not working
2524 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2526 With the patch, sasl_listmech() also gets called three times. But now
2527 SASL's mech_list option is set to the server_mech specified in the the
2528 authenticator. Or in other words, the answer from sasl_listmech()
2529 gets limited to just the mech you're testing for (which is different
2531 => the return list contains just 'plain' or 'login', 'cram-md5' or
2532 nothing depending on the value of ob->server_mech.
2534 I've just tested the patch: Authentication still works fine,
2535 unavailable mechs specified in the exim configuration are still
2536 caught, and the auth.log warnings about OTP are gone."
2538 PH/25 When debugging is enabled, the contents of the command line are added
2539 to the debugging output, even when log_selector=+arguments is not
2542 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2543 answer is "GNU", and only if the return is "GNU/something" is the answer
2546 PH/27 $acl_verify_message is now set immediately after the failure of a
2547 verification in an ACL, and so is available in subsequent modifiers. In
2548 particular, the message can be preserved by coding like this:
2550 warn !verify = sender
2551 set acl_m0 = $acl_verify_message
2553 Previously, $acl_verify_message was set only while expanding "message"
2554 and "log_message" when a very denied access.
2556 PH/28 Modified OS/os.c-Linux with
2558 -#ifndef OS_LOAD_AVERAGE
2559 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2561 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2562 nomenclature these days.)
2564 PH/29 Installed patch from the Sieve maintainer that adds the options
2565 sieve_useraddress and sieve_subaddress to the redirect router.
2567 PH/30 In these circumstances:
2568 . Two addresses routed to the same list of hosts;
2569 . First host does not offer TLS;
2570 . First host accepts first address;
2571 . First host gives temporary error to second address;
2572 . Second host offers TLS and a TLS session is established;
2573 . Second host accepts second address.
2574 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2575 and peerdn, if requested) that were in fact used only for the second
2578 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2579 attention to any local part prefix or suffix that was matched by the
2580 router that accepted the address. It now behaves in the same way as it
2581 does for delivery: the affixes are removed from the local part unless
2582 rcpt_include_affixes is set on the transport.
2584 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2585 timeout during the DATA phase of an incoming message.
2587 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2588 applied a patch sanctioned by the Sieve maintainer.
2590 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2591 the uid or gid is negative. A case of a negative gid caused this to be
2592 noticed. The fix allows for either to be negative.
2594 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2595 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2598 PH/36 The change PH/12 above was broken. Fixed it.
2600 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2601 the grounds that routing the same address twice would always produce the
2602 same answer. This might have been true once, but it is certainly no
2603 longer true now. Routing a child address may depend on the previous
2604 routing that produced that child. Some complicated redirection strategies
2605 went wrong when messages had multiple recipients, and made Exim's
2606 behaviour dependent on the order in which the addresses were given.
2608 I have moved the duplicate checking until after the routing is complete.
2609 Exim scans the addresses that are assigned to local and remote
2610 transports, and removes any duplicates. This means that more work will be
2611 done, as duplicates will always all be routed, but duplicates are
2612 presumably rare, so I don't expect this is of any significance.
2614 For deliveries to pipes, files, and autoreplies, the duplicate checking
2615 still happens during the routing process, since they are not going to be
2618 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2619 It corrects a timeout issue with spamd. This is Ian's comment: "The
2620 background is that sometimes spamd either never reads data from a
2621 connection it has accepted, or it never writes response data. The exiscan
2622 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2623 blindly assumes that writes won't block so it may never time out."
2625 PH/39 Allow G after quota size as well as K and M.
2627 PH/40 The value set for $authenticated_id in an authenticator may not contain
2628 binary zeroes or newlines because the value is written to log lines and
2629 to spool files. There was no check on this. Now the value is run through
2630 the string_printing() function so that such characters are converted to
2631 printable escape sequences.
2633 PH/41 $message_linecount is a new variable that contains the total number of
2634 lines in the message. Compare $body_linecount, which is the count for the
2637 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2638 bounce and delay warning messages, except in certain special cases, which
2641 (a) An SMTP error message from a remote host;
2642 (b) A message specified in a :fail: redirection;
2643 (c) A message specified in a "fail" command in a system filter;
2644 (d) A message specified in a FAIL return from the queryprogram router;
2645 (e) A message specified by the cannot_route_message router option.
2647 In these cases only, Exim does include the error details in bounce and
2648 warning messages. There are also a few cases where bland messages such
2649 as "unrouteable address" or "local delivery error" are given.
2651 PH/43 $value is now also set for the "else" part of a ${run expansion.
2653 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2654 being worked on, but at least Exim now implements the latest version to
2657 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2658 process to complete was treated as a delivery failure, a timeout while
2659 writing the message to the pipe was logged, but erroneously treated as a
2660 successful delivery. Such timeouts include transport filter timeouts. For
2661 consistency with the overall process timeout, these timeouts are now
2662 treated as errors, giving rise to delivery failures by default. However,
2663 there is now a new Boolean option for the pipe transport called
2664 timeout_defer, which, if set TRUE, converts the failures into defers for
2665 both kinds of timeout. A transport filter timeout is now identified in
2668 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2669 systems where "make" and "gmake" are different, calling "gmake" at top
2670 level broke things. I've arranged for the value of $(MAKE) to be passed
2671 from the Makefile to this script so that it can call the same version of
2675 A note about Exim versions 4.44 and 4.50
2676 ----------------------------------------
2678 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2679 changes of various kinds. As a consequence, a big documentation update was
2680 needed. This delayed the release for rather longer than seemed good, especially
2681 in the light of a couple of (minor) security issues. Therefore, the changes
2682 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2683 release. So 4.44 and 4.50 are in effect two different branches that both start
2686 I have left the 4.50 change log unchanged; it contains all the changes since
2687 4.43. The change log for 4.44 is below; many of its items are identical to
2688 those for 4.50. This seems to be the most sensible way to preserve the
2689 historical information.
2695 1. Minor wording change to the doc/README.SIEVE file.
2697 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2698 computation of the current number of files was incorrect.
2700 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2701 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2704 4. Give more explanation in the error message when the command for a transport
2705 filter fails to execute.
2707 5. There are several places where Exim runs a non-Exim command in a
2708 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2709 was being done only for the command run by the queryprogram router. It is
2710 now done for all such subprocesses. The other cases are: ${run, transport
2711 filters, and the commands run by the lmtp and pipe transports.
2713 6. Added CONFIGURE_GROUP build-time option.
2715 7. Some older OS have a limit of 256 on the maximum number of file
2716 descriptors. Exim was using setrlimit() to set 1000 as a large value
2717 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2718 systems. I've change it so that if it can't get 1000, it tries for 256.
2720 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2721 was an oversight, and furthermore, ever since the addition of extra
2722 controls (e.g. 4.43/32), the checks on when to allow different forms of
2723 "control" were broken. There should now be diagnostics for all cases when a
2724 control that does not make sense is encountered.
2726 9. Added the /retain_sender option to "control=submission".
2728 10. $recipients is now available in the predata ACL (oversight).
2730 11. Tidy the search cache before the fork to do a delivery from a message
2731 received from the command line. Otherwise the child will trigger a lookup
2732 failure and thereby defer the delivery if it tries to use (for example) a
2733 cached ldap connection that the parent has called unbind on.
2735 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2736 of $address_data from the recipient verification was clobbered by the
2737 sender verification.
2739 13. The value of address_data from a sender verification is now available in
2740 $sender_address_data in subsequent conditions in the ACL statement.
2742 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2744 15. Added a new option "connect=<time>" to callout options, to set a different
2747 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2748 was its contents. (It was OK if the option was not defined at all.)
2750 17. A "Completed" log line is now written for messages that are removed from
2751 the spool by the -Mrm option.
2753 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2754 information about exactly what failed.
2756 19. Added -dd to debug only the daemon process.
2758 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2759 handles timeouts, both on the server side and network timeouts. Renamed the
2760 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2762 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2763 It is now set to "smtps".
2765 22. $host_address is now set to the target address during the checking of
2766 ignore_target_hosts.
2768 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2769 being passed; this would have caused $sender_host_name to have been used if
2770 matching the list had actually called for a host name (not very likely,
2771 since this list is usually IP addresses). A host name is now passed as
2774 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2775 code that specifies a non-restarting handler (typically sigaction() in
2776 modern systems) in an attempt to fix a rare and obscure crash bug.
2778 25. Narrowed the window for a race in the daemon that could cause it to ignore
2779 SIGCHLD signals. This is not a major problem, because they are used only to
2780 wake it up if nothing else does.
2782 26. A malformed maildirsize file could cause Exim to calculate negative values
2783 for the mailbox size or file count. Odd effects could occur as a result.
2784 The maildirsize information is now recalculated if the size or filecount
2787 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2788 support for a long time. Removed HAVE_SYS_VFS_H.
2790 28. Installed the latest version of exipick from John Jetmore.
2792 29. In an address list, if the pattern was not a regular expression, an empty
2793 subject address (from a bounce message) matched only if the pattern was an
2794 empty string. Non-empty patterns were not even tested. This was the wrong
2795 because it is perfectly reasonable to use an empty address as part of a
2796 database query. An empty address is now tested by patterns that are
2797 lookups. However, all the other forms of pattern expect the subject to
2798 contain a local part and a domain, and therefore, for them, an empty
2799 address still always fails if the pattern is not itself empty.
2801 30. Exim went into a mad DNS loop when attempting to do a callout where the
2802 host was specified on an smtp transport, and looking it up yielded more
2803 than one IP address.
2805 31. Re-factored the code for checking spool and log partition space into a
2806 function that finds that data and another that does the check. The former
2807 is then used to implement four new variables: $spool_space, $log_space,
2808 $spool_inodes, and $log_inodes.
2810 32. The RFC2047 encoding function was originally intended for short strings
2811 such as real names; it was not keeping to the 75-character limit for
2812 encoded words that the RFC imposes. It now respects the limit, and
2813 generates multiple encoded words if necessary. To be on the safe side, I
2814 have increased the buffer size for the ${rfc2047: expansion operator from
2817 33. It is now permitted to omit both strings after an "if" condition; if the
2818 condition is true, the result is "true". As before, when the second string
2819 is omitted, a false condition yields an empty string. This makes it less
2820 cumbersome to write custom ACL and router conditions.
2822 34. Failure to deliver a bounce message always caused it to be frozen, even if
2823 there was an errors_to setting on the router. The errors_to setting is now
2826 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2827 canonical form (fully expanded) before being placed in
2828 $sender_host_address.
2830 36. The table in the code that translates DNS record types into text (T_A to
2831 "A" for instance) was missing entries for NS and CNAME. It is just possible
2832 that this could have caused confusion if both these types were looked up
2833 for the same domain, because the text type is used as part of Exim's
2834 per-process caching. But the chance of anyone hitting this buglet seems
2837 37. The dnsdb lookup has been extended in a number of ways.
2839 (1) There is a new type, "zns", which walks up the domain tree until it
2840 finds some nameserver records. It should be used with care.
2842 (2) There is a new type, "mxh", which is like "mx" except that it returns
2843 just the host names, not the priorities.
2845 (3) It is now possible to give a list of domains (or IP addresses) to be
2846 looked up. The behaviour when one of the lookups defers can be
2847 controlled by a keyword.
2849 (4) It is now possible to specify the separator character for use when
2850 multiple records are returned.
2852 38. The dnslists ACL condition has been extended: it is now possible to supply
2853 a list of IP addresses and/or domains to be looked up in a particular DNS
2856 39. Added log_selector=+queue_time_overall.
2858 40. When running the queue in the test harness, wait just a tad after forking a
2859 delivery process, to get repeatability of debugging output.
2861 41. Include certificate and key file names in error message when GnuTLS fails
2862 to set them up, because the GnuTLS error message doesn't include the name
2863 of the failing file when there is a problem reading it.
2865 42. Allow both -bf and -bF in the same test run.
2867 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2869 44. The "Exiscan patch" is now merged into the mainline Exim source.
2871 45. Sometimes the final signoff response after QUIT could fail to get
2872 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2873 < 0 before doing a fflush(). This bug looks as though it goes back to the
2874 introduction of TLS in release 3.20, but "sometimes" must have been rare
2875 because the tests only now provoked it.
2877 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2878 (this can affect the format of dates).
2880 47. exim_tidydb, when checking for the continued existence of a message for
2881 which it has found a message-specific retry record, was not finding
2882 messages that were in split spool directories. Consequently, it was
2883 deleting retry records that should have stayed in existence.
2885 48. Steve fixed some bugs in eximstats.
2887 49. The SPA authentication driver was not abandoning authentication and moving
2888 on to the next authenticator when an expansion was forced to fail,
2889 contradicting the general specification for all authenticators. Instead it
2890 was generating a temporary error. It now behaves as specified.
2892 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2893 (the order specifies the preference for clients). The order is now AES256,
2894 AES128, 3DES, ARCFOUR128.
2896 51. Small patch to Sieve code - explicitly set From: when generating an
2899 52. Exim crashed if a remote delivery caused a very long error message to be
2900 recorded - for instance if somebody sent an entire SpamAssassin report back
2901 as a large number of 550 error lines. This bug was coincidentally fixed by
2902 increasing the size of one of Exim's internal buffers (big_buffer) that
2903 happened as part of the Exiscan merge. However, to be on the safe side, I
2904 have made the code more robust (and fixed the comments that describe what
2907 53. Now that there can be additional text after "Completed" in log lines (if
2908 the queue_time_overall log selector is set), a one-byte patch to exigrep
2909 was needed to allow it to recognize "Completed" as not the last thing in
2912 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2913 patch that reportedly fixes this has been added. I am not expert enough to
2914 create a test for it. This is what the patch creator wrote:
2916 "I found a little strange behaviour of ldap code when working with
2917 Windows 2003 AD Domain, where users was placed in more than one
2918 Organization Units. When I tried to give exim partial DN, the exit code
2919 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2920 But simultaneously result of request was absolutely normal ldap result,
2921 so I produce this patch..."
2923 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2924 REFERENCE, so I have modified the code to exclude the patch when that macro
2927 55. Some experimental protocols are using DNS PTR records for new purposes. The
2928 keys for these records are domain names, not reversed IP addresses. The
2929 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2930 leaves it alone. Component reversal etc. now happens only for IP addresses.
2933 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2935 57. Double the size of the debug message buffer (to 2048) so that more of very
2936 long debug lines gets shown.
2938 58. The exicyclog utility now does better if the number of log files to keep
2939 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2941 59. Two changes related to the smtp_active_hostname option:
2943 (1) $smtp_active_hostname is now available as a variable.
2944 (2) The default for smtp_banner uses $smtp_active_hostname instead
2945 of $primary_hostname.
2947 60. The host_aton() function is supposed to be passed a string that is known
2948 to be a valid IP address. However, in the case of IPv6 addresses, it was
2949 not checking this. This is a hostage to fortune. Exim now panics and dies
2950 if the condition is not met. A case was found where this could be provoked
2951 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2952 components; fortuitously, this particular loophole had already been fixed
2953 by change 4.50/55 above.
2955 If there are any other similar loopholes, the new check in host_aton()
2956 itself should stop them being exploited. The report I received stated that
2957 data on the command line could provoke the exploit when Exim was running as
2958 exim, but did not say which command line option was involved. All I could
2959 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2960 running as the user.
2963 61. There was a buffer overflow vulnerability in the SPA authentication code
2964 (which came originally from the Samba project). I have added a test to the
2965 spa_base64_to_bits() function which I hope fixes it.
2968 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2969 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2971 63. The daemon start-up calls getloadavg() while still root for those OS that
2972 need the first call to be done as root, but it missed one case: when
2973 deliver_queue_load_max is set with deliver_drop_privilege. This is
2974 necessary for the benefit of the queue runner, because there is no re-exec
2975 when deliver_drop_privilege is set.
2977 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2978 This has been fixed.
2980 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2981 was in use, was not putting the data itself into the right store pool;
2982 consequently, it could be overwritten for a subsequent message in the same
2983 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2986 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2988 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2989 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2990 the function were treating the return as a boolean value, which happened to
2991 work because 0=false and not-0=true, but is not correct code.
2993 68. The host_aton() function was not handling scoped IPv6 addresses (those
2994 with, for example, "%eth0" on the end) correctly.
2996 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2997 negated items (that is, ~something) in unsigned ints. Some compilers
2998 apparently mutter when there is no cast.
3000 70. If an address verification called from an ACL failed, and did not produce a
3001 user-specific message (i.e. there was only a "system" message), nothing was
3002 put in $acl_verify_message. In this situation, it now puts the system
3005 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3006 session; change 4.31/43 added the unwanted input to the log line - except
3007 that it did not do this in the start of session case. It now does.
3009 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3010 This is wrong and can cause the other end to generate a synchronization
3011 error if it is another Exim or anything else that does the synchronization
3012 check. A QUIT command is no longer sent after a timeout.
3014 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3015 during host lookups.
3017 74. The defer_ok option of callout verification was not working if it was used
3018 when verifying addresses in header lines, that is, for this case:
3020 verify = header_sender/callout=defer_ok
3022 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3023 those file descriptors could be used for SMTP connections. If anything
3024 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3025 could be sent to the SMTP client, causing chaos. The daemon now opens
3026 stdin, stdout, and stderr to /dev/null when it puts itself into the
3029 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3030 log by default. The user can override this with suitable Perl magic.
3032 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3033 the log message when discard triggers, was not working for the DATA ACL or
3034 for the non-SMTP ACL.
3036 78. Error message wording change in sieve.c.
3038 79. If smtp_accept_max_per_host was set, the number of connections could be
3039 restricted to fewer than expected, because the daemon was trying to set up
3040 a new connection before checking whether the processes handling previous
3041 connections had finished. The check for completed processes is now done
3042 earlier. On busy systems, this bug wouldn't be noticed because something
3043 else would have woken the daemon, and it would have reaped the completed
3046 80. If a message was submitted locally by a user whose login name contained one
3047 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3048 It caused a spool format error. I have fixed the spool reading code. A
3049 related problem was that the "from" clause in the Received: line became
3050 illegal because of the space(s). It is now covered by ${quote_local_part.
3052 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3055 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3057 83. Patch from Sieve maintainer to fix unterminated string problem in
3058 "vacation" handling.
3060 84. Some minor changes to the Linux configuration files to help with other
3061 OS variants using glibc.
3063 85. One more patch for Sieve to update vacation handling to latest spec.
3066 ----------------------------------------------------
3067 See the note above about the 4.44 and 4.50 releases.
3068 ----------------------------------------------------
3074 1. Change 4.43/35 introduced a bug that caused file counts to be
3075 incorrectly computed when quota_filecount was set in an appendfile
3078 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3079 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3082 3. Give more explanation in the error message when the command for a transport
3083 filter fails to execute.
3085 4. There are several places where Exim runs a non-Exim command in a
3086 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3087 was being done only for the command run by the queryprogram router. It is
3088 now done for all such subprocesses. The other cases are: ${run, transport
3089 filters, and the commands run by the lmtp and pipe transports.
3091 5. Some older OS have a limit of 256 on the maximum number of file
3092 descriptors. Exim was using setrlimit() to set 1000 as a large value
3093 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3094 systems. I've change it so that if it can't get 1000, it tries for 256.
3096 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3097 was an oversight, and furthermore, ever since the addition of extra
3098 controls (e.g. 4.43/32), the checks on when to allow different forms of
3099 "control" were broken. There should now be diagnostics for all cases when a
3100 control that does not make sense is encountered.
3102 7. $recipients is now available in the predata ACL (oversight).
3104 8. Tidy the search cache before the fork to do a delivery from a message
3105 received from the command line. Otherwise the child will trigger a lookup
3106 failure and thereby defer the delivery if it tries to use (for example) a
3107 cached ldap connection that the parent has called unbind on.
3109 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3110 of $address_data from the recipient verification was clobbered by the
3111 sender verification.
3113 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3114 was its contents. (It was OK if the option was not defined at all.)
3116 11. A "Completed" log line is now written for messages that are removed from
3117 the spool by the -Mrm option.
3119 12. $host_address is now set to the target address during the checking of
3120 ignore_target_hosts.
3122 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3123 being passed; this would have caused $sender_host_name to have been used if
3124 matching the list had actually called for a host name (not very likely,
3125 since this list is usually IP addresses). A host name is now passed as
3128 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3129 code that specifies a non-restarting handler (typically sigaction() in
3130 modern systems) in an attempt to fix a rare and obscure crash bug.
3132 15. Narrowed the window for a race in the daemon that could cause it to ignore
3133 SIGCHLD signals. This is not a major problem, because they are used only to
3134 wake it up if nothing else does.
3136 16. A malformed maildirsize file could cause Exim to calculate negative values
3137 for the mailbox size or file count. Odd effects could occur as a result.
3138 The maildirsize information is now recalculated if the size or filecount
3141 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3142 support for a long time. Removed HAVE_SYS_VFS_H.
3144 18. Updated exipick to current release from John Jetmore.
3146 19. Allow an empty sender to be matched against a lookup in an address list.
3147 Previously the only cases considered were a regular expression, or an
3150 20. Exim went into a mad DNS lookup loop when doing a callout where the
3151 host was specified on the transport, if the DNS lookup yielded more than
3154 21. The RFC2047 encoding function was originally intended for short strings
3155 such as real names; it was not keeping to the 75-character limit for
3156 encoded words that the RFC imposes. It now respects the limit, and
3157 generates multiple encoded words if necessary. To be on the safe side, I
3158 have increased the buffer size for the ${rfc2047: expansion operator from
3161 22. Failure to deliver a bounce message always caused it to be frozen, even if
3162 there was an errors_to setting on the router. The errors_to setting is now
3165 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3166 canonical form (fully expanded) before being placed in
3167 $sender_host_address.
3169 24. Updated eximstats to version 1.33
3171 25. Include certificate and key file names in error message when GnuTLS fails
3172 to set them up, because the GnuTLS error message doesn't include the name
3173 of the failing file when there is a problem reading it.
3175 26. Expand error message when OpenSSL has problems setting up cert/key files.
3178 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3179 (this can affect the format of dates).
3181 28. exim_tidydb, when checking for the continued existence of a message for
3182 which it has found a message-specific retry record, was not finding
3183 messages that were in split spool directories. Consequently, it was
3184 deleting retry records that should have stayed in existence.
3186 29. eximstats updated to version 1.35
3187 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3188 1.35 - bugfix such that pie charts by volume are generated correctly
3190 30. The SPA authentication driver was not abandoning authentication and moving
3191 on to the next authenticator when an expansion was forced to fail,
3192 contradicting the general specification for all authenticators. Instead it
3193 was generating a temporary error. It now behaves as specified.
3195 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3196 (the order specifies the preference for clients). The order is now AES256,
3197 AES128, 3DES, ARCFOUR128.
3199 31. Small patch to Sieve code - explicitly set From: when generating an
3202 32. Exim crashed if a remote delivery caused a very long error message to be
3203 recorded - for instance if somebody sent an entire SpamAssassin report back
3204 as a large number of 550 error lines. This bug was coincidentally fixed by
3205 increasing the size of one of Exim's internal buffers (big_buffer) that
3206 happened as part of the Exiscan merge. However, to be on the safe side, I
3207 have made the code more robust (and fixed the comments that describe what
3210 33. Some experimental protocols are using DNS PTR records for new purposes. The
3211 keys for these records are domain names, not reversed IP addresses. The
3212 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3213 leaves it alone. Component reversal etc. now happens only for IP addresses.
3216 34. The host_aton() function is supposed to be passed a string that is known
3217 to be a valid IP address. However, in the case of IPv6 addresses, it was
3218 not checking this. This is a hostage to fortune. Exim now panics and dies
3219 if the condition is not met. A case was found where this could be provoked
3220 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3221 components; fortuitously, this particular loophole had already been fixed
3222 by change 4.50/55 or 4.44/33 above.
3224 If there are any other similar loopholes, the new check in host_aton()
3225 itself should stop them being exploited. The report I received stated that
3226 data on the command line could provoke the exploit when Exim was running as
3227 exim, but did not say which command line option was involved. All I could
3228 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3229 running as the user.
3232 35. There was a buffer overflow vulnerability in the SPA authentication code
3233 (which came originally from the Samba project). I have added a test to the
3234 spa_base64_to_bits() function which I hope fixes it.
3237 36. The daemon start-up calls getloadavg() while still root for those OS that
3238 need the first call to be done as root, but it missed one case: when
3239 deliver_queue_load_max is set with deliver_drop_privilege. This is
3240 necessary for the benefit of the queue runner, because there is no re-exec
3241 when deliver_drop_privilege is set.
3243 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3244 was in use, was not putting the data itself into the right store pool;
3245 consequently, it could be overwritten for a subsequent message in the same
3246 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3249 38. Sometimes the final signoff response after QUIT could fail to get
3250 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3251 < 0 before doing a fflush(). This bug looks as though it goes back to the
3252 introduction of TLS in release 3.20, but "sometimes" must have been rare
3253 because the tests only now provoked it.
3259 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3260 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3261 Now it returns an integer. A number of calls were still expecting a T/F
3262 return. Fortuitously, in all cases, the tests worked in OK situations,
3263 which is the norm. However, things would have gone wrong on any write
3264 failures on the smtp file descriptor. This function is used when sending
3265 messages over SMTP and also when doing verify callouts.
3267 2. When Exim is called to do synchronous delivery of a locally submitted
3268 message (the -odf or -odi options), it no longer closes stderr before doing
3271 3. Implemented the mua_wrapper option.
3273 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3275 5. Implemented the functions header_remove(), header_testname(),
3276 header_add_at_position(), and receive_remove_recipient(), and exported them
3279 6. If an ACL "warn" statement specified the addition of headers, Exim already
3280 inserted X-ACL-Warn: at the start if there was no header name. However, it
3281 was not making this test for the second and subsequent header lines if
3282 there were newlines in the string. This meant that an invalid header could
3283 be inserted if Exim was badly configured.
3285 7. Allow an ACL "warn" statement to add header lines at the start or after all
3286 the Received: headers, as well as at the end.
3288 8. Added the rcpt_4xx retry error code.
3290 9. Added postmaster_mailfrom=xxx to callout verification option.
3292 10. Added mailfrom=xxxx to the callout verification option, for verify=
3295 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3296 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3298 12. Inserted some casts to stop certain compilers warning when using pointer
3299 differences as field lengths or precisions in printf-type calls (mostly
3300 affecting debugging statements).
3302 13. Added optional readline() support for -be (dynamically loaded).
3304 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3305 same clock tick as a message's arrival, so that its received time was the
3306 same as the "first fail" time on the retry record, and that message
3307 remained on the queue past the ultimate address timeout, every queue runner
3308 would try a delivery (because it was past the ultimate address timeout) but
3309 after another failure, the ultimate address timeout, which should have then
3310 bounced the address, did not kick in. This was a "< instead of <=" error;
3311 in most cases the first failure would have been in the next clock tick
3312 after the received time, and all would be well.
3314 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3315 being recognized when the domain list was tested by the match_domain
3316 condition in an expansion string.
3318 16. Added the ${str2b64: operator.
3320 17. Exim was always calling setrlimit() to set a large limit for the number of
3321 processes, without checking whether the existing limit was already
3322 adequate. (It did check for the limit on file descriptors.) Furthermore,
3323 errors from getrlimit() and setrlimit() were being ignored. Now they are
3324 logged to the main and panic logs, but Exim does carry on, to try to do its
3325 job under whatever limits there are.
3327 18. Imported PCRE 5.0.
3329 19. Trivial typo in log message " temporarily refused connection" (the leading
3332 20. If the log selector return_path_on_delivery was set and an address was
3333 redirected to /dev/null, the delivery process crashed because it assumed
3334 that a return path would always be set for a "successful" delivery. In this
3335 case, the whole delivery is bypassed as an optimization, and therefore no
3338 21. Internal re-arrangement: the function for sending a challenge and reading
3339 a response while authentication was assuming a zero-terminated challenge
3340 string. It's now changed to take a pointer and a length, to allow for
3341 binary data in such strings.
3343 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3345 23. Exim was not respecting finduser_retries when seeking the login of the
3346 uid under which it was called; it was always trying 10 times. (The default
3347 setting of finduser_retries is zero.) Also, it was sleeping after the final
3348 failure, which is pointless.
3350 24. Implemented tls_on_connect_ports.
3352 25. Implemented acl_smtp_predata.
3354 26. If the domain in control=submission is set empty, Exim assumes that the
3355 authenticated id is a complete email address when it generates From: or
3356 Sender: header lines.
3358 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3359 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3360 chown and chgrp in /bin and hostname in /usr/bin.
3362 28. Exim was keeping the "process log" file open after each use, just as it
3363 does for the main log. This opens the possibility of it remaining open for
3364 long periods when the USR1 signal hits a daemon. Occasional processlog
3365 errors were reported, that could have been caused by this. Anyway, it seems
3366 much more sensible not to leave this file open at all, so that is what now
3369 29. The long-running daemon process does not normally write to the log once it
3370 has entered its main loop, and it closes the log before doing so. This is
3371 so that log files can straightforwardly be renamed and moved. However,
3372 there are a couple of unusual error situations where the daemon does write
3373 log entries, and I had neglected to close the log afterwards.
3375 30. The text of an SMTP error response that was received during a remote
3376 delivery was being truncated at 512 bytes. This is too short for some of
3377 the long messages that one sometimes sees. I've increased the limit to
3380 31. It is now possible to make retry rules that apply only when a message has a
3381 specific sender, in particular, an empty sender.
3383 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3384 it possible to be selective about when SMTP synchronization is enforced.
3386 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3388 32. Implemented hosts_connection_nolog.
3390 33. Added an ACL for QUIT.
3392 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3395 35. Added mailbox_size and mailbox_filecount to appendfile.
3397 36. Added control = no_multiline_responses to ACLs.
3399 37. There was a bug in the logic of the code that waits for the clock to tick
3400 in the case where the clock went backwards by a substantial amount such
3401 that the microsecond fraction of "now" was more than the microsecond
3402 fraction of "then" (but the whole seconds number was less).
3404 38. Added support for the libradius Radius client library this is found on
3405 FreeBSD (previously only the radiusclient library was supported).
3411 1. When certain lookups returned multiple values in the form name=value, the
3412 quoting of the values was not always being done properly. Specifically:
3413 (a) If the value started with a double quote, but contained no whitespace,
3415 (b) If the value contained whitespace other than a space character (i.e.
3416 tabs or newlines or carriage returns) it was not quoted.
3417 This fix has been applied to the mysql and pgsql lookups by writing a
3418 separate quoting function and calling it from the lookup code. The fix
3419 should probably also be applied to nisplus, ibase and oracle lookups, but
3420 since I cannot test any of those, I have not disturbed their existing code.
3422 2. A hit in the callout cache for a specific address caused a log line with no
3423 reason for rejecting RCPT. Now it says "Previous (cached) callout
3424 verification failure".
3426 3. There was an off-by-one bug in the queryprogram router. An over-long
3427 return line was truncated at 256 instead of 255 characters, thereby
3428 overflowing its buffer with the terminating zero. As well as fixing this, I
3429 have increased the buffer size to 1024 (and made a note to document this).
3431 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3432 when Exim is waiting for an SMTP response from a remote server, Exim
3433 restarts its select() call on the socket, thereby resetting its timeout.
3434 This is not a problem when such interrupts are rare. Somebody set up a cron
3435 job to run exiwhat every 2 minutes, which is less than the normal select()
3436 timeout (5 or 10 minutes). This meant that the select() timeout never
3437 kicked in because it was always reset. I have fixed this by comparing the
3438 time when an interrupt arrives with the time at the start of the first call
3439 to select(). If more time than the timeout has elapsed, the interrupt is
3440 treated as a timeout.
3442 5. Some internal re-factoring in preparation for the addition of Sieve
3443 extensions (by MH). In particular, the "personal" test is moved to a
3444 separate function, and given an option for scanning Cc: and Bcc: (which is
3445 not set for Exim filters).
3447 6. When Exim created an email address using the login of the caller as the
3448 local part (e.g. when creating a From: or Sender: header line), it was not
3449 quoting the local part when it contained special characters such as @.
3451 7. Installed new OpenBSD configuration files.
3453 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3454 try to make them clearer.
3456 9. Callout options, other than the timeout value, were being ignored when
3457 verifying sender addresses in header lines. For example, when using
3459 verify = header_sender/callout=no_cache
3461 the cache was (incorrectly) being used.
3463 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3464 only the Cygwin environment.
3466 11. When return_path_on_delivery was set as a log selector, if different remote
3467 addresses in the same message used different return paths and parallel
3468 remote delivery occurred, the wrong values would sometimes be logged.
3469 (Whenever a remote delivery process finished, the return path value from
3470 the most recently started remote delivery process was logged.)
3472 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3473 lines when AUTH and/or TLS are in use. This is the "received protocol"
3474 field. Exim used to use "asmtp" for authenticated SMTP, without any
3475 indication (in the protocol name) for TLS use. Now it follows the RFC and
3476 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3477 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3478 names appear in log lines as well as in Received: header lines.
3480 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3481 extensions, and comparison tests, and to fix some bugs.
3483 14. Changes to the "personal" filter test:
3485 (1) The test was buggy in that it was just doing the equivalent of
3486 "contains" tests on header lines. For example, if a user's address was
3487 anne@some.where, the "personal" test would incorrectly be true for
3489 To: susanne@some.where
3491 This test is now done by extracting each address from the header in turn,
3492 and checking the entire address. Other tests that are part of "personal"
3493 are now done using regular expressions (for example, to check local parts
3494 of addresses in From: header lines).
3496 (2) The list of non-personal local parts in From: addresses has been
3497 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3498 taken from the Sieve specification recommendations.
3500 (3) If the message contains any header line starting with "List-" it is
3501 treated as non-personal.
3503 (4) The test for "circular" in the Subject: header line has been removed
3504 because it now seems ill-conceived.
3506 15. Minor typos in src/EDITME comments corrected.
3508 16. Installed latest exipick from John Jetmore.
3510 17. If headers_add on a router specified a text string that was too long for
3511 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3512 of string_sprintf() is now avoided.
3514 18. $message_body_size was not set (it was always zero) when running the DATA
3515 ACL and the local_scan() function.
3517 19. For the "mail" command in an Exim filter, no default was being set for
3518 the once_repeat time, causing a random time value to be used if "once" was
3519 specified. (If the value happened to be <= 0, no repeat happened.) The
3520 default is now 0s, meaning "never repeat". The "vacation" command was OK
3521 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3522 (I found it when inspecting the code).
3524 20. There is now an overall timeout for performing a callout verification. It
3525 defaults to 4 times the callout timeout, which applies to individual SMTP
3526 commands during the callout. The overall timeout applies when there is more
3527 than one host that can be tried. The timeout is checked before trying the
3528 next host. This prevents very long delays if there are a large number of
3529 hosts and all are timing out (e.g. when the network connections are timing
3530 out). The value of the overall timeout can be changed by specifying an
3531 additional sub-option for "callout", called "maxwait". For example:
3533 verify = sender/callout=5s,maxwait=20s
3535 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3536 to the end before writing, but this should make it even safer).
3538 22. Exim was forgetting that it had advertised PIPELINING for the second and
3539 subsequent messages on an SMTP connection. It was also not resetting its
3540 memory on STARTTLS and an internal HELO.
3542 23. When Exim logs an SMTP synchronization error within a session, it now
3543 records whether PIPELINING has been advertised or not.
3545 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3546 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3547 rather than long int.
3549 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3551 26. Added the never_mail option to autoreply.
3557 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3558 crash if the getsockname() call failed; this can happen if a connection is
3559 closed very soon after it is established. The problem was simply in the
3560 order in which certain operations were done, causing Exim to try to write
3561 to the SMTP stream before it had set up the file descriptor. The bug has
3562 been fixed by making things happen in the correct order.
3568 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3569 before the connection was closed, thus losing the rejection response.
3571 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3572 some early Solaris releases, but causes trouble in current releases where
3573 socklen_t is defined.
3575 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3578 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3581 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3582 that releasing the top of it at the end releases what was used for sub-
3583 expansions (unless the block got too big). However, discard this block if
3584 the first thing is a variable or header, so that we can use its block when
3585 it is dynamic (useful for very large $message_headers, for example).
3587 6. Lookups now cache *every* query, not just the most recent. A new, separate
3588 store pool is used for this. It can be recovered when all lookup caches are
3589 flushed. Lookups now release memory at the end of their result strings.
3590 This has involved some general refactoring of the lookup sources.
3592 7. Some code has been added to the store_xxx() functions to reduce the amount
3593 of flapping under certain conditions.
3595 8. log_incoming_interface used to affect only the <= reception log lines. Now
3596 it causes the local interface and port to be added to several more SMTP log
3597 lines, for example "SMTP connection from", and rejection lines.
3599 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3601 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3603 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3604 could be overwritten at the end of the current message (or the start of a
3605 new message if it was set in a HELO ACL). The value is now preserved for
3606 the duration of the SMTP connection.
3608 12. If a transport had a headers_rewrite setting, and a matching header line
3609 contained an unqualified address, that address was qualified, even if it
3610 did not match any rewriting rules. The underlying bug was that the values
3611 of the flags that permit the existence of unqualified sender and recipient
3612 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3613 non-local messages, and by -bnq for local messages) were not being
3614 preserved with the message after it was received.
3616 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3617 "next input=" as part of the text comprising the host identity instead of
3618 the correct text. The code was using the same buffer for two different
3619 strings. However, depending on which order the printing function evaluated
3620 its arguments, the bug did not always show up. Under Linux, for example, my
3621 test suite worked just fine.
3623 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3624 doesn't work with some older versions of Perl. It has been changed to "my",
3625 which in any case is probably the better facility to use.
3627 15. A really picky compiler found some instances of statements for creating
3628 error messages that either had too many or two few arguments for the format
3631 16. The size of the buffer for calls to the DNS resolver has been increased
3632 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3633 for addresses that have a lot of PTR records. This alleviates a problem; it
3634 does not fully solve it.
3636 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3637 buffer now truncates the list and logs the incident, which is the same
3638 action as happens when Exim is looking up a host name and its aliases.
3639 Previously in this situation something unpredictable would happen;
3640 sometimes it was "internal error: store_reset failed".
3642 18. If a server dropped the connection unexpectedly when an Exim client was
3643 using GnuTLS and trying to read a response, the client delivery process
3644 crashed while trying to generate an error log message.
3646 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3647 string, for example:
3649 warn message = H1: something\nH2: something
3651 the text was added as a single header line from Exim's point of view
3652 though it ended up OK in the delivered message. However, searching for the
3653 second and subsequent header lines using $h_h2: did not work. This has been
3654 fixed. Similarly, if a system filter added multiple headers in this way,
3655 the routers could not see them.
3657 20. Expanded the error message when iplsearch is called with an invalid key to
3658 suggest using net-iplsearch in a host list.
3660 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3661 ACLs are no longer actually imposed (and a message to that effect is
3664 22. If a "gecos" field in a passwd entry contained escaped characters, in
3665 particular, if it contained a \" sequence, Exim got it wrong when building
3666 a From: or a Sender: header from that name. A second bug also caused
3667 incorrect handling when an unquoted " was present following a character
3668 that needed quoting.
3670 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3671 was not being matched caselessly.
3673 24. Arranged for all hyphens in the exim.8 source to be escaped with
3676 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3677 properly. Recipient callout cache records were still being keyed to include
3678 the sender, even when use_sender was set false. This led to far more
3679 callouts that were necessary. The sender is no longer included in the key
3680 when use_sender is false.
3682 26. Added "control = submission" modifier to ACLs.
3684 27. Added the ${base62d: operator to decode base 62 numbers.
3686 28. dnsdb lookups can now access SRV records.
3688 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3689 the configuration file.
3691 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3692 (-v) mode. This makes the output for a verbose queue run more intelligible.
3694 31. Added a use_postmaster feature to recipient callouts.
3696 32. Added the $body_zerocount variable, containing the number of binary zero
3697 bytes in the message body.
3699 33. The time of last modification of the "new" subdirectory is now used as the
3700 "mailbox time last read" when there is a quota error for a maildir
3703 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3705 35. Added +ignore_unknown as a special item in host lists.
3707 36. Code for decoding IPv6 addresses in host lists is now included, even if
3708 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3709 address was recognized as an IP address, but was then not correctly decoded
3710 into binary, causing unexpected and incorrect effects when compared with
3717 1. Very minor rewording of debugging text in manualroute to say "list of
3718 hosts" instead of "hostlist".
3720 2. If verify=header_syntax was set, and a header line with an unqualified
3721 address (no domain) and a large number of spaces between the end of the
3722 name and the colon was received, the reception process suffered a buffer
3723 overflow, and (when I tested it) crashed. This was caused by some obsolete
3724 code that should have been removed. The fix is to remove it!
3726 3. When running in the test harness, delay a bit after writing a bounce
3727 message to get a bit more predictability in the log output.
3729 4. Added a call to search_tidyup() just before forking a reception process. In
3730 theory, someone could use a lookup in the expansion of smtp_accept_max_
3731 per_host which, without the tidyup, could leave open a database connection.
3733 5. Added the variables $recipient_data and $sender_data which get set from a
3734 lookup success in an ACL "recipients" or "senders" condition, or a router
3735 "senders" option, similar to $domain_data and $local_part_data.
3737 6. Moved the writing of debug_print from before to after the "senders" test
3740 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3741 problems for message scanning, either using a data ACL, or using
3742 local_scan() because the Received: header was not generated till after they
3743 were called (in order to set the time as the time of reception completion).
3744 I have revised the way this works. The header is now generated after the
3745 body is received, but before the ACL or local_scan() are called. After they
3746 are run, the timestamp in the header is updated.
3752 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3753 before starting a queue runner without re-exec. This happened only when
3754 deliver_drop_privilege was set or when the Exim user was set to root. The
3755 effect of the bug was that timeouts during subsequent deliveries caused
3756 crashes instead of being properly handled. The handler is now left at its
3757 default (and expected) setting.
3759 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3760 message, again when deliver_drop_privilege is set or Exim is run as root.
3761 The bug described in (1) was not present in this case, but the tidying up
3762 of the other signals was missing. I have made the two cases consistent.
3764 3. The ignore_target_hosts setting on a manualroute router was being ignored
3765 for hosts that were looked up using the /MX notation.
3767 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3770 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3771 operated on the sender address. After changing the $sender_address to <>
3772 for the sender address verify, Exim was re-instated it as the original
3773 (before rewriting) address, but remembering that it had rewritten it, so it
3774 wasn't rewriting it again. This bug also had the effect of breaking the
3775 sender address verification caching when the sender address was rewritten.
3777 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3778 This has been changed so that if the ip literal address matches
3779 ignore_target_hosts, the router declines.
3781 7. Added expansion conditions match_domain, match_address, and match_local_
3782 part (NOT match_host).
3784 8. The placeholder for the Received: header didn't have a length field set.
3786 9. Added code to Exim itself and to exim_lock to test for a specific race
3787 condition that could lead to file corruption when using MBX delivery. The
3788 issue is with the lockfile that is created in /tmp. If this file is removed
3789 after a process has opened it but before that process has acquired a lock,
3790 there is the potential for a second process to recreate the file and also
3791 acquire a lock. This could lead to two Exim processes writing to the file
3792 at the same time. The added code performs the same test as UW imapd; it
3793 checks after acquiring the lock that its file descriptor still refers to
3794 the same named file.
3796 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3797 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3800 11. Added the smtp_active_hostname option. If used, this will typically be made
3801 to depend on the incoming interface address. Because $interface_address is
3802 not set up until the daemon has forked a reception process, error responses
3803 that can happen earlier (such as "too many connections") no longer contain
3806 12. If an expansion in a condition on a "warn" statement fails because a lookup
3807 defers, the "warn" statement is abandoned, and the next ACL statement is
3808 processed. Previously this caused the whole ACL to be aborted.
3810 13. Added the iplsearch lookup type.
3812 14. Added ident_timeout as a log selector.
3814 15. Added tls_certificate_verified as a log selector.
3816 16. Added a global option tls_require_ciphers (compare the smtp transport
3817 option of the same name). This controls incoming TLS connections.
3819 17. I finally figured out how to make tls_require_ciphers do a similar thing
3820 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3821 before starting the TLS session.
3823 18. Tabs are now shown as \t in -bP output.
3825 19. If the log selector return_path_on_delivery was set, Exim crashed when
3826 bouncing a message because it had too many Received: header lines.
3828 20. If two routers both had headers_remove settings, and the first one included
3829 a superfluous trailing colon, the final name in the first list and the
3830 first name in the second list were incorrectly joined into one item (with a
3831 colon in the middle).
3837 1. Added -C and -D options to the exinext utility, mainly to make it easier
3838 to include in the automated testing, but these could be helpful when
3839 multiple configurations are in use.
3841 2. The exinext utility was not formatting the output nicely when there was
3842 an alternate port involved in the retry record key, nor when there was a
3843 message id as well (for retries that were specific to a specific message
3844 and a specific host). It was also confused by IPv6 addresses, because of
3845 the additional colons they contain. I have fixed the IPv4 problem, and
3846 patched it up to do a reasonable job for IPv6.
3848 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3849 delivery, the log line now contains "pipelined" if PIPELINING was used.
3851 4. An SMTP transport process used to panic and die if the bind() call to set
3852 an explicit outgoing interface failed. This has been changed; it is now
3853 treated in the same way as a connect() failure.
3855 5. A reference to $sender_host_name in the part of a conditional expansion
3856 that was being skipped was still causing a DNS lookup. This no longer
3859 6. The def: expansion condition was not recognizing references to header lines
3860 that used bh_ and bheader_.
3862 7. Added the _cache feature to named lists.
3864 8. The code for checking quota_filecount in the appendfile transport was
3865 allowing one more file than it should have been.
3867 9. For compatibility with Sendmail, the command line option
3875 and sets the incoming protocol and host name (for trusted callers). The
3876 host name and its colon can be omitted when only the protocol is to be set.
3877 Note the Exim already has two private options, -pd and -ps, that refer to
3878 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3879 "s", but I don't think that's a major issue.
3881 10. A number of refactoring changes to the code, none of which should affect
3884 (a) The number of logging options was getting close to filling up the
3885 32-bit word that was used as a bit map. I have split them into two classes:
3886 those that are passed in the argument to log_write(), and those that are
3887 only ever tested independently outside of that function. These are now in
3888 separate 32-bit words, so there is plenty of room for expansion again.
3889 There is no change in the user interface or the logging behaviour.
3891 (b) When building, for example, log lines, the code previously used a
3892 macro that called string_cat() twice, in order to add two strings. This is
3893 not really sufficiently general. Furthermore, there was one instance where
3894 it was actually wrong because one of the argument was used twice, and in
3895 one call a function was used. (As it happened, calling the function twice
3896 did not affect the overall behaviour.) The macro has been replaced by a
3897 function that can join an arbitrary number of extra strings onto a growing
3900 (c) The code for expansion conditions now uses a table and a binary chop
3901 instead of a serial search (which was left over from when there were very
3902 few conditions). Also, it now recognizes conditions like "pam" even when
3903 the relevant support is not compiled in: a suitably worded error message is
3904 given if an attempt is made to use such a condition.
3906 11. Added ${time_interval:xxxxx}.
3908 12. A bug was causing one of the ddress fields not to be passed back correctly
3909 from remote delivery subprocesses. The field in question was not being
3910 subsequently used, so this caused to problems in practice.
3912 13. Added new log selectors queue_time and deliver_time.
3914 14. Might have fixed a bug in maildirsizefile handling that threw up
3915 "unexpected character" debug warnings, and recalculated the data
3916 unnecessarily. In any case, I expanded the warning message to give more
3919 15. Added the message "Restricted characters in address" to the statements in
3920 the default ACL that block characters like @ and % in local parts.
3922 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3923 Three changes have been made:
3925 (a) There was a serious bug; a negative response to MAIL caused the whole
3926 recipient domain to be cached as invalid, thereby blocking all messages
3927 to all local parts at the same domain, from all senders. This bug has
3928 been fixed. The domain is no longer cached after a negative response to
3929 MAIL if the sender used is not empty.
3931 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3934 (c) A new callout option, "use_sender" has been added for people who want
3935 the modified behaviour.
3941 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3944 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3945 indeed breaks things for older releases.
3947 3. Added additional logging to the case where there is a problem reading data
3948 from a filter that is running in a subprocess using a pipe, in order to
3949 try to track down a specific problem.
3951 4. Testing facility fudge: when running in the test harness and attempting
3952 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3953 getting "No route to host". Convert this to a timeout.
3955 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3958 6. Some OS don't have socklen_t but use size_t instead. This affects the
3959 fifth argument of getsockopt() amongst other things. This is now
3960 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3961 can be set for individual OS. I have set it for SunOS5, OSF1, and
3962 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3963 some earlier ones do not.
3965 7. Change 4.30/15 was not doing the test caselessly.
3967 8. The standard form for an IPv6 address literal was being rejected by address
3968 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3969 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3970 this, as well as the form without the "IPv6" on the front (but only when
3971 address literals are enabled, of course).
3973 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3975 10. Exim crashed if a message with an empty sender address specified by -f
3976 encountered a router with an errors_to setting. This could be provoked only
3977 by a command such as
3981 where an empty string was supplied; "<>" did not hit this bug.
3983 11. Installed PCRE release 4.5.
3985 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3986 remained set. It is now erased.
3988 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3989 times from message ids (which are base 36 rather than the normal 62).
3991 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3992 were being counted as actual protocol errors, and logged if the log
3993 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3994 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3995 and DATA following a set of rejected RCPTs do not count as protocol errors.
3996 In other words, Exim assumes they were pipelined, though this may not
3997 actually be the case. Of course, in all cases the client gets an
3998 appropriate error code.
4000 15. If a lookup fails in an ACL condition, a message about the failure may
4001 be available; it is used if testing the ACL cannot continue, because most
4002 such messages specify what the cause of the deferral is. However, some
4003 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4004 that caused an old message to be retained and used if a later statement
4005 caused a defer, replacing the real cause of the deferral.
4007 16. If an IP address had so many PTR records that the DNS lookup buffer
4008 was not large enough to hold them, Exim could crash while trying to process
4009 the truncated data. It now detects and logs this case.
4011 17. Further to 4.21/58, another change has been made: if (and only if) the
4012 first line of a message (the first header line) ends with CRLF, a bare LF
4013 in a subsequent header line has a space inserted after it, so as not to
4014 terminate the header.
4016 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4017 unnecessarily, used atoi() instead of strtol(), and didn't check the
4018 termination when getting file sizes from file names by regex.
4020 19. Completely re-implemented the support for maildirsize files, in the light
4021 of a number of problems with the previous contributed implementation
4022 (4.30/29). In particular:
4024 . If the quota is zero, the maildirsize file is maintained, but no quota is
4027 . If the maildir directory does not exist, it is created before any attempt
4028 to write a maildirsize file.
4030 . The quota value in the file is just a cache; if the quota is changed in
4031 the transport, the new value overrides.
4033 . A regular expression is available for excluding directories from the
4036 20. The autoreply transport checks the characters in options that define the
4037 message's headers; it allows continued headers, but it was checking with
4038 isspace() after an embedded newline instead of explicitly looking for a
4041 21. If all the "regular" hosts to which an address was routed had passed their
4042 expiry times, and had not reached their retry times, the address was
4043 bounced, even if fallback hosts were defined. Now Exim should go on to try
4046 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4047 equivalent code in the SMTP transport. Some hosts send humungous responses
4048 to HELO/EHLO, more than 1024 it seems.
4050 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4051 gives compiler warnings in some environments. I've now done it "properly",
4054 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4055 (because of problems with the built-in one) was declared to return uschar *
4056 instead of char *, causing compiler failure.
4058 25. Fixed a file descriptor leak when processing alias/forward files.
4060 26. Fixed a minor format string issue in dbfn.c.
4062 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4064 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4065 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4068 29. When a sender address is verified, it is cached, to save repeating the test
4069 when there is more than one recipient in a message. However, when the
4070 verification involves a callout, it is possible for different callout
4071 options to be set for different recipients. It is too complicated to keep
4072 track of this in the cache, so now Exim always runs a verification when a
4073 callout is required, relying on the callout cache for the optimization.
4074 The overhead is duplication of the address routing, but this should not be
4077 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4078 to be verified with callout=postmaster, and the main callout worked but the
4079 postmaster check failed, the verification correctly failed. However, if a
4080 subsequent RCPT command asked for sender verification *without* the
4081 postmaster check, incorrect caching caused this verification also to fail,
4084 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4085 it was not caching the DNS options (qualify_single, search_parents) that
4086 were used when the lookup failed. A subsequent lookup with different
4087 options therefore always gave the same answer, though there were cases
4088 where it should not have. (Example: a "domains = !$mx_any" option on a
4089 dnslookup router: the "domains" option is always processed without any
4090 widening, but the router might have qualify_single set.) Now Exim uses the
4091 cached value only when the same options are set.
4093 32. Added John Jetmore's "exipick" utility to the distribution.
4095 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4096 than a timeout (e.g. a certificate is required, and is not provided), an
4097 Exim server now closes the connection immediately. Previously it waited for
4098 the client to close - but if the client is SSL, it seems that they each
4099 wait for each other, leading to a delay before one of them times out.
4101 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4102 maintained 0.8.x compatibility because I don't think many are using it, and
4103 it is clearly obsolete.
4105 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4108 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4109 client certificate was expired. A simple patch fixes this, though I don't
4110 understand the full logic of why the verify callback is called multiple
4113 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4114 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4115 which causes problems with some clients (such as the Certicom SSL Plus
4116 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4117 disables the coutermeasure allowing Eudora to connect."
4119 38. Exim was not checking that a write() to a log file succeeded. This could
4120 lead to Bad Things if a log got too big, in particular if it hit a file
4121 size limit. Exim now panics and dies if it cannot write to a log file, just
4122 as it does if it cannot open a log file.
4124 39. Modified OS/Makefile-Linux so that it now contains
4126 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4128 The two -D definitions ensure that Exim is compiled with large file
4129 support, which makes it possible to handle log files that are bigger than
4132 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4133 instance) a domain was checked against a named list that involved a lookup,
4134 causing $domain_data to be set, then another domain was checked against the
4135 same list, then the first domain was re-checked, the value of $domain_data
4136 after the final check could be wrong. In particular, if the second check
4137 failed, it could be set empty. This bug probably also applied to
4140 41. The strip_trailing_dot option was not being applied to the address given
4141 with the -f command-line option.
4143 42. The code for reading a message's header from the spool was incrementing
4144 $received_count, but never initializing it. This meant that the value was
4145 incorrect (doubled) while delivering a message in the same process in which
4146 it was received. In the most common configuration of Exim, this never
4147 happens - a fresh exec is done - but it can happen when
4148 deliver_drop_privilege is set.
4150 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4151 it now includes up to 150 characters of the unexpected data in the log
4154 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4155 and building data strings. The size of both of these buffers was 10 000
4156 bytes - far larger than anybody would *ever* want, thought I. Needless to
4157 say, somebody hit the limit. I have increased the maximum line length to
4158 20 000 and the maximum data length of concatenated lines to 100 000. I have
4159 also fixed two bugs, because there was no checking on these buffers. Tsk,
4160 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4161 buffer is too small.
4163 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4164 lsearch lookups. Now it does.
4166 46. When parsing a route_list item in a manualroute router, a fixed-length
4167 buffer was used for the list of hosts. I made this 1024 bytes long,
4168 thinking that nobody would ever have a list of hosts that long. Wrong.
4169 Somebody had a whole pile of complicated expansion conditions, and the
4170 string was silently truncated, leading to an expansion error. It turns out
4171 that it is easier to change to an unlimited length (owing to other changes
4172 that have happened since this code was originally written) than to build
4173 structure for giving a limitation error. The length of the item that
4174 expands into the list of hosts is now unlimited.
4176 47. The lsearch lookup could not handle data where the length of text line was
4177 more than 4095 characters. Such lines were truncated, leading to shortened
4178 data being returned. It should now handle lines of any length.
4180 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4181 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4184 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4185 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4186 no longer happens in comment lines. A list of replacements is now placed
4187 at the head of all of the source files, except those whose only change is
4188 to replace PERL_COMMAND in the very first #! line.
4190 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4191 messages on the queue, with a bottom-up merge sort, using code contributed
4192 by Michael Haardt. This should make operations like -bp somewhat faster on
4193 large queues. It won't affect queue runners, except when queue_run_in_order
4196 51. Installed eximstats 1.31 in the distribution.
4198 52. Added support for SRV lookups to the dnslookup router.
4200 53. If an ACL referred to $message_body or $message_body_end, the value was not
4201 reset for any messages that followed in the same SMTP session.
4203 54. The store-handling optimization for building very long strings was not
4204 differentiating between the different store pools. I don't think this
4205 actually made any difference in practice, but I've tidied it.
4207 55. While running the routers to verify a sender address, $sender_address
4208 was still set to the sender address. This is wrong, because when routing to
4209 send a bounce to the sender, it would be empty. Therefore, I have changed
4210 it so that, while verifying a sender address, $sender_address is set to <>.
4211 (There is no change to what happens when verifying a recipient address.)
4213 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4214 target A or AAAA records (if not already returned) without resetting the
4215 qualify_single or search_parents options of the DNS resolver. These are
4216 inappropriate in this case because the targets of MX and SRV records must
4217 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4218 target that, when qualified, matched something in the local domain. These
4219 two options are now turned off when doing these lookups.
4221 57. It seems that at least some releases of Reiserfs (which does not have the
4222 concept of a fixed number of inodes) returns zero and not -1 for the
4223 number of available inodes. This interacted badly with check_spool_inodes,
4224 which assumed that -1 was the "no such thing" setting. What I have done is
4225 to check that the total number of inodes is greater than zero before doing
4226 the test of how many are available.
4228 58. When a "warn" ACL statement has a log_message modifier, the message is
4229 remembered, and not repeated. This is to avoid a lot of repetition when a
4230 message has many recipients that cause the same warning to be written.
4231 Howewer, Exim was preserving the list of already written lines for an
4232 entire SMTP session, which doesn't seem right. The memory is now reset if a
4233 new message is started.
4235 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4236 debugging output unless log_rewrite was also set.
4238 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4239 of (int)(handle) when we know that handle contains (void *)(-1).
4241 61. The Exim daemon panic-logs an error return when it closes the incoming
4242 connection. However "connection reset by peer" seems to be common, and
4243 isn't really an error worthy of noting specially, so that particular error
4246 62. When Exim is trying to find all the local interfaces, it used to panic and
4247 die if the ioctl to get the interface flags failed. However, it seems that
4248 on at least one OS (Solaris 9) it is possible to have an interface that is
4249 included in the list of interfaces, but for which you get a failure error
4250 for this call. This happens when the interface is not "plumbed" into a
4251 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4252 failure of the "get flags" call assumes that the interface is down.
4254 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4255 makes life easier for people who are doing arithmetic on fields extracted
4256 from dates, where you often get leading zeros that should not be
4257 interpreted as octal.
4259 64. Added qualify_domain to the redirect router, to override the global
4262 65. If a pathologically long header line contained very many addresses (the
4263 report of this problem mentioned 10 000) and each of them was rewritten,
4264 Exim could use up a very large amount of memory. (It kept on making new
4265 copies of the header line as it rewrote, and never released the old ones.)
4266 At the expense of a bit more processing, the header rewriting function has
4267 been changed so that it no longer eats memory in this way.
4269 66. The generation of the Received: header has been moved from the time that a
4270 message starts to be received, to the time that it finishes. The timestamp
4271 in the Received: header should now be very close to that of the <= log
4272 line. There are two side-effects of this change:
4274 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4275 logged header lines no longer include the local Received: line, because
4276 it has not yet been created. The same applies to a copy of the message
4277 that is returned to a non-SMTP sender when a message is rejected.
4279 (b) When a filter file is tested using -bf, no additional Received: header
4280 is added to the test message. After some thought, I decided that this
4283 This change does not affect the value of $received_for. It is still set
4284 after address rewriting, but before local_scan() is called.
4286 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4288 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4289 gave an unhelpful panic error message, and a defer error. I have managed to
4290 change this behaviour so that it now rejects any supplied certificate,
4291 which seems right, as the list of acceptable certificates is empty.
4293 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4294 gave an unhelpful defer error. I have not managed to make this reject any
4295 supplied certificates, but the error message it gives is "no certificate
4296 supplied", which is not helpful.
4298 70. exigrep's output now also includes lines that are not associated with any
4299 message, but which match the given pattern. Implemented by a patch from
4300 Martin Sluka, which also tidied up the Perl a bit.
4302 71. Recipient callout verification, like sender verification, was using <> in
4303 the MAIL FROM command. This isn't really the right thing, since the actual
4304 sender may affect whether the remote host accepts the recipient or not. I
4305 have changed it to use the actual sender in the callout; this means that
4306 the cache record is now keyed on a recipient/sender pair, not just the
4307 recipient address. There doesn't seem to be a real danger of callout loops,
4308 since a callout by the remote host to check the sender would use <>.
4309 [SEE ABOVE: changed after hitting problems.]
4311 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4312 temporary errors. However, in the case of such a code being given after
4313 the end of a data transmission (i.e. after ".") Exim was failing to write
4314 a retry record for the message. (Yes, there was some broken host that was
4315 actually sending 8xx at this point.)
4317 73. An unknown lookup type in a host list could cause Exim to panic-die when
4318 the list was checked. (An example that provoked this was putting <; in the
4319 middle of a list instead of at the start.) If this happened during a DATA
4320 ACL check, a -D file could be left lying around. This kind of configuration
4321 error no longer causes Exim to die; instead it causes a defer errror. The
4322 incident is still logged to the main and panic logs.
4324 74. Buglet left over from Exim 3 conversion. The message "too many messages
4325 in one connection" was written to the rejectlog but not the mainlog, except
4326 when address rewriting (yes!) was being logged.
4328 75. Added write_rejectlog option.
4330 76. When a system filter was run not as root (that is, when system_filter_user
4331 was set), the values of the $n variables were not being returned to the
4332 main process; thus, they were not subsequently available in the $sn
4335 77. Added +return_path_on_delivery log selector.
4337 78. A connection timeout was being treated differently from recipients deferred
4338 when testing hosts_max_try with a message that was older than the host's
4339 retry timeout. (The host should not be counted, thus allowing all hosts to
4340 be tried at least once before bouncing.) This may have been the cause of an
4341 occasionally reported bug whereby a message would remain on the queue
4342 longer than the retry timeout, but would be bounced if a delivery was
4343 forced. I say "may" because I never totally pinned down the problem;
4344 setting up timeout/retry tests is difficult. See also the next item.
4346 79. The ultimate address timeout was not being applied to errors that involved
4347 a combination of host plus message (for example, a timeout on a MAIL
4348 command). When an address resolved to a number of possible hosts, and they
4349 were not all tried for each delivery (e.g. because of hosts_max_try), a
4350 message could remain on the queue longer than the retry timeout.
4352 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4355 81. Fixed an obscure SMTP outgoing bug which required at least the following
4356 conditions: (a) there was another message waiting for the same server;
4357 (b) the server returned 5xx to all RCPT commands in the first message so
4358 that the message was not completed; (c) the server dropped the connection
4359 or gave a negative response to the RSET that Exim sends to abort the
4360 transaction. The observed case was a dropped connection after DATA that had
4361 been sent in pipelining mode. That is, the server had advertised PIPELINING
4362 but was not implementing it correctly. The effect of the bug was incorrect
4363 behaviour, such as trying another host, and this could lead to a crash.
4369 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4370 and daemon.c were passed as pointers to ints; they should have been
4371 pointers to socklen_t variables (which are typically unsigned ints).
4373 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4376 3. Fixed a really odd bug that affected only the testing scheme; patching a
4377 certain fixed string in the binary changed the value of another string that
4378 happened to be identical to the end of the original first string.
4380 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4381 name", it returns that address as the IP address. On some operating
4382 systems (e.g. Solaris), it also passes back the IP address string as the
4383 "host name". However, on others (e.g. Linux), it passes back an empty
4384 string. Exim wasn't checking for this, and was changing the host name to an
4385 empty string, assuming it had been canonicized.
4387 5. Although rare, it is permitted to have more than one PTR record for a given
4388 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4389 all the names associated with an address, because they do in Solaris.
4390 However, it seems that they do not in Linux for data that comes from the
4391 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4392 I found this out when I moved to a new Linux workstation and tried to run
4393 the Exim test suite.
4395 To get round this problem I have changed the code so that it now does its
4396 own call to the DNS to look up PTR records when searching for a host name.
4397 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4398 addresses that are only in /etc/hosts are still found.
4400 This behaviour is, however, controlled by an option called host_lookup_
4401 order, which defaults to "bydns:byaddr". If people want to use the other
4402 order, or indeed, just use one or the other means of lookup, they can
4403 specify it in this variable.
4405 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4406 some operating systems, this comes back from gethostbyaddr() as an empty
4407 string, and this is what Exim used to test for. However, it seems that in
4408 other systems, "." is yielded. Exim now tests for this case too.
4410 7. The values of check_spool_space and check_log_space are now held internally
4411 as a number of kilobytes instead of an absolute number of bytes. If a
4412 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4413 kilobyte. This means that much larger values can be stored.
4415 8. Exim monitor: an attempt to get the action menu when not actually pointing
4416 at a message produces an empty menu entitled "No message selected". This
4417 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4418 no entries in it ("Shell widget menu has zero width and/or height"). So I
4419 have added a single, blank menu entry in this case.
4421 9. Added ${quote_local_part.
4423 10. MIME decoding is now applied to the contents of Subject: header lines when
4426 11. Now that a reference to $sender_host_address automatically causes a reverse
4427 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4428 host lookup before query-style lookups in lists that might use this
4429 variable. This has therefore been abolished, and the "net-" prefix is no
4430 longer necessary for query-style lookups.
4432 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4433 have been a typo for LFLAGS, so it has been changed.
4435 13. The install script calls Exim with "-C /dev/null" in order to find the
4436 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4437 to be output. Howeve, since Exim outputs its version number before the
4438 error, it didn't break the script. It just looked ugly. I fixed this by
4439 always allowing "-C /dev/null" if the caller is root.
4441 14. Ignore overlarge ACL variable number when reading spool file - insurance
4442 against a later release with more variables having written the file.
4444 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4445 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4446 this, as well as the form without the "IPv6" on the front.
4448 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4449 OS/Makefile-Darwin file.
4451 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4452 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4454 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4455 variable. However, some OS use ints for inodes. I've added cast to long int
4456 to get rid of the compiler warning.
4458 19. I had forgotten to lock out "/../" in configuration file names when
4459 ALT_CONFIG_PREFIX was set.
4461 20. Routers used for verification do not need to specify transports. However,
4462 if such a router generated a host list, and callout was configured, Exim
4463 crashed, because it could not find a port number from the (non-existent)
4464 transport. It now assumes port 25 in this circumstance.
4466 21. Added the -t option to exigrep.
4468 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4469 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4470 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4471 them is set, LOOKUP_LSEARCH is forced.
4473 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4475 24. Added sender and host information to the "rejected by local_scan()" log
4476 line; previously there was no indication of these.
4478 25. Added .include_if_exists.
4480 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4481 Linux. It turns out that not all file systems support this. Apparently some
4482 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4483 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4484 not supported on the file descriptor, is now ignored when Exim is trying to
4485 sync a directory. This applies only to Linux.
4487 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4489 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4490 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4493 29. Added support for maildirsize files from supplied patch (modified a bit).
4495 30. The use of :fail: followed by an empty string could lead Exim to respond to
4496 sender verification failures with (e.g.):
4498 550 Verification failed for <xxx>
4499 550 Sender verify failed
4501 where the first response line was missing the '-' that indicates it is not
4502 the final line of the response.
4504 31. The loop for finding the name of the user that called Exim had a hardwired
4505 limit of 10; it now uses the value of finduser_retries, which is used for
4506 all other user lookups.
4508 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4511 33. Exim was neglecting to zero errno before one call of strtol() when
4512 expanding a string and expecting an integer value. On some systems this
4513 resulted in spurious "integer overflow" errors. Also, it was casting the
4514 result into an int without checking.
4516 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4517 did not work. The code looks as if it has *never* worked, though it appears
4518 to have been documented since at least releast 1.62. I have made it work.
4520 35. The "timeout_DNS" error in retry rules, also documented since at least
4521 1.62, also never worked. As it isn't clear exactly what this means, and
4522 clearly it isn't a major issue, I have abolished the feature by treating it
4523 as "timeout", and writing a warning to the main and panic logs.
4525 36. The display of retry rules for -brt wasn't always showing the error code
4528 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4529 timeout_connect_A, timeout_connect_MX.
4531 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4532 to the empty sender.
4534 39. The daemon was not analysing the content of -oX till after it had closed
4535 stderr and disconnected from the controlling terminal. This meant that any
4536 syntax errors were only noted on the panic log, and the return code from
4537 the command was 0. By re-arranging the code a little, I've made the
4538 decoding happen first, so such errors now appear on stderr, and the return
4539 code is 1. However, the actual setting up of the sockets still happens in
4540 the disconnected process, so errors there are still only recorded on the
4543 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4544 connections (as happens on some IP stacks) was logged at start up time as
4545 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4546 it from "IPv6 and IPv4", which means that two separate sockets are being
4549 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4550 says whether AF_INET or AF_INET6 was passed as an argument.
4552 42. Exiwhat output was messed up when time zones were included in log
4555 43. Exiwhat now gives more information about the daemon's listening ports,
4556 and whether -tls-on-connect was used.
4558 44. The "port" option of the smtp transport is now expanded.
4560 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4561 silently ignored. Now an error message is written to the main and panic
4564 46. There's a new ACL modifier called "logwrite" which writes to a log file
4565 as soon as it is encountered.
4567 47. Added $local_user_uid and $local_user_gid at routing time.
4569 48. Exim crashed when trying to verify a sender address that was being
4572 49. Exim was recognizing only a space character after ".include". It now also
4573 recognizes a tab character.
4575 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4576 extracting the relevant information from the specification. The man page no
4577 longer contains scrambled data for the -d option, and I've added a section
4578 at the front about calling Exim under different names.
4580 51. Added "extra_headers" argument to the "mail" command in filter files.
4582 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4585 53. Installed eximstats 1.29.
4587 54. Added transport_filter_timeout as a generic transport option.
4589 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4590 Cc: header lines. This was required by RFC 822, but it not required by RFC
4593 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4594 incoming messages that did not have them. Now it does so only if the
4595 message originates locally, that is, if there is no associated remote host
4596 address. When Resent- header lines are present, this applies to the Resent-
4597 lines rather than the non-Resent- lines.
4599 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4600 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4602 58. Messages for configuration errors now include the name of the main
4603 configuration file - useful now that there may be more than one file in a
4604 list (.included file names were always shown).
4606 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4607 for those rare installations that do not start the daemon as root or run it
4608 setuid root. I've cut out the call to initgroups() if the daemon is not
4611 60. The Exim user and group can now be bound into the binary as text strings
4612 that are looked up at the start of Exim's processing.
4614 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4616 62. Added $mailstore_basename variable.
4618 63. Installed patch to sieve.c from Michael Haardt.
4620 64. When Exim failed to open the panic log after failing to open the main log,
4621 the original message it was trying to log was written to stderr and debug
4622 output, but if they were not available (the usual case in production), it
4623 was lost. Now it is written to syslog before the two lines that record the
4624 failures to open the logs.
4626 65. Users' Exim filters run in subprocesses under the user's uid. It is
4627 possible for a "deliver" command or an alias in a "personal" command to
4628 provoke an address rewrite. If logging of address rewriting is configured,
4629 this fails because the process is not running as root or exim. There may be
4630 a better way of dealing with this, but for the moment (because 4.30 needs
4631 to be released), I have disabled address rewrite logging when running a
4632 filter in a non-root, non-exim process.
4638 1. The buildconfig auxiliary program wasn't quoting the value set for
4639 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4640 not defined. This bug was masked in 4.22 by the effect that was fixed in
4643 2. Some messages that were rejected after a message id was allocated were
4644 shown as "incomplete" by exigrep. It no longer does this for messages that
4645 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4647 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4648 have allow_domain_literals set, the ID did not get logged in the <= line.
4649 Domain literals are now always recognized in Message-ID: header lines.
4651 4. The first argument for a ${extract expansion item is the key name or field
4652 number. Leading and trailing spaces in this item were not being ignored,
4653 causing some misleading effects.
4655 5. When deliver_drop_privilege was set, single queue runner processes started
4656 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4657 same command in the process it spins off) were not dropping privilege.
4659 6. When the daemon running as "exim" started a queue runner, it always
4660 re-executed Exim in the spun-off process. This is a waste of effort when
4661 deliver_drop_privilege is set. The new process now just calls the
4662 queue-runner function directly.
4668 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4671 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4672 ignored. Though the use of -oP was forcing the writing of a pid file, it
4673 was always written to the default place.
4675 3. If the message "no IP address found for host xxxx" is generated during
4676 incoming verification, it is now followed by identification of the incoming
4677 connection (so you can more easily find what provoked it).
4679 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4681 5. Added some features to "harden" Exim a bit more against certain attacks:
4683 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4684 be put in Local/Makefile. This is like the never_users runtime option,
4685 but it cannot be overridden. The default setting is "root".
4687 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4688 prefix string with which any file named in a -C command line option
4691 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4692 is retained for -C and -D only if the caller of Exim is root. Without
4693 it, the exim user may also use -C and -D and retain privilege.
4695 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4696 command line option is disabled.
4698 6. Macro names set by the -D option must start with an upper case letter, just
4699 like macro names defined in the configuration file.
4701 7. Added "dereference=" facility to LDAP.
4703 8. Two instances of the typo "uknown" in the source files are fixed.
4705 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4706 the Configure-Makefile script screwed up while processing it.
4708 10. Incorporated PCRE 4.4.
4710 11. The SMTP synchronization check was not operating right at the start of an
4711 SMTP session. For example, it could not catch a HELO sent before the client
4712 waited for the greeting. There is now a check for outstanding input at the
4713 point when the greeting is written. Because of the duplex, asynchronous
4714 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4715 way, but not yet received, when the check is performed.
4717 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4718 on TCP/IP sockets, because this apparently causes some broken clients to
4721 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4722 unchanged) from the Cygwin maintainer.
4724 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4725 was supported instead of testing for mbx. Effectively a typo.
4727 15. The spa authenticator server code was not checking that the input it
4728 received was valid base64.
4730 16. The debug output line for the "set" modifier in ACLs was not showing the
4731 name of the variable that was being set.
4733 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4735 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4736 lookup on its own. Something else was needed to trigger the lookup. For
4737 example, a match in host_lookup or the need for a host name in a host list.
4738 Now, if $sender_host_name is referenced and the host name has not yet been
4739 looked up, a lookup is performed. If the lookup fails, the variable remains
4740 empty, and $host_lookup_failed is set to "1".
4742 19. Added "eqi" as a case-independent comparison operator.
4744 20. The saslauthd authentication condition could segfault if neither service
4745 nor realm was specified.
4747 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4748 error message that was logged was misleading, and incoming SMTP
4749 connections were dropped. The message is now more accurate, and temporary
4750 errors are given to SMTP connections.
4752 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4753 (except RSET) with a 421 error, until QUIT is received. However, it was
4754 failing to send a response to QUIT.
4756 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4757 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4758 ACL is now run after the helo_try_verify_hosts code.
4760 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4761 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4762 case-independent, so other case variants are also recognized). Apparently
4763 some systems use these upper case variants.
4765 25. If more than two messages were waiting for the same host, and a transport
4766 filter was specified for the transport, Exim sent two messages over the
4767 same TCP/IP connection, and then failed with "socket operation on non-
4768 socket" when it tried to send the third.
4770 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4772 27. The extern definition of crypt16() in expand.c was not being excluded when
4773 the OS had its own crypt16() function.
4775 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4776 as a preferred synonym for return_size_limit, both as an option and as an
4779 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4781 30. Changed the default configuration ACL to relax the local part checking rule
4782 for addresses that are not in any local domains. For these addresses,
4783 slashes and pipe symbols are allowed within local parts, but the sequence
4784 /../ is explicitly forbidden.
4786 31. SPA server authentication was not clearing the challenge buffer before
4789 32. log_message in a "warn" ACL statement was writing to the reject log as
4790 well as to the main log, which contradicts the documentation and doesn't
4791 seem right (because no rejection is happening). So I have stopped it.
4793 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4794 However, I am unable to do any testing of this.
4796 34. Fixed an infelicity in the appendfile transport. When checking directories
4797 for a mailbox, to see if any needed to be created, it was accidentally
4798 using path names with one or more superfluous leading slashes; tracing
4799 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4801 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4802 contents are added to the log line that is written for every discarded
4803 recipient. (Previously a log_message setting was ignored.)
4805 36. The ${quote: operator now quotes the string if it is empty.
4807 37. The install script runs exim in order to find its version number. If for
4808 some reason other than non-existence or emptiness, which it checks, it
4809 could not run './exim', it was installing it with an empty version number,
4810 i.e. as "exim-". This error state is now caught, and the installation is
4813 38. An argument was missing from the function that creates an error message
4814 when Exim fails to connect to the socket for saslauthd authentication.
4815 This could cause Exim to crash, or give a corrupted message.
4817 39. Added isip, isip4, and isip6 to ${if conditions.
4819 40. The ACL variables $acl_xx are now saved with the message, and can be
4820 accessed later in routers, transports, and filters.
4822 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4823 strings in the file are not string-expanded.
4825 42. If a MAIL command specified a SIZE value that was too large to fit into an
4826 int variable, the check against message_size_limit failed. Such values are
4827 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4828 day this will have to be increased, but I don't think I want to be around
4829 when emails are that large.
4836 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4837 iconv() is not standard in FreeBSD.
4839 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4840 IPv6 enabled. The observed symptom was a segmentation fault on return from
4841 the function os_common_find_running_interfaces() in src/os.c.
4843 3. In the check_special_case() function in daemon.c I had used "errno" as an
4844 argument name, which causes warnings on some systems. This was basically a
4845 typo, since it was named "eno" in the comments!
4847 4. The code that waits for the clock to tick (at a resolution of some fraction
4848 of a second) so as to ensure message-id uniqueness was always waiting for
4849 at least one whole tick, when it could have waited for less. [This is
4850 almost certainly not relevant at current processor speeds, where it is
4851 unlikely to ever wait at all. But we try to future-proof.]
4853 5. The function that sleeps for a time interval that includes fractions of a
4854 second contained a race. It did not block SIGALRM between setting the
4855 timer, and suspending (a couple of lines later). If the interval was short
4856 and the sigsuspend() was delayed until after it had expired, the suspension
4857 never ended. On busy systems this could lead to processes getting stuck for
4860 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4861 process, before it forks any delivery processes. The open lookup caching
4862 mechanism meant that the open file or database connection was passed into
4863 the delivery process. The problem was that delivery processes always tidy
4864 up cached lookup data. This could cause a problem for the next delivery
4865 process started by the queue runner, because the external queue runner
4866 process does not know about the closure. So the next delivery process
4867 still has data in the lookup cache. In the case of a file lookup, there was
4868 no problem because closing a file descriptor in a subprocess doesn't affect
4869 the parent. However, if the lookup was caching a connection to a database,
4870 the connection was closed, and the second delivery process was likely to
4871 see errors such as "PGSQL: query failed: server closed the connection
4872 unexpectedly". The problem has been fixed by closing all cached lookups
4873 in a queue runner before running a delivery process.
4875 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4876 seem to have the "const" qualifier which it has on other OS. I've
4879 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4880 *of the same type* (client or server) with the same public name that an
4881 error should be diagnosed.
4883 9. When Exim looked up a host name for an IP address, but failed to find the
4884 original IP address when looking up the host name (a safety check), it
4885 output the message "<ip address> does not match any IP for NULL", which was
4886 confusing, to say the least. The bug was that the host name should have
4887 appeared instead of "NULL".
4889 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4890 user that is built into the binary, and the -C or -D options is used, root
4891 privilege is dropped before the configuration file is read. In addition,
4892 logging is switched to stderr instead of the normal log files. If the
4893 configuration then re-defines the Exim user, the unprivileged environment
4894 is probably not what is expected, so Exim logs a panic warning message (but
4897 However, if deliver_drop_privilege is set, the unprivileged state may well
4898 be exactly what is intended, so the warning has been cut out in that case,
4899 and Exim is allowed to try to write to its normal log files.
4905 1. smtp_return_error_details was not giving details for temporary sender
4906 or receiver verification errors.
4908 2. Diagnose a configuration error if two authenticators have the same public
4911 3. Exim used not to create the message log file for a message until the first
4912 delivery attempt. This could be confusing when incoming messages were held
4913 for policy or load reasons. The message log file is now created at the time
4914 the message is received, and an initial "Received" line is written to it.
4916 4. The automatically generated man page for command line options had a minor
4917 bug that caused no ill effects; however, a more serious problem was that
4918 the procedure for building the man page automatically didn't always
4919 operate. Consequently, release 4.20 contains an out-of-date version. This
4920 shouldn't happen again.
4922 5. When building Exim with embedded Perl support, the script that builds the
4923 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4924 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4926 6. The freeze_tell option was not being used for messages that were frozen on
4927 arrival, either by an ACL or by local_scan().
4929 7. Added the smtp_incomplete_transaction log selector.
4931 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4932 was accepting AUTH without a new EHLO.
4934 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4935 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4938 10. Logging of TCP/IP connections (when configured) now happens in the main
4939 daemon process instead of the child process, so that the TCP/IP connection
4940 count is more accurate (but it can never be perfect).
4942 11. The use of "drop" in a nested ACL was not being handled correctly in the
4943 outer ACL. Now, if condition failure induced by the nested "drop" causes
4944 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4945 or "require"), the connection is dropped.
4947 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4948 that yield "discard" can now be used with an "accept" or a "discard" verb,
4949 but an error is generated for any others (because I can't see a useful way
4950 to define what should happen).
4952 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4953 are now processed in the same way as lines in the Exim configuration file.
4954 In particular, continuation lines are supported.
4956 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4958 15. Added -ti meaning -t -i.
4960 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4961 domains, and warn by logging if others are found.
4963 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4964 returned from GIFCONF when Exim is trying to find the list of interfaces on
4965 a host. The code in os.c has been modified to copy each ifreq to an aligned
4966 structure in all cases.
4968 Also, in some cases, the returned ifreq's were being copied to a 'struct
4969 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4970 means the last couple of bytes of an IPv6 address could be chopped if the
4971 ifreq contained only a normal sockaddr (14 bytes storage).
4973 18. Named domain lists were not supported in the hosts_treat_as_local option.
4974 An entry such as +xxxx was not recognized, and was treated as a literal
4977 19. Ensure that header lines added by a DATA ACL are included in the reject log
4978 if the ACL subsequently rejects the message.
4980 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4981 MD5 (which is deprecated).
4983 21. When testing a filter file using -bf, Exim was writing a message when it
4984 took the sender from a "From " line in the message, but it was not doing so
4985 when it took $return_path from a Return-Path: header line. It now does.
4987 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4988 with a valid header line field name (a series of printing characters
4989 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4991 23. Changed "disc" in the source to "disk" to conform to the documentation and
4992 the book and for uniformity.
4994 24. Ignore Sendmail's -Ooption=value command line item.
4996 25. When execve() failed while trying to run a command in a pipe transport,
4997 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4998 could be confused with a return value of 69 from the command itself. This
4999 has been changed to 127, the value the shell returns if it is asked to run
5000 a non-existent command. The wording for the related log line suggests a
5001 non-existent command as the problem.
5003 26. If received_header_text expands to an empty string, do not add a Received:
5004 header line to the message. (Well, it adds a token one on the spool, but
5005 marks it "old" so that it doesn't get used or transmitted.)
5007 27. Installed eximstats 1.28 (addition of -nt option).
5009 28. There was no check for failure on the call to getsockname() in the daemon
5010 code. This can fail if there is a shortage of resources on the system, with
5011 ENOMEM, for example. A temporary error is now given on failure.
5013 29. Contrary to the C standard, it seems that in some environments, the
5014 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5015 program. Exim now does this explicitly; it affects the formatting of
5016 timestamps using strftime().
5018 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5019 complaints. I've now initialized all the variables, to avoid this.
5021 32. Header lines added by a system filter were not being "seen" during
5022 transport-time rewrites.
5024 33. The info_callback() function passed to OpenSSL is set up with type void
5025 (*)(SSL *, int, int), as described somewhere. However, when calling the
5026 function (actually a macro) that sets it up, the type void(*)() is
5027 expected. I've put in a cast to prevent warnings from picky compilers.
5029 34. If a DNS black list lookup found a CNAME record, but there were no A
5030 records associated with the domain it pointed at, Exim crashed.
5032 35. If a DNS black list lookup returned more than one A record, Exim ignored
5033 all but the first. It now scans all returned addresses if a particular IP
5034 value is being sought. In this situation, the contents of the
5035 $dnslist_value variable are a list of all the addresses, separated by a
5038 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5039 to accept a host name and all its aliases if the forward lookup for any of
5040 them yielded the IP address of the incoming connection. Now it accepts only
5041 those names whose forward lookup yields the correct IP address. Any other
5042 names are discarded. This closes a loophole whereby a rogue DNS
5043 administrator could create reverse DNS records to break through a
5044 wildcarded host restriction in an ACL.
5046 37. If a user filter or a system filter that ran in a subprocess used any of
5047 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5048 the wrong values were passed to the pipe command ($thisaddress had the
5049 value of $0, $0 had the value of $1, etc). This bug was introduced by
5050 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5052 38. Improved the line breaking for long SMTP error messages from ACLs.
5053 Previously, if there was no break point between 40 and 75 characters, Exim
5054 left the rest of the message alone. Two changes have been made: (a) I've
5055 reduced the minimum length to 35 characters; (b) if it can't find a break
5056 point between 35 and 75 characters, it looks ahead and uses the first one
5057 that it finds. This may give the occasional overlong line, but at least the
5058 remaining text gets split now.
5060 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5061 file descriptors might be low, and that setting 1000 would always raise it.
5062 It turns out that in some environments, the limit is already over 1000 and
5063 that lowering it causes trouble. So now Exim takes care not to decrease it.
5065 40. When delivering a message, the value of $return_path is set to $sender_
5066 address at the start of routing (routers may change the value). By an
5067 oversight, this default was not being set up when an address was tested by
5068 -bt or -bv, which affected the outcome if any router or filter referred to
5071 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5072 log only when "message" or "log_message" are set. However, if one of the
5073 conditions was an address verification, or a call to a nested ACL, the
5074 messages generated by the underlying test were being passed through. This
5075 no longer happens. The underlying message is available in $acl_verify_
5076 message for both "message" and "log_message" expansions, so it can be
5077 passed through if needed.
5079 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5080 new expansion $bh_ to give the encoded byte string without charset
5081 translation. Translation happens only if iconv() is available; HAVE_ICONV
5082 indicates this at build time. HEADERS_CHARSET gives the charset to
5083 translate to; headers_charset can change it in the configuration, and
5084 "headers charset" can change it in an individual filter file.
5086 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5087 that creates RFC 2047 encoded "words" labels them as that charset instead
5088 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5089 expansion operator; (ii) when Exim creates a From: line for a local
5090 message; (iii) when a header line is rewritten to include a "phrase" part.
5092 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5093 buggy, causing it to skip the first lines of messages whose message ID
5094 ended in 'D'. This would not have bitten before Exim release 4.14, because
5095 message IDs were unlikely to end in 'D' before then. The effect was to have
5096 incorrect size information for certain domains.
5098 45. #include "config.h" was missing at the start of the crypt16.c module. This
5099 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5102 46. If there was a timeout during a "random" callout check, Exim treated it as
5103 a failure of the random address, and carried on sending RSET and the real
5104 address. If the delay was just some slowness somewhere, the response to the
5105 original RCPT would be taken as a response to RSET and so on, causing
5106 mayhem of various kinds.
5108 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5109 when I implemented it. It didn't allow for the fact that some option values
5110 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5111 the right test for positive values.
5113 48. Domain names in DNS records are case-independent. Exim always looks them up
5114 in lower case. Some resolvers return domain names in exactly the case they
5115 appear in the zone file, that is, they may contain uppercase letters. Not
5116 all resolvers do this - some return always lower case. Exim was treating a
5117 change of case by a resolver as a change of domain, similar to a widening
5118 of a domain abbreviation. This triggered its re-routing code and so it was
5119 trying to route what was effectively the same domain again. This normally
5120 caused routing to fail (because the router wouldn't handle the domain
5121 twice). Now Exim checks for this case specially, and just changes the
5122 casing of the domain that it ultimately uses when it transmits the message
5125 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5128 50. If a filter generated a file delivery with a non-absolute name (possible if
5129 no home directory exists for the router), the forbid_file option was not
5132 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5133 the existing equality matching.
5135 52. Exim was using ints instead of ino_t variables in some places where it was
5136 dealing with inode numbers.
5138 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5139 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5140 called TMPDIR, and if it finds it is different, it changes its value.
5142 54. The smtp_printf() function is now made available to local_scan() so
5143 additional output lines can be written before returning. There is also an
5144 smtp_fflush() function to enable the detection of a dropped connection.
5145 The variables smtp_input and smtp_batched_input are exported to
5148 55. Changed the default runtime configuration: the message "Unknown user"
5149 has been removed from the ACL, and instead placed on the localuser router,
5150 using the cannot_route_message feature. This means that any verification
5151 failures that generate their own messages won't get overridden. Similarly,
5152 the "Unrouteable address" message that was in the ACL for unverifiable
5153 relay addresses has also been removed.
5155 56. Added hosts_avoid_esmtp to the smtp transport.
5157 57. The exicyclog script was not checking for the esoteric option
5158 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5159 will work only if exicyclog is run under the appropriate euid.
5161 58. Following a discussion on the list, the rules by which Exim recognises line
5162 endings on incoming messages have been changed. The -dropcr and drop_cr
5163 options are now no-ops, retained only for backwards compatibility. The
5164 following line terminators are recognized: LF CRLF CR. However, special
5165 processing applies to CR:
5167 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5168 nor a local message in the state where . is a terminator.
5170 (ii) If a bare CR is encountered in a header line, an extra space is added
5171 after the line terminator so as not to end the header. The reasoning
5172 behind this is that bare CRs in header lines are most likely either
5173 to be mistakes, or people trying to play silly games.
5175 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5176 was being incorrectly given as 18 bytes larger than it should have been.
5177 This is a VOB (very old bug).
5179 60. This may never have affected anything current, but just in case it has:
5180 When the local host is found other than at the start of a list of hosts,
5181 the local host, those with the same MX, and any that follow, are discarded.
5182 When the list in question was part of a longer list of hosts, the following
5183 hosts (not currently being processed) were also being discarded. This no
5184 longer happens. I'm not sure if this situation could ever has previously
5187 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5190 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5191 header. This is something that is recommended in a new Internet Draft, and
5192 is something that is documented as being done by Sendmail. There are two
5193 possible values. For messages generated by the autoreply transport, Exim
5196 Auto-Submitted: auto-replied
5198 whereas for all other generated messages (e.g. bounces) it adds
5200 Auto-Submitted: auto-generated
5202 63. The "personal" condition in filters now includes a test for the
5203 Auto-Submitted: header. If it contains the string "auto-" the message it
5204 not considered personal.
5206 64. Added rcpt_include_affixes as a generic transport option.
5208 65. Added queue_only_override (default true).
5210 66. Added the syslog_duplication option.
5212 67. If what should have been the first header line of a message consisted of
5213 a space followed by a colon, Exim was mis-interpreting it as a header line.
5214 It isn't of course - it is syntactically invalid and should therefore be
5215 treated as the start of the message body. The misbehaviour could have
5216 caused a number of strange effects, including loss of data in subsequent
5217 header lines, and spool format errors.
5219 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5220 client host had authenticated. This control can now be exercised by an ACL
5221 for more flexibility.
5223 69. By default, callouts do not happen when testing with -bh. There is now a
5224 variant, -bhc, which does actually run the callout code, including
5225 consulting and updating the callout cache.
5227 70. Added support for saslauthd authentication, courtesy of Alexander
5230 71. If statvfs() failed on the spool or log directories while checking their
5231 size for availability, Exim confusingly gave the error "space shortage".
5232 Furthermore, in debugging mode it crashed with a floating point exception.
5233 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5234 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5235 serious problem, Exim now writes to the main and panic logs when this
5236 happens, with details of the failure. It then refuses to accept the
5237 incoming message, giving the message "spool directory problem" or "log
5238 directory problem" with a 421 code for SMTP messages.
5240 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5241 0, 1, and 2 exist, because some OS complain for execs without them (see
5242 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5243 descriptors. However, the code omitted to check that the open succeeded,
5244 causing mysterious errors if for some reason the permissions on /dev/null
5245 got screwed. Now Exim writes a message to the main and panic logs, and
5246 bombs out if it can't open /dev/null.
5248 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5249 interact so that it is all more flexible. It is supposed to remain
5250 backwards compatible. Also added extra_local_interfaces.
5252 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5253 to bomb out with an assertion failure - to the client this appears as a
5254 connection drop. This problem occurs in the part of the code that was taken
5255 from the Samba project. Fortunately, the assertion is in a very simple
5256 function, so I have fixed this by reproducing the function inline in the
5257 one place where it is called, and arranging for authentication to fail
5258 instead of killing the process with assert().
5260 75. The SPA client code was not working when the server requested OEM rather
5261 than Unicode encoding.
5263 76. Added code to make require_files with a specific uid setting more usable in
5264 the case where statting the file as root fails - usually a non-root-mounted
5265 NFS file system. When this happens and the failure is EACCES, Exim now
5266 forks a subprocess and does the per-uid checking as the relevant uid.
5268 77. Added process_log_path.
5270 78. If log_file_path was not explicitly set, a setting of check_log_space or
5271 check_log_inodes was ignored.
5273 79. If a space check for the spool or log partitions fails, the incident is now
5274 logged. Of course, in the latter case the data may get lost...
5276 80. Added the %p formatting code to string_format() so that it can be used to
5277 print addresses in debug_print(). Adjusted all the address printing in the
5278 debugging in store.c to use %p rather than %d.
5280 81. There was a concern that a line of code in smtp_in.c could overflow a
5281 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5282 initially expressed, the concern was not well-founded, because trailing
5283 spaces are removed early. However, if the trailing spaces were followed by
5284 a NULL, they did not get removed, so the overflow was possible. Two fixes
5287 (a) I re-wrote the offending code in a cleaner fashion.
5288 (b) If an incoming SMTP command contains a NULL character, it is rejected
5291 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5292 runs initgroups(), so that if the Exim user is in any additional groups,
5293 they will be used during message reception.
5299 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob