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.
51 TF/01 Failure to get a lock on a hints database can have serious
52 consequences so log it to the panic log.
54 TF/02 Log LMTP confirmation messages in the same way as SMTP,
55 controlled using the smtp_confirmation log selector.
57 TF/03 Include the error message when we fail to unlink a spool file.
59 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
60 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
61 for maintaining out-of-tree patches for some time.
63 PP/01 Bugzilla 139: Documentation and portability issues.
64 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
65 Handle per-OS dynamic-module compilation flags.
67 PP/02 Let /dev/null have normal permissions.
68 The 4.73 fixes were a little too stringent and complained about the
69 permissions on /dev/null. Exempt it from some checks.
70 Reported by Andreas M. Kirchwitz.
72 PP/03 Report version information for many libraries, including
73 Exim version information for dynamically loaded libraries. Created
74 version.h, now support a version extension string for distributors
75 who patch heavily. Dynamic module ABI change.
77 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
78 privilege escalation vulnerability whereby the Exim run-time user
79 can cause root to append content of the attacker's choosing to
82 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
85 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
86 If dropping privileges for untrusted macros, we disabled normal logging
87 on the basis that it would fail; for the Exim run-time user, this is not
88 the case, and it resulted in successful deliveries going unlogged.
89 Fixed. Reported by Andreas Metzler.
95 PP/01 Date: & Message-Id: revert to normally being appended to a message,
96 only prepend for the Resent-* case. Fixes regression introduced in
97 Exim 4.70 by NM/22 for Bugzilla 607.
99 PP/02 Include check_rfc2047_length in configure.default because we're seeing
100 increasing numbers of administrators be bitten by this.
102 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
104 PP/03 Bugzilla 994: added openssl_options main configuration option.
106 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
108 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
110 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
112 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
113 without a peer certificate, leading to a segfault because of an
114 assumption that peers always have certificates. Be a little more
115 paranoid. Problem reported by Martin Tscholak.
117 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
118 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
119 NB: ClamAV planning to remove STREAM in "middle of 2010".
120 CL also introduces -bmalware, various -d+acl logging additions and
121 more caution in buffer sizes.
123 PP/09 Implemented reverse_ip expansion operator.
125 PP/10 Bugzilla 937: provide a "debug" ACL control.
127 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
129 PP/12 Bugzilla 973: Implement --version.
131 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
133 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
135 PP/15 Bugzilla 816: support multiple condition rules on Routers.
137 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
138 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
139 ignore trailing whitespace.
141 JJ/02 prevent non-panic DKIM error from being sent to paniclog
143 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
146 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
147 Notification from Dr Andrew Aitchison.
149 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
150 ExtendedDetectionInfo response format.
151 Notification from John Horne.
153 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
156 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
157 XSL and documented dependency on system catalogs, with examples of how
160 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
163 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
164 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
165 configuration file which is writeable by the Exim user or group.
167 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
168 of configuration files to cover files specified with the -C option if
169 they are going to be used with root privileges, not just the default
172 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
173 option (effectively making it always true).
175 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
176 files to be used while preserving root privileges.
178 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
179 that rogue child processes cannot use them.
181 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
182 run-time user, instead of root.
184 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
185 Exim run-time user without dropping privileges.
187 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
188 result string, instead of calling string_vformat() twice with the same
191 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
192 for other users. Others should always drop root privileges if they use
193 -C on the command line, even for a whitelisted configure file.
195 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
197 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
203 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
204 $data_path, and $header_path variables; fixed documentation bugs and
207 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
208 exipick to access non-standard spools, including the "frozen" queue
211 NM/01 Bugzilla 965: Support mysql stored procedures.
212 Patch from Alain Williams
214 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
216 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
217 Patch from Andreas Metzler
219 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
220 Patch from Kirill Miazine
222 NM/05 Bugzilla 671: Added umask to procmail example.
224 JJ/03 installed exipick 20100323.0, fixing doc bug
226 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
227 directory. Notification and patch from Dan Rosenberg.
229 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
231 TK/02 Improve log output when DKIM signing operation fails.
233 MH/01 Treat the transport option dkim_domain as a colon separated
234 list, not as a single string, and sign the message with each element,
235 omitting multiple occurences of the same signer.
237 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
238 Bugzilla 985, 986. Patch by Simon Arlott
240 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
241 Patch by Simon Arlott
243 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
244 MBX locking. Notification from Dan Rosenberg.
250 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
252 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
254 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
256 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
258 NM/04 Bugzilla 915: Flag broken perl installation during build.
264 TK/01 Added patch by Johannes Berg that expands the main option
265 "spamd_address" if it starts with a dollar sign.
267 TK/02 Write list of recipients to X-Envelope-Sender header when building
268 the mbox-format spool file for content scanning (suggested by Jakob
271 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
272 (http://www.dcc-servers.net/) support via dccifd. Activated by
273 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
275 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
276 by Mark Daniel Reidel <mr@df.eu>.
278 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
279 When building exim an external PCRE library is now needed -
280 PCRE is a system library on the majority of modern systems.
281 See entry on PCRE_LIBS in EDITME file.
283 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
284 conversation. Added nologin parameter to request.
285 Patch contributed by Kirill Miazine.
287 TF/01 Do not log submission mode rewrites if they do not change the address.
289 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
291 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
292 log files in place. Contributed by Roberto Lima.
294 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
296 TF/03 Bugzilla 615: When checking the local_parts router precondition
297 after a local_part_suffix or local_part_prefix option, Exim now
298 does not use the address's named list lookup cache, since this
299 contains cached lookups for the whole local part.
301 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
302 Robert Millan. Documentation is in experimental-spec.txt.
304 TF/04 Bugzilla 668: Fix parallel build (make -j).
306 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
308 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
309 Patch provided by Jan Srzednicki.
311 TF/05 Leading white space used to be stripped from $spam_report which
312 wrecked the formatting. Now it is preserved.
314 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
315 that they are available at delivery time.
317 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
319 TF/08 TLS error reporting now respects the incoming_interface and
320 incoming_port log selectors.
322 TF/09 Produce a more useful error message if an SMTP transport's hosts
323 setting expands to an empty string.
325 NM/06 Bugzilla 744: EXPN did not work under TLS.
326 Patch provided by Phil Pennock.
328 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
329 Patch provided by Richard Godbee.
331 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
332 acl_smtp_notquit, added index entry.
334 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
335 Patch provided by Eugene Bujak.
337 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
338 accept(). Patch provided by Maxim Dounin.
340 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
341 Patch provided by Phil Pennock.
343 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
345 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
346 Patch provided by Brad "anomie" Jorsch.
348 NM/14 Bugzilla 622: Added timeout setting to transport filter.
349 Patch provided by Dean Brooks.
351 TK/05 Add native DKIM support (does not depend on external libraries).
353 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
354 Patch provided by Graeme Fowler.
356 NM/16 Bugzilla 851: Documentation example syntax fix.
358 NM/17 Changed NOTICE file to remove references to embedded PCRE.
360 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
363 NM/19 Bugzilla 745: TLS version reporting.
364 Patch provided by Phil Pennock.
366 NM/20 Bugzilla 167: bool: condition support.
367 Patch provided by Phil Pennock.
369 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
370 clients. Patch provided by Phil Pennock.
372 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
373 Patch provided by Brad "anomie" Jorsch.
375 NM/23 Bugzilla 687: Fix misparses in eximstats.
376 Patch provided by Heiko Schlittermann.
378 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
379 Patch provided by Heiko Schlittermann.
381 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
382 plus update to original patch.
384 NM/26 Bugzilla 799: Documentation correction for ratelimit.
386 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
387 Patch provided by David Brownlee.
389 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
391 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
393 NM/30 Bugzilla 888: TLS documentation bugfixes.
395 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
397 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
398 Unlike the original bugzilla I have changed all shell scripts in src tree.
400 NM/33 Bugzilla 898: Transport filter timeout fix.
401 Patch by Todd Rinaldo.
403 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
404 Patch by Serge Demonchaux.
406 NM/35 Bugzilla 39: Base64 decode bug fixes.
407 Patch by Jakob Hirsch.
409 NM/36 Bugzilla 909: Correct connect() call in dcc code.
411 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
413 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
415 NM/39 Bugzilla 911: Fixed MakeLinks build script.
421 TK/01 Add preliminary DKIM support. Currently requires a forked version of
422 ALT-N's libdkim that I have put here:
423 http://duncanthrax.net/exim-experimental/
425 Note to Michael Haardt: I had to rename some vars in sieve.c. They
426 were called 'true' and it seems that C99 defines that as a reserved
427 keyword to be used with 'bool' variable types. That means you could
428 not include C99-style headers which use bools without triggering
429 build errors in sieve.c.
431 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
432 as mailq or other aliases. Changed the --help handling significantly
433 to do whats expected. exim_usage() emits usage/help information.
435 SC/01 Added the -bylocaldomain option to eximstats.
437 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
439 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
441 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
447 PH/01 Another patch from the Sieve maintainer.
449 PH/02 When an IPv6 address is converted to a string for single-key lookup
450 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
451 dots are used instead of colons so that keys in lsearch files need not
452 contain colons. This was done some time before quoting was made available
453 in lsearch files. However, iplsearch files do require colons in IPv6 keys
454 (notated using the quote facility) so as to distinguish them from IPv4
455 keys. This meant that lookups for IP addresses in host lists did not work
456 for iplsearch lookups.
458 This has been fixed by arranging for IPv6 addresses to be expressed with
459 colons if the lookup type is iplsearch. This is not incompatible, because
460 previously such lookups could never work.
462 The situation is now rather anomolous, since one *can* have colons in
463 ordinary lsearch keys. However, making the change in all cases is
464 incompatible and would probably break a number of configurations.
466 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
469 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
470 conversion specification without a maximum field width, thereby enabling
471 a rogue spamd server to cause a buffer overflow. While nobody in their
472 right mind would setup Exim to query an untrusted spamd server, an
473 attacker that gains access to a server running spamd could potentially
474 exploit this vulnerability to run arbitrary code as the Exim user.
476 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
477 $primary_hostname instead of what libspf2 thinks the hosts name is.
479 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
480 a directory entry by the name of the lookup key. Previously, if a
481 symlink pointed to a non-existing file or a file in a directory that
482 Exim lacked permissions to read, a lookup for a key matching that
483 symlink would fail. Now it is enough that a matching directory entry
484 exists, symlink or not. (Bugzilla 503.)
486 PH/03 The body_linecount and body_zerocount variables are now exported in the
489 PH/04 Added the $dnslist_matched variable.
491 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
492 This means they are set thereafter only if the connection becomes
495 PH/06 Added the client_condition to authenticators so that some can be skipped
496 by clients under certain conditions.
498 PH/07 The error message for a badly-placed control=no_multiline_responses left
499 "_responses" off the end of the name.
501 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
503 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
504 (without spaces) instead of just copying the configuration text.
506 PH/10 Added the /noupdate option to the ratelimit ACL condition.
508 PH/11 Added $max_received_linelength.
510 PH/12 Added +ignore_defer and +include_defer to host lists.
512 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
513 way in which PCRE > 7.0 is built.
515 PH/14 Implemented queue_only_load_latch.
517 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
518 MAIL command. The effect was to mangle the value on 64-bit systems.
520 PH/16 Another patch from the Sieve maintainer.
522 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
524 PH/18 If a system quota error occurred while trying to create the file for
525 a maildir delivery, the message "Mailbox is full" was not appended to the
526 bounce if the delivery eventually timed out. Change 4.67/27 below applied
527 only to a quota excession during the actual writing of the file.
529 PH/19 It seems that peer DN values may contain newlines (and other non-printing
530 characters?) which causes problems in log lines. The DN values are now
531 passed through string_printing() before being added to log lines.
533 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
534 and InterBase are left for another time.)
536 PH/21 Added message_body_newlines option.
538 PH/22 Guard against possible overflow in moan_check_errorcopy().
540 PH/23 POSIX allows open() to be a macro; guard against that.
542 PH/24 If the recipient of an error message contained an @ in the local part
543 (suitably quoted, of course), incorrect values were put in $domain and
544 $local_part during the evaluation of errors_copy.
550 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
551 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
554 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
555 issue a MAIL command.
557 PH/02 In an ACL statement such as
559 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
561 if a client was not listed at all, or was listed with a value other than
562 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
563 the condition was not true (as it should be), so access was not denied.
564 The bug was that the ! inversion was incorrectly passed on to the second
565 item. This has been fixed.
567 PH/03 Added additional dnslists conditions == and =& which are different from
568 = and & when the dns lookup returns more than one IP address.
570 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
571 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
573 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
574 FSYNC, which compiles an option called disable_fsync that allows for
575 bypassing fsync(). The documentation is heavily laced with warnings.
577 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
579 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
580 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
581 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
582 including adding "make clean"; (3) Added -fPIC when compiling the test
583 dynamically loaded module, to get rid of a warning.
585 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
586 message fails, move_frozen_messages = true and ignore_bounce_errors_after
587 = 0s. The bug is otherwise harmless.
589 PH/07 There was a bug in the dovecot authenticator such that the value of
590 $auth1 could be overwritten, and so not correctly preserved, after a
591 successful authentication. This usually meant that the value preserved by
592 the server_setid option was incorrect.
594 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
596 PH/09 Installed PCRE release 7.0.
598 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
599 run for batched SMTP input. It is now run at the start of every message
600 in the batch. While fixing this I discovered that the process information
601 (output by running exiwhat) was not always getting set for -bs and -bS
602 input. This is fixed, and it now also says "batched" for BSMTP.
604 PH/11 Added control=no_pipelining.
606 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
607 patch, slightly modified), and move the expansion of helo_data till after
608 the connection is made in the smtp transport (so it can use these
611 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
613 PH/14 Added log_selector = +pid.
615 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
617 PH/16 Add ${if forany and ${if forall.
619 PH/17 Added dsn_from option to vary the From: line in DSNs.
621 PH/18 Flush SMTP output before performing a callout, unless control =
622 no_callout_flush is set.
624 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
625 was true (the default) a successful delivery failed to delete the retry
626 item, thus causing premature timeout of the address. The bug is now
629 PH/20 Added hosts_avoid_pipelining to the smtp transport.
631 PH/21 Long custom messages for fakedefer and fakereject are now split up
632 into multiline reponses in the same way that messages for "deny" and
633 other ACL rejections are.
635 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
636 with slight modification.
638 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
639 draft, changing the syntax and factoring some duplicate code".
641 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
642 for deliveries of the second and subsequent messages over the same SMTP
645 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
646 ${reduce, with only minor "tidies".
648 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
650 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
651 expansion side effects.
653 PH/27 When a message times out after an over-quota error from an Exim-imposed
654 quota, the bounce message says "mailbox is full". This message was not
655 being given when it was a system quota that was exceeded. It now should
658 MH/03 Made $recipients available in local_scan(). local_scan() already has
659 better access to the recipient list through recipients_list[], but
660 $recipients can be useful in postmaster-provided expansion strings.
662 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
663 in the case of a MAIL command with additional options following the
664 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
665 were accidentally chopped off.
667 PH/29 SMTP synchronization checks are implemented when a command is read -
668 there is a check that no more input is waiting when there shouldn't be
669 any. However, for some commands, a delay in an ACL can mean that it is
670 some time before the response is written. In this time, more input might
671 arrive, invalidly. So now there are extra checks after an ACL has run for
672 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
673 pipelining has not been advertised.
675 PH/30 MH's patch to allow iscntrl() characters to be list separators.
677 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
678 returned in the SMTP response when smtp_return_error_details was false.
681 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
682 instead of the C I/O that was originally supplied, because problems were
685 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
686 Exim which did not show up earlier: it was assuming that a call to
687 SSL_CTX_set_info_callback() might give an error value. In fact, there is
688 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
689 was a macro that became an assignment, so it seemed to work. This has
690 changed to a proper function call with a void return, hence the compile
691 error. Exim's code has been fixed.
693 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
696 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
698 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
700 PH/37 If a message is not accepted after it has had an id assigned (e.g.
701 because it turns out to be too big or there is a timeout) there is no
702 "Completed" line in the log. When some messages of this type were
703 selected by exigrep, they were listed as "not completed". Others were
704 picked up by some special patterns. I have improved the selection
705 criteria to be more general.
707 PH/38 The host_find_failed option in the manualroute router can now be set
708 to "ignore", to completely ignore a host whose IP address cannot be
709 found. If all hosts are ignored, the behaviour is controlled by the new
710 host_all_ignored option.
712 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
713 homing or because of multiple MX records with /mx) generated more than
714 one IP address, and the following item turned out to be the local host,
715 all the secondary addresses of the first item were incorrectly removed
716 from the list, along with the local host and any following hosts (which
717 is what is supposed to happen).
719 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
720 whoever called Exim into the -H file. In the case of the daemon it was
721 behaving confusingly. When first started, it used values for whoever
722 started the daemon, but after a SIGHUP it used the Exim user (because it
723 calls itself on a restart). I have changed the code so that it now always
726 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
727 message are rejected with the same error (e.g. no authentication or bad
728 sender address), and a DATA command is nevertheless sent (as can happen
729 with PIPELINING or a stupid MUA), the error message that was given to the
730 RCPT commands is included in the rejection of the DATA command. This is
731 intended to be helpful for MUAs that show only the final error to their
734 PH/42 Another patch from the Sieve maintainer.
736 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
737 Eximstats - Fixed some broken HTML links and added missing column headers
739 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
740 columns for Rejects, Temp Rejects, Ham, and Spam rows.
742 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
744 PH/43 Yet another patch from the Sieve maintainer.
746 PH/44 I found a way to check for a TCP/IP connection going away before sending
747 the response to the final '.' that terminates a message, but only in the
748 case where the client has not sent further data following the '.'
749 (unfortunately, this is allowed). However, in many cases there won't be
750 any further data because there won't be any more messages to send. A call
751 to select() can be used: if it shows that the input is "ready", there is
752 either input waiting, or the socket has been closed. An attempt to read
753 the next input character can distinguish the two cases. Previously, Exim
754 would have sent an OK response which the client would never have see.
755 This could lead to message repetition. This fix should cure that, at
756 least in a lot of common cases.
758 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
759 advertised in response to EHLO.
765 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
766 fixed by 4.65/MH/01 (is this a record?) are fixed:
768 (i) An empty string was always treated as zero by the numeric comparison
769 operators. This behaviour has been restored.
771 (ii) It is documented that the numeric comparison operators always treat
772 their arguments as decimal numbers. This was broken in that numbers
773 starting with 0 were being interpreted as octal.
775 While fixing these problems I realized that there was another issue that
776 hadn't been noticed. Values of message_size_limit (both the global option
777 and the transport option) were treated as octal if they started with 0.
778 The documentation was vague. These values are now always treated as
779 decimal, and I will make that clear in the documentation.
785 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
786 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
789 MH/01 Don't check that the operands of numeric comparison operators are
790 integers when their expansion is in "skipping" mode (fixes bug
791 introduced by 4.64-PH/07).
793 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
794 child addresses, Exim now panics and dies. Previously, because the count
795 is held in a short int, deliveries were likely to be lost. As such a
796 large number of recipients for a single message is ridiculous
797 (performance will be very, very poor), I have chosen to impose a limit
798 rather than extend the field.
804 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
805 leftover -K file (the existence of which was triggered by #402).
806 While we were at it, introduced process PID as part of the -K
807 filename. This should rule out race conditions when creating
810 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
811 processing considerably. Previous code took too long for large mails,
812 triggering a timeout which in turn triggers #401.
814 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
815 in the DK code in transports.c. sendfile() is not really portable,
816 hence the _LINUX specificness.
818 TF/01 In the add_headers option to the mail command in an Exim filter,
819 there was a bug that Exim would claim a syntax error in any
820 header after the first one which had an odd number of characters
823 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
824 callout verification, Exim cached a "reject" for the entire domain. This
825 is correct for most verifications, but it is not correct for a recipient
826 verification with use_sender or use_postmaster set, because in that case
827 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
828 case of MAIL FROM:<> rejection from other early rejections (e.g.
829 rejection of HELO). When verifying a recipient using a non-null MAIL
830 address, the cache is ignored if it shows MAIL FROM:<> rejection.
831 Whatever the result of the callout, the value of the domain cache is
832 left unchanged (for any other kind of callout, getting as far as trying
833 RCPT means that the domain itself is ok).
835 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
838 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
839 manifest itself as EPIPE rather than ECONNECT. When tidying away a
840 session, the daemon ignores ECONNECT errors and logs others; it now
841 ignores EPIPE as well.
843 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
844 (quoted-printable decoding).
846 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
847 later the small subsequent patch to fix an introduced bug.
849 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
851 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
853 PH/08 An error is now given if message_size_limit is specified negative.
855 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
856 to be given (somewhat) arbitrary names.
858 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
861 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
862 miscellaneous code fixes
864 PH/10 Added the log_reject_target ACL modifier to specify where to log
867 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
868 hostname. This is wrong, because it relates to the incoming message (and
869 probably the interface on which it is arriving) and not to the outgoing
870 callout (which could be using a different interface). This has been
871 changed to use the value of the helo_data option from the smtp transport
872 instead - this is what is used when a message is actually being sent. If
873 there is no remote transport (possible with a router that sets up host
874 addresses), $smtp_active_hostname is used.
876 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
877 tweaks were necessary in order to get it to work (see also 21 below):
878 (a) The code assumed that strncpy() returns a negative number on buffer
879 overflow, which isn't the case. Replaced with Exim's string_format()
881 (b) There were several signed/unsigned issues. I just did the minimum
882 hacking in of casts. There is scope for a larger refactoring.
883 (c) The code used strcasecmp() which is not a standard C function.
884 Replaced with Exim's strcmpic() function.
885 (d) The code set only $1; it now sets $auth1 as well.
886 (e) A simple test gave the error "authentication client didn't specify
887 service in request". It would seem that Dovecot has changed its
888 interface. Fortunately there's a specification; I followed it and
889 changed what the client sends and it appears to be working now.
891 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
894 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
895 address A is aliased to B and C, where B exists and C does not. Without
896 -v the output is "A verified" because verification stops after a
897 successful redirection if more than one address is generated. However,
898 with -v the child addresses are also verified. Exim was outputting "A
899 failed to verify" and then showing the successful verification for C,
900 with its parentage. It now outputs "B failed to verify", showing B's
901 parentage before showing the successful verification of C.
903 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
904 look up a TXT record in a specific list after matching in a combined
907 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
908 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
909 they consult the DNS. I had assumed they would set it the way they
910 wanted; and indeed my experiments on Linux seem to show that in some
911 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
912 To be on the safe side, however, I have now made the interface to
913 host_find_byname() similar to host_find_bydns(), with an argument
914 containing the DNS resolver options. The host_find_byname() function now
915 sets these options at its start, just as host_find_bydns() does. The smtp
916 transport options dns_qualify_single and dns_search_parents are passed to
917 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
918 of host_find_byname() use the default settings of RES_DEFNAMES
919 (qualify_single) but not RES_DNSRCH (search_parents).
921 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
922 spool_read_header() do less string testing, by means of a preliminary
923 switch on the second character of optional "-foo" lines. (This is
924 overdue, caused by the large number of possibilities that now exist.
925 Originally there were few.) While I was there, I also converted the
926 str(n)cmp tests so they don't re-test the leading "-" and the first
927 character, in the hope this might squeeze out yet more improvement.
929 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
930 flag allowing group syntax was set by the header_syntax check but not
931 turned off, possible causing trouble later; (2) The flag was not being
932 set at all for the header_verify test, causing "group"-style headers to
933 be rejected. I have now set it in this case, and also caused header_
934 verify to ignore an empty address taken from a group. While doing this, I
935 came across some other cases where the code for allowing group syntax
936 while scanning a header line wasn't quite right (mostly, not resetting
937 the flag correctly in the right place). These bugs could have caused
938 trouble for malformed header lines. I hope it is now all correct.
940 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
941 with the "reply" argument non-NULL. The code, however (which originally
942 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
943 but it didn't always do it. This confused somebody who was copying the
944 code for some other use. I have removed all the tests.
946 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
947 feature that was used to support insecure browsers during the U.S. crypto
948 embargo. It requires special client support, and Exim is probably the
949 only MTA that supported it -- and would never use it because real RSA is
950 always available. This code has been removed, because it had the bad
951 effect of slowing Exim down by computing (never used) parameters for the
952 RSA_EXPORT functionality.
954 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
955 authenticator to fail if there's a tab character in the incoming data
956 (there should never be unless someone is messing about, as it's supposed
957 to be base64-encoded). Also added, on Timo's advice, the "secured" option
958 if the connection is using TLS or if the remote IP is the same as the
959 local IP, and the "valid-client-cert option" if a client certificate has
962 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
963 authenticators. This can be used for authorization after authentication
964 succeeds. (In the case of plaintext, it servers for both authentication
967 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
968 if any retry times were supplied.
970 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
971 connection, where there is no client IP address to check. In this
972 situation, the verify now always succeeds.
974 PH/25 Applied John Jetmore's -Mset patch.
976 PH/26 Added -bem to be like -Mset, but loading a message from a file.
978 PH/27 In a string expansion for a processed (not raw) header when multiple
979 headers of the same name were present, leading whitespace was being
980 removed from all of them, but trailing whitespace was being removed only
981 from the last one. Now trailing whitespace is removed from each header
982 before concatenation. Completely empty headers in a concatenation (as
985 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
986 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
988 PH/29 [Removed. This was a change that I later backed out, and forgot to
989 correct the ChangeLog entry (that I had efficiently created) before
990 committing the later change.]
992 PH/30 Exim was sometimes attempting to deliver messages that had suffered
993 address errors (4xx response to RCPT) over the same connection as other
994 messages routed to the same hosts. Such deliveries are always "forced",
995 so retry times are not inspected. This resulted in far too many retries
996 for the affected addresses. The effect occurred only when there were more
997 hosts than the hosts_max_try setting in the smtp transport when it had
998 the 4xx errors. Those hosts that it had tried were not added to the list
999 of hosts for which the message was waiting, so if all were tried, there
1000 was no problem. Two fixes have been applied:
1002 (i) If there are any address or message errors in an SMTP delivery, none
1003 of the hosts (tried or untried) are now added to the list of hosts
1004 for which the message is waiting, so the message should not be a
1005 candidate for sending over the same connection that was used for a
1006 successful delivery of some other message. This seems entirely
1007 reasonable: after all the message is NOT "waiting for some host".
1008 This is so "obvious" that I'm not sure why it wasn't done
1009 previously. Hope I haven't missed anything, but it can't do any
1010 harm, as the worst effect is to miss an optimization.
1012 (ii) If, despite (i), such a delivery is accidentally attempted, the
1013 routing retry time is respected, so at least it doesn't keep
1014 hammering the server.
1016 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1017 in ${readsocket because some servers need this prod.
1019 PH/32 Added some extra debug output when updating a wait-xxx database.
1021 PH/33 The hint "could be header name not terminated by colon", which has been
1022 given for certain expansion errors for a long time, was not being given
1023 for the ${if def:h_colon_omitted{... case.
1025 PH/34 The spec says: "With one important exception, whenever a domain list is
1026 being scanned, $domain contains the subject domain." There was at least
1027 one case where this was not true.
1029 PH/35 The error "getsockname() failed: connection reset by peer" was being
1030 written to the panic log as well as the main log, but it isn't really
1031 panic-worthy as it just means the connection died rather early on. I have
1032 removed the panic log writing for the ECONNRESET error when getsockname()
1035 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1036 runs only) independently of the message's sender address. This meant
1037 that, if the 4xx error was in fact related to the sender, a different
1038 message to the same recipient with a different sender could confuse
1039 things. In particualar, this can happen when sending to a greylisting
1040 server, but other circumstances could also provoke similar problems.
1041 I have changed the default so that the retry time for these errors is now
1042 based a combination of the sender and recipient addresses. This change
1043 can be overridden by setting address_retry_include_sender=false in the
1046 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1047 remote server are returned as part of bounce messages. This was not
1048 happening for LMTP over a pipe (the lmtp transport), but now it is the
1049 same for both kinds of LMTP.
1051 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1052 in header lines that were in fact CNAMEs. This is no longer the case.
1054 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1055 and queue runs started by the daemon processed all messages. This has
1056 been fixed so that -R and -S can now usefully be given with -q<time>.
1058 PH/40 Import PCRE release 6.7 (fixes some bugs).
1060 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1062 PH/42 Give an error if -q is specified more than once.
1064 PH/43 Renamed the variables $interface_address and $interface_port as
1065 $received_ip_address and $received_port, to make it clear that these
1066 values apply to message reception, and not to the outgoing interface when
1067 a message is delivered. (The old names remain recognized, of course.)
1069 PH/44 There was no timeout on the connect() call when using a Unix domain
1070 socket in the ${readsocket expansion. There now is.
1072 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1073 be meaningful with "accept".
1075 SC/01 Eximstats V1.43
1076 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1078 SC/02 Eximstats V1.44
1079 Use a glob alias rather than an array ref in the generated
1080 parser. This improves both readability and performance.
1082 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1083 Collect SpamAssassin and rejection statistics.
1084 Don't display local sender or destination tables unless
1085 there is data to show.
1086 Added average volumes into the top table text output.
1088 SC/04 Eximstats V1.46
1089 Collect data on the number of addresses (recipients)
1090 as well as the number of messages.
1092 SC/05 Eximstats V1.47
1093 Added 'Message too big' to the list of mail rejection
1094 reasons (thanks to Marco Gaiarin).
1096 SC/06 Eximstats V1.48
1097 Mainlog lines which have GMT offsets and are too short to
1098 have a flag are now skipped.
1100 SC/07 Eximstats V1.49 (Alain Williams)
1101 Added the -emptyok flag.
1103 SC/08 Eximstats V1.50
1104 Fixes for obtaining the IP address from reject messages.
1106 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1107 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1108 whitesspace changes from 4.64-PH/27
1110 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1113 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1114 are found, allow negative numbers in numeric criteria)
1116 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1118 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1119 to match changes made in 4.64-PH/43
1121 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1123 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1124 "domain" is actually an IP address, and force a failure. This locks out
1125 those revolvers/nameservers that support "A-for-A" lookups, in
1126 contravention of the specifications.
1128 PH/48 When a host name was looked up from an IP address, and the subsequent
1129 forward lookup of the name timed out, the host name was left in
1130 $sender_host_name, contrary to the specification.
1132 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1133 restricted to single-key lookups, Exim was not diagnosing an error if
1134 * or *@ was used with a query-style lookup.
1136 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1138 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1139 long ago, but noone interested enough thought of it. Let's just say that
1140 the "1.1" means that there are some new functions that weren't there at
1141 some point in the past.
1143 PH/51 Error processing for expansion failure of helo_data from an smtp
1144 transport during callout processing was broken.
1146 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1147 tested/used via the -bh/-bhc/-bs options.
1149 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1150 bug, fixed in subsequent PCRE releases).
1152 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1153 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1155 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1161 SC/01 Use a glob alias rather than an array ref in eximstats generated
1162 parser. This improves both readability and performance.
1164 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1165 Don't display local sender or destination tables in eximstats unless
1166 there is data to show.
1167 Added average volumes into the eximstats top table text output.
1169 SC/03 Collect data on the number of addresses (recipients) as well
1170 as the number of messages in eximstats.
1172 TF/01 Correct an error in the documentation for the redirect router. Exim
1173 does (usually) call initgroups() when daemonizing.
1175 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1176 with consistent privilege compared to when running as a daemon.
1178 TF/03 Note in the spec that $authenticated_id is not set for local
1179 submissions from trusted users.
1181 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1182 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1184 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1185 by adding some example configuration directives to the default
1186 configuration file. A little bit of work is required to uncomment the
1187 directives and define how usernames and passwords are checked, but
1188 there is now a framework to start from.
1190 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1191 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1192 without this. I don't know how relevant this is to other LDAP libraries.
1194 PH/02 Add the verb name to the "unknown ACL verb" error.
1196 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1198 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1200 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1201 directory not expanded when it should be if an expanded home directory
1202 was set for the address (which is overridden by the transport).
1204 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1207 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1208 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1209 because it is too late at that time, and has no effect.
1211 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1212 security issue with \' (bugzilla #107). I could not use the
1213 PQescapeStringConn() function, because it needs a PGconn value as one of
1216 PH/08 When testing addresses using -bt, indicate those final addresses that
1217 are duplicates that would not cause an additional delivery. At least one
1218 person was confused, thinking that -bt output corresponded to deliveries.
1219 (Suppressing duplicates isn't a good idea as you lose the information
1220 about possibly different redirections that led to the duplicates.)
1222 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1223 systems where poll() doesn't work, in particular OS X.
1225 PH/10 Added more information to debugging output for retry time not reached.
1227 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1228 operations in malware.c.
1230 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1233 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1234 syslog_duplication set false, log lines that would normally be written
1235 both the the main log and to the reject log were not written to syslog at
1238 PH/14 In the default configuration, change the use of "message" in ACL warn
1239 statements to "add_header".
1241 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1242 not followed by a command (e.g. "seen endif").
1244 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1245 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1248 PH/17 Added extra conditions to the default value of delay_warning_condition
1252 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1253 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1254 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1257 The Auto-Submitted: and various List- headers are standardised, whereas I
1258 don't think Precedence: ever was.
1260 PH/18 Refactored debugging code in route_finduser() to show more information,
1261 in particular, the error code if getpwnam() issues one.
1263 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1264 This is apparently needed in addition to the PH/07 change above to avoid
1265 any possible encoding problems.
1267 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1268 but not after initializing Perl.
1270 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1271 output them only if debugging. By default they are written stderr,
1272 apparently, which is not desirable.
1274 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1277 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1280 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1282 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1283 authenticated or an ident call has been made. Suppress the default
1284 values for $authenticated_id and $authenticated_sender (but permit -oMai
1285 and -oMas) when testing with -bh.
1287 PH/24 Re-jigged the order of the tests in the default configuration so that the
1288 tests for valid domains and recipients precede the DNS black list and CSA
1289 tests, on the grounds that those ones are more expensive.
1291 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1292 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1293 This bug exists in every version of Exim that I still have, right back to
1296 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1297 However, an attempt to turn on case-sensitivity in a regex key by
1298 including (?-i) didn't work because the subject string was already
1299 lowercased, and the effects were non-intuitive. It turns out that a
1300 one-line patch can be used to allow (?-i) to work as expected.
1306 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1307 other effects) broke the use of negated acl sub-conditions.
1309 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1312 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1313 "Deny" causes Exim to reject the incoming connection with a 554 error.
1314 Unfortunately, if there is a major crisis, such as a disk failure,
1315 tcp-wrappers gives "deny", whereas what one would like would be some
1316 kind of temporary error. A kludge has been added to help with this.
1317 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1318 554 error is used if errno is still zero or contains ENOENT (which occurs
1319 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1322 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1324 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1325 errors. Otherwise a message that provokes a temporary error (when other
1326 messages do not) can cause a whole host to time out.
1328 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1329 addresses were routed directly to files or pipes from a redirect router.
1330 File deliveries just didn't batch; pipe deliveries might have suffered
1333 PH/06 A failure to get a lock for a hints database would erroneously always say
1334 "Failed to get write lock", even when it was really a read lock.
1336 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1337 of 0600. This has been changed to use the value of the lockfile_mode
1338 option (which defaults to 0600).
1340 PH/08 Applied small patch from the Sieve maintainer.
1342 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1343 folder from quota calculations, a direct delivery into this folder messed
1344 up the contents of the maildirsize file. This was because the regex was
1345 used only to exclude .Trash (or whatever) when the size of the mailbox
1346 was calculated. There was no check that a delivery was happening into an
1347 excluded directory. This bug has been fixed by ignoring all quota
1348 processing for deliveries into excluded directories.
1350 PH/10 Added the maildirfolder_create_regex option to appendfile.
1356 PH/01 The code for finding all the local interface addresses on a FreeBSD
1357 system running IPv6 was broken. This may well have applied to all BSD
1358 systems, as well as to others that have similar system calls. The broken
1359 code found IPv4 interfaces correctly, but gave incorrect values for the
1360 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1361 that it would not match correctly against @[] and not recognize the IPv6
1364 PH/02 The ipliteral router was not recognizing addresses of the form user@
1365 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1367 PH/03 Added disable_ipv6.
1369 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1370 decoded form, because it is most often used to construct To: headers
1371 lines in autoreplies, and the decoded form may well be syntactically
1372 invalid. However, $reply_address has leading white space removed, and all
1373 newlines turned into spaces so that the autoreply transport does not
1376 PH/05 If group was specified without a user on a router, and no group or user
1377 was specified on a transport, the group from the router was ignored.
1379 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1380 for visible compile-time settings that can be used to change these
1381 numbers, for those that want even more. Backwards compatibility with old
1382 spool files has been maintained. However, going back to a previous Exim
1383 release will lost any variables that are in spool files.
1385 PH/07 Two small changes when running in the test harness: increase delay when
1386 passing a TCP/IP connection to a new process, in case the original
1387 process has to generate a bounce, and remove special handling of
1388 127.0.0.2 (sic), which is no longer necessary.
1390 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1391 be the same on different OS.
1393 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1396 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1397 whether --show-vars was specified or not
1399 JJ/02 exipick: Added support for new ACL variable spool format introduced
1402 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1403 syntactically invalid From: or Reply-to: line, and a filter used this to
1404 generate an autoreply, and therefore failed to obtain an address for the
1405 autoreply, Exim could try to deliver to a non-existent relative file
1406 name, causing unrelated and misleading errors. What now happens is that
1407 it logs this as a hard delivery error, but does not attempt to create a
1410 PH/11 The exinext utility has a -C option for testing purposes, but although
1411 the given file was scanned by exinext itself; it wasn't being passed on
1412 when Exim was called.
1414 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1415 an end-of-file indication when reading a command response.
1417 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1418 compiled. In many other places in Exim, IPv6 addresses are always
1419 recognized, so I have changed this. It also means that IPv4 domain
1420 literals of the form [IPV4:n.n.n.n] are now always recognized.
1422 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1423 used if the router is not running as root, for example, when verifying at
1424 ACL time, or when using -bh. The debugging output from this situation was
1425 non-existent - all you got was a failure to exec. I have made two
1428 (a) Failures to set uid/gid, the current directory, or a process leader
1429 in a subprocess such as that created by queryprogram now generate
1430 suitable debugging ouput when -d is set.
1432 (b) The queryprogram router detects when it is not running as root,
1433 outputs suitable debugging information if -d is set, and then runs
1434 the subprocess without attempting to change uid/gid.
1436 PH/15 Minor change to Makefile for building test_host (undocumented testing
1439 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1440 additional section of a DNS packet that returns MX or SRV records.
1441 Instead, it always explicitly searches for A/AAAA records. This avoids
1442 major problems that occur when a DNS server includes only records of one
1443 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1444 fixed another bug: if SRV records were looked up and the corresponding
1445 address records were *not* found in the additional section, the port
1446 values from the SRV records were lost.
1448 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1449 using the correct key (the original address) when searching the retry
1450 rules in order to find which one to use for generating the retry hint.
1452 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1453 adding the default one. Similarly, if it contains a Reply-To: header, the
1454 errors_reply_to option, if set, is not used.
1456 PH/19 When calculating a retry time, Exim used to measure the "time since
1457 failure" by looking at the "first failed" field in the retry record. Now
1458 it does not use this if it is later than than the arrival time of the
1459 message. Instead it uses the arrival time. This makes for better
1460 behaviour in cases where some deliveries succeed, thus re-setting the
1461 "first failed" field. An example is a quota failure for a huge message
1462 when small messages continue to be delivered. Without this change, the
1463 "time since failure" will always be short, possible causing more frequent
1464 delivery attempts for the huge message than are intended.
1465 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1467 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1468 $1, $2, $3) because the numerical variables can be reset during some
1469 expansion items (e.g. "match"), thereby losing the authentication data.
1471 PH/21 Make -bV show the size of off_t variables so that the test suite can
1472 decide whether to run tests for quotas > 2G.
1474 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1475 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1476 filecount value is greater than 2G or if a quota value is greater than 2G
1477 on a system where the size of off_t is not greater than 4, a panic error
1480 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1481 never match. The debug and -bh output now contains an explicit error
1482 message indicating a malformed IPv4 address or mask.
1484 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1485 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1486 PH/23 above applies.
1488 PH/25 Do not write to syslog when running in the test harness. The only
1489 occasion when this arises is a failure to open the main or panic logs
1490 (for which there is an explicit test).
1492 PH/26 Added the /no_tell option to "control=freeze".
1494 PH/27 If a host name lookup failed very early in a connection, for example, if
1495 the IP address matched host_lookup and the reverse lookup yielded a name
1496 that did not have a forward lookup, an error message of the form "no IP
1497 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1498 could be logged. Now it outputs the IP address instead of "NULL".
1500 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1501 allows the sender and the authenticated sender to be set when
1502 submitting a message from within Exim. Since child_open_exim() is
1503 documented for local_scan(), the new function should be too.
1505 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1506 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1507 results in an empty string is now treated as unset.
1509 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1511 PH/31 Added sender_verify_fail logging option.
1513 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1514 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1515 tidied the source and removed it altogether.
1517 PH/33 When a queue run was abandoned because the load average was too high, a
1518 log line was always written; now it is written only if the queue_run log
1519 selector is set. In addition, the log line for abandonment now contains
1520 information about the queue run such as the pid. This is always present
1521 in "start" and "stop" lines but was omitted from the "abandon" line.
1523 PH/34 Omit spaces between a header name and the colon in the error message that
1524 is given when verify = headers_syntax fails (if there are lots of them,
1525 the message gets confusing).
1527 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1528 names, as there are now some PTR records that contain slashes. This check
1529 is only to protect against broken name servers that fall over on strange
1530 characters, so the fact that it applies to all lookups doesn't matter.
1532 PH/36 Now that the new test suite is complete, we can remove some of the
1533 special code in Exim that was needed for the old test suite. For example,
1534 sorting DNS records because real resolvers return them in an arbitrary
1535 order. The new test suite's fake resolver always returns records in the
1538 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1539 bounces) except when queue_only is set, to avoid logging races between
1540 the different processes.
1542 PH/38 Panic-die if .include specifies a non-absolute path.
1544 PH/39 A tweak to the "H" retry rule from its user.
1546 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1547 a label. They prevented compilation on older perls.
1549 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1550 a warning to be raised on newish perls.
1552 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1553 on queue. Changes to match documented behaviour of showing count of
1554 messages matching specified criteria.
1556 PH/40 Changed the default ident timeout from 30s to 5s.
1558 PH/41 Added support for the use of login_cap features, on those BSD systems
1559 that have them, for controlling the resources used by pipe deliveries.
1561 PH/42 The content-scanning code uses fopen() to create files in which to put
1562 message data. Previously it was not paying any attention to the mode of
1563 the files. Exim runs with umask(0) because the rest of the code creates
1564 files with open(), and sets the required mode explicitly. Thus, these
1565 files were ending up world-writeable. This was not a big issue, because,
1566 being within the spool directory, they were not world-accessible. I have
1567 created a function called modefopen, which takes an additional mode
1568 argument. It sets umask(777), creates the file, chmods it to the required
1569 mode, then resets the umask. All the relevant calls to fopen() in the
1570 content scanning code have been changed to use this function.
1572 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1573 to 24 hours. This avoids potential overflow problems when processing G
1574 and H retry rules. I suspect nobody ever tinkers with this value.
1576 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1578 PH/45 When the plaintext authenticator is running as a client, the server's
1579 challenges are checked to ensure they are valid base64 strings. By
1580 default, the authentication attempt is cancelled if an invalid string is
1581 received. Setting client_ignore_invalid_base64 true ignores these errors.
1582 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1583 they are received. Thus, the responses can be made to depend on the
1584 challenges. If an invalid string is ignored, an empty string is placed in
1587 PH/46 Messages that are created by the autoreply transport now contains a
1588 References: header, in accordance with RFCs 2822 and 3834.
1590 PH/47 Added authenticated_sender_force to the smtp transport.
1592 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1594 PH/49 Installed latest patch from the Sieve maintainer.
1596 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1597 was also set, the appendfile transport was unnecessarily scanning a
1598 directory of message files (e.g. for maildir delivery) to find the count
1599 of files (along with the size), even though it did not need this
1600 information. It now does the scan only if it needs to find either the
1601 size of the count of files.
1603 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1605 PH/52 Two bugs concerned with error handling when the smtp transport is
1608 (i) Exim was not creating retry information for temporary errors given
1609 for individual recipients after the DATA command when the smtp transport
1610 was used in LMTP mode. This meant that they could be retried too
1611 frequently, and not timed out correctly.
1613 (ii) Exim was setting the flag that allows error details to be returned
1614 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1615 recipients that were returned after the DATA command.
1617 PH/53 This is related to PH/52, but is more general: for any failing address,
1618 when detailed error information was permitted to be returned to the
1619 sender, but the error was temporary, then after the final timeout, only
1620 "retry timeout exceeded" was returned. Now it returns the full error as
1621 well as "retry timeout exceeded".
1623 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1624 do this, and (what is worse) MTAs that accept it.
1626 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1627 will now be deprecated.
1629 PH/56 New os.c-cygwin from the Cygwin maintainer.
1631 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1632 formats (previously only available in exim formats via -bpr, -bpru,
1633 and -bpra. Now also available in native and exiqgrep formats)
1635 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1636 with very large, slow to parse queues
1638 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1640 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1642 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1643 responses to authentication challenges, though it was showing the
1644 challenges; (ii) I've removed the CR characters from the debug output for
1647 PH/58 Allow for the insertion of a newline as well as a space when a string
1648 is turned into more than one encoded-word during RFC 2047 encoding. The
1649 Sieve code now uses this.
1651 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1652 data_4xx, lost_connection, tls_required.
1654 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1655 message was being sent as an SMTP response.
1657 PH/61 Add -l and -k options to exicyclog.
1659 PH/62 When verifying, if an address was redirected to one new address, so that
1660 verification continued, and the new address failed or deferred after
1661 having set something in $address_data, the value of $address_data was not
1662 passed back to the ACL. This was different to the case when no
1663 redirection occurred. The value is now passed back in both cases.
1665 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1666 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1667 use login_cap.h, so on its own it isn't the distinguishing feature. The
1668 new name refers directly to the setclassresources() function.
1670 PH/65 Added configuration files for NetBSD3.
1672 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1674 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1675 is preferred over IPv4.
1677 PH/68 The bounce_return_message and bounce_return_body options were not being
1678 honoured for bounces generated during the reception of non-SMTP messages.
1679 In particular, this applied to messages rejected by the ACL. This bug has
1680 been fixed. However, if bounce_return_message is true and bounce_return_
1681 body is false, the headers that are returned for a non-SMTP message
1682 include only those that have been read before the error was detected.
1683 (In the case of an ACL rejection, they have all been read.)
1685 PH/69 The HTML version of the specification is now built in a directory called
1686 spec_html instead of spec.html, because the latter looks like a path with
1687 a MIME-type, and this confuses some software.
1689 PH/70 Catch two compiler warnings in sieve.c.
1691 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1692 function verify_get_ident() calls ip_connect() to connect a socket, but
1693 if the "connect()" function timed out, ip_connect() used to close the
1694 socket. However, verify_get_ident() also closes the socket later, and in
1695 between Exim writes to the log, which may get opened at this point. When
1696 the socket was closed in ip_connect(), the log could get the same file
1697 descriptor number as the socket. This naturally causes chaos. The fix is
1698 not to close the socket in ip_connect(); the socket should be closed by
1699 the function that creates it. There was only one place in the code where
1700 this was missing, in the iplookup router, which I don't think anybody now
1701 uses, but I've fixed it anyway.
1703 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1704 well as to direct DNS lookups. Otherwise the handling of names in host
1705 lists is inconsistent and therefore confusing.
1711 PH/01 Two changes to the default runtime configuration:
1713 (1) Move the checks for relay_from_hosts and authenticated clients from
1714 after to before the (commented out) DNS black list checks.
1716 (2) Add control=submission to the relay_from_hosts and authenticated
1717 clients checks, on the grounds that messages accepted by these
1718 statements are most likely to be submissions.
1720 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1722 (1) Generate an error if the third argument for the ${prvs expansion is
1725 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1728 (3) Reset the variables that are obtained from the first argument of
1729 ${prvscheck and used in the second argument before leaving the code,
1730 because their memory is reclaimed, so using them afterwards may do
1733 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1734 one (it's much easier than Tom thought :-).
1736 (5) Because of (4), we can now allow for the use of $prvscheck_result
1737 inside the third argument.
1739 PH/03 For some reason, the default setting of PATH when running a command from
1740 a pipe transport was just "/usr/bin". I have changed it to
1743 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1744 anything to be listed in the output from -bV.
1746 PH/05 When a filter generated an autoreply, the entire To: header line was
1747 quoted in the delivery log line, like this:
1749 => >A.N.Other <ano@some.domain> <original@ddress> ...
1751 This has been changed so that it extracts the operative address. There
1752 may be more than one such address. If so, they are comma-separated, like
1755 => >ano@some.domain,ona@other.domain <original@ddress> ...
1757 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1758 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1759 not being looked up in the rDNS to get a host name, Exim was showing the
1760 IP address twice in Received: lines, even though the IP addresses were
1761 identical. For example:
1763 Received: from [1.2.3.4] (helo=[1.2.3.4])
1765 However, if the real host name was known, it was omitting the HELO data
1766 if it matched the actual IP address. This has been tidied up so that it
1767 doesn't show the same IP address twice.
1769 PH/07 When both +timestamp and +memory debugging was on, the value given by
1770 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1771 called by the debug printing, thereby overwriting the timestamp buffer.
1772 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1774 PH/08 When the original message was included in an autoreply transport, it
1775 always said "this is a copy of the message, including all the headers",
1776 even if body_only or headers_only was set. It now gives an appropriate
1779 PH/09 Applied a patch from the Sieve maintainer which:
1781 o fixes some comments
1782 o adds the (disabled) notify extension core
1783 o adds some debug output for the result of if/elsif tests
1784 o points to the current vacation draft in the documentation
1785 and documents the missing references header update
1789 o fixes a bug in processing the envelope test (when testing
1790 multiple envelope elements, the last element determinted the
1793 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1794 Electronic Mail") by including:
1796 Auto-submitted: auto-generated
1798 in the messages that it generates (bounce messages and others, such as
1799 warnings). In the case of bounce messages for non-SMTP mesages, there was
1800 also a typo: it was using "Auto_submitted" (underscore instead of
1801 hyphen). Since every message generated by Exim is necessarily in response
1802 to another message, thes have all been changed to:
1804 Auto-Submitted: auto-replied
1806 in accordance with these statements in the RFC:
1808 The auto-replied keyword:
1810 - SHOULD be used on messages sent in direct response to another
1811 message by an automatic process,
1813 - MUST NOT be used on manually-generated messages,
1815 - MAY be used on Delivery Status Notifications (DSNs) and Message
1816 Disposition Notifications (MDNs),
1818 - MUST NOT be used on messages generated by automatic or periodic
1819 processes, except for messages which are automatic responses to
1822 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1823 to the default Received: header definition.
1825 PH/12 Added log selector acl_warn_skipped (default on).
1827 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1828 variables because (a) they are in the wrong storage pool and (b) even if
1829 they were copied, it wouldn't work properly because of the caching.
1831 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1832 checking when decoding. Apparently there are clients that generate
1833 overlong encoded strings. Why am I not surprised?
1835 PH/15 If the first argument of "${if match_address" was not empty, but did not
1836 contain an "@" character, Exim crashed. Now it writes a panic log message
1837 and treats the condition as false.
1839 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1841 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1842 extension "envelope-auth". The code is finished and in agreement with
1843 other implementations, but there is no documentation so far and in fact,
1844 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1845 not changing the active code.
1847 Print executed "if" and "elsif" statements when debugging is used. This
1848 helps a great deal to understand what a filter does.
1850 Document more things not specified clearly in RFC3028. I had all this
1851 sorted out, when out of a sudden new issues came to my mind. Oops."
1853 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1856 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1857 canonical form (as documented). However, after a host name lookup from
1858 the IP address, check_host() was doing a simple string comparison with
1859 addresses acquired from the DNS when checking that the found name did
1860 have the original IP as one of its addresses. Since any found IPv6
1861 addresses are likely to be in abbreviated form, the comparison could
1862 fail. Luckily, there already exists a function for doing the comparison
1863 by converting both addresses to binary, so now that is used instead of
1864 the text comparison.
1866 PH/20 There was another similar case to PH/19, when a complete host name was
1867 given in a host list; looking up its IP address could give an abbreviated
1868 form, whereas the current host's name might or might not be abbreviated.
1869 The same fix has been applied.
1875 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1876 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1879 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1880 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1882 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1884 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1885 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1886 Also, for "Auto-Submitted", treat anything other than "no" as
1887 non-personal, in accordance with RFC 3834. (Previously it treated
1888 anything starting "auto-" as non-personal.)
1890 TF/01 The control=submission/name=... option had a problem with syntax
1891 errors if the name included a slash character. The /name= option
1892 now slurps the rest of the string, so it can include any characters
1893 but it must come last in the list of options (after /sender_retain
1896 PH/05 Some modifications to the interface to the fake nameserver for the new
1904 TK/01 Added the "success_on_redirect" address verification option. See
1905 NewStuff for rationale and an example.
1907 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1909 PH/02 Patch to exigrep to allow it to work on syslog lines.
1911 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1912 fread() to skip over the body file's header line, because in Cygwin the
1913 header line is locked and is inaccessible.
1915 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1916 co-exist for some time) to make it clear that it is the Exim ID that is
1917 referenced, not the Message-ID: header line.
1919 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1920 string_format() function, because snprintf() does not exist on all
1923 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1924 ${stat: expansion item.
1926 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1927 protocol synchronization error", to keep the pedants happy.
1929 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1930 well as for IRIX systems, when gcc is being used. See the host.c source
1933 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1935 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1938 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1939 transport and to the smtp transport in LMTP mode.
1941 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1943 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1944 run a filter in a subprocess. This could lead to confusion in subsequent
1945 lookups in the parent process. There should also be a search_tidyup() at
1946 the end of the subprocess.
1948 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1949 only if the host matched helo_try_verify_hosts, which caused the
1950 verification to occur when the EHLO/HELO command was issued. The ACL just
1951 tested the remembered result. Now, if a previous verification attempt has
1952 not happened, "verify = helo" does it there and then.
1954 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1956 TK/03 Fix log output including CR from clamd.
1958 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1959 exist provoked a memory error which could cause a segfault.
1961 PH/15 Installed PCRE 6.2
1963 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1965 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1966 of the problem. Specifically, suggested +O2 rather than +O1 for the
1969 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1971 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1972 record, the port was not correctly logged when the outgoing_port log
1973 selector was set (it logged the transort's default port).
1975 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1976 fallback_hosts, and "hosts" in the smtp transport.
1978 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1979 host errors such as "Connection refused".
1981 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1982 authentication with radiusclient 0.4.9:
1984 - Error returned from rc_read_config was caught wrongly
1985 - Username/password not passed on to radius server due to wrong length.
1987 The presumption is that some radiusclient API changes for 4.51/PH/17
1988 were not taken care of correctly. The code is still untested by me (my
1989 Linux distribution still has 0.3.2 of radiusclient), but it was
1990 contributed by a Radius user.
1992 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1993 expanding the "port" option of the smtp transport.
1995 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1996 while reading a MIME header. Thanks to Tom Hughes for a patch.
1998 PH/24 Include config.h inside local_scan.h so that configuration settings are
2001 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2002 that in an ACL for RCPT (for example), you can examine exactly what was
2005 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2006 commands, but it was not correctly comparing the address with the actual
2007 client host address. Thus, it would show the EHLO address in Received:
2008 header lines when this was not necessary.
2010 PH/27 Added the % operator to ${eval:}.
2012 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2013 it should be ignoring failures (because with -C, for example, it has lost
2014 privilege). It wasn't ignoring creation failures other than "already
2017 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2020 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2021 because an input line was too long, either on its own, or by virtue of
2022 too many continuations, the temporary file was not being removed, and the
2023 return code was incorrect.
2025 PH/31 Missing "BOOL" in function definition in filtertest.c.
2027 PH/32 Applied Sieve patches from the maintainer.
2029 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2031 PH/33 Added "verify = not_blind".
2033 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2034 Local/Makefile (with some defaults set). These are used in built scripts
2035 such as exicyclog, but they have never been used in the exim_install
2036 script (though there are many overriding facilities there). I have
2037 arranged that the exim_install script now takes note of these two
2040 PH/35 Installed configuration files for Dragonfly.
2042 PH/36 When a locally submitted message by a trusted user did not contain a
2043 From: header, and the sender address was obtained from -f or from an SMTP
2044 MAIL command, and the trusted user did not use -F to supply a sender
2045 name, $originator_name was incorrectly used when constructing a From:
2046 header. Furthermore, $originator_name was used for submission mode
2047 messages from external hosts without From: headers in a similar way,
2048 which is clearly wrong.
2050 PH/37 Added control=suppress_local_fixups.
2052 PH/38 When log_selector = +received_sender was set, and the addition of the
2053 sender made the log line's construction buffer exactly full, or one byte
2054 less than full, an overflow happened when the terminating "\n" was
2057 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2058 when the result of a list match is failure because a DNS lookup failed.
2060 PH/40 RM_COMMAND is now used in the building process.
2062 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2063 the "build-* directories that it finds.
2065 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2066 address in a domain literal was a prefix of an interface address.
2068 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2069 when verifying a sender address, unless rewrite_headers is false.
2071 PH/44 Wrote a long comment about why errors_to addresses are verified as
2072 recipients, not senders.
2074 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2075 the ratelimit ACL was added.
2077 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2079 PH/46 Added extra information about PostgreSQL errors to the error string.
2081 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2082 avoiding the need to install special zones in a real server. This is
2083 backwards compatible; if it can't find the fake resolver, it drops back.
2084 Thus, both old and new test suites can be run.
2086 TF/02 Added util/ratelimit.pl
2088 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2089 clock is set back in time.
2091 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2092 Candler <B.Candler@pobox.com>.
2094 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2095 OK for addresses that are the result of redirections.
2097 PH/48 A number of further additions for the benefit of the new test suite,
2098 including a fake gethostbyname() that interfaces to the fake DNS resolver
2101 TF/06 The fix for widen_domains has also been applied to qualify_single and
2102 search_parents which are the other dnslookup options that can cause
2105 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2108 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2110 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2111 service" if it encountered an MX record with an empty target hostname.
2112 The message is now "an MX or SRV record indicated no SMTP service".
2114 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2115 if the DNS of the sending site is misconfigured. This is quite a
2116 common situation. This change restores the behaviour of treating a
2117 helo verification defer as a failure.
2119 PH/51 If self=fail was set on a router, the bounce message did not include the
2120 actual error message.
2126 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2128 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2129 command itself ended in error, the underlying message about the transport
2130 filter timeout was being overwritten with the pipe command error. Now the
2131 underlying error message should be appended to the second error message.
2133 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2135 PH/02 Reduce the amount of output that "make" produces by default. Full output
2136 can still be requested.
2138 PH/03 The warning log line about a condition test deferring for a "warn" verb
2139 was being output only once per connection, rather than after each
2140 occurrence (because it was using the same function as for successful
2141 "warn" verbs). This seems wrong, so I have changed it.
2143 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2144 it should not have, which might have caused a crash in the right
2145 circumstances, but probably never did.
2147 PH/04 Installed a modified version of Tony Finch's patch to make submission
2148 mode fix the return path as well as the Sender: header line, and to
2149 add a /name= option so that you can make the user's friendly name appear
2152 TF/03 Added the control = fakedefer ACL modifier.
2154 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2155 Mark Lowes for thorough testing.
2157 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2159 TK/03 Merged latest SRS patch from Miles Wilton.
2161 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2162 with the definition in sysexits.h (which is #included earlier).
2163 Fortunately, Exim does not actually use EX_OK. The code used to try to
2164 preserve the sysexits.h value, by assumimg that macro definitions were
2165 scanned for macro replacements. I have been disabused of this notion,
2166 so now the code just undefines EX_OK before #including unistd.h.
2168 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2169 in the smtp transport. When a block could not be written in a single
2170 write() function, the timeout was being re-applied to each part-write.
2171 This seems wrong - if the receiver was accepting one byte at a time it
2172 would take for ever. The timeout is now adjusted when this happens. It
2173 doesn't have to be particularly precise.
2175 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2176 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2178 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2179 without a domain if the check to <postmaster@domain> fails.
2181 SC/01 Eximstats: added -xls and the ability to specify output files
2182 (patch written by Frank Heydlauf).
2184 SC/02 Eximstats: use FileHandles for outputing results.
2186 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2188 SC/04 Eximstats: fixed display of large numbers with -nvr option
2190 SC/05 Eximstats: fixed merging of reports with empty tables.
2192 SC/06 Eximstats: added the -include_original_destination flag
2194 SC/07 Eximstats: removed tabs and trailing whitespace.
2196 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2198 TK/06 MBOX spool code: Add real "From " MBOX separator line
2199 so the .eml file is really in mbox format (even though
2200 most programs do not really care). Patch from Alex Miller.
2202 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2203 The latter is generated from $received_to and is only set if the
2204 message has one envelope recipient. SA can use these headers,
2205 obviously out-of-the-box. Patch from Alex Miller.
2207 PH/08 The ${def test on a variable was returning false if the variable's
2208 value was "0", contrary to what the specification has always said!
2209 The result should be true unless the variable is empty.
2211 PH/09 The syntax error of a character other than { following "${if
2212 def:variable_name" (after optional whitespace) was not being diagnosed.
2213 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2214 accidental colon was present, for example, could give incorrect results.
2216 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2217 result is used (not including appendfile, where other changes are about
2220 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2221 This involved changing a lot of size variables from int to off_t. It
2222 should work with maildirs and everything.
2224 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2225 spamd dying while we are connected to it.
2227 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2230 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2231 function for BDB 4.3.
2233 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2235 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2236 the sources for PCRE have been re-arranged and more files are now
2239 PH/15 The code I had for printing potentially long long variables in PH/11
2240 above was not the best (it lost precision). The length of off_t variables
2241 is now inspected at build time, and an appropriate printing format (%ld
2242 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2243 to be "long long int" or "long int". This is needed for the internal
2244 formatting function string_vformat().
2246 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2247 the configuration file to be ":syslog", then the script "guesses" where
2248 the logs files are, rather than using the compiled in default. In our
2249 case the guess is not the same as the compiled default, so the script
2250 suddenly stopped working when I started to use syslog. The patch checks
2251 to see if log_file_path is "". If so, it attempts to read it from exim
2252 with no configuration file to get the compiled in version, before it
2253 falls back to the previous guessing code."
2255 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2256 implementing BATV in an Exim configuration. See NewStuff for the gory
2259 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2260 Makefile that are specific to HP-UX.
2262 PH/18 If the "use_postmaster" option was set for a recipient callout together
2263 with the "random" option, the postmaster address was used as the MAIL
2264 FROM address for the random test, but not for the subsequent recipient
2265 test. It is now used for both.
2267 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2268 patch removes a few documentation additions to RFC 3028, because the
2269 latest draft now contains them. It adds the new en;ascii-case comparator
2270 and a new error check for 8bit text in MIME parts. Comparator and
2271 require names are now matched exactly. I enabled the subaddress
2272 extension, but it is not well tested yet (read: it works for me)."
2274 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2275 rework some of the code of TK/09 above to avoid the hardwired use of
2276 "%lld" and "long long". Replaced the call to snprintf() with a call to
2279 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2280 records point to non-existent hosts", "retry timeout exceeded", and
2281 "retry time not reached for any host after a long failure period".
2283 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2284 experimental DomainKeys support:
2286 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2287 (2) On an error such as an illegally used "control", the wrong name for
2288 the control was given.
2290 These problems did NOT occur unless DomainKeys support was compiled.
2292 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2294 PH/24 Added ${if match_ip condition.
2296 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2297 databases so that it will be absolutely obvious if a crash occurs in the
2298 DB library. This is a regular occurrence (often caused by mis-matched
2301 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2302 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2303 picked up on a user's system that detects such things. There doesn't seem
2304 to be a gcc warning option for this - only an attribute that has to be
2305 put on the function's prototype. It seems that in Fedora Core 4 they have
2306 set this on a number of new functions. No doubt there will be more in due
2309 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2310 specify a transport. However, if an address that was verified by such a
2311 router was the subject of a callout, Exim crashed because it tried to
2312 read the rcpt_include_affixes from the non-existent transport. Now it
2313 just assumes that the setting of that option is false. This bug was
2314 introduced by 4.51/PH/31.
2316 PH/28 Changed -d+all to exclude +memory, because that information is very
2317 rarely of interest, but it makes the output a lot bigger. People tend to
2318 do -d+all out of habit.
2320 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2321 code in os-type was giving problems when libc.so lives in lib64, like on
2324 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2325 aren't the modern standard, and it seems that some systems' include files
2326 don't always have them. Exim was already checking for some of the newer
2327 ones like T_AAAA, and defining it itself. I've added checks for all the
2328 record types that Exim uses.
2330 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2331 not automatically generating a new one, as it is supposed to. This
2332 prevented TLS from working. If the file did exist, but contained invalid
2333 data, a new version was generated, as expected. It was only the case of a
2334 non-existent file that was broken.
2336 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2337 with a change in libdomainkeys > 0.64.
2339 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2340 from DNS. If the selector record carries the flag, it now has
2341 precedence over the domain-wide flag.
2343 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2345 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2346 the use of a port name that isn't defined in /etc/services) occurred, the
2347 message was deferred as in a normal delivery, and thus remained on the
2348 spool, instead of being failed because of the mua_wrapper setting. This
2349 is now fixed, and I tidied up some of the mua_wrapper messages at the
2352 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2353 the messages in a hash of arrays rather than using individual hashes.
2354 This is a bit cleaner and results in dramatic memory savings, albeit
2355 at a slight CPU cost.
2357 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2358 as requested by Marc Sherman.
2360 SC/10 Eximstats: added histograms for user specified patterns as requested
2363 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2365 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2366 fopen() in the content-scanning modules that did not already have it.
2372 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2373 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2375 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2377 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2379 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2380 to test Sieve filters that use "vacation".
2382 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2383 that changes the way the GnuTLS parameters are stored in the cache file.
2384 The new format can be generated externally. For backward compatibility,
2385 if the data in the cache doesn't make sense, Exim assumes it has read an
2386 old-format file, and it generates new data and writes a new file. This
2387 means that you can't go back to an older release without removing the
2390 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2391 work if there are any delivery delays because "one_time" forces the
2392 parent to be marked "delivered", so its unseen clone is never tried
2393 again. For this reason, Exim now forbids the simultaneous setting of
2396 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2397 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2398 entry if you want to know the details. The fix, however, neglected to
2399 consider the case when local delivery batching is involved. The test for
2400 "previously delivered" was not happening when checking to see if an
2401 address could be batched with a previous (undelivered) one; under
2402 certain circumstances this could lead to multiple deliveries to the same
2405 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2406 in its include files, and this causes problems building Exim.
2408 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2409 header_syntax) but Exim was just ignoring anything given after a slash.
2410 In particular, this caused confusion with an attempt to use "verify =
2411 reverse_host_lookup/defer_ok". An error is now given when options are
2412 supplied for verify items that do not have them. (Maybe reverse_host_
2413 lookup should have a defer_ok option, but that's a different point.)
2415 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2416 defined by RFC 821) to 2048, because there were problems with some AUTH
2417 commands, and RFC 1869 says the size should be increased for extended
2418 SMTP commands that take arguments.
2420 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2423 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2424 "unknown" error; now it says that the functionality isn't in the binary.
2426 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2427 an address' error message when a string expansion fails (syntax or
2428 whatever). Otherwise the password may appear in the log. Following change
2429 PH/42 below, there is no longer a chance of it appearing in a bounce
2432 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2434 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2435 of its addresses was ever tried. (Bugzilla bug #2.)
2437 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2438 the result incorrectly in the debug output. (It correctly added a newline
2439 to what was transported.)
2441 TF/01 Added $received_time.
2443 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2444 commented out examples of how to interface to a virus scanner and to
2445 SpamAssassin. Also added commented examples of av_scanner and
2446 spamd_address settings.
2448 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2449 and controls are allowed in which ACLs. There were a couple of minor
2450 errors. Some of the entries in the conditions table (which is a table of
2451 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2452 negation of where the condition IS allowed.
2454 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2456 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2457 header file does not have a version number, so I've had to invent a new
2458 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2459 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2460 radiusclient), but it was contributed by a Radius user.
2462 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2463 files or directories, for OpenSSL.
2465 PH/19 When an Exim process that is running as root has to create an Exim log
2466 file, it does so in a subprocess that runs as exim:exim so as to get the
2467 ownership right at creation (otherwise, other Exim processes might see
2468 the file with the wrong ownership). There was no test for failure of this
2469 fork() call, which would lead to the process getting stuck as it waited
2470 for a non-existent subprocess. Forks do occasionally fail when resources
2471 run out. I reviewed all the other calls to fork(); they all seem to check
2474 PH/20 When checking for unexpected SMTP input at connect time (before writing
2475 the banner), Exim was not dealing correctly with a non-positive return
2476 from the read() function. If the client had disconnected by this time,
2477 the result was a log entry for a synchronization error with an empty
2478 string after "input=" when read() returned zero. If read() returned -1
2479 (an event I could not check), uninitialized data bytes were printed.
2480 There were reports of junk text (parts of files, etc) appearing after
2483 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2485 PH/22 Added support for macro redefinition, and (re)definition in between
2486 driver and ACL definitions.
2488 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2489 forgetting to use the resulting value; it was using the unexpanded value.
2491 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2492 hadn't been configured. The fix is from Juergen Kreileder, who
2493 understands it better than I do:
2495 "Here's what I see happening with three configured cyrus_sasl
2496 authenticators configured (plain, login, cram-md5):
2498 On startup auth_cyrus_sasl_init() gets called for each of these.
2499 This means three calls to sasl_listmech() without a specified mech_list.
2500 => SASL tests which mechs of all available mechs actually work
2501 => three warnings about OTP not working
2502 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2504 With the patch, sasl_listmech() also gets called three times. But now
2505 SASL's mech_list option is set to the server_mech specified in the the
2506 authenticator. Or in other words, the answer from sasl_listmech()
2507 gets limited to just the mech you're testing for (which is different
2509 => the return list contains just 'plain' or 'login', 'cram-md5' or
2510 nothing depending on the value of ob->server_mech.
2512 I've just tested the patch: Authentication still works fine,
2513 unavailable mechs specified in the exim configuration are still
2514 caught, and the auth.log warnings about OTP are gone."
2516 PH/25 When debugging is enabled, the contents of the command line are added
2517 to the debugging output, even when log_selector=+arguments is not
2520 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2521 answer is "GNU", and only if the return is "GNU/something" is the answer
2524 PH/27 $acl_verify_message is now set immediately after the failure of a
2525 verification in an ACL, and so is available in subsequent modifiers. In
2526 particular, the message can be preserved by coding like this:
2528 warn !verify = sender
2529 set acl_m0 = $acl_verify_message
2531 Previously, $acl_verify_message was set only while expanding "message"
2532 and "log_message" when a very denied access.
2534 PH/28 Modified OS/os.c-Linux with
2536 -#ifndef OS_LOAD_AVERAGE
2537 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2539 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2540 nomenclature these days.)
2542 PH/29 Installed patch from the Sieve maintainer that adds the options
2543 sieve_useraddress and sieve_subaddress to the redirect router.
2545 PH/30 In these circumstances:
2546 . Two addresses routed to the same list of hosts;
2547 . First host does not offer TLS;
2548 . First host accepts first address;
2549 . First host gives temporary error to second address;
2550 . Second host offers TLS and a TLS session is established;
2551 . Second host accepts second address.
2552 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2553 and peerdn, if requested) that were in fact used only for the second
2556 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2557 attention to any local part prefix or suffix that was matched by the
2558 router that accepted the address. It now behaves in the same way as it
2559 does for delivery: the affixes are removed from the local part unless
2560 rcpt_include_affixes is set on the transport.
2562 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2563 timeout during the DATA phase of an incoming message.
2565 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2566 applied a patch sanctioned by the Sieve maintainer.
2568 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2569 the uid or gid is negative. A case of a negative gid caused this to be
2570 noticed. The fix allows for either to be negative.
2572 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2573 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2576 PH/36 The change PH/12 above was broken. Fixed it.
2578 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2579 the grounds that routing the same address twice would always produce the
2580 same answer. This might have been true once, but it is certainly no
2581 longer true now. Routing a child address may depend on the previous
2582 routing that produced that child. Some complicated redirection strategies
2583 went wrong when messages had multiple recipients, and made Exim's
2584 behaviour dependent on the order in which the addresses were given.
2586 I have moved the duplicate checking until after the routing is complete.
2587 Exim scans the addresses that are assigned to local and remote
2588 transports, and removes any duplicates. This means that more work will be
2589 done, as duplicates will always all be routed, but duplicates are
2590 presumably rare, so I don't expect this is of any significance.
2592 For deliveries to pipes, files, and autoreplies, the duplicate checking
2593 still happens during the routing process, since they are not going to be
2596 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2597 It corrects a timeout issue with spamd. This is Ian's comment: "The
2598 background is that sometimes spamd either never reads data from a
2599 connection it has accepted, or it never writes response data. The exiscan
2600 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2601 blindly assumes that writes won't block so it may never time out."
2603 PH/39 Allow G after quota size as well as K and M.
2605 PH/40 The value set for $authenticated_id in an authenticator may not contain
2606 binary zeroes or newlines because the value is written to log lines and
2607 to spool files. There was no check on this. Now the value is run through
2608 the string_printing() function so that such characters are converted to
2609 printable escape sequences.
2611 PH/41 $message_linecount is a new variable that contains the total number of
2612 lines in the message. Compare $body_linecount, which is the count for the
2615 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2616 bounce and delay warning messages, except in certain special cases, which
2619 (a) An SMTP error message from a remote host;
2620 (b) A message specified in a :fail: redirection;
2621 (c) A message specified in a "fail" command in a system filter;
2622 (d) A message specified in a FAIL return from the queryprogram router;
2623 (e) A message specified by the cannot_route_message router option.
2625 In these cases only, Exim does include the error details in bounce and
2626 warning messages. There are also a few cases where bland messages such
2627 as "unrouteable address" or "local delivery error" are given.
2629 PH/43 $value is now also set for the "else" part of a ${run expansion.
2631 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2632 being worked on, but at least Exim now implements the latest version to
2635 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2636 process to complete was treated as a delivery failure, a timeout while
2637 writing the message to the pipe was logged, but erroneously treated as a
2638 successful delivery. Such timeouts include transport filter timeouts. For
2639 consistency with the overall process timeout, these timeouts are now
2640 treated as errors, giving rise to delivery failures by default. However,
2641 there is now a new Boolean option for the pipe transport called
2642 timeout_defer, which, if set TRUE, converts the failures into defers for
2643 both kinds of timeout. A transport filter timeout is now identified in
2646 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2647 systems where "make" and "gmake" are different, calling "gmake" at top
2648 level broke things. I've arranged for the value of $(MAKE) to be passed
2649 from the Makefile to this script so that it can call the same version of
2653 A note about Exim versions 4.44 and 4.50
2654 ----------------------------------------
2656 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2657 changes of various kinds. As a consequence, a big documentation update was
2658 needed. This delayed the release for rather longer than seemed good, especially
2659 in the light of a couple of (minor) security issues. Therefore, the changes
2660 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2661 release. So 4.44 and 4.50 are in effect two different branches that both start
2664 I have left the 4.50 change log unchanged; it contains all the changes since
2665 4.43. The change log for 4.44 is below; many of its items are identical to
2666 those for 4.50. This seems to be the most sensible way to preserve the
2667 historical information.
2673 1. Minor wording change to the doc/README.SIEVE file.
2675 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2676 computation of the current number of files was incorrect.
2678 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2679 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2682 4. Give more explanation in the error message when the command for a transport
2683 filter fails to execute.
2685 5. There are several places where Exim runs a non-Exim command in a
2686 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2687 was being done only for the command run by the queryprogram router. It is
2688 now done for all such subprocesses. The other cases are: ${run, transport
2689 filters, and the commands run by the lmtp and pipe transports.
2691 6. Added CONFIGURE_GROUP build-time option.
2693 7. Some older OS have a limit of 256 on the maximum number of file
2694 descriptors. Exim was using setrlimit() to set 1000 as a large value
2695 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2696 systems. I've change it so that if it can't get 1000, it tries for 256.
2698 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2699 was an oversight, and furthermore, ever since the addition of extra
2700 controls (e.g. 4.43/32), the checks on when to allow different forms of
2701 "control" were broken. There should now be diagnostics for all cases when a
2702 control that does not make sense is encountered.
2704 9. Added the /retain_sender option to "control=submission".
2706 10. $recipients is now available in the predata ACL (oversight).
2708 11. Tidy the search cache before the fork to do a delivery from a message
2709 received from the command line. Otherwise the child will trigger a lookup
2710 failure and thereby defer the delivery if it tries to use (for example) a
2711 cached ldap connection that the parent has called unbind on.
2713 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2714 of $address_data from the recipient verification was clobbered by the
2715 sender verification.
2717 13. The value of address_data from a sender verification is now available in
2718 $sender_address_data in subsequent conditions in the ACL statement.
2720 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2722 15. Added a new option "connect=<time>" to callout options, to set a different
2725 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2726 was its contents. (It was OK if the option was not defined at all.)
2728 17. A "Completed" log line is now written for messages that are removed from
2729 the spool by the -Mrm option.
2731 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2732 information about exactly what failed.
2734 19. Added -dd to debug only the daemon process.
2736 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2737 handles timeouts, both on the server side and network timeouts. Renamed the
2738 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2740 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2741 It is now set to "smtps".
2743 22. $host_address is now set to the target address during the checking of
2744 ignore_target_hosts.
2746 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2747 being passed; this would have caused $sender_host_name to have been used if
2748 matching the list had actually called for a host name (not very likely,
2749 since this list is usually IP addresses). A host name is now passed as
2752 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2753 code that specifies a non-restarting handler (typically sigaction() in
2754 modern systems) in an attempt to fix a rare and obscure crash bug.
2756 25. Narrowed the window for a race in the daemon that could cause it to ignore
2757 SIGCHLD signals. This is not a major problem, because they are used only to
2758 wake it up if nothing else does.
2760 26. A malformed maildirsize file could cause Exim to calculate negative values
2761 for the mailbox size or file count. Odd effects could occur as a result.
2762 The maildirsize information is now recalculated if the size or filecount
2765 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2766 support for a long time. Removed HAVE_SYS_VFS_H.
2768 28. Installed the latest version of exipick from John Jetmore.
2770 29. In an address list, if the pattern was not a regular expression, an empty
2771 subject address (from a bounce message) matched only if the pattern was an
2772 empty string. Non-empty patterns were not even tested. This was the wrong
2773 because it is perfectly reasonable to use an empty address as part of a
2774 database query. An empty address is now tested by patterns that are
2775 lookups. However, all the other forms of pattern expect the subject to
2776 contain a local part and a domain, and therefore, for them, an empty
2777 address still always fails if the pattern is not itself empty.
2779 30. Exim went into a mad DNS loop when attempting to do a callout where the
2780 host was specified on an smtp transport, and looking it up yielded more
2781 than one IP address.
2783 31. Re-factored the code for checking spool and log partition space into a
2784 function that finds that data and another that does the check. The former
2785 is then used to implement four new variables: $spool_space, $log_space,
2786 $spool_inodes, and $log_inodes.
2788 32. The RFC2047 encoding function was originally intended for short strings
2789 such as real names; it was not keeping to the 75-character limit for
2790 encoded words that the RFC imposes. It now respects the limit, and
2791 generates multiple encoded words if necessary. To be on the safe side, I
2792 have increased the buffer size for the ${rfc2047: expansion operator from
2795 33. It is now permitted to omit both strings after an "if" condition; if the
2796 condition is true, the result is "true". As before, when the second string
2797 is omitted, a false condition yields an empty string. This makes it less
2798 cumbersome to write custom ACL and router conditions.
2800 34. Failure to deliver a bounce message always caused it to be frozen, even if
2801 there was an errors_to setting on the router. The errors_to setting is now
2804 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2805 canonical form (fully expanded) before being placed in
2806 $sender_host_address.
2808 36. The table in the code that translates DNS record types into text (T_A to
2809 "A" for instance) was missing entries for NS and CNAME. It is just possible
2810 that this could have caused confusion if both these types were looked up
2811 for the same domain, because the text type is used as part of Exim's
2812 per-process caching. But the chance of anyone hitting this buglet seems
2815 37. The dnsdb lookup has been extended in a number of ways.
2817 (1) There is a new type, "zns", which walks up the domain tree until it
2818 finds some nameserver records. It should be used with care.
2820 (2) There is a new type, "mxh", which is like "mx" except that it returns
2821 just the host names, not the priorities.
2823 (3) It is now possible to give a list of domains (or IP addresses) to be
2824 looked up. The behaviour when one of the lookups defers can be
2825 controlled by a keyword.
2827 (4) It is now possible to specify the separator character for use when
2828 multiple records are returned.
2830 38. The dnslists ACL condition has been extended: it is now possible to supply
2831 a list of IP addresses and/or domains to be looked up in a particular DNS
2834 39. Added log_selector=+queue_time_overall.
2836 40. When running the queue in the test harness, wait just a tad after forking a
2837 delivery process, to get repeatability of debugging output.
2839 41. Include certificate and key file names in error message when GnuTLS fails
2840 to set them up, because the GnuTLS error message doesn't include the name
2841 of the failing file when there is a problem reading it.
2843 42. Allow both -bf and -bF in the same test run.
2845 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2847 44. The "Exiscan patch" is now merged into the mainline Exim source.
2849 45. Sometimes the final signoff response after QUIT could fail to get
2850 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2851 < 0 before doing a fflush(). This bug looks as though it goes back to the
2852 introduction of TLS in release 3.20, but "sometimes" must have been rare
2853 because the tests only now provoked it.
2855 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2856 (this can affect the format of dates).
2858 47. exim_tidydb, when checking for the continued existence of a message for
2859 which it has found a message-specific retry record, was not finding
2860 messages that were in split spool directories. Consequently, it was
2861 deleting retry records that should have stayed in existence.
2863 48. Steve fixed some bugs in eximstats.
2865 49. The SPA authentication driver was not abandoning authentication and moving
2866 on to the next authenticator when an expansion was forced to fail,
2867 contradicting the general specification for all authenticators. Instead it
2868 was generating a temporary error. It now behaves as specified.
2870 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2871 (the order specifies the preference for clients). The order is now AES256,
2872 AES128, 3DES, ARCFOUR128.
2874 51. Small patch to Sieve code - explicitly set From: when generating an
2877 52. Exim crashed if a remote delivery caused a very long error message to be
2878 recorded - for instance if somebody sent an entire SpamAssassin report back
2879 as a large number of 550 error lines. This bug was coincidentally fixed by
2880 increasing the size of one of Exim's internal buffers (big_buffer) that
2881 happened as part of the Exiscan merge. However, to be on the safe side, I
2882 have made the code more robust (and fixed the comments that describe what
2885 53. Now that there can be additional text after "Completed" in log lines (if
2886 the queue_time_overall log selector is set), a one-byte patch to exigrep
2887 was needed to allow it to recognize "Completed" as not the last thing in
2890 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2891 patch that reportedly fixes this has been added. I am not expert enough to
2892 create a test for it. This is what the patch creator wrote:
2894 "I found a little strange behaviour of ldap code when working with
2895 Windows 2003 AD Domain, where users was placed in more than one
2896 Organization Units. When I tried to give exim partial DN, the exit code
2897 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2898 But simultaneously result of request was absolutely normal ldap result,
2899 so I produce this patch..."
2901 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2902 REFERENCE, so I have modified the code to exclude the patch when that macro
2905 55. Some experimental protocols are using DNS PTR records for new purposes. The
2906 keys for these records are domain names, not reversed IP addresses. The
2907 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2908 leaves it alone. Component reversal etc. now happens only for IP addresses.
2911 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2913 57. Double the size of the debug message buffer (to 2048) so that more of very
2914 long debug lines gets shown.
2916 58. The exicyclog utility now does better if the number of log files to keep
2917 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2919 59. Two changes related to the smtp_active_hostname option:
2921 (1) $smtp_active_hostname is now available as a variable.
2922 (2) The default for smtp_banner uses $smtp_active_hostname instead
2923 of $primary_hostname.
2925 60. The host_aton() function is supposed to be passed a string that is known
2926 to be a valid IP address. However, in the case of IPv6 addresses, it was
2927 not checking this. This is a hostage to fortune. Exim now panics and dies
2928 if the condition is not met. A case was found where this could be provoked
2929 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2930 components; fortuitously, this particular loophole had already been fixed
2931 by change 4.50/55 above.
2933 If there are any other similar loopholes, the new check in host_aton()
2934 itself should stop them being exploited. The report I received stated that
2935 data on the command line could provoke the exploit when Exim was running as
2936 exim, but did not say which command line option was involved. All I could
2937 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2938 running as the user.
2941 61. There was a buffer overflow vulnerability in the SPA authentication code
2942 (which came originally from the Samba project). I have added a test to the
2943 spa_base64_to_bits() function which I hope fixes it.
2946 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2947 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2949 63. The daemon start-up calls getloadavg() while still root for those OS that
2950 need the first call to be done as root, but it missed one case: when
2951 deliver_queue_load_max is set with deliver_drop_privilege. This is
2952 necessary for the benefit of the queue runner, because there is no re-exec
2953 when deliver_drop_privilege is set.
2955 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2956 This has been fixed.
2958 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2959 was in use, was not putting the data itself into the right store pool;
2960 consequently, it could be overwritten for a subsequent message in the same
2961 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2964 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2966 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2967 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2968 the function were treating the return as a boolean value, which happened to
2969 work because 0=false and not-0=true, but is not correct code.
2971 68. The host_aton() function was not handling scoped IPv6 addresses (those
2972 with, for example, "%eth0" on the end) correctly.
2974 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2975 negated items (that is, ~something) in unsigned ints. Some compilers
2976 apparently mutter when there is no cast.
2978 70. If an address verification called from an ACL failed, and did not produce a
2979 user-specific message (i.e. there was only a "system" message), nothing was
2980 put in $acl_verify_message. In this situation, it now puts the system
2983 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2984 session; change 4.31/43 added the unwanted input to the log line - except
2985 that it did not do this in the start of session case. It now does.
2987 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2988 This is wrong and can cause the other end to generate a synchronization
2989 error if it is another Exim or anything else that does the synchronization
2990 check. A QUIT command is no longer sent after a timeout.
2992 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2993 during host lookups.
2995 74. The defer_ok option of callout verification was not working if it was used
2996 when verifying addresses in header lines, that is, for this case:
2998 verify = header_sender/callout=defer_ok
3000 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3001 those file descriptors could be used for SMTP connections. If anything
3002 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3003 could be sent to the SMTP client, causing chaos. The daemon now opens
3004 stdin, stdout, and stderr to /dev/null when it puts itself into the
3007 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3008 log by default. The user can override this with suitable Perl magic.
3010 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3011 the log message when discard triggers, was not working for the DATA ACL or
3012 for the non-SMTP ACL.
3014 78. Error message wording change in sieve.c.
3016 79. If smtp_accept_max_per_host was set, the number of connections could be
3017 restricted to fewer than expected, because the daemon was trying to set up
3018 a new connection before checking whether the processes handling previous
3019 connections had finished. The check for completed processes is now done
3020 earlier. On busy systems, this bug wouldn't be noticed because something
3021 else would have woken the daemon, and it would have reaped the completed
3024 80. If a message was submitted locally by a user whose login name contained one
3025 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3026 It caused a spool format error. I have fixed the spool reading code. A
3027 related problem was that the "from" clause in the Received: line became
3028 illegal because of the space(s). It is now covered by ${quote_local_part.
3030 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3033 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3035 83. Patch from Sieve maintainer to fix unterminated string problem in
3036 "vacation" handling.
3038 84. Some minor changes to the Linux configuration files to help with other
3039 OS variants using glibc.
3041 85. One more patch for Sieve to update vacation handling to latest spec.
3044 ----------------------------------------------------
3045 See the note above about the 4.44 and 4.50 releases.
3046 ----------------------------------------------------
3052 1. Change 4.43/35 introduced a bug that caused file counts to be
3053 incorrectly computed when quota_filecount was set in an appendfile
3056 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3057 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3060 3. Give more explanation in the error message when the command for a transport
3061 filter fails to execute.
3063 4. There are several places where Exim runs a non-Exim command in a
3064 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3065 was being done only for the command run by the queryprogram router. It is
3066 now done for all such subprocesses. The other cases are: ${run, transport
3067 filters, and the commands run by the lmtp and pipe transports.
3069 5. Some older OS have a limit of 256 on the maximum number of file
3070 descriptors. Exim was using setrlimit() to set 1000 as a large value
3071 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3072 systems. I've change it so that if it can't get 1000, it tries for 256.
3074 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3075 was an oversight, and furthermore, ever since the addition of extra
3076 controls (e.g. 4.43/32), the checks on when to allow different forms of
3077 "control" were broken. There should now be diagnostics for all cases when a
3078 control that does not make sense is encountered.
3080 7. $recipients is now available in the predata ACL (oversight).
3082 8. Tidy the search cache before the fork to do a delivery from a message
3083 received from the command line. Otherwise the child will trigger a lookup
3084 failure and thereby defer the delivery if it tries to use (for example) a
3085 cached ldap connection that the parent has called unbind on.
3087 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3088 of $address_data from the recipient verification was clobbered by the
3089 sender verification.
3091 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3092 was its contents. (It was OK if the option was not defined at all.)
3094 11. A "Completed" log line is now written for messages that are removed from
3095 the spool by the -Mrm option.
3097 12. $host_address is now set to the target address during the checking of
3098 ignore_target_hosts.
3100 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3101 being passed; this would have caused $sender_host_name to have been used if
3102 matching the list had actually called for a host name (not very likely,
3103 since this list is usually IP addresses). A host name is now passed as
3106 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3107 code that specifies a non-restarting handler (typically sigaction() in
3108 modern systems) in an attempt to fix a rare and obscure crash bug.
3110 15. Narrowed the window for a race in the daemon that could cause it to ignore
3111 SIGCHLD signals. This is not a major problem, because they are used only to
3112 wake it up if nothing else does.
3114 16. A malformed maildirsize file could cause Exim to calculate negative values
3115 for the mailbox size or file count. Odd effects could occur as a result.
3116 The maildirsize information is now recalculated if the size or filecount
3119 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3120 support for a long time. Removed HAVE_SYS_VFS_H.
3122 18. Updated exipick to current release from John Jetmore.
3124 19. Allow an empty sender to be matched against a lookup in an address list.
3125 Previously the only cases considered were a regular expression, or an
3128 20. Exim went into a mad DNS lookup loop when doing a callout where the
3129 host was specified on the transport, if the DNS lookup yielded more than
3132 21. The RFC2047 encoding function was originally intended for short strings
3133 such as real names; it was not keeping to the 75-character limit for
3134 encoded words that the RFC imposes. It now respects the limit, and
3135 generates multiple encoded words if necessary. To be on the safe side, I
3136 have increased the buffer size for the ${rfc2047: expansion operator from
3139 22. Failure to deliver a bounce message always caused it to be frozen, even if
3140 there was an errors_to setting on the router. The errors_to setting is now
3143 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3144 canonical form (fully expanded) before being placed in
3145 $sender_host_address.
3147 24. Updated eximstats to version 1.33
3149 25. Include certificate and key file names in error message when GnuTLS fails
3150 to set them up, because the GnuTLS error message doesn't include the name
3151 of the failing file when there is a problem reading it.
3153 26. Expand error message when OpenSSL has problems setting up cert/key files.
3156 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3157 (this can affect the format of dates).
3159 28. exim_tidydb, when checking for the continued existence of a message for
3160 which it has found a message-specific retry record, was not finding
3161 messages that were in split spool directories. Consequently, it was
3162 deleting retry records that should have stayed in existence.
3164 29. eximstats updated to version 1.35
3165 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3166 1.35 - bugfix such that pie charts by volume are generated correctly
3168 30. The SPA authentication driver was not abandoning authentication and moving
3169 on to the next authenticator when an expansion was forced to fail,
3170 contradicting the general specification for all authenticators. Instead it
3171 was generating a temporary error. It now behaves as specified.
3173 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3174 (the order specifies the preference for clients). The order is now AES256,
3175 AES128, 3DES, ARCFOUR128.
3177 31. Small patch to Sieve code - explicitly set From: when generating an
3180 32. Exim crashed if a remote delivery caused a very long error message to be
3181 recorded - for instance if somebody sent an entire SpamAssassin report back
3182 as a large number of 550 error lines. This bug was coincidentally fixed by
3183 increasing the size of one of Exim's internal buffers (big_buffer) that
3184 happened as part of the Exiscan merge. However, to be on the safe side, I
3185 have made the code more robust (and fixed the comments that describe what
3188 33. Some experimental protocols are using DNS PTR records for new purposes. The
3189 keys for these records are domain names, not reversed IP addresses. The
3190 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3191 leaves it alone. Component reversal etc. now happens only for IP addresses.
3194 34. The host_aton() function is supposed to be passed a string that is known
3195 to be a valid IP address. However, in the case of IPv6 addresses, it was
3196 not checking this. This is a hostage to fortune. Exim now panics and dies
3197 if the condition is not met. A case was found where this could be provoked
3198 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3199 components; fortuitously, this particular loophole had already been fixed
3200 by change 4.50/55 or 4.44/33 above.
3202 If there are any other similar loopholes, the new check in host_aton()
3203 itself should stop them being exploited. The report I received stated that
3204 data on the command line could provoke the exploit when Exim was running as
3205 exim, but did not say which command line option was involved. All I could
3206 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3207 running as the user.
3210 35. There was a buffer overflow vulnerability in the SPA authentication code
3211 (which came originally from the Samba project). I have added a test to the
3212 spa_base64_to_bits() function which I hope fixes it.
3215 36. The daemon start-up calls getloadavg() while still root for those OS that
3216 need the first call to be done as root, but it missed one case: when
3217 deliver_queue_load_max is set with deliver_drop_privilege. This is
3218 necessary for the benefit of the queue runner, because there is no re-exec
3219 when deliver_drop_privilege is set.
3221 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3222 was in use, was not putting the data itself into the right store pool;
3223 consequently, it could be overwritten for a subsequent message in the same
3224 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3227 38. Sometimes the final signoff response after QUIT could fail to get
3228 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3229 < 0 before doing a fflush(). This bug looks as though it goes back to the
3230 introduction of TLS in release 3.20, but "sometimes" must have been rare
3231 because the tests only now provoked it.
3237 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3238 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3239 Now it returns an integer. A number of calls were still expecting a T/F
3240 return. Fortuitously, in all cases, the tests worked in OK situations,
3241 which is the norm. However, things would have gone wrong on any write
3242 failures on the smtp file descriptor. This function is used when sending
3243 messages over SMTP and also when doing verify callouts.
3245 2. When Exim is called to do synchronous delivery of a locally submitted
3246 message (the -odf or -odi options), it no longer closes stderr before doing
3249 3. Implemented the mua_wrapper option.
3251 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3253 5. Implemented the functions header_remove(), header_testname(),
3254 header_add_at_position(), and receive_remove_recipient(), and exported them
3257 6. If an ACL "warn" statement specified the addition of headers, Exim already
3258 inserted X-ACL-Warn: at the start if there was no header name. However, it
3259 was not making this test for the second and subsequent header lines if
3260 there were newlines in the string. This meant that an invalid header could
3261 be inserted if Exim was badly configured.
3263 7. Allow an ACL "warn" statement to add header lines at the start or after all
3264 the Received: headers, as well as at the end.
3266 8. Added the rcpt_4xx retry error code.
3268 9. Added postmaster_mailfrom=xxx to callout verification option.
3270 10. Added mailfrom=xxxx to the callout verification option, for verify=
3273 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3274 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3276 12. Inserted some casts to stop certain compilers warning when using pointer
3277 differences as field lengths or precisions in printf-type calls (mostly
3278 affecting debugging statements).
3280 13. Added optional readline() support for -be (dynamically loaded).
3282 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3283 same clock tick as a message's arrival, so that its received time was the
3284 same as the "first fail" time on the retry record, and that message
3285 remained on the queue past the ultimate address timeout, every queue runner
3286 would try a delivery (because it was past the ultimate address timeout) but
3287 after another failure, the ultimate address timeout, which should have then
3288 bounced the address, did not kick in. This was a "< instead of <=" error;
3289 in most cases the first failure would have been in the next clock tick
3290 after the received time, and all would be well.
3292 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3293 being recognized when the domain list was tested by the match_domain
3294 condition in an expansion string.
3296 16. Added the ${str2b64: operator.
3298 17. Exim was always calling setrlimit() to set a large limit for the number of
3299 processes, without checking whether the existing limit was already
3300 adequate. (It did check for the limit on file descriptors.) Furthermore,
3301 errors from getrlimit() and setrlimit() were being ignored. Now they are
3302 logged to the main and panic logs, but Exim does carry on, to try to do its
3303 job under whatever limits there are.
3305 18. Imported PCRE 5.0.
3307 19. Trivial typo in log message " temporarily refused connection" (the leading
3310 20. If the log selector return_path_on_delivery was set and an address was
3311 redirected to /dev/null, the delivery process crashed because it assumed
3312 that a return path would always be set for a "successful" delivery. In this
3313 case, the whole delivery is bypassed as an optimization, and therefore no
3316 21. Internal re-arrangement: the function for sending a challenge and reading
3317 a response while authentication was assuming a zero-terminated challenge
3318 string. It's now changed to take a pointer and a length, to allow for
3319 binary data in such strings.
3321 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3323 23. Exim was not respecting finduser_retries when seeking the login of the
3324 uid under which it was called; it was always trying 10 times. (The default
3325 setting of finduser_retries is zero.) Also, it was sleeping after the final
3326 failure, which is pointless.
3328 24. Implemented tls_on_connect_ports.
3330 25. Implemented acl_smtp_predata.
3332 26. If the domain in control=submission is set empty, Exim assumes that the
3333 authenticated id is a complete email address when it generates From: or
3334 Sender: header lines.
3336 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3337 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3338 chown and chgrp in /bin and hostname in /usr/bin.
3340 28. Exim was keeping the "process log" file open after each use, just as it
3341 does for the main log. This opens the possibility of it remaining open for
3342 long periods when the USR1 signal hits a daemon. Occasional processlog
3343 errors were reported, that could have been caused by this. Anyway, it seems
3344 much more sensible not to leave this file open at all, so that is what now
3347 29. The long-running daemon process does not normally write to the log once it
3348 has entered its main loop, and it closes the log before doing so. This is
3349 so that log files can straightforwardly be renamed and moved. However,
3350 there are a couple of unusual error situations where the daemon does write
3351 log entries, and I had neglected to close the log afterwards.
3353 30. The text of an SMTP error response that was received during a remote
3354 delivery was being truncated at 512 bytes. This is too short for some of
3355 the long messages that one sometimes sees. I've increased the limit to
3358 31. It is now possible to make retry rules that apply only when a message has a
3359 specific sender, in particular, an empty sender.
3361 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3362 it possible to be selective about when SMTP synchronization is enforced.
3364 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3366 32. Implemented hosts_connection_nolog.
3368 33. Added an ACL for QUIT.
3370 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3373 35. Added mailbox_size and mailbox_filecount to appendfile.
3375 36. Added control = no_multiline_responses to ACLs.
3377 37. There was a bug in the logic of the code that waits for the clock to tick
3378 in the case where the clock went backwards by a substantial amount such
3379 that the microsecond fraction of "now" was more than the microsecond
3380 fraction of "then" (but the whole seconds number was less).
3382 38. Added support for the libradius Radius client library this is found on
3383 FreeBSD (previously only the radiusclient library was supported).
3389 1. When certain lookups returned multiple values in the form name=value, the
3390 quoting of the values was not always being done properly. Specifically:
3391 (a) If the value started with a double quote, but contained no whitespace,
3393 (b) If the value contained whitespace other than a space character (i.e.
3394 tabs or newlines or carriage returns) it was not quoted.
3395 This fix has been applied to the mysql and pgsql lookups by writing a
3396 separate quoting function and calling it from the lookup code. The fix
3397 should probably also be applied to nisplus, ibase and oracle lookups, but
3398 since I cannot test any of those, I have not disturbed their existing code.
3400 2. A hit in the callout cache for a specific address caused a log line with no
3401 reason for rejecting RCPT. Now it says "Previous (cached) callout
3402 verification failure".
3404 3. There was an off-by-one bug in the queryprogram router. An over-long
3405 return line was truncated at 256 instead of 255 characters, thereby
3406 overflowing its buffer with the terminating zero. As well as fixing this, I
3407 have increased the buffer size to 1024 (and made a note to document this).
3409 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3410 when Exim is waiting for an SMTP response from a remote server, Exim
3411 restarts its select() call on the socket, thereby resetting its timeout.
3412 This is not a problem when such interrupts are rare. Somebody set up a cron
3413 job to run exiwhat every 2 minutes, which is less than the normal select()
3414 timeout (5 or 10 minutes). This meant that the select() timeout never
3415 kicked in because it was always reset. I have fixed this by comparing the
3416 time when an interrupt arrives with the time at the start of the first call
3417 to select(). If more time than the timeout has elapsed, the interrupt is
3418 treated as a timeout.
3420 5. Some internal re-factoring in preparation for the addition of Sieve
3421 extensions (by MH). In particular, the "personal" test is moved to a
3422 separate function, and given an option for scanning Cc: and Bcc: (which is
3423 not set for Exim filters).
3425 6. When Exim created an email address using the login of the caller as the
3426 local part (e.g. when creating a From: or Sender: header line), it was not
3427 quoting the local part when it contained special characters such as @.
3429 7. Installed new OpenBSD configuration files.
3431 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3432 try to make them clearer.
3434 9. Callout options, other than the timeout value, were being ignored when
3435 verifying sender addresses in header lines. For example, when using
3437 verify = header_sender/callout=no_cache
3439 the cache was (incorrectly) being used.
3441 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3442 only the Cygwin environment.
3444 11. When return_path_on_delivery was set as a log selector, if different remote
3445 addresses in the same message used different return paths and parallel
3446 remote delivery occurred, the wrong values would sometimes be logged.
3447 (Whenever a remote delivery process finished, the return path value from
3448 the most recently started remote delivery process was logged.)
3450 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3451 lines when AUTH and/or TLS are in use. This is the "received protocol"
3452 field. Exim used to use "asmtp" for authenticated SMTP, without any
3453 indication (in the protocol name) for TLS use. Now it follows the RFC and
3454 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3455 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3456 names appear in log lines as well as in Received: header lines.
3458 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3459 extensions, and comparison tests, and to fix some bugs.
3461 14. Changes to the "personal" filter test:
3463 (1) The test was buggy in that it was just doing the equivalent of
3464 "contains" tests on header lines. For example, if a user's address was
3465 anne@some.where, the "personal" test would incorrectly be true for
3467 To: susanne@some.where
3469 This test is now done by extracting each address from the header in turn,
3470 and checking the entire address. Other tests that are part of "personal"
3471 are now done using regular expressions (for example, to check local parts
3472 of addresses in From: header lines).
3474 (2) The list of non-personal local parts in From: addresses has been
3475 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3476 taken from the Sieve specification recommendations.
3478 (3) If the message contains any header line starting with "List-" it is
3479 treated as non-personal.
3481 (4) The test for "circular" in the Subject: header line has been removed
3482 because it now seems ill-conceived.
3484 15. Minor typos in src/EDITME comments corrected.
3486 16. Installed latest exipick from John Jetmore.
3488 17. If headers_add on a router specified a text string that was too long for
3489 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3490 of string_sprintf() is now avoided.
3492 18. $message_body_size was not set (it was always zero) when running the DATA
3493 ACL and the local_scan() function.
3495 19. For the "mail" command in an Exim filter, no default was being set for
3496 the once_repeat time, causing a random time value to be used if "once" was
3497 specified. (If the value happened to be <= 0, no repeat happened.) The
3498 default is now 0s, meaning "never repeat". The "vacation" command was OK
3499 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3500 (I found it when inspecting the code).
3502 20. There is now an overall timeout for performing a callout verification. It
3503 defaults to 4 times the callout timeout, which applies to individual SMTP
3504 commands during the callout. The overall timeout applies when there is more
3505 than one host that can be tried. The timeout is checked before trying the
3506 next host. This prevents very long delays if there are a large number of
3507 hosts and all are timing out (e.g. when the network connections are timing
3508 out). The value of the overall timeout can be changed by specifying an
3509 additional sub-option for "callout", called "maxwait". For example:
3511 verify = sender/callout=5s,maxwait=20s
3513 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3514 to the end before writing, but this should make it even safer).
3516 22. Exim was forgetting that it had advertised PIPELINING for the second and
3517 subsequent messages on an SMTP connection. It was also not resetting its
3518 memory on STARTTLS and an internal HELO.
3520 23. When Exim logs an SMTP synchronization error within a session, it now
3521 records whether PIPELINING has been advertised or not.
3523 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3524 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3525 rather than long int.
3527 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3529 26. Added the never_mail option to autoreply.
3535 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3536 crash if the getsockname() call failed; this can happen if a connection is
3537 closed very soon after it is established. The problem was simply in the
3538 order in which certain operations were done, causing Exim to try to write
3539 to the SMTP stream before it had set up the file descriptor. The bug has
3540 been fixed by making things happen in the correct order.
3546 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3547 before the connection was closed, thus losing the rejection response.
3549 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3550 some early Solaris releases, but causes trouble in current releases where
3551 socklen_t is defined.
3553 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3556 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3559 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3560 that releasing the top of it at the end releases what was used for sub-
3561 expansions (unless the block got too big). However, discard this block if
3562 the first thing is a variable or header, so that we can use its block when
3563 it is dynamic (useful for very large $message_headers, for example).
3565 6. Lookups now cache *every* query, not just the most recent. A new, separate
3566 store pool is used for this. It can be recovered when all lookup caches are
3567 flushed. Lookups now release memory at the end of their result strings.
3568 This has involved some general refactoring of the lookup sources.
3570 7. Some code has been added to the store_xxx() functions to reduce the amount
3571 of flapping under certain conditions.
3573 8. log_incoming_interface used to affect only the <= reception log lines. Now
3574 it causes the local interface and port to be added to several more SMTP log
3575 lines, for example "SMTP connection from", and rejection lines.
3577 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3579 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3581 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3582 could be overwritten at the end of the current message (or the start of a
3583 new message if it was set in a HELO ACL). The value is now preserved for
3584 the duration of the SMTP connection.
3586 12. If a transport had a headers_rewrite setting, and a matching header line
3587 contained an unqualified address, that address was qualified, even if it
3588 did not match any rewriting rules. The underlying bug was that the values
3589 of the flags that permit the existence of unqualified sender and recipient
3590 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3591 non-local messages, and by -bnq for local messages) were not being
3592 preserved with the message after it was received.
3594 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3595 "next input=" as part of the text comprising the host identity instead of
3596 the correct text. The code was using the same buffer for two different
3597 strings. However, depending on which order the printing function evaluated
3598 its arguments, the bug did not always show up. Under Linux, for example, my
3599 test suite worked just fine.
3601 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3602 doesn't work with some older versions of Perl. It has been changed to "my",
3603 which in any case is probably the better facility to use.
3605 15. A really picky compiler found some instances of statements for creating
3606 error messages that either had too many or two few arguments for the format
3609 16. The size of the buffer for calls to the DNS resolver has been increased
3610 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3611 for addresses that have a lot of PTR records. This alleviates a problem; it
3612 does not fully solve it.
3614 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3615 buffer now truncates the list and logs the incident, which is the same
3616 action as happens when Exim is looking up a host name and its aliases.
3617 Previously in this situation something unpredictable would happen;
3618 sometimes it was "internal error: store_reset failed".
3620 18. If a server dropped the connection unexpectedly when an Exim client was
3621 using GnuTLS and trying to read a response, the client delivery process
3622 crashed while trying to generate an error log message.
3624 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3625 string, for example:
3627 warn message = H1: something\nH2: something
3629 the text was added as a single header line from Exim's point of view
3630 though it ended up OK in the delivered message. However, searching for the
3631 second and subsequent header lines using $h_h2: did not work. This has been
3632 fixed. Similarly, if a system filter added multiple headers in this way,
3633 the routers could not see them.
3635 20. Expanded the error message when iplsearch is called with an invalid key to
3636 suggest using net-iplsearch in a host list.
3638 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3639 ACLs are no longer actually imposed (and a message to that effect is
3642 22. If a "gecos" field in a passwd entry contained escaped characters, in
3643 particular, if it contained a \" sequence, Exim got it wrong when building
3644 a From: or a Sender: header from that name. A second bug also caused
3645 incorrect handling when an unquoted " was present following a character
3646 that needed quoting.
3648 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3649 was not being matched caselessly.
3651 24. Arranged for all hyphens in the exim.8 source to be escaped with
3654 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3655 properly. Recipient callout cache records were still being keyed to include
3656 the sender, even when use_sender was set false. This led to far more
3657 callouts that were necessary. The sender is no longer included in the key
3658 when use_sender is false.
3660 26. Added "control = submission" modifier to ACLs.
3662 27. Added the ${base62d: operator to decode base 62 numbers.
3664 28. dnsdb lookups can now access SRV records.
3666 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3667 the configuration file.
3669 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3670 (-v) mode. This makes the output for a verbose queue run more intelligible.
3672 31. Added a use_postmaster feature to recipient callouts.
3674 32. Added the $body_zerocount variable, containing the number of binary zero
3675 bytes in the message body.
3677 33. The time of last modification of the "new" subdirectory is now used as the
3678 "mailbox time last read" when there is a quota error for a maildir
3681 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3683 35. Added +ignore_unknown as a special item in host lists.
3685 36. Code for decoding IPv6 addresses in host lists is now included, even if
3686 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3687 address was recognized as an IP address, but was then not correctly decoded
3688 into binary, causing unexpected and incorrect effects when compared with
3695 1. Very minor rewording of debugging text in manualroute to say "list of
3696 hosts" instead of "hostlist".
3698 2. If verify=header_syntax was set, and a header line with an unqualified
3699 address (no domain) and a large number of spaces between the end of the
3700 name and the colon was received, the reception process suffered a buffer
3701 overflow, and (when I tested it) crashed. This was caused by some obsolete
3702 code that should have been removed. The fix is to remove it!
3704 3. When running in the test harness, delay a bit after writing a bounce
3705 message to get a bit more predictability in the log output.
3707 4. Added a call to search_tidyup() just before forking a reception process. In
3708 theory, someone could use a lookup in the expansion of smtp_accept_max_
3709 per_host which, without the tidyup, could leave open a database connection.
3711 5. Added the variables $recipient_data and $sender_data which get set from a
3712 lookup success in an ACL "recipients" or "senders" condition, or a router
3713 "senders" option, similar to $domain_data and $local_part_data.
3715 6. Moved the writing of debug_print from before to after the "senders" test
3718 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3719 problems for message scanning, either using a data ACL, or using
3720 local_scan() because the Received: header was not generated till after they
3721 were called (in order to set the time as the time of reception completion).
3722 I have revised the way this works. The header is now generated after the
3723 body is received, but before the ACL or local_scan() are called. After they
3724 are run, the timestamp in the header is updated.
3730 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3731 before starting a queue runner without re-exec. This happened only when
3732 deliver_drop_privilege was set or when the Exim user was set to root. The
3733 effect of the bug was that timeouts during subsequent deliveries caused
3734 crashes instead of being properly handled. The handler is now left at its
3735 default (and expected) setting.
3737 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3738 message, again when deliver_drop_privilege is set or Exim is run as root.
3739 The bug described in (1) was not present in this case, but the tidying up
3740 of the other signals was missing. I have made the two cases consistent.
3742 3. The ignore_target_hosts setting on a manualroute router was being ignored
3743 for hosts that were looked up using the /MX notation.
3745 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3748 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3749 operated on the sender address. After changing the $sender_address to <>
3750 for the sender address verify, Exim was re-instated it as the original
3751 (before rewriting) address, but remembering that it had rewritten it, so it
3752 wasn't rewriting it again. This bug also had the effect of breaking the
3753 sender address verification caching when the sender address was rewritten.
3755 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3756 This has been changed so that if the ip literal address matches
3757 ignore_target_hosts, the router declines.
3759 7. Added expansion conditions match_domain, match_address, and match_local_
3760 part (NOT match_host).
3762 8. The placeholder for the Received: header didn't have a length field set.
3764 9. Added code to Exim itself and to exim_lock to test for a specific race
3765 condition that could lead to file corruption when using MBX delivery. The
3766 issue is with the lockfile that is created in /tmp. If this file is removed
3767 after a process has opened it but before that process has acquired a lock,
3768 there is the potential for a second process to recreate the file and also
3769 acquire a lock. This could lead to two Exim processes writing to the file
3770 at the same time. The added code performs the same test as UW imapd; it
3771 checks after acquiring the lock that its file descriptor still refers to
3772 the same named file.
3774 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3775 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3778 11. Added the smtp_active_hostname option. If used, this will typically be made
3779 to depend on the incoming interface address. Because $interface_address is
3780 not set up until the daemon has forked a reception process, error responses
3781 that can happen earlier (such as "too many connections") no longer contain
3784 12. If an expansion in a condition on a "warn" statement fails because a lookup
3785 defers, the "warn" statement is abandoned, and the next ACL statement is
3786 processed. Previously this caused the whole ACL to be aborted.
3788 13. Added the iplsearch lookup type.
3790 14. Added ident_timeout as a log selector.
3792 15. Added tls_certificate_verified as a log selector.
3794 16. Added a global option tls_require_ciphers (compare the smtp transport
3795 option of the same name). This controls incoming TLS connections.
3797 17. I finally figured out how to make tls_require_ciphers do a similar thing
3798 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3799 before starting the TLS session.
3801 18. Tabs are now shown as \t in -bP output.
3803 19. If the log selector return_path_on_delivery was set, Exim crashed when
3804 bouncing a message because it had too many Received: header lines.
3806 20. If two routers both had headers_remove settings, and the first one included
3807 a superfluous trailing colon, the final name in the first list and the
3808 first name in the second list were incorrectly joined into one item (with a
3809 colon in the middle).
3815 1. Added -C and -D options to the exinext utility, mainly to make it easier
3816 to include in the automated testing, but these could be helpful when
3817 multiple configurations are in use.
3819 2. The exinext utility was not formatting the output nicely when there was
3820 an alternate port involved in the retry record key, nor when there was a
3821 message id as well (for retries that were specific to a specific message
3822 and a specific host). It was also confused by IPv6 addresses, because of
3823 the additional colons they contain. I have fixed the IPv4 problem, and
3824 patched it up to do a reasonable job for IPv6.
3826 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3827 delivery, the log line now contains "pipelined" if PIPELINING was used.
3829 4. An SMTP transport process used to panic and die if the bind() call to set
3830 an explicit outgoing interface failed. This has been changed; it is now
3831 treated in the same way as a connect() failure.
3833 5. A reference to $sender_host_name in the part of a conditional expansion
3834 that was being skipped was still causing a DNS lookup. This no longer
3837 6. The def: expansion condition was not recognizing references to header lines
3838 that used bh_ and bheader_.
3840 7. Added the _cache feature to named lists.
3842 8. The code for checking quota_filecount in the appendfile transport was
3843 allowing one more file than it should have been.
3845 9. For compatibility with Sendmail, the command line option
3853 and sets the incoming protocol and host name (for trusted callers). The
3854 host name and its colon can be omitted when only the protocol is to be set.
3855 Note the Exim already has two private options, -pd and -ps, that refer to
3856 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3857 "s", but I don't think that's a major issue.
3859 10. A number of refactoring changes to the code, none of which should affect
3862 (a) The number of logging options was getting close to filling up the
3863 32-bit word that was used as a bit map. I have split them into two classes:
3864 those that are passed in the argument to log_write(), and those that are
3865 only ever tested independently outside of that function. These are now in
3866 separate 32-bit words, so there is plenty of room for expansion again.
3867 There is no change in the user interface or the logging behaviour.
3869 (b) When building, for example, log lines, the code previously used a
3870 macro that called string_cat() twice, in order to add two strings. This is
3871 not really sufficiently general. Furthermore, there was one instance where
3872 it was actually wrong because one of the argument was used twice, and in
3873 one call a function was used. (As it happened, calling the function twice
3874 did not affect the overall behaviour.) The macro has been replaced by a
3875 function that can join an arbitrary number of extra strings onto a growing
3878 (c) The code for expansion conditions now uses a table and a binary chop
3879 instead of a serial search (which was left over from when there were very
3880 few conditions). Also, it now recognizes conditions like "pam" even when
3881 the relevant support is not compiled in: a suitably worded error message is
3882 given if an attempt is made to use such a condition.
3884 11. Added ${time_interval:xxxxx}.
3886 12. A bug was causing one of the ddress fields not to be passed back correctly
3887 from remote delivery subprocesses. The field in question was not being
3888 subsequently used, so this caused to problems in practice.
3890 13. Added new log selectors queue_time and deliver_time.
3892 14. Might have fixed a bug in maildirsizefile handling that threw up
3893 "unexpected character" debug warnings, and recalculated the data
3894 unnecessarily. In any case, I expanded the warning message to give more
3897 15. Added the message "Restricted characters in address" to the statements in
3898 the default ACL that block characters like @ and % in local parts.
3900 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3901 Three changes have been made:
3903 (a) There was a serious bug; a negative response to MAIL caused the whole
3904 recipient domain to be cached as invalid, thereby blocking all messages
3905 to all local parts at the same domain, from all senders. This bug has
3906 been fixed. The domain is no longer cached after a negative response to
3907 MAIL if the sender used is not empty.
3909 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3912 (c) A new callout option, "use_sender" has been added for people who want
3913 the modified behaviour.
3919 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3922 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3923 indeed breaks things for older releases.
3925 3. Added additional logging to the case where there is a problem reading data
3926 from a filter that is running in a subprocess using a pipe, in order to
3927 try to track down a specific problem.
3929 4. Testing facility fudge: when running in the test harness and attempting
3930 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3931 getting "No route to host". Convert this to a timeout.
3933 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3936 6. Some OS don't have socklen_t but use size_t instead. This affects the
3937 fifth argument of getsockopt() amongst other things. This is now
3938 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3939 can be set for individual OS. I have set it for SunOS5, OSF1, and
3940 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3941 some earlier ones do not.
3943 7. Change 4.30/15 was not doing the test caselessly.
3945 8. The standard form for an IPv6 address literal was being rejected by address
3946 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3947 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3948 this, as well as the form without the "IPv6" on the front (but only when
3949 address literals are enabled, of course).
3951 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3953 10. Exim crashed if a message with an empty sender address specified by -f
3954 encountered a router with an errors_to setting. This could be provoked only
3955 by a command such as
3959 where an empty string was supplied; "<>" did not hit this bug.
3961 11. Installed PCRE release 4.5.
3963 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3964 remained set. It is now erased.
3966 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3967 times from message ids (which are base 36 rather than the normal 62).
3969 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3970 were being counted as actual protocol errors, and logged if the log
3971 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3972 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3973 and DATA following a set of rejected RCPTs do not count as protocol errors.
3974 In other words, Exim assumes they were pipelined, though this may not
3975 actually be the case. Of course, in all cases the client gets an
3976 appropriate error code.
3978 15. If a lookup fails in an ACL condition, a message about the failure may
3979 be available; it is used if testing the ACL cannot continue, because most
3980 such messages specify what the cause of the deferral is. However, some
3981 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3982 that caused an old message to be retained and used if a later statement
3983 caused a defer, replacing the real cause of the deferral.
3985 16. If an IP address had so many PTR records that the DNS lookup buffer
3986 was not large enough to hold them, Exim could crash while trying to process
3987 the truncated data. It now detects and logs this case.
3989 17. Further to 4.21/58, another change has been made: if (and only if) the
3990 first line of a message (the first header line) ends with CRLF, a bare LF
3991 in a subsequent header line has a space inserted after it, so as not to
3992 terminate the header.
3994 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3995 unnecessarily, used atoi() instead of strtol(), and didn't check the
3996 termination when getting file sizes from file names by regex.
3998 19. Completely re-implemented the support for maildirsize files, in the light
3999 of a number of problems with the previous contributed implementation
4000 (4.30/29). In particular:
4002 . If the quota is zero, the maildirsize file is maintained, but no quota is
4005 . If the maildir directory does not exist, it is created before any attempt
4006 to write a maildirsize file.
4008 . The quota value in the file is just a cache; if the quota is changed in
4009 the transport, the new value overrides.
4011 . A regular expression is available for excluding directories from the
4014 20. The autoreply transport checks the characters in options that define the
4015 message's headers; it allows continued headers, but it was checking with
4016 isspace() after an embedded newline instead of explicitly looking for a
4019 21. If all the "regular" hosts to which an address was routed had passed their
4020 expiry times, and had not reached their retry times, the address was
4021 bounced, even if fallback hosts were defined. Now Exim should go on to try
4024 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4025 equivalent code in the SMTP transport. Some hosts send humungous responses
4026 to HELO/EHLO, more than 1024 it seems.
4028 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4029 gives compiler warnings in some environments. I've now done it "properly",
4032 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4033 (because of problems with the built-in one) was declared to return uschar *
4034 instead of char *, causing compiler failure.
4036 25. Fixed a file descriptor leak when processing alias/forward files.
4038 26. Fixed a minor format string issue in dbfn.c.
4040 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4042 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4043 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4046 29. When a sender address is verified, it is cached, to save repeating the test
4047 when there is more than one recipient in a message. However, when the
4048 verification involves a callout, it is possible for different callout
4049 options to be set for different recipients. It is too complicated to keep
4050 track of this in the cache, so now Exim always runs a verification when a
4051 callout is required, relying on the callout cache for the optimization.
4052 The overhead is duplication of the address routing, but this should not be
4055 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4056 to be verified with callout=postmaster, and the main callout worked but the
4057 postmaster check failed, the verification correctly failed. However, if a
4058 subsequent RCPT command asked for sender verification *without* the
4059 postmaster check, incorrect caching caused this verification also to fail,
4062 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4063 it was not caching the DNS options (qualify_single, search_parents) that
4064 were used when the lookup failed. A subsequent lookup with different
4065 options therefore always gave the same answer, though there were cases
4066 where it should not have. (Example: a "domains = !$mx_any" option on a
4067 dnslookup router: the "domains" option is always processed without any
4068 widening, but the router might have qualify_single set.) Now Exim uses the
4069 cached value only when the same options are set.
4071 32. Added John Jetmore's "exipick" utility to the distribution.
4073 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4074 than a timeout (e.g. a certificate is required, and is not provided), an
4075 Exim server now closes the connection immediately. Previously it waited for
4076 the client to close - but if the client is SSL, it seems that they each
4077 wait for each other, leading to a delay before one of them times out.
4079 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4080 maintained 0.8.x compatibility because I don't think many are using it, and
4081 it is clearly obsolete.
4083 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4086 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4087 client certificate was expired. A simple patch fixes this, though I don't
4088 understand the full logic of why the verify callback is called multiple
4091 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4092 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4093 which causes problems with some clients (such as the Certicom SSL Plus
4094 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4095 disables the coutermeasure allowing Eudora to connect."
4097 38. Exim was not checking that a write() to a log file succeeded. This could
4098 lead to Bad Things if a log got too big, in particular if it hit a file
4099 size limit. Exim now panics and dies if it cannot write to a log file, just
4100 as it does if it cannot open a log file.
4102 39. Modified OS/Makefile-Linux so that it now contains
4104 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4106 The two -D definitions ensure that Exim is compiled with large file
4107 support, which makes it possible to handle log files that are bigger than
4110 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4111 instance) a domain was checked against a named list that involved a lookup,
4112 causing $domain_data to be set, then another domain was checked against the
4113 same list, then the first domain was re-checked, the value of $domain_data
4114 after the final check could be wrong. In particular, if the second check
4115 failed, it could be set empty. This bug probably also applied to
4118 41. The strip_trailing_dot option was not being applied to the address given
4119 with the -f command-line option.
4121 42. The code for reading a message's header from the spool was incrementing
4122 $received_count, but never initializing it. This meant that the value was
4123 incorrect (doubled) while delivering a message in the same process in which
4124 it was received. In the most common configuration of Exim, this never
4125 happens - a fresh exec is done - but it can happen when
4126 deliver_drop_privilege is set.
4128 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4129 it now includes up to 150 characters of the unexpected data in the log
4132 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4133 and building data strings. The size of both of these buffers was 10 000
4134 bytes - far larger than anybody would *ever* want, thought I. Needless to
4135 say, somebody hit the limit. I have increased the maximum line length to
4136 20 000 and the maximum data length of concatenated lines to 100 000. I have
4137 also fixed two bugs, because there was no checking on these buffers. Tsk,
4138 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4139 buffer is too small.
4141 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4142 lsearch lookups. Now it does.
4144 46. When parsing a route_list item in a manualroute router, a fixed-length
4145 buffer was used for the list of hosts. I made this 1024 bytes long,
4146 thinking that nobody would ever have a list of hosts that long. Wrong.
4147 Somebody had a whole pile of complicated expansion conditions, and the
4148 string was silently truncated, leading to an expansion error. It turns out
4149 that it is easier to change to an unlimited length (owing to other changes
4150 that have happened since this code was originally written) than to build
4151 structure for giving a limitation error. The length of the item that
4152 expands into the list of hosts is now unlimited.
4154 47. The lsearch lookup could not handle data where the length of text line was
4155 more than 4095 characters. Such lines were truncated, leading to shortened
4156 data being returned. It should now handle lines of any length.
4158 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4159 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4162 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4163 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4164 no longer happens in comment lines. A list of replacements is now placed
4165 at the head of all of the source files, except those whose only change is
4166 to replace PERL_COMMAND in the very first #! line.
4168 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4169 messages on the queue, with a bottom-up merge sort, using code contributed
4170 by Michael Haardt. This should make operations like -bp somewhat faster on
4171 large queues. It won't affect queue runners, except when queue_run_in_order
4174 51. Installed eximstats 1.31 in the distribution.
4176 52. Added support for SRV lookups to the dnslookup router.
4178 53. If an ACL referred to $message_body or $message_body_end, the value was not
4179 reset for any messages that followed in the same SMTP session.
4181 54. The store-handling optimization for building very long strings was not
4182 differentiating between the different store pools. I don't think this
4183 actually made any difference in practice, but I've tidied it.
4185 55. While running the routers to verify a sender address, $sender_address
4186 was still set to the sender address. This is wrong, because when routing to
4187 send a bounce to the sender, it would be empty. Therefore, I have changed
4188 it so that, while verifying a sender address, $sender_address is set to <>.
4189 (There is no change to what happens when verifying a recipient address.)
4191 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4192 target A or AAAA records (if not already returned) without resetting the
4193 qualify_single or search_parents options of the DNS resolver. These are
4194 inappropriate in this case because the targets of MX and SRV records must
4195 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4196 target that, when qualified, matched something in the local domain. These
4197 two options are now turned off when doing these lookups.
4199 57. It seems that at least some releases of Reiserfs (which does not have the
4200 concept of a fixed number of inodes) returns zero and not -1 for the
4201 number of available inodes. This interacted badly with check_spool_inodes,
4202 which assumed that -1 was the "no such thing" setting. What I have done is
4203 to check that the total number of inodes is greater than zero before doing
4204 the test of how many are available.
4206 58. When a "warn" ACL statement has a log_message modifier, the message is
4207 remembered, and not repeated. This is to avoid a lot of repetition when a
4208 message has many recipients that cause the same warning to be written.
4209 Howewer, Exim was preserving the list of already written lines for an
4210 entire SMTP session, which doesn't seem right. The memory is now reset if a
4211 new message is started.
4213 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4214 debugging output unless log_rewrite was also set.
4216 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4217 of (int)(handle) when we know that handle contains (void *)(-1).
4219 61. The Exim daemon panic-logs an error return when it closes the incoming
4220 connection. However "connection reset by peer" seems to be common, and
4221 isn't really an error worthy of noting specially, so that particular error
4224 62. When Exim is trying to find all the local interfaces, it used to panic and
4225 die if the ioctl to get the interface flags failed. However, it seems that
4226 on at least one OS (Solaris 9) it is possible to have an interface that is
4227 included in the list of interfaces, but for which you get a failure error
4228 for this call. This happens when the interface is not "plumbed" into a
4229 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4230 failure of the "get flags" call assumes that the interface is down.
4232 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4233 makes life easier for people who are doing arithmetic on fields extracted
4234 from dates, where you often get leading zeros that should not be
4235 interpreted as octal.
4237 64. Added qualify_domain to the redirect router, to override the global
4240 65. If a pathologically long header line contained very many addresses (the
4241 report of this problem mentioned 10 000) and each of them was rewritten,
4242 Exim could use up a very large amount of memory. (It kept on making new
4243 copies of the header line as it rewrote, and never released the old ones.)
4244 At the expense of a bit more processing, the header rewriting function has
4245 been changed so that it no longer eats memory in this way.
4247 66. The generation of the Received: header has been moved from the time that a
4248 message starts to be received, to the time that it finishes. The timestamp
4249 in the Received: header should now be very close to that of the <= log
4250 line. There are two side-effects of this change:
4252 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4253 logged header lines no longer include the local Received: line, because
4254 it has not yet been created. The same applies to a copy of the message
4255 that is returned to a non-SMTP sender when a message is rejected.
4257 (b) When a filter file is tested using -bf, no additional Received: header
4258 is added to the test message. After some thought, I decided that this
4261 This change does not affect the value of $received_for. It is still set
4262 after address rewriting, but before local_scan() is called.
4264 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4266 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4267 gave an unhelpful panic error message, and a defer error. I have managed to
4268 change this behaviour so that it now rejects any supplied certificate,
4269 which seems right, as the list of acceptable certificates is empty.
4271 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4272 gave an unhelpful defer error. I have not managed to make this reject any
4273 supplied certificates, but the error message it gives is "no certificate
4274 supplied", which is not helpful.
4276 70. exigrep's output now also includes lines that are not associated with any
4277 message, but which match the given pattern. Implemented by a patch from
4278 Martin Sluka, which also tidied up the Perl a bit.
4280 71. Recipient callout verification, like sender verification, was using <> in
4281 the MAIL FROM command. This isn't really the right thing, since the actual
4282 sender may affect whether the remote host accepts the recipient or not. I
4283 have changed it to use the actual sender in the callout; this means that
4284 the cache record is now keyed on a recipient/sender pair, not just the
4285 recipient address. There doesn't seem to be a real danger of callout loops,
4286 since a callout by the remote host to check the sender would use <>.
4287 [SEE ABOVE: changed after hitting problems.]
4289 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4290 temporary errors. However, in the case of such a code being given after
4291 the end of a data transmission (i.e. after ".") Exim was failing to write
4292 a retry record for the message. (Yes, there was some broken host that was
4293 actually sending 8xx at this point.)
4295 73. An unknown lookup type in a host list could cause Exim to panic-die when
4296 the list was checked. (An example that provoked this was putting <; in the
4297 middle of a list instead of at the start.) If this happened during a DATA
4298 ACL check, a -D file could be left lying around. This kind of configuration
4299 error no longer causes Exim to die; instead it causes a defer errror. The
4300 incident is still logged to the main and panic logs.
4302 74. Buglet left over from Exim 3 conversion. The message "too many messages
4303 in one connection" was written to the rejectlog but not the mainlog, except
4304 when address rewriting (yes!) was being logged.
4306 75. Added write_rejectlog option.
4308 76. When a system filter was run not as root (that is, when system_filter_user
4309 was set), the values of the $n variables were not being returned to the
4310 main process; thus, they were not subsequently available in the $sn
4313 77. Added +return_path_on_delivery log selector.
4315 78. A connection timeout was being treated differently from recipients deferred
4316 when testing hosts_max_try with a message that was older than the host's
4317 retry timeout. (The host should not be counted, thus allowing all hosts to
4318 be tried at least once before bouncing.) This may have been the cause of an
4319 occasionally reported bug whereby a message would remain on the queue
4320 longer than the retry timeout, but would be bounced if a delivery was
4321 forced. I say "may" because I never totally pinned down the problem;
4322 setting up timeout/retry tests is difficult. See also the next item.
4324 79. The ultimate address timeout was not being applied to errors that involved
4325 a combination of host plus message (for example, a timeout on a MAIL
4326 command). When an address resolved to a number of possible hosts, and they
4327 were not all tried for each delivery (e.g. because of hosts_max_try), a
4328 message could remain on the queue longer than the retry timeout.
4330 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4333 81. Fixed an obscure SMTP outgoing bug which required at least the following
4334 conditions: (a) there was another message waiting for the same server;
4335 (b) the server returned 5xx to all RCPT commands in the first message so
4336 that the message was not completed; (c) the server dropped the connection
4337 or gave a negative response to the RSET that Exim sends to abort the
4338 transaction. The observed case was a dropped connection after DATA that had
4339 been sent in pipelining mode. That is, the server had advertised PIPELINING
4340 but was not implementing it correctly. The effect of the bug was incorrect
4341 behaviour, such as trying another host, and this could lead to a crash.
4347 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4348 and daemon.c were passed as pointers to ints; they should have been
4349 pointers to socklen_t variables (which are typically unsigned ints).
4351 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4354 3. Fixed a really odd bug that affected only the testing scheme; patching a
4355 certain fixed string in the binary changed the value of another string that
4356 happened to be identical to the end of the original first string.
4358 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4359 name", it returns that address as the IP address. On some operating
4360 systems (e.g. Solaris), it also passes back the IP address string as the
4361 "host name". However, on others (e.g. Linux), it passes back an empty
4362 string. Exim wasn't checking for this, and was changing the host name to an
4363 empty string, assuming it had been canonicized.
4365 5. Although rare, it is permitted to have more than one PTR record for a given
4366 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4367 all the names associated with an address, because they do in Solaris.
4368 However, it seems that they do not in Linux for data that comes from the
4369 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4370 I found this out when I moved to a new Linux workstation and tried to run
4371 the Exim test suite.
4373 To get round this problem I have changed the code so that it now does its
4374 own call to the DNS to look up PTR records when searching for a host name.
4375 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4376 addresses that are only in /etc/hosts are still found.
4378 This behaviour is, however, controlled by an option called host_lookup_
4379 order, which defaults to "bydns:byaddr". If people want to use the other
4380 order, or indeed, just use one or the other means of lookup, they can
4381 specify it in this variable.
4383 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4384 some operating systems, this comes back from gethostbyaddr() as an empty
4385 string, and this is what Exim used to test for. However, it seems that in
4386 other systems, "." is yielded. Exim now tests for this case too.
4388 7. The values of check_spool_space and check_log_space are now held internally
4389 as a number of kilobytes instead of an absolute number of bytes. If a
4390 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4391 kilobyte. This means that much larger values can be stored.
4393 8. Exim monitor: an attempt to get the action menu when not actually pointing
4394 at a message produces an empty menu entitled "No message selected". This
4395 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4396 no entries in it ("Shell widget menu has zero width and/or height"). So I
4397 have added a single, blank menu entry in this case.
4399 9. Added ${quote_local_part.
4401 10. MIME decoding is now applied to the contents of Subject: header lines when
4404 11. Now that a reference to $sender_host_address automatically causes a reverse
4405 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4406 host lookup before query-style lookups in lists that might use this
4407 variable. This has therefore been abolished, and the "net-" prefix is no
4408 longer necessary for query-style lookups.
4410 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4411 have been a typo for LFLAGS, so it has been changed.
4413 13. The install script calls Exim with "-C /dev/null" in order to find the
4414 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4415 to be output. Howeve, since Exim outputs its version number before the
4416 error, it didn't break the script. It just looked ugly. I fixed this by
4417 always allowing "-C /dev/null" if the caller is root.
4419 14. Ignore overlarge ACL variable number when reading spool file - insurance
4420 against a later release with more variables having written the file.
4422 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4423 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4424 this, as well as the form without the "IPv6" on the front.
4426 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4427 OS/Makefile-Darwin file.
4429 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4430 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4432 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4433 variable. However, some OS use ints for inodes. I've added cast to long int
4434 to get rid of the compiler warning.
4436 19. I had forgotten to lock out "/../" in configuration file names when
4437 ALT_CONFIG_PREFIX was set.
4439 20. Routers used for verification do not need to specify transports. However,
4440 if such a router generated a host list, and callout was configured, Exim
4441 crashed, because it could not find a port number from the (non-existent)
4442 transport. It now assumes port 25 in this circumstance.
4444 21. Added the -t option to exigrep.
4446 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4447 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4448 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4449 them is set, LOOKUP_LSEARCH is forced.
4451 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4453 24. Added sender and host information to the "rejected by local_scan()" log
4454 line; previously there was no indication of these.
4456 25. Added .include_if_exists.
4458 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4459 Linux. It turns out that not all file systems support this. Apparently some
4460 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4461 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4462 not supported on the file descriptor, is now ignored when Exim is trying to
4463 sync a directory. This applies only to Linux.
4465 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4467 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4468 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4471 29. Added support for maildirsize files from supplied patch (modified a bit).
4473 30. The use of :fail: followed by an empty string could lead Exim to respond to
4474 sender verification failures with (e.g.):
4476 550 Verification failed for <xxx>
4477 550 Sender verify failed
4479 where the first response line was missing the '-' that indicates it is not
4480 the final line of the response.
4482 31. The loop for finding the name of the user that called Exim had a hardwired
4483 limit of 10; it now uses the value of finduser_retries, which is used for
4484 all other user lookups.
4486 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4489 33. Exim was neglecting to zero errno before one call of strtol() when
4490 expanding a string and expecting an integer value. On some systems this
4491 resulted in spurious "integer overflow" errors. Also, it was casting the
4492 result into an int without checking.
4494 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4495 did not work. The code looks as if it has *never* worked, though it appears
4496 to have been documented since at least releast 1.62. I have made it work.
4498 35. The "timeout_DNS" error in retry rules, also documented since at least
4499 1.62, also never worked. As it isn't clear exactly what this means, and
4500 clearly it isn't a major issue, I have abolished the feature by treating it
4501 as "timeout", and writing a warning to the main and panic logs.
4503 36. The display of retry rules for -brt wasn't always showing the error code
4506 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4507 timeout_connect_A, timeout_connect_MX.
4509 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4510 to the empty sender.
4512 39. The daemon was not analysing the content of -oX till after it had closed
4513 stderr and disconnected from the controlling terminal. This meant that any
4514 syntax errors were only noted on the panic log, and the return code from
4515 the command was 0. By re-arranging the code a little, I've made the
4516 decoding happen first, so such errors now appear on stderr, and the return
4517 code is 1. However, the actual setting up of the sockets still happens in
4518 the disconnected process, so errors there are still only recorded on the
4521 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4522 connections (as happens on some IP stacks) was logged at start up time as
4523 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4524 it from "IPv6 and IPv4", which means that two separate sockets are being
4527 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4528 says whether AF_INET or AF_INET6 was passed as an argument.
4530 42. Exiwhat output was messed up when time zones were included in log
4533 43. Exiwhat now gives more information about the daemon's listening ports,
4534 and whether -tls-on-connect was used.
4536 44. The "port" option of the smtp transport is now expanded.
4538 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4539 silently ignored. Now an error message is written to the main and panic
4542 46. There's a new ACL modifier called "logwrite" which writes to a log file
4543 as soon as it is encountered.
4545 47. Added $local_user_uid and $local_user_gid at routing time.
4547 48. Exim crashed when trying to verify a sender address that was being
4550 49. Exim was recognizing only a space character after ".include". It now also
4551 recognizes a tab character.
4553 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4554 extracting the relevant information from the specification. The man page no
4555 longer contains scrambled data for the -d option, and I've added a section
4556 at the front about calling Exim under different names.
4558 51. Added "extra_headers" argument to the "mail" command in filter files.
4560 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4563 53. Installed eximstats 1.29.
4565 54. Added transport_filter_timeout as a generic transport option.
4567 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4568 Cc: header lines. This was required by RFC 822, but it not required by RFC
4571 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4572 incoming messages that did not have them. Now it does so only if the
4573 message originates locally, that is, if there is no associated remote host
4574 address. When Resent- header lines are present, this applies to the Resent-
4575 lines rather than the non-Resent- lines.
4577 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4578 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4580 58. Messages for configuration errors now include the name of the main
4581 configuration file - useful now that there may be more than one file in a
4582 list (.included file names were always shown).
4584 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4585 for those rare installations that do not start the daemon as root or run it
4586 setuid root. I've cut out the call to initgroups() if the daemon is not
4589 60. The Exim user and group can now be bound into the binary as text strings
4590 that are looked up at the start of Exim's processing.
4592 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4594 62. Added $mailstore_basename variable.
4596 63. Installed patch to sieve.c from Michael Haardt.
4598 64. When Exim failed to open the panic log after failing to open the main log,
4599 the original message it was trying to log was written to stderr and debug
4600 output, but if they were not available (the usual case in production), it
4601 was lost. Now it is written to syslog before the two lines that record the
4602 failures to open the logs.
4604 65. Users' Exim filters run in subprocesses under the user's uid. It is
4605 possible for a "deliver" command or an alias in a "personal" command to
4606 provoke an address rewrite. If logging of address rewriting is configured,
4607 this fails because the process is not running as root or exim. There may be
4608 a better way of dealing with this, but for the moment (because 4.30 needs
4609 to be released), I have disabled address rewrite logging when running a
4610 filter in a non-root, non-exim process.
4616 1. The buildconfig auxiliary program wasn't quoting the value set for
4617 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4618 not defined. This bug was masked in 4.22 by the effect that was fixed in
4621 2. Some messages that were rejected after a message id was allocated were
4622 shown as "incomplete" by exigrep. It no longer does this for messages that
4623 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4625 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4626 have allow_domain_literals set, the ID did not get logged in the <= line.
4627 Domain literals are now always recognized in Message-ID: header lines.
4629 4. The first argument for a ${extract expansion item is the key name or field
4630 number. Leading and trailing spaces in this item were not being ignored,
4631 causing some misleading effects.
4633 5. When deliver_drop_privilege was set, single queue runner processes started
4634 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4635 same command in the process it spins off) were not dropping privilege.
4637 6. When the daemon running as "exim" started a queue runner, it always
4638 re-executed Exim in the spun-off process. This is a waste of effort when
4639 deliver_drop_privilege is set. The new process now just calls the
4640 queue-runner function directly.
4646 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4649 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4650 ignored. Though the use of -oP was forcing the writing of a pid file, it
4651 was always written to the default place.
4653 3. If the message "no IP address found for host xxxx" is generated during
4654 incoming verification, it is now followed by identification of the incoming
4655 connection (so you can more easily find what provoked it).
4657 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4659 5. Added some features to "harden" Exim a bit more against certain attacks:
4661 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4662 be put in Local/Makefile. This is like the never_users runtime option,
4663 but it cannot be overridden. The default setting is "root".
4665 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4666 prefix string with which any file named in a -C command line option
4669 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4670 is retained for -C and -D only if the caller of Exim is root. Without
4671 it, the exim user may also use -C and -D and retain privilege.
4673 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4674 command line option is disabled.
4676 6. Macro names set by the -D option must start with an upper case letter, just
4677 like macro names defined in the configuration file.
4679 7. Added "dereference=" facility to LDAP.
4681 8. Two instances of the typo "uknown" in the source files are fixed.
4683 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4684 the Configure-Makefile script screwed up while processing it.
4686 10. Incorporated PCRE 4.4.
4688 11. The SMTP synchronization check was not operating right at the start of an
4689 SMTP session. For example, it could not catch a HELO sent before the client
4690 waited for the greeting. There is now a check for outstanding input at the
4691 point when the greeting is written. Because of the duplex, asynchronous
4692 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4693 way, but not yet received, when the check is performed.
4695 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4696 on TCP/IP sockets, because this apparently causes some broken clients to
4699 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4700 unchanged) from the Cygwin maintainer.
4702 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4703 was supported instead of testing for mbx. Effectively a typo.
4705 15. The spa authenticator server code was not checking that the input it
4706 received was valid base64.
4708 16. The debug output line for the "set" modifier in ACLs was not showing the
4709 name of the variable that was being set.
4711 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4713 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4714 lookup on its own. Something else was needed to trigger the lookup. For
4715 example, a match in host_lookup or the need for a host name in a host list.
4716 Now, if $sender_host_name is referenced and the host name has not yet been
4717 looked up, a lookup is performed. If the lookup fails, the variable remains
4718 empty, and $host_lookup_failed is set to "1".
4720 19. Added "eqi" as a case-independent comparison operator.
4722 20. The saslauthd authentication condition could segfault if neither service
4723 nor realm was specified.
4725 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4726 error message that was logged was misleading, and incoming SMTP
4727 connections were dropped. The message is now more accurate, and temporary
4728 errors are given to SMTP connections.
4730 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4731 (except RSET) with a 421 error, until QUIT is received. However, it was
4732 failing to send a response to QUIT.
4734 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4735 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4736 ACL is now run after the helo_try_verify_hosts code.
4738 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4739 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4740 case-independent, so other case variants are also recognized). Apparently
4741 some systems use these upper case variants.
4743 25. If more than two messages were waiting for the same host, and a transport
4744 filter was specified for the transport, Exim sent two messages over the
4745 same TCP/IP connection, and then failed with "socket operation on non-
4746 socket" when it tried to send the third.
4748 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4750 27. The extern definition of crypt16() in expand.c was not being excluded when
4751 the OS had its own crypt16() function.
4753 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4754 as a preferred synonym for return_size_limit, both as an option and as an
4757 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4759 30. Changed the default configuration ACL to relax the local part checking rule
4760 for addresses that are not in any local domains. For these addresses,
4761 slashes and pipe symbols are allowed within local parts, but the sequence
4762 /../ is explicitly forbidden.
4764 31. SPA server authentication was not clearing the challenge buffer before
4767 32. log_message in a "warn" ACL statement was writing to the reject log as
4768 well as to the main log, which contradicts the documentation and doesn't
4769 seem right (because no rejection is happening). So I have stopped it.
4771 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4772 However, I am unable to do any testing of this.
4774 34. Fixed an infelicity in the appendfile transport. When checking directories
4775 for a mailbox, to see if any needed to be created, it was accidentally
4776 using path names with one or more superfluous leading slashes; tracing
4777 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4779 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4780 contents are added to the log line that is written for every discarded
4781 recipient. (Previously a log_message setting was ignored.)
4783 36. The ${quote: operator now quotes the string if it is empty.
4785 37. The install script runs exim in order to find its version number. If for
4786 some reason other than non-existence or emptiness, which it checks, it
4787 could not run './exim', it was installing it with an empty version number,
4788 i.e. as "exim-". This error state is now caught, and the installation is
4791 38. An argument was missing from the function that creates an error message
4792 when Exim fails to connect to the socket for saslauthd authentication.
4793 This could cause Exim to crash, or give a corrupted message.
4795 39. Added isip, isip4, and isip6 to ${if conditions.
4797 40. The ACL variables $acl_xx are now saved with the message, and can be
4798 accessed later in routers, transports, and filters.
4800 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4801 strings in the file are not string-expanded.
4803 42. If a MAIL command specified a SIZE value that was too large to fit into an
4804 int variable, the check against message_size_limit failed. Such values are
4805 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4806 day this will have to be increased, but I don't think I want to be around
4807 when emails are that large.
4814 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4815 iconv() is not standard in FreeBSD.
4817 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4818 IPv6 enabled. The observed symptom was a segmentation fault on return from
4819 the function os_common_find_running_interfaces() in src/os.c.
4821 3. In the check_special_case() function in daemon.c I had used "errno" as an
4822 argument name, which causes warnings on some systems. This was basically a
4823 typo, since it was named "eno" in the comments!
4825 4. The code that waits for the clock to tick (at a resolution of some fraction
4826 of a second) so as to ensure message-id uniqueness was always waiting for
4827 at least one whole tick, when it could have waited for less. [This is
4828 almost certainly not relevant at current processor speeds, where it is
4829 unlikely to ever wait at all. But we try to future-proof.]
4831 5. The function that sleeps for a time interval that includes fractions of a
4832 second contained a race. It did not block SIGALRM between setting the
4833 timer, and suspending (a couple of lines later). If the interval was short
4834 and the sigsuspend() was delayed until after it had expired, the suspension
4835 never ended. On busy systems this could lead to processes getting stuck for
4838 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4839 process, before it forks any delivery processes. The open lookup caching
4840 mechanism meant that the open file or database connection was passed into
4841 the delivery process. The problem was that delivery processes always tidy
4842 up cached lookup data. This could cause a problem for the next delivery
4843 process started by the queue runner, because the external queue runner
4844 process does not know about the closure. So the next delivery process
4845 still has data in the lookup cache. In the case of a file lookup, there was
4846 no problem because closing a file descriptor in a subprocess doesn't affect
4847 the parent. However, if the lookup was caching a connection to a database,
4848 the connection was closed, and the second delivery process was likely to
4849 see errors such as "PGSQL: query failed: server closed the connection
4850 unexpectedly". The problem has been fixed by closing all cached lookups
4851 in a queue runner before running a delivery process.
4853 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4854 seem to have the "const" qualifier which it has on other OS. I've
4857 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4858 *of the same type* (client or server) with the same public name that an
4859 error should be diagnosed.
4861 9. When Exim looked up a host name for an IP address, but failed to find the
4862 original IP address when looking up the host name (a safety check), it
4863 output the message "<ip address> does not match any IP for NULL", which was
4864 confusing, to say the least. The bug was that the host name should have
4865 appeared instead of "NULL".
4867 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4868 user that is built into the binary, and the -C or -D options is used, root
4869 privilege is dropped before the configuration file is read. In addition,
4870 logging is switched to stderr instead of the normal log files. If the
4871 configuration then re-defines the Exim user, the unprivileged environment
4872 is probably not what is expected, so Exim logs a panic warning message (but
4875 However, if deliver_drop_privilege is set, the unprivileged state may well
4876 be exactly what is intended, so the warning has been cut out in that case,
4877 and Exim is allowed to try to write to its normal log files.
4883 1. smtp_return_error_details was not giving details for temporary sender
4884 or receiver verification errors.
4886 2. Diagnose a configuration error if two authenticators have the same public
4889 3. Exim used not to create the message log file for a message until the first
4890 delivery attempt. This could be confusing when incoming messages were held
4891 for policy or load reasons. The message log file is now created at the time
4892 the message is received, and an initial "Received" line is written to it.
4894 4. The automatically generated man page for command line options had a minor
4895 bug that caused no ill effects; however, a more serious problem was that
4896 the procedure for building the man page automatically didn't always
4897 operate. Consequently, release 4.20 contains an out-of-date version. This
4898 shouldn't happen again.
4900 5. When building Exim with embedded Perl support, the script that builds the
4901 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4902 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4904 6. The freeze_tell option was not being used for messages that were frozen on
4905 arrival, either by an ACL or by local_scan().
4907 7. Added the smtp_incomplete_transaction log selector.
4909 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4910 was accepting AUTH without a new EHLO.
4912 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4913 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4916 10. Logging of TCP/IP connections (when configured) now happens in the main
4917 daemon process instead of the child process, so that the TCP/IP connection
4918 count is more accurate (but it can never be perfect).
4920 11. The use of "drop" in a nested ACL was not being handled correctly in the
4921 outer ACL. Now, if condition failure induced by the nested "drop" causes
4922 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4923 or "require"), the connection is dropped.
4925 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4926 that yield "discard" can now be used with an "accept" or a "discard" verb,
4927 but an error is generated for any others (because I can't see a useful way
4928 to define what should happen).
4930 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4931 are now processed in the same way as lines in the Exim configuration file.
4932 In particular, continuation lines are supported.
4934 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4936 15. Added -ti meaning -t -i.
4938 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4939 domains, and warn by logging if others are found.
4941 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4942 returned from GIFCONF when Exim is trying to find the list of interfaces on
4943 a host. The code in os.c has been modified to copy each ifreq to an aligned
4944 structure in all cases.
4946 Also, in some cases, the returned ifreq's were being copied to a 'struct
4947 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4948 means the last couple of bytes of an IPv6 address could be chopped if the
4949 ifreq contained only a normal sockaddr (14 bytes storage).
4951 18. Named domain lists were not supported in the hosts_treat_as_local option.
4952 An entry such as +xxxx was not recognized, and was treated as a literal
4955 19. Ensure that header lines added by a DATA ACL are included in the reject log
4956 if the ACL subsequently rejects the message.
4958 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4959 MD5 (which is deprecated).
4961 21. When testing a filter file using -bf, Exim was writing a message when it
4962 took the sender from a "From " line in the message, but it was not doing so
4963 when it took $return_path from a Return-Path: header line. It now does.
4965 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4966 with a valid header line field name (a series of printing characters
4967 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4969 23. Changed "disc" in the source to "disk" to conform to the documentation and
4970 the book and for uniformity.
4972 24. Ignore Sendmail's -Ooption=value command line item.
4974 25. When execve() failed while trying to run a command in a pipe transport,
4975 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4976 could be confused with a return value of 69 from the command itself. This
4977 has been changed to 127, the value the shell returns if it is asked to run
4978 a non-existent command. The wording for the related log line suggests a
4979 non-existent command as the problem.
4981 26. If received_header_text expands to an empty string, do not add a Received:
4982 header line to the message. (Well, it adds a token one on the spool, but
4983 marks it "old" so that it doesn't get used or transmitted.)
4985 27. Installed eximstats 1.28 (addition of -nt option).
4987 28. There was no check for failure on the call to getsockname() in the daemon
4988 code. This can fail if there is a shortage of resources on the system, with
4989 ENOMEM, for example. A temporary error is now given on failure.
4991 29. Contrary to the C standard, it seems that in some environments, the
4992 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4993 program. Exim now does this explicitly; it affects the formatting of
4994 timestamps using strftime().
4996 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4997 complaints. I've now initialized all the variables, to avoid this.
4999 32. Header lines added by a system filter were not being "seen" during
5000 transport-time rewrites.
5002 33. The info_callback() function passed to OpenSSL is set up with type void
5003 (*)(SSL *, int, int), as described somewhere. However, when calling the
5004 function (actually a macro) that sets it up, the type void(*)() is
5005 expected. I've put in a cast to prevent warnings from picky compilers.
5007 34. If a DNS black list lookup found a CNAME record, but there were no A
5008 records associated with the domain it pointed at, Exim crashed.
5010 35. If a DNS black list lookup returned more than one A record, Exim ignored
5011 all but the first. It now scans all returned addresses if a particular IP
5012 value is being sought. In this situation, the contents of the
5013 $dnslist_value variable are a list of all the addresses, separated by a
5016 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5017 to accept a host name and all its aliases if the forward lookup for any of
5018 them yielded the IP address of the incoming connection. Now it accepts only
5019 those names whose forward lookup yields the correct IP address. Any other
5020 names are discarded. This closes a loophole whereby a rogue DNS
5021 administrator could create reverse DNS records to break through a
5022 wildcarded host restriction in an ACL.
5024 37. If a user filter or a system filter that ran in a subprocess used any of
5025 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5026 the wrong values were passed to the pipe command ($thisaddress had the
5027 value of $0, $0 had the value of $1, etc). This bug was introduced by
5028 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5030 38. Improved the line breaking for long SMTP error messages from ACLs.
5031 Previously, if there was no break point between 40 and 75 characters, Exim
5032 left the rest of the message alone. Two changes have been made: (a) I've
5033 reduced the minimum length to 35 characters; (b) if it can't find a break
5034 point between 35 and 75 characters, it looks ahead and uses the first one
5035 that it finds. This may give the occasional overlong line, but at least the
5036 remaining text gets split now.
5038 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5039 file descriptors might be low, and that setting 1000 would always raise it.
5040 It turns out that in some environments, the limit is already over 1000 and
5041 that lowering it causes trouble. So now Exim takes care not to decrease it.
5043 40. When delivering a message, the value of $return_path is set to $sender_
5044 address at the start of routing (routers may change the value). By an
5045 oversight, this default was not being set up when an address was tested by
5046 -bt or -bv, which affected the outcome if any router or filter referred to
5049 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5050 log only when "message" or "log_message" are set. However, if one of the
5051 conditions was an address verification, or a call to a nested ACL, the
5052 messages generated by the underlying test were being passed through. This
5053 no longer happens. The underlying message is available in $acl_verify_
5054 message for both "message" and "log_message" expansions, so it can be
5055 passed through if needed.
5057 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5058 new expansion $bh_ to give the encoded byte string without charset
5059 translation. Translation happens only if iconv() is available; HAVE_ICONV
5060 indicates this at build time. HEADERS_CHARSET gives the charset to
5061 translate to; headers_charset can change it in the configuration, and
5062 "headers charset" can change it in an individual filter file.
5064 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5065 that creates RFC 2047 encoded "words" labels them as that charset instead
5066 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5067 expansion operator; (ii) when Exim creates a From: line for a local
5068 message; (iii) when a header line is rewritten to include a "phrase" part.
5070 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5071 buggy, causing it to skip the first lines of messages whose message ID
5072 ended in 'D'. This would not have bitten before Exim release 4.14, because
5073 message IDs were unlikely to end in 'D' before then. The effect was to have
5074 incorrect size information for certain domains.
5076 45. #include "config.h" was missing at the start of the crypt16.c module. This
5077 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5080 46. If there was a timeout during a "random" callout check, Exim treated it as
5081 a failure of the random address, and carried on sending RSET and the real
5082 address. If the delay was just some slowness somewhere, the response to the
5083 original RCPT would be taken as a response to RSET and so on, causing
5084 mayhem of various kinds.
5086 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5087 when I implemented it. It didn't allow for the fact that some option values
5088 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5089 the right test for positive values.
5091 48. Domain names in DNS records are case-independent. Exim always looks them up
5092 in lower case. Some resolvers return domain names in exactly the case they
5093 appear in the zone file, that is, they may contain uppercase letters. Not
5094 all resolvers do this - some return always lower case. Exim was treating a
5095 change of case by a resolver as a change of domain, similar to a widening
5096 of a domain abbreviation. This triggered its re-routing code and so it was
5097 trying to route what was effectively the same domain again. This normally
5098 caused routing to fail (because the router wouldn't handle the domain
5099 twice). Now Exim checks for this case specially, and just changes the
5100 casing of the domain that it ultimately uses when it transmits the message
5103 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5106 50. If a filter generated a file delivery with a non-absolute name (possible if
5107 no home directory exists for the router), the forbid_file option was not
5110 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5111 the existing equality matching.
5113 52. Exim was using ints instead of ino_t variables in some places where it was
5114 dealing with inode numbers.
5116 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5117 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5118 called TMPDIR, and if it finds it is different, it changes its value.
5120 54. The smtp_printf() function is now made available to local_scan() so
5121 additional output lines can be written before returning. There is also an
5122 smtp_fflush() function to enable the detection of a dropped connection.
5123 The variables smtp_input and smtp_batched_input are exported to
5126 55. Changed the default runtime configuration: the message "Unknown user"
5127 has been removed from the ACL, and instead placed on the localuser router,
5128 using the cannot_route_message feature. This means that any verification
5129 failures that generate their own messages won't get overridden. Similarly,
5130 the "Unrouteable address" message that was in the ACL for unverifiable
5131 relay addresses has also been removed.
5133 56. Added hosts_avoid_esmtp to the smtp transport.
5135 57. The exicyclog script was not checking for the esoteric option
5136 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5137 will work only if exicyclog is run under the appropriate euid.
5139 58. Following a discussion on the list, the rules by which Exim recognises line
5140 endings on incoming messages have been changed. The -dropcr and drop_cr
5141 options are now no-ops, retained only for backwards compatibility. The
5142 following line terminators are recognized: LF CRLF CR. However, special
5143 processing applies to CR:
5145 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5146 nor a local message in the state where . is a terminator.
5148 (ii) If a bare CR is encountered in a header line, an extra space is added
5149 after the line terminator so as not to end the header. The reasoning
5150 behind this is that bare CRs in header lines are most likely either
5151 to be mistakes, or people trying to play silly games.
5153 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5154 was being incorrectly given as 18 bytes larger than it should have been.
5155 This is a VOB (very old bug).
5157 60. This may never have affected anything current, but just in case it has:
5158 When the local host is found other than at the start of a list of hosts,
5159 the local host, those with the same MX, and any that follow, are discarded.
5160 When the list in question was part of a longer list of hosts, the following
5161 hosts (not currently being processed) were also being discarded. This no
5162 longer happens. I'm not sure if this situation could ever has previously
5165 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5168 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5169 header. This is something that is recommended in a new Internet Draft, and
5170 is something that is documented as being done by Sendmail. There are two
5171 possible values. For messages generated by the autoreply transport, Exim
5174 Auto-Submitted: auto-replied
5176 whereas for all other generated messages (e.g. bounces) it adds
5178 Auto-Submitted: auto-generated
5180 63. The "personal" condition in filters now includes a test for the
5181 Auto-Submitted: header. If it contains the string "auto-" the message it
5182 not considered personal.
5184 64. Added rcpt_include_affixes as a generic transport option.
5186 65. Added queue_only_override (default true).
5188 66. Added the syslog_duplication option.
5190 67. If what should have been the first header line of a message consisted of
5191 a space followed by a colon, Exim was mis-interpreting it as a header line.
5192 It isn't of course - it is syntactically invalid and should therefore be
5193 treated as the start of the message body. The misbehaviour could have
5194 caused a number of strange effects, including loss of data in subsequent
5195 header lines, and spool format errors.
5197 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5198 client host had authenticated. This control can now be exercised by an ACL
5199 for more flexibility.
5201 69. By default, callouts do not happen when testing with -bh. There is now a
5202 variant, -bhc, which does actually run the callout code, including
5203 consulting and updating the callout cache.
5205 70. Added support for saslauthd authentication, courtesy of Alexander
5208 71. If statvfs() failed on the spool or log directories while checking their
5209 size for availability, Exim confusingly gave the error "space shortage".
5210 Furthermore, in debugging mode it crashed with a floating point exception.
5211 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5212 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5213 serious problem, Exim now writes to the main and panic logs when this
5214 happens, with details of the failure. It then refuses to accept the
5215 incoming message, giving the message "spool directory problem" or "log
5216 directory problem" with a 421 code for SMTP messages.
5218 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5219 0, 1, and 2 exist, because some OS complain for execs without them (see
5220 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5221 descriptors. However, the code omitted to check that the open succeeded,
5222 causing mysterious errors if for some reason the permissions on /dev/null
5223 got screwed. Now Exim writes a message to the main and panic logs, and
5224 bombs out if it can't open /dev/null.
5226 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5227 interact so that it is all more flexible. It is supposed to remain
5228 backwards compatible. Also added extra_local_interfaces.
5230 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5231 to bomb out with an assertion failure - to the client this appears as a
5232 connection drop. This problem occurs in the part of the code that was taken
5233 from the Samba project. Fortunately, the assertion is in a very simple
5234 function, so I have fixed this by reproducing the function inline in the
5235 one place where it is called, and arranging for authentication to fail
5236 instead of killing the process with assert().
5238 75. The SPA client code was not working when the server requested OEM rather
5239 than Unicode encoding.
5241 76. Added code to make require_files with a specific uid setting more usable in
5242 the case where statting the file as root fails - usually a non-root-mounted
5243 NFS file system. When this happens and the failure is EACCES, Exim now
5244 forks a subprocess and does the per-uid checking as the relevant uid.
5246 77. Added process_log_path.
5248 78. If log_file_path was not explicitly set, a setting of check_log_space or
5249 check_log_inodes was ignored.
5251 79. If a space check for the spool or log partitions fails, the incident is now
5252 logged. Of course, in the latter case the data may get lost...
5254 80. Added the %p formatting code to string_format() so that it can be used to
5255 print addresses in debug_print(). Adjusted all the address printing in the
5256 debugging in store.c to use %p rather than %d.
5258 81. There was a concern that a line of code in smtp_in.c could overflow a
5259 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5260 initially expressed, the concern was not well-founded, because trailing
5261 spaces are removed early. However, if the trailing spaces were followed by
5262 a NULL, they did not get removed, so the overflow was possible. Two fixes
5265 (a) I re-wrote the offending code in a cleaner fashion.
5266 (b) If an incoming SMTP command contains a NULL character, it is rejected
5269 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5270 runs initgroups(), so that if the Exim user is in any additional groups,
5271 they will be used during message reception.
5277 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob