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.
30 TF/01 Failure to get a lock on a hints database can have serious
31 consequences so log it to the panic log.
33 TF/02 Log LMTP confirmation messages in the same way as SMTP,
34 controlled using the smtp_confirmation log selector.
36 TF/03 Include the error message when we fail to unlink a spool file.
38 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
39 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
40 for maintaining out-of-tree patches for some time.
42 PP/01 Bugzilla 139: Documentation and portability issues.
43 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
44 Handle per-OS dynamic-module compilation flags.
46 PP/02 Let /dev/null have normal permissions.
47 The 4.73 fixes were a little too stringent and complained about the
48 permissions on /dev/null. Exempt it from some checks.
49 Reported by Andreas M. Kirchwitz.
51 PP/03 Report version information for many libraries, including
52 Exim version information for dynamically loaded libraries. Created
53 version.h, now support a version extension string for distributors
54 who patch heavily. Dynamic module ABI change.
56 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
57 privilege escalation vulnerability whereby the Exim run-time user
58 can cause root to append content of the attacker's choosing to
61 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
64 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
65 If dropping privileges for untrusted macros, we disabled normal logging
66 on the basis that it would fail; for the Exim run-time user, this is not
67 the case, and it resulted in successful deliveries going unlogged.
68 Fixed. Reported by Andreas Metzler.
74 PP/01 Date: & Message-Id: revert to normally being appended to a message,
75 only prepend for the Resent-* case. Fixes regression introduced in
76 Exim 4.70 by NM/22 for Bugzilla 607.
78 PP/02 Include check_rfc2047_length in configure.default because we're seeing
79 increasing numbers of administrators be bitten by this.
81 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
83 PP/03 Bugzilla 994: added openssl_options main configuration option.
85 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
87 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
89 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
91 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
92 without a peer certificate, leading to a segfault because of an
93 assumption that peers always have certificates. Be a little more
94 paranoid. Problem reported by Martin Tscholak.
96 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
97 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
98 NB: ClamAV planning to remove STREAM in "middle of 2010".
99 CL also introduces -bmalware, various -d+acl logging additions and
100 more caution in buffer sizes.
102 PP/09 Implemented reverse_ip expansion operator.
104 PP/10 Bugzilla 937: provide a "debug" ACL control.
106 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
108 PP/12 Bugzilla 973: Implement --version.
110 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
112 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
114 PP/15 Bugzilla 816: support multiple condition rules on Routers.
116 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
117 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
118 ignore trailing whitespace.
120 JJ/02 prevent non-panic DKIM error from being sent to paniclog
122 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
125 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
126 Notification from Dr Andrew Aitchison.
128 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
129 ExtendedDetectionInfo response format.
130 Notification from John Horne.
132 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
135 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
136 XSL and documented dependency on system catalogs, with examples of how
139 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
142 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
143 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
144 configuration file which is writeable by the Exim user or group.
146 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
147 of configuration files to cover files specified with the -C option if
148 they are going to be used with root privileges, not just the default
151 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
152 option (effectively making it always true).
154 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
155 files to be used while preserving root privileges.
157 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
158 that rogue child processes cannot use them.
160 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
161 run-time user, instead of root.
163 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
164 Exim run-time user without dropping privileges.
166 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
167 result string, instead of calling string_vformat() twice with the same
170 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
171 for other users. Others should always drop root privileges if they use
172 -C on the command line, even for a whitelisted configure file.
174 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
176 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
182 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
183 $data_path, and $header_path variables; fixed documentation bugs and
186 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
187 exipick to access non-standard spools, including the "frozen" queue
190 NM/01 Bugzilla 965: Support mysql stored procedures.
191 Patch from Alain Williams
193 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
195 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
196 Patch from Andreas Metzler
198 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
199 Patch from Kirill Miazine
201 NM/05 Bugzilla 671: Added umask to procmail example.
203 JJ/03 installed exipick 20100323.0, fixing doc bug
205 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
206 directory. Notification and patch from Dan Rosenberg.
208 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
210 TK/02 Improve log output when DKIM signing operation fails.
212 MH/01 Treat the transport option dkim_domain as a colon separated
213 list, not as a single string, and sign the message with each element,
214 omitting multiple occurences of the same signer.
216 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
217 Bugzilla 985, 986. Patch by Simon Arlott
219 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
220 Patch by Simon Arlott
222 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
223 MBX locking. Notification from Dan Rosenberg.
229 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
231 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
233 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
235 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
237 NM/04 Bugzilla 915: Flag broken perl installation during build.
243 TK/01 Added patch by Johannes Berg that expands the main option
244 "spamd_address" if it starts with a dollar sign.
246 TK/02 Write list of recipients to X-Envelope-Sender header when building
247 the mbox-format spool file for content scanning (suggested by Jakob
250 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
251 (http://www.dcc-servers.net/) support via dccifd. Activated by
252 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
254 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
255 by Mark Daniel Reidel <mr@df.eu>.
257 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
258 When building exim an external PCRE library is now needed -
259 PCRE is a system library on the majority of modern systems.
260 See entry on PCRE_LIBS in EDITME file.
262 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
263 conversation. Added nologin parameter to request.
264 Patch contributed by Kirill Miazine.
266 TF/01 Do not log submission mode rewrites if they do not change the address.
268 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
270 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
271 log files in place. Contributed by Roberto Lima.
273 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
275 TF/03 Bugzilla 615: When checking the local_parts router precondition
276 after a local_part_suffix or local_part_prefix option, Exim now
277 does not use the address's named list lookup cache, since this
278 contains cached lookups for the whole local part.
280 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
281 Robert Millan. Documentation is in experimental-spec.txt.
283 TF/04 Bugzilla 668: Fix parallel build (make -j).
285 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
287 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
288 Patch provided by Jan Srzednicki.
290 TF/05 Leading white space used to be stripped from $spam_report which
291 wrecked the formatting. Now it is preserved.
293 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
294 that they are available at delivery time.
296 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
298 TF/08 TLS error reporting now respects the incoming_interface and
299 incoming_port log selectors.
301 TF/09 Produce a more useful error message if an SMTP transport's hosts
302 setting expands to an empty string.
304 NM/06 Bugzilla 744: EXPN did not work under TLS.
305 Patch provided by Phil Pennock.
307 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
308 Patch provided by Richard Godbee.
310 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
311 acl_smtp_notquit, added index entry.
313 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
314 Patch provided by Eugene Bujak.
316 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
317 accept(). Patch provided by Maxim Dounin.
319 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
320 Patch provided by Phil Pennock.
322 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
324 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
325 Patch provided by Brad "anomie" Jorsch.
327 NM/14 Bugzilla 622: Added timeout setting to transport filter.
328 Patch provided by Dean Brooks.
330 TK/05 Add native DKIM support (does not depend on external libraries).
332 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
333 Patch provided by Graeme Fowler.
335 NM/16 Bugzilla 851: Documentation example syntax fix.
337 NM/17 Changed NOTICE file to remove references to embedded PCRE.
339 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
342 NM/19 Bugzilla 745: TLS version reporting.
343 Patch provided by Phil Pennock.
345 NM/20 Bugzilla 167: bool: condition support.
346 Patch provided by Phil Pennock.
348 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
349 clients. Patch provided by Phil Pennock.
351 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
352 Patch provided by Brad "anomie" Jorsch.
354 NM/23 Bugzilla 687: Fix misparses in eximstats.
355 Patch provided by Heiko Schlittermann.
357 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
358 Patch provided by Heiko Schlittermann.
360 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
361 plus update to original patch.
363 NM/26 Bugzilla 799: Documentation correction for ratelimit.
365 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
366 Patch provided by David Brownlee.
368 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
370 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
372 NM/30 Bugzilla 888: TLS documentation bugfixes.
374 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
376 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
377 Unlike the original bugzilla I have changed all shell scripts in src tree.
379 NM/33 Bugzilla 898: Transport filter timeout fix.
380 Patch by Todd Rinaldo.
382 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
383 Patch by Serge Demonchaux.
385 NM/35 Bugzilla 39: Base64 decode bug fixes.
386 Patch by Jakob Hirsch.
388 NM/36 Bugzilla 909: Correct connect() call in dcc code.
390 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
392 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
394 NM/39 Bugzilla 911: Fixed MakeLinks build script.
400 TK/01 Add preliminary DKIM support. Currently requires a forked version of
401 ALT-N's libdkim that I have put here:
402 http://duncanthrax.net/exim-experimental/
404 Note to Michael Haardt: I had to rename some vars in sieve.c. They
405 were called 'true' and it seems that C99 defines that as a reserved
406 keyword to be used with 'bool' variable types. That means you could
407 not include C99-style headers which use bools without triggering
408 build errors in sieve.c.
410 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
411 as mailq or other aliases. Changed the --help handling significantly
412 to do whats expected. exim_usage() emits usage/help information.
414 SC/01 Added the -bylocaldomain option to eximstats.
416 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
418 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
420 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
426 PH/01 Another patch from the Sieve maintainer.
428 PH/02 When an IPv6 address is converted to a string for single-key lookup
429 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
430 dots are used instead of colons so that keys in lsearch files need not
431 contain colons. This was done some time before quoting was made available
432 in lsearch files. However, iplsearch files do require colons in IPv6 keys
433 (notated using the quote facility) so as to distinguish them from IPv4
434 keys. This meant that lookups for IP addresses in host lists did not work
435 for iplsearch lookups.
437 This has been fixed by arranging for IPv6 addresses to be expressed with
438 colons if the lookup type is iplsearch. This is not incompatible, because
439 previously such lookups could never work.
441 The situation is now rather anomolous, since one *can* have colons in
442 ordinary lsearch keys. However, making the change in all cases is
443 incompatible and would probably break a number of configurations.
445 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
448 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
449 conversion specification without a maximum field width, thereby enabling
450 a rogue spamd server to cause a buffer overflow. While nobody in their
451 right mind would setup Exim to query an untrusted spamd server, an
452 attacker that gains access to a server running spamd could potentially
453 exploit this vulnerability to run arbitrary code as the Exim user.
455 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
456 $primary_hostname instead of what libspf2 thinks the hosts name is.
458 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
459 a directory entry by the name of the lookup key. Previously, if a
460 symlink pointed to a non-existing file or a file in a directory that
461 Exim lacked permissions to read, a lookup for a key matching that
462 symlink would fail. Now it is enough that a matching directory entry
463 exists, symlink or not. (Bugzilla 503.)
465 PH/03 The body_linecount and body_zerocount variables are now exported in the
468 PH/04 Added the $dnslist_matched variable.
470 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
471 This means they are set thereafter only if the connection becomes
474 PH/06 Added the client_condition to authenticators so that some can be skipped
475 by clients under certain conditions.
477 PH/07 The error message for a badly-placed control=no_multiline_responses left
478 "_responses" off the end of the name.
480 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
482 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
483 (without spaces) instead of just copying the configuration text.
485 PH/10 Added the /noupdate option to the ratelimit ACL condition.
487 PH/11 Added $max_received_linelength.
489 PH/12 Added +ignore_defer and +include_defer to host lists.
491 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
492 way in which PCRE > 7.0 is built.
494 PH/14 Implemented queue_only_load_latch.
496 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
497 MAIL command. The effect was to mangle the value on 64-bit systems.
499 PH/16 Another patch from the Sieve maintainer.
501 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
503 PH/18 If a system quota error occurred while trying to create the file for
504 a maildir delivery, the message "Mailbox is full" was not appended to the
505 bounce if the delivery eventually timed out. Change 4.67/27 below applied
506 only to a quota excession during the actual writing of the file.
508 PH/19 It seems that peer DN values may contain newlines (and other non-printing
509 characters?) which causes problems in log lines. The DN values are now
510 passed through string_printing() before being added to log lines.
512 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
513 and InterBase are left for another time.)
515 PH/21 Added message_body_newlines option.
517 PH/22 Guard against possible overflow in moan_check_errorcopy().
519 PH/23 POSIX allows open() to be a macro; guard against that.
521 PH/24 If the recipient of an error message contained an @ in the local part
522 (suitably quoted, of course), incorrect values were put in $domain and
523 $local_part during the evaluation of errors_copy.
529 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
530 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
533 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
534 issue a MAIL command.
536 PH/02 In an ACL statement such as
538 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
540 if a client was not listed at all, or was listed with a value other than
541 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
542 the condition was not true (as it should be), so access was not denied.
543 The bug was that the ! inversion was incorrectly passed on to the second
544 item. This has been fixed.
546 PH/03 Added additional dnslists conditions == and =& which are different from
547 = and & when the dns lookup returns more than one IP address.
549 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
550 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
552 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
553 FSYNC, which compiles an option called disable_fsync that allows for
554 bypassing fsync(). The documentation is heavily laced with warnings.
556 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
558 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
559 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
560 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
561 including adding "make clean"; (3) Added -fPIC when compiling the test
562 dynamically loaded module, to get rid of a warning.
564 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
565 message fails, move_frozen_messages = true and ignore_bounce_errors_after
566 = 0s. The bug is otherwise harmless.
568 PH/07 There was a bug in the dovecot authenticator such that the value of
569 $auth1 could be overwritten, and so not correctly preserved, after a
570 successful authentication. This usually meant that the value preserved by
571 the server_setid option was incorrect.
573 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
575 PH/09 Installed PCRE release 7.0.
577 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
578 run for batched SMTP input. It is now run at the start of every message
579 in the batch. While fixing this I discovered that the process information
580 (output by running exiwhat) was not always getting set for -bs and -bS
581 input. This is fixed, and it now also says "batched" for BSMTP.
583 PH/11 Added control=no_pipelining.
585 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
586 patch, slightly modified), and move the expansion of helo_data till after
587 the connection is made in the smtp transport (so it can use these
590 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
592 PH/14 Added log_selector = +pid.
594 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
596 PH/16 Add ${if forany and ${if forall.
598 PH/17 Added dsn_from option to vary the From: line in DSNs.
600 PH/18 Flush SMTP output before performing a callout, unless control =
601 no_callout_flush is set.
603 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
604 was true (the default) a successful delivery failed to delete the retry
605 item, thus causing premature timeout of the address. The bug is now
608 PH/20 Added hosts_avoid_pipelining to the smtp transport.
610 PH/21 Long custom messages for fakedefer and fakereject are now split up
611 into multiline reponses in the same way that messages for "deny" and
612 other ACL rejections are.
614 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
615 with slight modification.
617 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
618 draft, changing the syntax and factoring some duplicate code".
620 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
621 for deliveries of the second and subsequent messages over the same SMTP
624 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
625 ${reduce, with only minor "tidies".
627 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
629 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
630 expansion side effects.
632 PH/27 When a message times out after an over-quota error from an Exim-imposed
633 quota, the bounce message says "mailbox is full". This message was not
634 being given when it was a system quota that was exceeded. It now should
637 MH/03 Made $recipients available in local_scan(). local_scan() already has
638 better access to the recipient list through recipients_list[], but
639 $recipients can be useful in postmaster-provided expansion strings.
641 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
642 in the case of a MAIL command with additional options following the
643 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
644 were accidentally chopped off.
646 PH/29 SMTP synchronization checks are implemented when a command is read -
647 there is a check that no more input is waiting when there shouldn't be
648 any. However, for some commands, a delay in an ACL can mean that it is
649 some time before the response is written. In this time, more input might
650 arrive, invalidly. So now there are extra checks after an ACL has run for
651 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
652 pipelining has not been advertised.
654 PH/30 MH's patch to allow iscntrl() characters to be list separators.
656 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
657 returned in the SMTP response when smtp_return_error_details was false.
660 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
661 instead of the C I/O that was originally supplied, because problems were
664 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
665 Exim which did not show up earlier: it was assuming that a call to
666 SSL_CTX_set_info_callback() might give an error value. In fact, there is
667 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
668 was a macro that became an assignment, so it seemed to work. This has
669 changed to a proper function call with a void return, hence the compile
670 error. Exim's code has been fixed.
672 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
675 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
677 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
679 PH/37 If a message is not accepted after it has had an id assigned (e.g.
680 because it turns out to be too big or there is a timeout) there is no
681 "Completed" line in the log. When some messages of this type were
682 selected by exigrep, they were listed as "not completed". Others were
683 picked up by some special patterns. I have improved the selection
684 criteria to be more general.
686 PH/38 The host_find_failed option in the manualroute router can now be set
687 to "ignore", to completely ignore a host whose IP address cannot be
688 found. If all hosts are ignored, the behaviour is controlled by the new
689 host_all_ignored option.
691 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
692 homing or because of multiple MX records with /mx) generated more than
693 one IP address, and the following item turned out to be the local host,
694 all the secondary addresses of the first item were incorrectly removed
695 from the list, along with the local host and any following hosts (which
696 is what is supposed to happen).
698 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
699 whoever called Exim into the -H file. In the case of the daemon it was
700 behaving confusingly. When first started, it used values for whoever
701 started the daemon, but after a SIGHUP it used the Exim user (because it
702 calls itself on a restart). I have changed the code so that it now always
705 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
706 message are rejected with the same error (e.g. no authentication or bad
707 sender address), and a DATA command is nevertheless sent (as can happen
708 with PIPELINING or a stupid MUA), the error message that was given to the
709 RCPT commands is included in the rejection of the DATA command. This is
710 intended to be helpful for MUAs that show only the final error to their
713 PH/42 Another patch from the Sieve maintainer.
715 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
716 Eximstats - Fixed some broken HTML links and added missing column headers
718 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
719 columns for Rejects, Temp Rejects, Ham, and Spam rows.
721 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
723 PH/43 Yet another patch from the Sieve maintainer.
725 PH/44 I found a way to check for a TCP/IP connection going away before sending
726 the response to the final '.' that terminates a message, but only in the
727 case where the client has not sent further data following the '.'
728 (unfortunately, this is allowed). However, in many cases there won't be
729 any further data because there won't be any more messages to send. A call
730 to select() can be used: if it shows that the input is "ready", there is
731 either input waiting, or the socket has been closed. An attempt to read
732 the next input character can distinguish the two cases. Previously, Exim
733 would have sent an OK response which the client would never have see.
734 This could lead to message repetition. This fix should cure that, at
735 least in a lot of common cases.
737 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
738 advertised in response to EHLO.
744 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
745 fixed by 4.65/MH/01 (is this a record?) are fixed:
747 (i) An empty string was always treated as zero by the numeric comparison
748 operators. This behaviour has been restored.
750 (ii) It is documented that the numeric comparison operators always treat
751 their arguments as decimal numbers. This was broken in that numbers
752 starting with 0 were being interpreted as octal.
754 While fixing these problems I realized that there was another issue that
755 hadn't been noticed. Values of message_size_limit (both the global option
756 and the transport option) were treated as octal if they started with 0.
757 The documentation was vague. These values are now always treated as
758 decimal, and I will make that clear in the documentation.
764 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
765 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
768 MH/01 Don't check that the operands of numeric comparison operators are
769 integers when their expansion is in "skipping" mode (fixes bug
770 introduced by 4.64-PH/07).
772 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
773 child addresses, Exim now panics and dies. Previously, because the count
774 is held in a short int, deliveries were likely to be lost. As such a
775 large number of recipients for a single message is ridiculous
776 (performance will be very, very poor), I have chosen to impose a limit
777 rather than extend the field.
783 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
784 leftover -K file (the existence of which was triggered by #402).
785 While we were at it, introduced process PID as part of the -K
786 filename. This should rule out race conditions when creating
789 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
790 processing considerably. Previous code took too long for large mails,
791 triggering a timeout which in turn triggers #401.
793 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
794 in the DK code in transports.c. sendfile() is not really portable,
795 hence the _LINUX specificness.
797 TF/01 In the add_headers option to the mail command in an Exim filter,
798 there was a bug that Exim would claim a syntax error in any
799 header after the first one which had an odd number of characters
802 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
803 callout verification, Exim cached a "reject" for the entire domain. This
804 is correct for most verifications, but it is not correct for a recipient
805 verification with use_sender or use_postmaster set, because in that case
806 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
807 case of MAIL FROM:<> rejection from other early rejections (e.g.
808 rejection of HELO). When verifying a recipient using a non-null MAIL
809 address, the cache is ignored if it shows MAIL FROM:<> rejection.
810 Whatever the result of the callout, the value of the domain cache is
811 left unchanged (for any other kind of callout, getting as far as trying
812 RCPT means that the domain itself is ok).
814 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
817 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
818 manifest itself as EPIPE rather than ECONNECT. When tidying away a
819 session, the daemon ignores ECONNECT errors and logs others; it now
820 ignores EPIPE as well.
822 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
823 (quoted-printable decoding).
825 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
826 later the small subsequent patch to fix an introduced bug.
828 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
830 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
832 PH/08 An error is now given if message_size_limit is specified negative.
834 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
835 to be given (somewhat) arbitrary names.
837 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
840 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
841 miscellaneous code fixes
843 PH/10 Added the log_reject_target ACL modifier to specify where to log
846 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
847 hostname. This is wrong, because it relates to the incoming message (and
848 probably the interface on which it is arriving) and not to the outgoing
849 callout (which could be using a different interface). This has been
850 changed to use the value of the helo_data option from the smtp transport
851 instead - this is what is used when a message is actually being sent. If
852 there is no remote transport (possible with a router that sets up host
853 addresses), $smtp_active_hostname is used.
855 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
856 tweaks were necessary in order to get it to work (see also 21 below):
857 (a) The code assumed that strncpy() returns a negative number on buffer
858 overflow, which isn't the case. Replaced with Exim's string_format()
860 (b) There were several signed/unsigned issues. I just did the minimum
861 hacking in of casts. There is scope for a larger refactoring.
862 (c) The code used strcasecmp() which is not a standard C function.
863 Replaced with Exim's strcmpic() function.
864 (d) The code set only $1; it now sets $auth1 as well.
865 (e) A simple test gave the error "authentication client didn't specify
866 service in request". It would seem that Dovecot has changed its
867 interface. Fortunately there's a specification; I followed it and
868 changed what the client sends and it appears to be working now.
870 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
873 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
874 address A is aliased to B and C, where B exists and C does not. Without
875 -v the output is "A verified" because verification stops after a
876 successful redirection if more than one address is generated. However,
877 with -v the child addresses are also verified. Exim was outputting "A
878 failed to verify" and then showing the successful verification for C,
879 with its parentage. It now outputs "B failed to verify", showing B's
880 parentage before showing the successful verification of C.
882 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
883 look up a TXT record in a specific list after matching in a combined
886 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
887 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
888 they consult the DNS. I had assumed they would set it the way they
889 wanted; and indeed my experiments on Linux seem to show that in some
890 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
891 To be on the safe side, however, I have now made the interface to
892 host_find_byname() similar to host_find_bydns(), with an argument
893 containing the DNS resolver options. The host_find_byname() function now
894 sets these options at its start, just as host_find_bydns() does. The smtp
895 transport options dns_qualify_single and dns_search_parents are passed to
896 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
897 of host_find_byname() use the default settings of RES_DEFNAMES
898 (qualify_single) but not RES_DNSRCH (search_parents).
900 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
901 spool_read_header() do less string testing, by means of a preliminary
902 switch on the second character of optional "-foo" lines. (This is
903 overdue, caused by the large number of possibilities that now exist.
904 Originally there were few.) While I was there, I also converted the
905 str(n)cmp tests so they don't re-test the leading "-" and the first
906 character, in the hope this might squeeze out yet more improvement.
908 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
909 flag allowing group syntax was set by the header_syntax check but not
910 turned off, possible causing trouble later; (2) The flag was not being
911 set at all for the header_verify test, causing "group"-style headers to
912 be rejected. I have now set it in this case, and also caused header_
913 verify to ignore an empty address taken from a group. While doing this, I
914 came across some other cases where the code for allowing group syntax
915 while scanning a header line wasn't quite right (mostly, not resetting
916 the flag correctly in the right place). These bugs could have caused
917 trouble for malformed header lines. I hope it is now all correct.
919 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
920 with the "reply" argument non-NULL. The code, however (which originally
921 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
922 but it didn't always do it. This confused somebody who was copying the
923 code for some other use. I have removed all the tests.
925 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
926 feature that was used to support insecure browsers during the U.S. crypto
927 embargo. It requires special client support, and Exim is probably the
928 only MTA that supported it -- and would never use it because real RSA is
929 always available. This code has been removed, because it had the bad
930 effect of slowing Exim down by computing (never used) parameters for the
931 RSA_EXPORT functionality.
933 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
934 authenticator to fail if there's a tab character in the incoming data
935 (there should never be unless someone is messing about, as it's supposed
936 to be base64-encoded). Also added, on Timo's advice, the "secured" option
937 if the connection is using TLS or if the remote IP is the same as the
938 local IP, and the "valid-client-cert option" if a client certificate has
941 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
942 authenticators. This can be used for authorization after authentication
943 succeeds. (In the case of plaintext, it servers for both authentication
946 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
947 if any retry times were supplied.
949 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
950 connection, where there is no client IP address to check. In this
951 situation, the verify now always succeeds.
953 PH/25 Applied John Jetmore's -Mset patch.
955 PH/26 Added -bem to be like -Mset, but loading a message from a file.
957 PH/27 In a string expansion for a processed (not raw) header when multiple
958 headers of the same name were present, leading whitespace was being
959 removed from all of them, but trailing whitespace was being removed only
960 from the last one. Now trailing whitespace is removed from each header
961 before concatenation. Completely empty headers in a concatenation (as
964 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
965 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
967 PH/29 [Removed. This was a change that I later backed out, and forgot to
968 correct the ChangeLog entry (that I had efficiently created) before
969 committing the later change.]
971 PH/30 Exim was sometimes attempting to deliver messages that had suffered
972 address errors (4xx response to RCPT) over the same connection as other
973 messages routed to the same hosts. Such deliveries are always "forced",
974 so retry times are not inspected. This resulted in far too many retries
975 for the affected addresses. The effect occurred only when there were more
976 hosts than the hosts_max_try setting in the smtp transport when it had
977 the 4xx errors. Those hosts that it had tried were not added to the list
978 of hosts for which the message was waiting, so if all were tried, there
979 was no problem. Two fixes have been applied:
981 (i) If there are any address or message errors in an SMTP delivery, none
982 of the hosts (tried or untried) are now added to the list of hosts
983 for which the message is waiting, so the message should not be a
984 candidate for sending over the same connection that was used for a
985 successful delivery of some other message. This seems entirely
986 reasonable: after all the message is NOT "waiting for some host".
987 This is so "obvious" that I'm not sure why it wasn't done
988 previously. Hope I haven't missed anything, but it can't do any
989 harm, as the worst effect is to miss an optimization.
991 (ii) If, despite (i), such a delivery is accidentally attempted, the
992 routing retry time is respected, so at least it doesn't keep
993 hammering the server.
995 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
996 in ${readsocket because some servers need this prod.
998 PH/32 Added some extra debug output when updating a wait-xxx database.
1000 PH/33 The hint "could be header name not terminated by colon", which has been
1001 given for certain expansion errors for a long time, was not being given
1002 for the ${if def:h_colon_omitted{... case.
1004 PH/34 The spec says: "With one important exception, whenever a domain list is
1005 being scanned, $domain contains the subject domain." There was at least
1006 one case where this was not true.
1008 PH/35 The error "getsockname() failed: connection reset by peer" was being
1009 written to the panic log as well as the main log, but it isn't really
1010 panic-worthy as it just means the connection died rather early on. I have
1011 removed the panic log writing for the ECONNRESET error when getsockname()
1014 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1015 runs only) independently of the message's sender address. This meant
1016 that, if the 4xx error was in fact related to the sender, a different
1017 message to the same recipient with a different sender could confuse
1018 things. In particualar, this can happen when sending to a greylisting
1019 server, but other circumstances could also provoke similar problems.
1020 I have changed the default so that the retry time for these errors is now
1021 based a combination of the sender and recipient addresses. This change
1022 can be overridden by setting address_retry_include_sender=false in the
1025 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1026 remote server are returned as part of bounce messages. This was not
1027 happening for LMTP over a pipe (the lmtp transport), but now it is the
1028 same for both kinds of LMTP.
1030 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1031 in header lines that were in fact CNAMEs. This is no longer the case.
1033 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1034 and queue runs started by the daemon processed all messages. This has
1035 been fixed so that -R and -S can now usefully be given with -q<time>.
1037 PH/40 Import PCRE release 6.7 (fixes some bugs).
1039 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1041 PH/42 Give an error if -q is specified more than once.
1043 PH/43 Renamed the variables $interface_address and $interface_port as
1044 $received_ip_address and $received_port, to make it clear that these
1045 values apply to message reception, and not to the outgoing interface when
1046 a message is delivered. (The old names remain recognized, of course.)
1048 PH/44 There was no timeout on the connect() call when using a Unix domain
1049 socket in the ${readsocket expansion. There now is.
1051 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1052 be meaningful with "accept".
1054 SC/01 Eximstats V1.43
1055 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1057 SC/02 Eximstats V1.44
1058 Use a glob alias rather than an array ref in the generated
1059 parser. This improves both readability and performance.
1061 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1062 Collect SpamAssassin and rejection statistics.
1063 Don't display local sender or destination tables unless
1064 there is data to show.
1065 Added average volumes into the top table text output.
1067 SC/04 Eximstats V1.46
1068 Collect data on the number of addresses (recipients)
1069 as well as the number of messages.
1071 SC/05 Eximstats V1.47
1072 Added 'Message too big' to the list of mail rejection
1073 reasons (thanks to Marco Gaiarin).
1075 SC/06 Eximstats V1.48
1076 Mainlog lines which have GMT offsets and are too short to
1077 have a flag are now skipped.
1079 SC/07 Eximstats V1.49 (Alain Williams)
1080 Added the -emptyok flag.
1082 SC/08 Eximstats V1.50
1083 Fixes for obtaining the IP address from reject messages.
1085 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1086 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1087 whitesspace changes from 4.64-PH/27
1089 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1092 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1093 are found, allow negative numbers in numeric criteria)
1095 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1097 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1098 to match changes made in 4.64-PH/43
1100 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1102 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1103 "domain" is actually an IP address, and force a failure. This locks out
1104 those revolvers/nameservers that support "A-for-A" lookups, in
1105 contravention of the specifications.
1107 PH/48 When a host name was looked up from an IP address, and the subsequent
1108 forward lookup of the name timed out, the host name was left in
1109 $sender_host_name, contrary to the specification.
1111 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1112 restricted to single-key lookups, Exim was not diagnosing an error if
1113 * or *@ was used with a query-style lookup.
1115 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1117 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1118 long ago, but noone interested enough thought of it. Let's just say that
1119 the "1.1" means that there are some new functions that weren't there at
1120 some point in the past.
1122 PH/51 Error processing for expansion failure of helo_data from an smtp
1123 transport during callout processing was broken.
1125 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1126 tested/used via the -bh/-bhc/-bs options.
1128 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1129 bug, fixed in subsequent PCRE releases).
1131 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1132 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1134 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1140 SC/01 Use a glob alias rather than an array ref in eximstats generated
1141 parser. This improves both readability and performance.
1143 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1144 Don't display local sender or destination tables in eximstats unless
1145 there is data to show.
1146 Added average volumes into the eximstats top table text output.
1148 SC/03 Collect data on the number of addresses (recipients) as well
1149 as the number of messages in eximstats.
1151 TF/01 Correct an error in the documentation for the redirect router. Exim
1152 does (usually) call initgroups() when daemonizing.
1154 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1155 with consistent privilege compared to when running as a daemon.
1157 TF/03 Note in the spec that $authenticated_id is not set for local
1158 submissions from trusted users.
1160 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1161 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1163 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1164 by adding some example configuration directives to the default
1165 configuration file. A little bit of work is required to uncomment the
1166 directives and define how usernames and passwords are checked, but
1167 there is now a framework to start from.
1169 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1170 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1171 without this. I don't know how relevant this is to other LDAP libraries.
1173 PH/02 Add the verb name to the "unknown ACL verb" error.
1175 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1177 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1179 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1180 directory not expanded when it should be if an expanded home directory
1181 was set for the address (which is overridden by the transport).
1183 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1186 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1187 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1188 because it is too late at that time, and has no effect.
1190 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1191 security issue with \' (bugzilla #107). I could not use the
1192 PQescapeStringConn() function, because it needs a PGconn value as one of
1195 PH/08 When testing addresses using -bt, indicate those final addresses that
1196 are duplicates that would not cause an additional delivery. At least one
1197 person was confused, thinking that -bt output corresponded to deliveries.
1198 (Suppressing duplicates isn't a good idea as you lose the information
1199 about possibly different redirections that led to the duplicates.)
1201 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1202 systems where poll() doesn't work, in particular OS X.
1204 PH/10 Added more information to debugging output for retry time not reached.
1206 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1207 operations in malware.c.
1209 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1212 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1213 syslog_duplication set false, log lines that would normally be written
1214 both the the main log and to the reject log were not written to syslog at
1217 PH/14 In the default configuration, change the use of "message" in ACL warn
1218 statements to "add_header".
1220 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1221 not followed by a command (e.g. "seen endif").
1223 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1224 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1227 PH/17 Added extra conditions to the default value of delay_warning_condition
1231 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1232 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1233 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1236 The Auto-Submitted: and various List- headers are standardised, whereas I
1237 don't think Precedence: ever was.
1239 PH/18 Refactored debugging code in route_finduser() to show more information,
1240 in particular, the error code if getpwnam() issues one.
1242 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1243 This is apparently needed in addition to the PH/07 change above to avoid
1244 any possible encoding problems.
1246 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1247 but not after initializing Perl.
1249 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1250 output them only if debugging. By default they are written stderr,
1251 apparently, which is not desirable.
1253 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1256 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1259 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1261 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1262 authenticated or an ident call has been made. Suppress the default
1263 values for $authenticated_id and $authenticated_sender (but permit -oMai
1264 and -oMas) when testing with -bh.
1266 PH/24 Re-jigged the order of the tests in the default configuration so that the
1267 tests for valid domains and recipients precede the DNS black list and CSA
1268 tests, on the grounds that those ones are more expensive.
1270 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1271 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1272 This bug exists in every version of Exim that I still have, right back to
1275 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1276 However, an attempt to turn on case-sensitivity in a regex key by
1277 including (?-i) didn't work because the subject string was already
1278 lowercased, and the effects were non-intuitive. It turns out that a
1279 one-line patch can be used to allow (?-i) to work as expected.
1285 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1286 other effects) broke the use of negated acl sub-conditions.
1288 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1291 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1292 "Deny" causes Exim to reject the incoming connection with a 554 error.
1293 Unfortunately, if there is a major crisis, such as a disk failure,
1294 tcp-wrappers gives "deny", whereas what one would like would be some
1295 kind of temporary error. A kludge has been added to help with this.
1296 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1297 554 error is used if errno is still zero or contains ENOENT (which occurs
1298 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1301 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1303 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1304 errors. Otherwise a message that provokes a temporary error (when other
1305 messages do not) can cause a whole host to time out.
1307 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1308 addresses were routed directly to files or pipes from a redirect router.
1309 File deliveries just didn't batch; pipe deliveries might have suffered
1312 PH/06 A failure to get a lock for a hints database would erroneously always say
1313 "Failed to get write lock", even when it was really a read lock.
1315 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1316 of 0600. This has been changed to use the value of the lockfile_mode
1317 option (which defaults to 0600).
1319 PH/08 Applied small patch from the Sieve maintainer.
1321 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1322 folder from quota calculations, a direct delivery into this folder messed
1323 up the contents of the maildirsize file. This was because the regex was
1324 used only to exclude .Trash (or whatever) when the size of the mailbox
1325 was calculated. There was no check that a delivery was happening into an
1326 excluded directory. This bug has been fixed by ignoring all quota
1327 processing for deliveries into excluded directories.
1329 PH/10 Added the maildirfolder_create_regex option to appendfile.
1335 PH/01 The code for finding all the local interface addresses on a FreeBSD
1336 system running IPv6 was broken. This may well have applied to all BSD
1337 systems, as well as to others that have similar system calls. The broken
1338 code found IPv4 interfaces correctly, but gave incorrect values for the
1339 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1340 that it would not match correctly against @[] and not recognize the IPv6
1343 PH/02 The ipliteral router was not recognizing addresses of the form user@
1344 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1346 PH/03 Added disable_ipv6.
1348 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1349 decoded form, because it is most often used to construct To: headers
1350 lines in autoreplies, and the decoded form may well be syntactically
1351 invalid. However, $reply_address has leading white space removed, and all
1352 newlines turned into spaces so that the autoreply transport does not
1355 PH/05 If group was specified without a user on a router, and no group or user
1356 was specified on a transport, the group from the router was ignored.
1358 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1359 for visible compile-time settings that can be used to change these
1360 numbers, for those that want even more. Backwards compatibility with old
1361 spool files has been maintained. However, going back to a previous Exim
1362 release will lost any variables that are in spool files.
1364 PH/07 Two small changes when running in the test harness: increase delay when
1365 passing a TCP/IP connection to a new process, in case the original
1366 process has to generate a bounce, and remove special handling of
1367 127.0.0.2 (sic), which is no longer necessary.
1369 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1370 be the same on different OS.
1372 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1375 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1376 whether --show-vars was specified or not
1378 JJ/02 exipick: Added support for new ACL variable spool format introduced
1381 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1382 syntactically invalid From: or Reply-to: line, and a filter used this to
1383 generate an autoreply, and therefore failed to obtain an address for the
1384 autoreply, Exim could try to deliver to a non-existent relative file
1385 name, causing unrelated and misleading errors. What now happens is that
1386 it logs this as a hard delivery error, but does not attempt to create a
1389 PH/11 The exinext utility has a -C option for testing purposes, but although
1390 the given file was scanned by exinext itself; it wasn't being passed on
1391 when Exim was called.
1393 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1394 an end-of-file indication when reading a command response.
1396 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1397 compiled. In many other places in Exim, IPv6 addresses are always
1398 recognized, so I have changed this. It also means that IPv4 domain
1399 literals of the form [IPV4:n.n.n.n] are now always recognized.
1401 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1402 used if the router is not running as root, for example, when verifying at
1403 ACL time, or when using -bh. The debugging output from this situation was
1404 non-existent - all you got was a failure to exec. I have made two
1407 (a) Failures to set uid/gid, the current directory, or a process leader
1408 in a subprocess such as that created by queryprogram now generate
1409 suitable debugging ouput when -d is set.
1411 (b) The queryprogram router detects when it is not running as root,
1412 outputs suitable debugging information if -d is set, and then runs
1413 the subprocess without attempting to change uid/gid.
1415 PH/15 Minor change to Makefile for building test_host (undocumented testing
1418 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1419 additional section of a DNS packet that returns MX or SRV records.
1420 Instead, it always explicitly searches for A/AAAA records. This avoids
1421 major problems that occur when a DNS server includes only records of one
1422 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1423 fixed another bug: if SRV records were looked up and the corresponding
1424 address records were *not* found in the additional section, the port
1425 values from the SRV records were lost.
1427 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1428 using the correct key (the original address) when searching the retry
1429 rules in order to find which one to use for generating the retry hint.
1431 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1432 adding the default one. Similarly, if it contains a Reply-To: header, the
1433 errors_reply_to option, if set, is not used.
1435 PH/19 When calculating a retry time, Exim used to measure the "time since
1436 failure" by looking at the "first failed" field in the retry record. Now
1437 it does not use this if it is later than than the arrival time of the
1438 message. Instead it uses the arrival time. This makes for better
1439 behaviour in cases where some deliveries succeed, thus re-setting the
1440 "first failed" field. An example is a quota failure for a huge message
1441 when small messages continue to be delivered. Without this change, the
1442 "time since failure" will always be short, possible causing more frequent
1443 delivery attempts for the huge message than are intended.
1444 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1446 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1447 $1, $2, $3) because the numerical variables can be reset during some
1448 expansion items (e.g. "match"), thereby losing the authentication data.
1450 PH/21 Make -bV show the size of off_t variables so that the test suite can
1451 decide whether to run tests for quotas > 2G.
1453 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1454 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1455 filecount value is greater than 2G or if a quota value is greater than 2G
1456 on a system where the size of off_t is not greater than 4, a panic error
1459 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1460 never match. The debug and -bh output now contains an explicit error
1461 message indicating a malformed IPv4 address or mask.
1463 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1464 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1465 PH/23 above applies.
1467 PH/25 Do not write to syslog when running in the test harness. The only
1468 occasion when this arises is a failure to open the main or panic logs
1469 (for which there is an explicit test).
1471 PH/26 Added the /no_tell option to "control=freeze".
1473 PH/27 If a host name lookup failed very early in a connection, for example, if
1474 the IP address matched host_lookup and the reverse lookup yielded a name
1475 that did not have a forward lookup, an error message of the form "no IP
1476 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1477 could be logged. Now it outputs the IP address instead of "NULL".
1479 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1480 allows the sender and the authenticated sender to be set when
1481 submitting a message from within Exim. Since child_open_exim() is
1482 documented for local_scan(), the new function should be too.
1484 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1485 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1486 results in an empty string is now treated as unset.
1488 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1490 PH/31 Added sender_verify_fail logging option.
1492 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1493 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1494 tidied the source and removed it altogether.
1496 PH/33 When a queue run was abandoned because the load average was too high, a
1497 log line was always written; now it is written only if the queue_run log
1498 selector is set. In addition, the log line for abandonment now contains
1499 information about the queue run such as the pid. This is always present
1500 in "start" and "stop" lines but was omitted from the "abandon" line.
1502 PH/34 Omit spaces between a header name and the colon in the error message that
1503 is given when verify = headers_syntax fails (if there are lots of them,
1504 the message gets confusing).
1506 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1507 names, as there are now some PTR records that contain slashes. This check
1508 is only to protect against broken name servers that fall over on strange
1509 characters, so the fact that it applies to all lookups doesn't matter.
1511 PH/36 Now that the new test suite is complete, we can remove some of the
1512 special code in Exim that was needed for the old test suite. For example,
1513 sorting DNS records because real resolvers return them in an arbitrary
1514 order. The new test suite's fake resolver always returns records in the
1517 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1518 bounces) except when queue_only is set, to avoid logging races between
1519 the different processes.
1521 PH/38 Panic-die if .include specifies a non-absolute path.
1523 PH/39 A tweak to the "H" retry rule from its user.
1525 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1526 a label. They prevented compilation on older perls.
1528 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1529 a warning to be raised on newish perls.
1531 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1532 on queue. Changes to match documented behaviour of showing count of
1533 messages matching specified criteria.
1535 PH/40 Changed the default ident timeout from 30s to 5s.
1537 PH/41 Added support for the use of login_cap features, on those BSD systems
1538 that have them, for controlling the resources used by pipe deliveries.
1540 PH/42 The content-scanning code uses fopen() to create files in which to put
1541 message data. Previously it was not paying any attention to the mode of
1542 the files. Exim runs with umask(0) because the rest of the code creates
1543 files with open(), and sets the required mode explicitly. Thus, these
1544 files were ending up world-writeable. This was not a big issue, because,
1545 being within the spool directory, they were not world-accessible. I have
1546 created a function called modefopen, which takes an additional mode
1547 argument. It sets umask(777), creates the file, chmods it to the required
1548 mode, then resets the umask. All the relevant calls to fopen() in the
1549 content scanning code have been changed to use this function.
1551 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1552 to 24 hours. This avoids potential overflow problems when processing G
1553 and H retry rules. I suspect nobody ever tinkers with this value.
1555 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1557 PH/45 When the plaintext authenticator is running as a client, the server's
1558 challenges are checked to ensure they are valid base64 strings. By
1559 default, the authentication attempt is cancelled if an invalid string is
1560 received. Setting client_ignore_invalid_base64 true ignores these errors.
1561 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1562 they are received. Thus, the responses can be made to depend on the
1563 challenges. If an invalid string is ignored, an empty string is placed in
1566 PH/46 Messages that are created by the autoreply transport now contains a
1567 References: header, in accordance with RFCs 2822 and 3834.
1569 PH/47 Added authenticated_sender_force to the smtp transport.
1571 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1573 PH/49 Installed latest patch from the Sieve maintainer.
1575 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1576 was also set, the appendfile transport was unnecessarily scanning a
1577 directory of message files (e.g. for maildir delivery) to find the count
1578 of files (along with the size), even though it did not need this
1579 information. It now does the scan only if it needs to find either the
1580 size of the count of files.
1582 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1584 PH/52 Two bugs concerned with error handling when the smtp transport is
1587 (i) Exim was not creating retry information for temporary errors given
1588 for individual recipients after the DATA command when the smtp transport
1589 was used in LMTP mode. This meant that they could be retried too
1590 frequently, and not timed out correctly.
1592 (ii) Exim was setting the flag that allows error details to be returned
1593 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1594 recipients that were returned after the DATA command.
1596 PH/53 This is related to PH/52, but is more general: for any failing address,
1597 when detailed error information was permitted to be returned to the
1598 sender, but the error was temporary, then after the final timeout, only
1599 "retry timeout exceeded" was returned. Now it returns the full error as
1600 well as "retry timeout exceeded".
1602 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1603 do this, and (what is worse) MTAs that accept it.
1605 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1606 will now be deprecated.
1608 PH/56 New os.c-cygwin from the Cygwin maintainer.
1610 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1611 formats (previously only available in exim formats via -bpr, -bpru,
1612 and -bpra. Now also available in native and exiqgrep formats)
1614 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1615 with very large, slow to parse queues
1617 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1619 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1621 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1622 responses to authentication challenges, though it was showing the
1623 challenges; (ii) I've removed the CR characters from the debug output for
1626 PH/58 Allow for the insertion of a newline as well as a space when a string
1627 is turned into more than one encoded-word during RFC 2047 encoding. The
1628 Sieve code now uses this.
1630 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1631 data_4xx, lost_connection, tls_required.
1633 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1634 message was being sent as an SMTP response.
1636 PH/61 Add -l and -k options to exicyclog.
1638 PH/62 When verifying, if an address was redirected to one new address, so that
1639 verification continued, and the new address failed or deferred after
1640 having set something in $address_data, the value of $address_data was not
1641 passed back to the ACL. This was different to the case when no
1642 redirection occurred. The value is now passed back in both cases.
1644 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1645 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1646 use login_cap.h, so on its own it isn't the distinguishing feature. The
1647 new name refers directly to the setclassresources() function.
1649 PH/65 Added configuration files for NetBSD3.
1651 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1653 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1654 is preferred over IPv4.
1656 PH/68 The bounce_return_message and bounce_return_body options were not being
1657 honoured for bounces generated during the reception of non-SMTP messages.
1658 In particular, this applied to messages rejected by the ACL. This bug has
1659 been fixed. However, if bounce_return_message is true and bounce_return_
1660 body is false, the headers that are returned for a non-SMTP message
1661 include only those that have been read before the error was detected.
1662 (In the case of an ACL rejection, they have all been read.)
1664 PH/69 The HTML version of the specification is now built in a directory called
1665 spec_html instead of spec.html, because the latter looks like a path with
1666 a MIME-type, and this confuses some software.
1668 PH/70 Catch two compiler warnings in sieve.c.
1670 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1671 function verify_get_ident() calls ip_connect() to connect a socket, but
1672 if the "connect()" function timed out, ip_connect() used to close the
1673 socket. However, verify_get_ident() also closes the socket later, and in
1674 between Exim writes to the log, which may get opened at this point. When
1675 the socket was closed in ip_connect(), the log could get the same file
1676 descriptor number as the socket. This naturally causes chaos. The fix is
1677 not to close the socket in ip_connect(); the socket should be closed by
1678 the function that creates it. There was only one place in the code where
1679 this was missing, in the iplookup router, which I don't think anybody now
1680 uses, but I've fixed it anyway.
1682 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1683 well as to direct DNS lookups. Otherwise the handling of names in host
1684 lists is inconsistent and therefore confusing.
1690 PH/01 Two changes to the default runtime configuration:
1692 (1) Move the checks for relay_from_hosts and authenticated clients from
1693 after to before the (commented out) DNS black list checks.
1695 (2) Add control=submission to the relay_from_hosts and authenticated
1696 clients checks, on the grounds that messages accepted by these
1697 statements are most likely to be submissions.
1699 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1701 (1) Generate an error if the third argument for the ${prvs expansion is
1704 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1707 (3) Reset the variables that are obtained from the first argument of
1708 ${prvscheck and used in the second argument before leaving the code,
1709 because their memory is reclaimed, so using them afterwards may do
1712 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1713 one (it's much easier than Tom thought :-).
1715 (5) Because of (4), we can now allow for the use of $prvscheck_result
1716 inside the third argument.
1718 PH/03 For some reason, the default setting of PATH when running a command from
1719 a pipe transport was just "/usr/bin". I have changed it to
1722 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1723 anything to be listed in the output from -bV.
1725 PH/05 When a filter generated an autoreply, the entire To: header line was
1726 quoted in the delivery log line, like this:
1728 => >A.N.Other <ano@some.domain> <original@ddress> ...
1730 This has been changed so that it extracts the operative address. There
1731 may be more than one such address. If so, they are comma-separated, like
1734 => >ano@some.domain,ona@other.domain <original@ddress> ...
1736 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1737 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1738 not being looked up in the rDNS to get a host name, Exim was showing the
1739 IP address twice in Received: lines, even though the IP addresses were
1740 identical. For example:
1742 Received: from [1.2.3.4] (helo=[1.2.3.4])
1744 However, if the real host name was known, it was omitting the HELO data
1745 if it matched the actual IP address. This has been tidied up so that it
1746 doesn't show the same IP address twice.
1748 PH/07 When both +timestamp and +memory debugging was on, the value given by
1749 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1750 called by the debug printing, thereby overwriting the timestamp buffer.
1751 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1753 PH/08 When the original message was included in an autoreply transport, it
1754 always said "this is a copy of the message, including all the headers",
1755 even if body_only or headers_only was set. It now gives an appropriate
1758 PH/09 Applied a patch from the Sieve maintainer which:
1760 o fixes some comments
1761 o adds the (disabled) notify extension core
1762 o adds some debug output for the result of if/elsif tests
1763 o points to the current vacation draft in the documentation
1764 and documents the missing references header update
1768 o fixes a bug in processing the envelope test (when testing
1769 multiple envelope elements, the last element determinted the
1772 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1773 Electronic Mail") by including:
1775 Auto-submitted: auto-generated
1777 in the messages that it generates (bounce messages and others, such as
1778 warnings). In the case of bounce messages for non-SMTP mesages, there was
1779 also a typo: it was using "Auto_submitted" (underscore instead of
1780 hyphen). Since every message generated by Exim is necessarily in response
1781 to another message, thes have all been changed to:
1783 Auto-Submitted: auto-replied
1785 in accordance with these statements in the RFC:
1787 The auto-replied keyword:
1789 - SHOULD be used on messages sent in direct response to another
1790 message by an automatic process,
1792 - MUST NOT be used on manually-generated messages,
1794 - MAY be used on Delivery Status Notifications (DSNs) and Message
1795 Disposition Notifications (MDNs),
1797 - MUST NOT be used on messages generated by automatic or periodic
1798 processes, except for messages which are automatic responses to
1801 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1802 to the default Received: header definition.
1804 PH/12 Added log selector acl_warn_skipped (default on).
1806 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1807 variables because (a) they are in the wrong storage pool and (b) even if
1808 they were copied, it wouldn't work properly because of the caching.
1810 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1811 checking when decoding. Apparently there are clients that generate
1812 overlong encoded strings. Why am I not surprised?
1814 PH/15 If the first argument of "${if match_address" was not empty, but did not
1815 contain an "@" character, Exim crashed. Now it writes a panic log message
1816 and treats the condition as false.
1818 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1820 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1821 extension "envelope-auth". The code is finished and in agreement with
1822 other implementations, but there is no documentation so far and in fact,
1823 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1824 not changing the active code.
1826 Print executed "if" and "elsif" statements when debugging is used. This
1827 helps a great deal to understand what a filter does.
1829 Document more things not specified clearly in RFC3028. I had all this
1830 sorted out, when out of a sudden new issues came to my mind. Oops."
1832 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1835 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1836 canonical form (as documented). However, after a host name lookup from
1837 the IP address, check_host() was doing a simple string comparison with
1838 addresses acquired from the DNS when checking that the found name did
1839 have the original IP as one of its addresses. Since any found IPv6
1840 addresses are likely to be in abbreviated form, the comparison could
1841 fail. Luckily, there already exists a function for doing the comparison
1842 by converting both addresses to binary, so now that is used instead of
1843 the text comparison.
1845 PH/20 There was another similar case to PH/19, when a complete host name was
1846 given in a host list; looking up its IP address could give an abbreviated
1847 form, whereas the current host's name might or might not be abbreviated.
1848 The same fix has been applied.
1854 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1855 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1858 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1859 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1861 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1863 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1864 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1865 Also, for "Auto-Submitted", treat anything other than "no" as
1866 non-personal, in accordance with RFC 3834. (Previously it treated
1867 anything starting "auto-" as non-personal.)
1869 TF/01 The control=submission/name=... option had a problem with syntax
1870 errors if the name included a slash character. The /name= option
1871 now slurps the rest of the string, so it can include any characters
1872 but it must come last in the list of options (after /sender_retain
1875 PH/05 Some modifications to the interface to the fake nameserver for the new
1883 TK/01 Added the "success_on_redirect" address verification option. See
1884 NewStuff for rationale and an example.
1886 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1888 PH/02 Patch to exigrep to allow it to work on syslog lines.
1890 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1891 fread() to skip over the body file's header line, because in Cygwin the
1892 header line is locked and is inaccessible.
1894 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1895 co-exist for some time) to make it clear that it is the Exim ID that is
1896 referenced, not the Message-ID: header line.
1898 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1899 string_format() function, because snprintf() does not exist on all
1902 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1903 ${stat: expansion item.
1905 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1906 protocol synchronization error", to keep the pedants happy.
1908 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1909 well as for IRIX systems, when gcc is being used. See the host.c source
1912 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1914 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1917 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1918 transport and to the smtp transport in LMTP mode.
1920 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1922 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1923 run a filter in a subprocess. This could lead to confusion in subsequent
1924 lookups in the parent process. There should also be a search_tidyup() at
1925 the end of the subprocess.
1927 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1928 only if the host matched helo_try_verify_hosts, which caused the
1929 verification to occur when the EHLO/HELO command was issued. The ACL just
1930 tested the remembered result. Now, if a previous verification attempt has
1931 not happened, "verify = helo" does it there and then.
1933 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1935 TK/03 Fix log output including CR from clamd.
1937 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1938 exist provoked a memory error which could cause a segfault.
1940 PH/15 Installed PCRE 6.2
1942 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1944 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1945 of the problem. Specifically, suggested +O2 rather than +O1 for the
1948 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1950 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1951 record, the port was not correctly logged when the outgoing_port log
1952 selector was set (it logged the transort's default port).
1954 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1955 fallback_hosts, and "hosts" in the smtp transport.
1957 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1958 host errors such as "Connection refused".
1960 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1961 authentication with radiusclient 0.4.9:
1963 - Error returned from rc_read_config was caught wrongly
1964 - Username/password not passed on to radius server due to wrong length.
1966 The presumption is that some radiusclient API changes for 4.51/PH/17
1967 were not taken care of correctly. The code is still untested by me (my
1968 Linux distribution still has 0.3.2 of radiusclient), but it was
1969 contributed by a Radius user.
1971 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1972 expanding the "port" option of the smtp transport.
1974 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1975 while reading a MIME header. Thanks to Tom Hughes for a patch.
1977 PH/24 Include config.h inside local_scan.h so that configuration settings are
1980 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1981 that in an ACL for RCPT (for example), you can examine exactly what was
1984 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1985 commands, but it was not correctly comparing the address with the actual
1986 client host address. Thus, it would show the EHLO address in Received:
1987 header lines when this was not necessary.
1989 PH/27 Added the % operator to ${eval:}.
1991 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1992 it should be ignoring failures (because with -C, for example, it has lost
1993 privilege). It wasn't ignoring creation failures other than "already
1996 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1999 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2000 because an input line was too long, either on its own, or by virtue of
2001 too many continuations, the temporary file was not being removed, and the
2002 return code was incorrect.
2004 PH/31 Missing "BOOL" in function definition in filtertest.c.
2006 PH/32 Applied Sieve patches from the maintainer.
2008 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2010 PH/33 Added "verify = not_blind".
2012 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2013 Local/Makefile (with some defaults set). These are used in built scripts
2014 such as exicyclog, but they have never been used in the exim_install
2015 script (though there are many overriding facilities there). I have
2016 arranged that the exim_install script now takes note of these two
2019 PH/35 Installed configuration files for Dragonfly.
2021 PH/36 When a locally submitted message by a trusted user did not contain a
2022 From: header, and the sender address was obtained from -f or from an SMTP
2023 MAIL command, and the trusted user did not use -F to supply a sender
2024 name, $originator_name was incorrectly used when constructing a From:
2025 header. Furthermore, $originator_name was used for submission mode
2026 messages from external hosts without From: headers in a similar way,
2027 which is clearly wrong.
2029 PH/37 Added control=suppress_local_fixups.
2031 PH/38 When log_selector = +received_sender was set, and the addition of the
2032 sender made the log line's construction buffer exactly full, or one byte
2033 less than full, an overflow happened when the terminating "\n" was
2036 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2037 when the result of a list match is failure because a DNS lookup failed.
2039 PH/40 RM_COMMAND is now used in the building process.
2041 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2042 the "build-* directories that it finds.
2044 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2045 address in a domain literal was a prefix of an interface address.
2047 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2048 when verifying a sender address, unless rewrite_headers is false.
2050 PH/44 Wrote a long comment about why errors_to addresses are verified as
2051 recipients, not senders.
2053 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2054 the ratelimit ACL was added.
2056 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2058 PH/46 Added extra information about PostgreSQL errors to the error string.
2060 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2061 avoiding the need to install special zones in a real server. This is
2062 backwards compatible; if it can't find the fake resolver, it drops back.
2063 Thus, both old and new test suites can be run.
2065 TF/02 Added util/ratelimit.pl
2067 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2068 clock is set back in time.
2070 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2071 Candler <B.Candler@pobox.com>.
2073 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2074 OK for addresses that are the result of redirections.
2076 PH/48 A number of further additions for the benefit of the new test suite,
2077 including a fake gethostbyname() that interfaces to the fake DNS resolver
2080 TF/06 The fix for widen_domains has also been applied to qualify_single and
2081 search_parents which are the other dnslookup options that can cause
2084 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2087 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2089 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2090 service" if it encountered an MX record with an empty target hostname.
2091 The message is now "an MX or SRV record indicated no SMTP service".
2093 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2094 if the DNS of the sending site is misconfigured. This is quite a
2095 common situation. This change restores the behaviour of treating a
2096 helo verification defer as a failure.
2098 PH/51 If self=fail was set on a router, the bounce message did not include the
2099 actual error message.
2105 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2107 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2108 command itself ended in error, the underlying message about the transport
2109 filter timeout was being overwritten with the pipe command error. Now the
2110 underlying error message should be appended to the second error message.
2112 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2114 PH/02 Reduce the amount of output that "make" produces by default. Full output
2115 can still be requested.
2117 PH/03 The warning log line about a condition test deferring for a "warn" verb
2118 was being output only once per connection, rather than after each
2119 occurrence (because it was using the same function as for successful
2120 "warn" verbs). This seems wrong, so I have changed it.
2122 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2123 it should not have, which might have caused a crash in the right
2124 circumstances, but probably never did.
2126 PH/04 Installed a modified version of Tony Finch's patch to make submission
2127 mode fix the return path as well as the Sender: header line, and to
2128 add a /name= option so that you can make the user's friendly name appear
2131 TF/03 Added the control = fakedefer ACL modifier.
2133 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2134 Mark Lowes for thorough testing.
2136 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2138 TK/03 Merged latest SRS patch from Miles Wilton.
2140 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2141 with the definition in sysexits.h (which is #included earlier).
2142 Fortunately, Exim does not actually use EX_OK. The code used to try to
2143 preserve the sysexits.h value, by assumimg that macro definitions were
2144 scanned for macro replacements. I have been disabused of this notion,
2145 so now the code just undefines EX_OK before #including unistd.h.
2147 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2148 in the smtp transport. When a block could not be written in a single
2149 write() function, the timeout was being re-applied to each part-write.
2150 This seems wrong - if the receiver was accepting one byte at a time it
2151 would take for ever. The timeout is now adjusted when this happens. It
2152 doesn't have to be particularly precise.
2154 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2155 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2157 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2158 without a domain if the check to <postmaster@domain> fails.
2160 SC/01 Eximstats: added -xls and the ability to specify output files
2161 (patch written by Frank Heydlauf).
2163 SC/02 Eximstats: use FileHandles for outputing results.
2165 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2167 SC/04 Eximstats: fixed display of large numbers with -nvr option
2169 SC/05 Eximstats: fixed merging of reports with empty tables.
2171 SC/06 Eximstats: added the -include_original_destination flag
2173 SC/07 Eximstats: removed tabs and trailing whitespace.
2175 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2177 TK/06 MBOX spool code: Add real "From " MBOX separator line
2178 so the .eml file is really in mbox format (even though
2179 most programs do not really care). Patch from Alex Miller.
2181 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2182 The latter is generated from $received_to and is only set if the
2183 message has one envelope recipient. SA can use these headers,
2184 obviously out-of-the-box. Patch from Alex Miller.
2186 PH/08 The ${def test on a variable was returning false if the variable's
2187 value was "0", contrary to what the specification has always said!
2188 The result should be true unless the variable is empty.
2190 PH/09 The syntax error of a character other than { following "${if
2191 def:variable_name" (after optional whitespace) was not being diagnosed.
2192 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2193 accidental colon was present, for example, could give incorrect results.
2195 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2196 result is used (not including appendfile, where other changes are about
2199 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2200 This involved changing a lot of size variables from int to off_t. It
2201 should work with maildirs and everything.
2203 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2204 spamd dying while we are connected to it.
2206 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2209 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2210 function for BDB 4.3.
2212 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2214 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2215 the sources for PCRE have been re-arranged and more files are now
2218 PH/15 The code I had for printing potentially long long variables in PH/11
2219 above was not the best (it lost precision). The length of off_t variables
2220 is now inspected at build time, and an appropriate printing format (%ld
2221 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2222 to be "long long int" or "long int". This is needed for the internal
2223 formatting function string_vformat().
2225 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2226 the configuration file to be ":syslog", then the script "guesses" where
2227 the logs files are, rather than using the compiled in default. In our
2228 case the guess is not the same as the compiled default, so the script
2229 suddenly stopped working when I started to use syslog. The patch checks
2230 to see if log_file_path is "". If so, it attempts to read it from exim
2231 with no configuration file to get the compiled in version, before it
2232 falls back to the previous guessing code."
2234 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2235 implementing BATV in an Exim configuration. See NewStuff for the gory
2238 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2239 Makefile that are specific to HP-UX.
2241 PH/18 If the "use_postmaster" option was set for a recipient callout together
2242 with the "random" option, the postmaster address was used as the MAIL
2243 FROM address for the random test, but not for the subsequent recipient
2244 test. It is now used for both.
2246 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2247 patch removes a few documentation additions to RFC 3028, because the
2248 latest draft now contains them. It adds the new en;ascii-case comparator
2249 and a new error check for 8bit text in MIME parts. Comparator and
2250 require names are now matched exactly. I enabled the subaddress
2251 extension, but it is not well tested yet (read: it works for me)."
2253 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2254 rework some of the code of TK/09 above to avoid the hardwired use of
2255 "%lld" and "long long". Replaced the call to snprintf() with a call to
2258 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2259 records point to non-existent hosts", "retry timeout exceeded", and
2260 "retry time not reached for any host after a long failure period".
2262 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2263 experimental DomainKeys support:
2265 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2266 (2) On an error such as an illegally used "control", the wrong name for
2267 the control was given.
2269 These problems did NOT occur unless DomainKeys support was compiled.
2271 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2273 PH/24 Added ${if match_ip condition.
2275 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2276 databases so that it will be absolutely obvious if a crash occurs in the
2277 DB library. This is a regular occurrence (often caused by mis-matched
2280 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2281 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2282 picked up on a user's system that detects such things. There doesn't seem
2283 to be a gcc warning option for this - only an attribute that has to be
2284 put on the function's prototype. It seems that in Fedora Core 4 they have
2285 set this on a number of new functions. No doubt there will be more in due
2288 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2289 specify a transport. However, if an address that was verified by such a
2290 router was the subject of a callout, Exim crashed because it tried to
2291 read the rcpt_include_affixes from the non-existent transport. Now it
2292 just assumes that the setting of that option is false. This bug was
2293 introduced by 4.51/PH/31.
2295 PH/28 Changed -d+all to exclude +memory, because that information is very
2296 rarely of interest, but it makes the output a lot bigger. People tend to
2297 do -d+all out of habit.
2299 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2300 code in os-type was giving problems when libc.so lives in lib64, like on
2303 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2304 aren't the modern standard, and it seems that some systems' include files
2305 don't always have them. Exim was already checking for some of the newer
2306 ones like T_AAAA, and defining it itself. I've added checks for all the
2307 record types that Exim uses.
2309 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2310 not automatically generating a new one, as it is supposed to. This
2311 prevented TLS from working. If the file did exist, but contained invalid
2312 data, a new version was generated, as expected. It was only the case of a
2313 non-existent file that was broken.
2315 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2316 with a change in libdomainkeys > 0.64.
2318 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2319 from DNS. If the selector record carries the flag, it now has
2320 precedence over the domain-wide flag.
2322 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2324 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2325 the use of a port name that isn't defined in /etc/services) occurred, the
2326 message was deferred as in a normal delivery, and thus remained on the
2327 spool, instead of being failed because of the mua_wrapper setting. This
2328 is now fixed, and I tidied up some of the mua_wrapper messages at the
2331 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2332 the messages in a hash of arrays rather than using individual hashes.
2333 This is a bit cleaner and results in dramatic memory savings, albeit
2334 at a slight CPU cost.
2336 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2337 as requested by Marc Sherman.
2339 SC/10 Eximstats: added histograms for user specified patterns as requested
2342 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2344 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2345 fopen() in the content-scanning modules that did not already have it.
2351 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2352 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2354 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2356 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2358 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2359 to test Sieve filters that use "vacation".
2361 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2362 that changes the way the GnuTLS parameters are stored in the cache file.
2363 The new format can be generated externally. For backward compatibility,
2364 if the data in the cache doesn't make sense, Exim assumes it has read an
2365 old-format file, and it generates new data and writes a new file. This
2366 means that you can't go back to an older release without removing the
2369 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2370 work if there are any delivery delays because "one_time" forces the
2371 parent to be marked "delivered", so its unseen clone is never tried
2372 again. For this reason, Exim now forbids the simultaneous setting of
2375 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2376 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2377 entry if you want to know the details. The fix, however, neglected to
2378 consider the case when local delivery batching is involved. The test for
2379 "previously delivered" was not happening when checking to see if an
2380 address could be batched with a previous (undelivered) one; under
2381 certain circumstances this could lead to multiple deliveries to the same
2384 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2385 in its include files, and this causes problems building Exim.
2387 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2388 header_syntax) but Exim was just ignoring anything given after a slash.
2389 In particular, this caused confusion with an attempt to use "verify =
2390 reverse_host_lookup/defer_ok". An error is now given when options are
2391 supplied for verify items that do not have them. (Maybe reverse_host_
2392 lookup should have a defer_ok option, but that's a different point.)
2394 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2395 defined by RFC 821) to 2048, because there were problems with some AUTH
2396 commands, and RFC 1869 says the size should be increased for extended
2397 SMTP commands that take arguments.
2399 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2402 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2403 "unknown" error; now it says that the functionality isn't in the binary.
2405 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2406 an address' error message when a string expansion fails (syntax or
2407 whatever). Otherwise the password may appear in the log. Following change
2408 PH/42 below, there is no longer a chance of it appearing in a bounce
2411 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2413 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2414 of its addresses was ever tried. (Bugzilla bug #2.)
2416 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2417 the result incorrectly in the debug output. (It correctly added a newline
2418 to what was transported.)
2420 TF/01 Added $received_time.
2422 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2423 commented out examples of how to interface to a virus scanner and to
2424 SpamAssassin. Also added commented examples of av_scanner and
2425 spamd_address settings.
2427 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2428 and controls are allowed in which ACLs. There were a couple of minor
2429 errors. Some of the entries in the conditions table (which is a table of
2430 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2431 negation of where the condition IS allowed.
2433 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2435 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2436 header file does not have a version number, so I've had to invent a new
2437 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2438 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2439 radiusclient), but it was contributed by a Radius user.
2441 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2442 files or directories, for OpenSSL.
2444 PH/19 When an Exim process that is running as root has to create an Exim log
2445 file, it does so in a subprocess that runs as exim:exim so as to get the
2446 ownership right at creation (otherwise, other Exim processes might see
2447 the file with the wrong ownership). There was no test for failure of this
2448 fork() call, which would lead to the process getting stuck as it waited
2449 for a non-existent subprocess. Forks do occasionally fail when resources
2450 run out. I reviewed all the other calls to fork(); they all seem to check
2453 PH/20 When checking for unexpected SMTP input at connect time (before writing
2454 the banner), Exim was not dealing correctly with a non-positive return
2455 from the read() function. If the client had disconnected by this time,
2456 the result was a log entry for a synchronization error with an empty
2457 string after "input=" when read() returned zero. If read() returned -1
2458 (an event I could not check), uninitialized data bytes were printed.
2459 There were reports of junk text (parts of files, etc) appearing after
2462 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2464 PH/22 Added support for macro redefinition, and (re)definition in between
2465 driver and ACL definitions.
2467 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2468 forgetting to use the resulting value; it was using the unexpanded value.
2470 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2471 hadn't been configured. The fix is from Juergen Kreileder, who
2472 understands it better than I do:
2474 "Here's what I see happening with three configured cyrus_sasl
2475 authenticators configured (plain, login, cram-md5):
2477 On startup auth_cyrus_sasl_init() gets called for each of these.
2478 This means three calls to sasl_listmech() without a specified mech_list.
2479 => SASL tests which mechs of all available mechs actually work
2480 => three warnings about OTP not working
2481 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2483 With the patch, sasl_listmech() also gets called three times. But now
2484 SASL's mech_list option is set to the server_mech specified in the the
2485 authenticator. Or in other words, the answer from sasl_listmech()
2486 gets limited to just the mech you're testing for (which is different
2488 => the return list contains just 'plain' or 'login', 'cram-md5' or
2489 nothing depending on the value of ob->server_mech.
2491 I've just tested the patch: Authentication still works fine,
2492 unavailable mechs specified in the exim configuration are still
2493 caught, and the auth.log warnings about OTP are gone."
2495 PH/25 When debugging is enabled, the contents of the command line are added
2496 to the debugging output, even when log_selector=+arguments is not
2499 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2500 answer is "GNU", and only if the return is "GNU/something" is the answer
2503 PH/27 $acl_verify_message is now set immediately after the failure of a
2504 verification in an ACL, and so is available in subsequent modifiers. In
2505 particular, the message can be preserved by coding like this:
2507 warn !verify = sender
2508 set acl_m0 = $acl_verify_message
2510 Previously, $acl_verify_message was set only while expanding "message"
2511 and "log_message" when a very denied access.
2513 PH/28 Modified OS/os.c-Linux with
2515 -#ifndef OS_LOAD_AVERAGE
2516 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2518 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2519 nomenclature these days.)
2521 PH/29 Installed patch from the Sieve maintainer that adds the options
2522 sieve_useraddress and sieve_subaddress to the redirect router.
2524 PH/30 In these circumstances:
2525 . Two addresses routed to the same list of hosts;
2526 . First host does not offer TLS;
2527 . First host accepts first address;
2528 . First host gives temporary error to second address;
2529 . Second host offers TLS and a TLS session is established;
2530 . Second host accepts second address.
2531 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2532 and peerdn, if requested) that were in fact used only for the second
2535 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2536 attention to any local part prefix or suffix that was matched by the
2537 router that accepted the address. It now behaves in the same way as it
2538 does for delivery: the affixes are removed from the local part unless
2539 rcpt_include_affixes is set on the transport.
2541 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2542 timeout during the DATA phase of an incoming message.
2544 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2545 applied a patch sanctioned by the Sieve maintainer.
2547 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2548 the uid or gid is negative. A case of a negative gid caused this to be
2549 noticed. The fix allows for either to be negative.
2551 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2552 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2555 PH/36 The change PH/12 above was broken. Fixed it.
2557 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2558 the grounds that routing the same address twice would always produce the
2559 same answer. This might have been true once, but it is certainly no
2560 longer true now. Routing a child address may depend on the previous
2561 routing that produced that child. Some complicated redirection strategies
2562 went wrong when messages had multiple recipients, and made Exim's
2563 behaviour dependent on the order in which the addresses were given.
2565 I have moved the duplicate checking until after the routing is complete.
2566 Exim scans the addresses that are assigned to local and remote
2567 transports, and removes any duplicates. This means that more work will be
2568 done, as duplicates will always all be routed, but duplicates are
2569 presumably rare, so I don't expect this is of any significance.
2571 For deliveries to pipes, files, and autoreplies, the duplicate checking
2572 still happens during the routing process, since they are not going to be
2575 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2576 It corrects a timeout issue with spamd. This is Ian's comment: "The
2577 background is that sometimes spamd either never reads data from a
2578 connection it has accepted, or it never writes response data. The exiscan
2579 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2580 blindly assumes that writes won't block so it may never time out."
2582 PH/39 Allow G after quota size as well as K and M.
2584 PH/40 The value set for $authenticated_id in an authenticator may not contain
2585 binary zeroes or newlines because the value is written to log lines and
2586 to spool files. There was no check on this. Now the value is run through
2587 the string_printing() function so that such characters are converted to
2588 printable escape sequences.
2590 PH/41 $message_linecount is a new variable that contains the total number of
2591 lines in the message. Compare $body_linecount, which is the count for the
2594 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2595 bounce and delay warning messages, except in certain special cases, which
2598 (a) An SMTP error message from a remote host;
2599 (b) A message specified in a :fail: redirection;
2600 (c) A message specified in a "fail" command in a system filter;
2601 (d) A message specified in a FAIL return from the queryprogram router;
2602 (e) A message specified by the cannot_route_message router option.
2604 In these cases only, Exim does include the error details in bounce and
2605 warning messages. There are also a few cases where bland messages such
2606 as "unrouteable address" or "local delivery error" are given.
2608 PH/43 $value is now also set for the "else" part of a ${run expansion.
2610 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2611 being worked on, but at least Exim now implements the latest version to
2614 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2615 process to complete was treated as a delivery failure, a timeout while
2616 writing the message to the pipe was logged, but erroneously treated as a
2617 successful delivery. Such timeouts include transport filter timeouts. For
2618 consistency with the overall process timeout, these timeouts are now
2619 treated as errors, giving rise to delivery failures by default. However,
2620 there is now a new Boolean option for the pipe transport called
2621 timeout_defer, which, if set TRUE, converts the failures into defers for
2622 both kinds of timeout. A transport filter timeout is now identified in
2625 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2626 systems where "make" and "gmake" are different, calling "gmake" at top
2627 level broke things. I've arranged for the value of $(MAKE) to be passed
2628 from the Makefile to this script so that it can call the same version of
2632 A note about Exim versions 4.44 and 4.50
2633 ----------------------------------------
2635 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2636 changes of various kinds. As a consequence, a big documentation update was
2637 needed. This delayed the release for rather longer than seemed good, especially
2638 in the light of a couple of (minor) security issues. Therefore, the changes
2639 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2640 release. So 4.44 and 4.50 are in effect two different branches that both start
2643 I have left the 4.50 change log unchanged; it contains all the changes since
2644 4.43. The change log for 4.44 is below; many of its items are identical to
2645 those for 4.50. This seems to be the most sensible way to preserve the
2646 historical information.
2652 1. Minor wording change to the doc/README.SIEVE file.
2654 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2655 computation of the current number of files was incorrect.
2657 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2658 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2661 4. Give more explanation in the error message when the command for a transport
2662 filter fails to execute.
2664 5. There are several places where Exim runs a non-Exim command in a
2665 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2666 was being done only for the command run by the queryprogram router. It is
2667 now done for all such subprocesses. The other cases are: ${run, transport
2668 filters, and the commands run by the lmtp and pipe transports.
2670 6. Added CONFIGURE_GROUP build-time option.
2672 7. Some older OS have a limit of 256 on the maximum number of file
2673 descriptors. Exim was using setrlimit() to set 1000 as a large value
2674 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2675 systems. I've change it so that if it can't get 1000, it tries for 256.
2677 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2678 was an oversight, and furthermore, ever since the addition of extra
2679 controls (e.g. 4.43/32), the checks on when to allow different forms of
2680 "control" were broken. There should now be diagnostics for all cases when a
2681 control that does not make sense is encountered.
2683 9. Added the /retain_sender option to "control=submission".
2685 10. $recipients is now available in the predata ACL (oversight).
2687 11. Tidy the search cache before the fork to do a delivery from a message
2688 received from the command line. Otherwise the child will trigger a lookup
2689 failure and thereby defer the delivery if it tries to use (for example) a
2690 cached ldap connection that the parent has called unbind on.
2692 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2693 of $address_data from the recipient verification was clobbered by the
2694 sender verification.
2696 13. The value of address_data from a sender verification is now available in
2697 $sender_address_data in subsequent conditions in the ACL statement.
2699 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2701 15. Added a new option "connect=<time>" to callout options, to set a different
2704 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2705 was its contents. (It was OK if the option was not defined at all.)
2707 17. A "Completed" log line is now written for messages that are removed from
2708 the spool by the -Mrm option.
2710 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2711 information about exactly what failed.
2713 19. Added -dd to debug only the daemon process.
2715 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2716 handles timeouts, both on the server side and network timeouts. Renamed the
2717 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2719 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2720 It is now set to "smtps".
2722 22. $host_address is now set to the target address during the checking of
2723 ignore_target_hosts.
2725 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2726 being passed; this would have caused $sender_host_name to have been used if
2727 matching the list had actually called for a host name (not very likely,
2728 since this list is usually IP addresses). A host name is now passed as
2731 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2732 code that specifies a non-restarting handler (typically sigaction() in
2733 modern systems) in an attempt to fix a rare and obscure crash bug.
2735 25. Narrowed the window for a race in the daemon that could cause it to ignore
2736 SIGCHLD signals. This is not a major problem, because they are used only to
2737 wake it up if nothing else does.
2739 26. A malformed maildirsize file could cause Exim to calculate negative values
2740 for the mailbox size or file count. Odd effects could occur as a result.
2741 The maildirsize information is now recalculated if the size or filecount
2744 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2745 support for a long time. Removed HAVE_SYS_VFS_H.
2747 28. Installed the latest version of exipick from John Jetmore.
2749 29. In an address list, if the pattern was not a regular expression, an empty
2750 subject address (from a bounce message) matched only if the pattern was an
2751 empty string. Non-empty patterns were not even tested. This was the wrong
2752 because it is perfectly reasonable to use an empty address as part of a
2753 database query. An empty address is now tested by patterns that are
2754 lookups. However, all the other forms of pattern expect the subject to
2755 contain a local part and a domain, and therefore, for them, an empty
2756 address still always fails if the pattern is not itself empty.
2758 30. Exim went into a mad DNS loop when attempting to do a callout where the
2759 host was specified on an smtp transport, and looking it up yielded more
2760 than one IP address.
2762 31. Re-factored the code for checking spool and log partition space into a
2763 function that finds that data and another that does the check. The former
2764 is then used to implement four new variables: $spool_space, $log_space,
2765 $spool_inodes, and $log_inodes.
2767 32. The RFC2047 encoding function was originally intended for short strings
2768 such as real names; it was not keeping to the 75-character limit for
2769 encoded words that the RFC imposes. It now respects the limit, and
2770 generates multiple encoded words if necessary. To be on the safe side, I
2771 have increased the buffer size for the ${rfc2047: expansion operator from
2774 33. It is now permitted to omit both strings after an "if" condition; if the
2775 condition is true, the result is "true". As before, when the second string
2776 is omitted, a false condition yields an empty string. This makes it less
2777 cumbersome to write custom ACL and router conditions.
2779 34. Failure to deliver a bounce message always caused it to be frozen, even if
2780 there was an errors_to setting on the router. The errors_to setting is now
2783 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2784 canonical form (fully expanded) before being placed in
2785 $sender_host_address.
2787 36. The table in the code that translates DNS record types into text (T_A to
2788 "A" for instance) was missing entries for NS and CNAME. It is just possible
2789 that this could have caused confusion if both these types were looked up
2790 for the same domain, because the text type is used as part of Exim's
2791 per-process caching. But the chance of anyone hitting this buglet seems
2794 37. The dnsdb lookup has been extended in a number of ways.
2796 (1) There is a new type, "zns", which walks up the domain tree until it
2797 finds some nameserver records. It should be used with care.
2799 (2) There is a new type, "mxh", which is like "mx" except that it returns
2800 just the host names, not the priorities.
2802 (3) It is now possible to give a list of domains (or IP addresses) to be
2803 looked up. The behaviour when one of the lookups defers can be
2804 controlled by a keyword.
2806 (4) It is now possible to specify the separator character for use when
2807 multiple records are returned.
2809 38. The dnslists ACL condition has been extended: it is now possible to supply
2810 a list of IP addresses and/or domains to be looked up in a particular DNS
2813 39. Added log_selector=+queue_time_overall.
2815 40. When running the queue in the test harness, wait just a tad after forking a
2816 delivery process, to get repeatability of debugging output.
2818 41. Include certificate and key file names in error message when GnuTLS fails
2819 to set them up, because the GnuTLS error message doesn't include the name
2820 of the failing file when there is a problem reading it.
2822 42. Allow both -bf and -bF in the same test run.
2824 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2826 44. The "Exiscan patch" is now merged into the mainline Exim source.
2828 45. Sometimes the final signoff response after QUIT could fail to get
2829 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2830 < 0 before doing a fflush(). This bug looks as though it goes back to the
2831 introduction of TLS in release 3.20, but "sometimes" must have been rare
2832 because the tests only now provoked it.
2834 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2835 (this can affect the format of dates).
2837 47. exim_tidydb, when checking for the continued existence of a message for
2838 which it has found a message-specific retry record, was not finding
2839 messages that were in split spool directories. Consequently, it was
2840 deleting retry records that should have stayed in existence.
2842 48. Steve fixed some bugs in eximstats.
2844 49. The SPA authentication driver was not abandoning authentication and moving
2845 on to the next authenticator when an expansion was forced to fail,
2846 contradicting the general specification for all authenticators. Instead it
2847 was generating a temporary error. It now behaves as specified.
2849 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2850 (the order specifies the preference for clients). The order is now AES256,
2851 AES128, 3DES, ARCFOUR128.
2853 51. Small patch to Sieve code - explicitly set From: when generating an
2856 52. Exim crashed if a remote delivery caused a very long error message to be
2857 recorded - for instance if somebody sent an entire SpamAssassin report back
2858 as a large number of 550 error lines. This bug was coincidentally fixed by
2859 increasing the size of one of Exim's internal buffers (big_buffer) that
2860 happened as part of the Exiscan merge. However, to be on the safe side, I
2861 have made the code more robust (and fixed the comments that describe what
2864 53. Now that there can be additional text after "Completed" in log lines (if
2865 the queue_time_overall log selector is set), a one-byte patch to exigrep
2866 was needed to allow it to recognize "Completed" as not the last thing in
2869 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2870 patch that reportedly fixes this has been added. I am not expert enough to
2871 create a test for it. This is what the patch creator wrote:
2873 "I found a little strange behaviour of ldap code when working with
2874 Windows 2003 AD Domain, where users was placed in more than one
2875 Organization Units. When I tried to give exim partial DN, the exit code
2876 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2877 But simultaneously result of request was absolutely normal ldap result,
2878 so I produce this patch..."
2880 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2881 REFERENCE, so I have modified the code to exclude the patch when that macro
2884 55. Some experimental protocols are using DNS PTR records for new purposes. The
2885 keys for these records are domain names, not reversed IP addresses. The
2886 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2887 leaves it alone. Component reversal etc. now happens only for IP addresses.
2890 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2892 57. Double the size of the debug message buffer (to 2048) so that more of very
2893 long debug lines gets shown.
2895 58. The exicyclog utility now does better if the number of log files to keep
2896 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2898 59. Two changes related to the smtp_active_hostname option:
2900 (1) $smtp_active_hostname is now available as a variable.
2901 (2) The default for smtp_banner uses $smtp_active_hostname instead
2902 of $primary_hostname.
2904 60. The host_aton() function is supposed to be passed a string that is known
2905 to be a valid IP address. However, in the case of IPv6 addresses, it was
2906 not checking this. This is a hostage to fortune. Exim now panics and dies
2907 if the condition is not met. A case was found where this could be provoked
2908 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2909 components; fortuitously, this particular loophole had already been fixed
2910 by change 4.50/55 above.
2912 If there are any other similar loopholes, the new check in host_aton()
2913 itself should stop them being exploited. The report I received stated that
2914 data on the command line could provoke the exploit when Exim was running as
2915 exim, but did not say which command line option was involved. All I could
2916 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2917 running as the user.
2920 61. There was a buffer overflow vulnerability in the SPA authentication code
2921 (which came originally from the Samba project). I have added a test to the
2922 spa_base64_to_bits() function which I hope fixes it.
2925 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2926 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2928 63. The daemon start-up calls getloadavg() while still root for those OS that
2929 need the first call to be done as root, but it missed one case: when
2930 deliver_queue_load_max is set with deliver_drop_privilege. This is
2931 necessary for the benefit of the queue runner, because there is no re-exec
2932 when deliver_drop_privilege is set.
2934 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2935 This has been fixed.
2937 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2938 was in use, was not putting the data itself into the right store pool;
2939 consequently, it could be overwritten for a subsequent message in the same
2940 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2943 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2945 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2946 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2947 the function were treating the return as a boolean value, which happened to
2948 work because 0=false and not-0=true, but is not correct code.
2950 68. The host_aton() function was not handling scoped IPv6 addresses (those
2951 with, for example, "%eth0" on the end) correctly.
2953 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2954 negated items (that is, ~something) in unsigned ints. Some compilers
2955 apparently mutter when there is no cast.
2957 70. If an address verification called from an ACL failed, and did not produce a
2958 user-specific message (i.e. there was only a "system" message), nothing was
2959 put in $acl_verify_message. In this situation, it now puts the system
2962 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2963 session; change 4.31/43 added the unwanted input to the log line - except
2964 that it did not do this in the start of session case. It now does.
2966 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2967 This is wrong and can cause the other end to generate a synchronization
2968 error if it is another Exim or anything else that does the synchronization
2969 check. A QUIT command is no longer sent after a timeout.
2971 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2972 during host lookups.
2974 74. The defer_ok option of callout verification was not working if it was used
2975 when verifying addresses in header lines, that is, for this case:
2977 verify = header_sender/callout=defer_ok
2979 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2980 those file descriptors could be used for SMTP connections. If anything
2981 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2982 could be sent to the SMTP client, causing chaos. The daemon now opens
2983 stdin, stdout, and stderr to /dev/null when it puts itself into the
2986 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2987 log by default. The user can override this with suitable Perl magic.
2989 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2990 the log message when discard triggers, was not working for the DATA ACL or
2991 for the non-SMTP ACL.
2993 78. Error message wording change in sieve.c.
2995 79. If smtp_accept_max_per_host was set, the number of connections could be
2996 restricted to fewer than expected, because the daemon was trying to set up
2997 a new connection before checking whether the processes handling previous
2998 connections had finished. The check for completed processes is now done
2999 earlier. On busy systems, this bug wouldn't be noticed because something
3000 else would have woken the daemon, and it would have reaped the completed
3003 80. If a message was submitted locally by a user whose login name contained one
3004 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3005 It caused a spool format error. I have fixed the spool reading code. A
3006 related problem was that the "from" clause in the Received: line became
3007 illegal because of the space(s). It is now covered by ${quote_local_part.
3009 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3012 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3014 83. Patch from Sieve maintainer to fix unterminated string problem in
3015 "vacation" handling.
3017 84. Some minor changes to the Linux configuration files to help with other
3018 OS variants using glibc.
3020 85. One more patch for Sieve to update vacation handling to latest spec.
3023 ----------------------------------------------------
3024 See the note above about the 4.44 and 4.50 releases.
3025 ----------------------------------------------------
3031 1. Change 4.43/35 introduced a bug that caused file counts to be
3032 incorrectly computed when quota_filecount was set in an appendfile
3035 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3036 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3039 3. Give more explanation in the error message when the command for a transport
3040 filter fails to execute.
3042 4. There are several places where Exim runs a non-Exim command in a
3043 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3044 was being done only for the command run by the queryprogram router. It is
3045 now done for all such subprocesses. The other cases are: ${run, transport
3046 filters, and the commands run by the lmtp and pipe transports.
3048 5. Some older OS have a limit of 256 on the maximum number of file
3049 descriptors. Exim was using setrlimit() to set 1000 as a large value
3050 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3051 systems. I've change it so that if it can't get 1000, it tries for 256.
3053 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3054 was an oversight, and furthermore, ever since the addition of extra
3055 controls (e.g. 4.43/32), the checks on when to allow different forms of
3056 "control" were broken. There should now be diagnostics for all cases when a
3057 control that does not make sense is encountered.
3059 7. $recipients is now available in the predata ACL (oversight).
3061 8. Tidy the search cache before the fork to do a delivery from a message
3062 received from the command line. Otherwise the child will trigger a lookup
3063 failure and thereby defer the delivery if it tries to use (for example) a
3064 cached ldap connection that the parent has called unbind on.
3066 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3067 of $address_data from the recipient verification was clobbered by the
3068 sender verification.
3070 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3071 was its contents. (It was OK if the option was not defined at all.)
3073 11. A "Completed" log line is now written for messages that are removed from
3074 the spool by the -Mrm option.
3076 12. $host_address is now set to the target address during the checking of
3077 ignore_target_hosts.
3079 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3080 being passed; this would have caused $sender_host_name to have been used if
3081 matching the list had actually called for a host name (not very likely,
3082 since this list is usually IP addresses). A host name is now passed as
3085 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3086 code that specifies a non-restarting handler (typically sigaction() in
3087 modern systems) in an attempt to fix a rare and obscure crash bug.
3089 15. Narrowed the window for a race in the daemon that could cause it to ignore
3090 SIGCHLD signals. This is not a major problem, because they are used only to
3091 wake it up if nothing else does.
3093 16. A malformed maildirsize file could cause Exim to calculate negative values
3094 for the mailbox size or file count. Odd effects could occur as a result.
3095 The maildirsize information is now recalculated if the size or filecount
3098 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3099 support for a long time. Removed HAVE_SYS_VFS_H.
3101 18. Updated exipick to current release from John Jetmore.
3103 19. Allow an empty sender to be matched against a lookup in an address list.
3104 Previously the only cases considered were a regular expression, or an
3107 20. Exim went into a mad DNS lookup loop when doing a callout where the
3108 host was specified on the transport, if the DNS lookup yielded more than
3111 21. The RFC2047 encoding function was originally intended for short strings
3112 such as real names; it was not keeping to the 75-character limit for
3113 encoded words that the RFC imposes. It now respects the limit, and
3114 generates multiple encoded words if necessary. To be on the safe side, I
3115 have increased the buffer size for the ${rfc2047: expansion operator from
3118 22. Failure to deliver a bounce message always caused it to be frozen, even if
3119 there was an errors_to setting on the router. The errors_to setting is now
3122 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3123 canonical form (fully expanded) before being placed in
3124 $sender_host_address.
3126 24. Updated eximstats to version 1.33
3128 25. Include certificate and key file names in error message when GnuTLS fails
3129 to set them up, because the GnuTLS error message doesn't include the name
3130 of the failing file when there is a problem reading it.
3132 26. Expand error message when OpenSSL has problems setting up cert/key files.
3135 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3136 (this can affect the format of dates).
3138 28. exim_tidydb, when checking for the continued existence of a message for
3139 which it has found a message-specific retry record, was not finding
3140 messages that were in split spool directories. Consequently, it was
3141 deleting retry records that should have stayed in existence.
3143 29. eximstats updated to version 1.35
3144 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3145 1.35 - bugfix such that pie charts by volume are generated correctly
3147 30. The SPA authentication driver was not abandoning authentication and moving
3148 on to the next authenticator when an expansion was forced to fail,
3149 contradicting the general specification for all authenticators. Instead it
3150 was generating a temporary error. It now behaves as specified.
3152 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3153 (the order specifies the preference for clients). The order is now AES256,
3154 AES128, 3DES, ARCFOUR128.
3156 31. Small patch to Sieve code - explicitly set From: when generating an
3159 32. Exim crashed if a remote delivery caused a very long error message to be
3160 recorded - for instance if somebody sent an entire SpamAssassin report back
3161 as a large number of 550 error lines. This bug was coincidentally fixed by
3162 increasing the size of one of Exim's internal buffers (big_buffer) that
3163 happened as part of the Exiscan merge. However, to be on the safe side, I
3164 have made the code more robust (and fixed the comments that describe what
3167 33. Some experimental protocols are using DNS PTR records for new purposes. The
3168 keys for these records are domain names, not reversed IP addresses. The
3169 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3170 leaves it alone. Component reversal etc. now happens only for IP addresses.
3173 34. The host_aton() function is supposed to be passed a string that is known
3174 to be a valid IP address. However, in the case of IPv6 addresses, it was
3175 not checking this. This is a hostage to fortune. Exim now panics and dies
3176 if the condition is not met. A case was found where this could be provoked
3177 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3178 components; fortuitously, this particular loophole had already been fixed
3179 by change 4.50/55 or 4.44/33 above.
3181 If there are any other similar loopholes, the new check in host_aton()
3182 itself should stop them being exploited. The report I received stated that
3183 data on the command line could provoke the exploit when Exim was running as
3184 exim, but did not say which command line option was involved. All I could
3185 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3186 running as the user.
3189 35. There was a buffer overflow vulnerability in the SPA authentication code
3190 (which came originally from the Samba project). I have added a test to the
3191 spa_base64_to_bits() function which I hope fixes it.
3194 36. The daemon start-up calls getloadavg() while still root for those OS that
3195 need the first call to be done as root, but it missed one case: when
3196 deliver_queue_load_max is set with deliver_drop_privilege. This is
3197 necessary for the benefit of the queue runner, because there is no re-exec
3198 when deliver_drop_privilege is set.
3200 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3201 was in use, was not putting the data itself into the right store pool;
3202 consequently, it could be overwritten for a subsequent message in the same
3203 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3206 38. Sometimes the final signoff response after QUIT could fail to get
3207 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3208 < 0 before doing a fflush(). This bug looks as though it goes back to the
3209 introduction of TLS in release 3.20, but "sometimes" must have been rare
3210 because the tests only now provoked it.
3216 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3217 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3218 Now it returns an integer. A number of calls were still expecting a T/F
3219 return. Fortuitously, in all cases, the tests worked in OK situations,
3220 which is the norm. However, things would have gone wrong on any write
3221 failures on the smtp file descriptor. This function is used when sending
3222 messages over SMTP and also when doing verify callouts.
3224 2. When Exim is called to do synchronous delivery of a locally submitted
3225 message (the -odf or -odi options), it no longer closes stderr before doing
3228 3. Implemented the mua_wrapper option.
3230 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3232 5. Implemented the functions header_remove(), header_testname(),
3233 header_add_at_position(), and receive_remove_recipient(), and exported them
3236 6. If an ACL "warn" statement specified the addition of headers, Exim already
3237 inserted X-ACL-Warn: at the start if there was no header name. However, it
3238 was not making this test for the second and subsequent header lines if
3239 there were newlines in the string. This meant that an invalid header could
3240 be inserted if Exim was badly configured.
3242 7. Allow an ACL "warn" statement to add header lines at the start or after all
3243 the Received: headers, as well as at the end.
3245 8. Added the rcpt_4xx retry error code.
3247 9. Added postmaster_mailfrom=xxx to callout verification option.
3249 10. Added mailfrom=xxxx to the callout verification option, for verify=
3252 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3253 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3255 12. Inserted some casts to stop certain compilers warning when using pointer
3256 differences as field lengths or precisions in printf-type calls (mostly
3257 affecting debugging statements).
3259 13. Added optional readline() support for -be (dynamically loaded).
3261 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3262 same clock tick as a message's arrival, so that its received time was the
3263 same as the "first fail" time on the retry record, and that message
3264 remained on the queue past the ultimate address timeout, every queue runner
3265 would try a delivery (because it was past the ultimate address timeout) but
3266 after another failure, the ultimate address timeout, which should have then
3267 bounced the address, did not kick in. This was a "< instead of <=" error;
3268 in most cases the first failure would have been in the next clock tick
3269 after the received time, and all would be well.
3271 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3272 being recognized when the domain list was tested by the match_domain
3273 condition in an expansion string.
3275 16. Added the ${str2b64: operator.
3277 17. Exim was always calling setrlimit() to set a large limit for the number of
3278 processes, without checking whether the existing limit was already
3279 adequate. (It did check for the limit on file descriptors.) Furthermore,
3280 errors from getrlimit() and setrlimit() were being ignored. Now they are
3281 logged to the main and panic logs, but Exim does carry on, to try to do its
3282 job under whatever limits there are.
3284 18. Imported PCRE 5.0.
3286 19. Trivial typo in log message " temporarily refused connection" (the leading
3289 20. If the log selector return_path_on_delivery was set and an address was
3290 redirected to /dev/null, the delivery process crashed because it assumed
3291 that a return path would always be set for a "successful" delivery. In this
3292 case, the whole delivery is bypassed as an optimization, and therefore no
3295 21. Internal re-arrangement: the function for sending a challenge and reading
3296 a response while authentication was assuming a zero-terminated challenge
3297 string. It's now changed to take a pointer and a length, to allow for
3298 binary data in such strings.
3300 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3302 23. Exim was not respecting finduser_retries when seeking the login of the
3303 uid under which it was called; it was always trying 10 times. (The default
3304 setting of finduser_retries is zero.) Also, it was sleeping after the final
3305 failure, which is pointless.
3307 24. Implemented tls_on_connect_ports.
3309 25. Implemented acl_smtp_predata.
3311 26. If the domain in control=submission is set empty, Exim assumes that the
3312 authenticated id is a complete email address when it generates From: or
3313 Sender: header lines.
3315 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3316 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3317 chown and chgrp in /bin and hostname in /usr/bin.
3319 28. Exim was keeping the "process log" file open after each use, just as it
3320 does for the main log. This opens the possibility of it remaining open for
3321 long periods when the USR1 signal hits a daemon. Occasional processlog
3322 errors were reported, that could have been caused by this. Anyway, it seems
3323 much more sensible not to leave this file open at all, so that is what now
3326 29. The long-running daemon process does not normally write to the log once it
3327 has entered its main loop, and it closes the log before doing so. This is
3328 so that log files can straightforwardly be renamed and moved. However,
3329 there are a couple of unusual error situations where the daemon does write
3330 log entries, and I had neglected to close the log afterwards.
3332 30. The text of an SMTP error response that was received during a remote
3333 delivery was being truncated at 512 bytes. This is too short for some of
3334 the long messages that one sometimes sees. I've increased the limit to
3337 31. It is now possible to make retry rules that apply only when a message has a
3338 specific sender, in particular, an empty sender.
3340 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3341 it possible to be selective about when SMTP synchronization is enforced.
3343 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3345 32. Implemented hosts_connection_nolog.
3347 33. Added an ACL for QUIT.
3349 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3352 35. Added mailbox_size and mailbox_filecount to appendfile.
3354 36. Added control = no_multiline_responses to ACLs.
3356 37. There was a bug in the logic of the code that waits for the clock to tick
3357 in the case where the clock went backwards by a substantial amount such
3358 that the microsecond fraction of "now" was more than the microsecond
3359 fraction of "then" (but the whole seconds number was less).
3361 38. Added support for the libradius Radius client library this is found on
3362 FreeBSD (previously only the radiusclient library was supported).
3368 1. When certain lookups returned multiple values in the form name=value, the
3369 quoting of the values was not always being done properly. Specifically:
3370 (a) If the value started with a double quote, but contained no whitespace,
3372 (b) If the value contained whitespace other than a space character (i.e.
3373 tabs or newlines or carriage returns) it was not quoted.
3374 This fix has been applied to the mysql and pgsql lookups by writing a
3375 separate quoting function and calling it from the lookup code. The fix
3376 should probably also be applied to nisplus, ibase and oracle lookups, but
3377 since I cannot test any of those, I have not disturbed their existing code.
3379 2. A hit in the callout cache for a specific address caused a log line with no
3380 reason for rejecting RCPT. Now it says "Previous (cached) callout
3381 verification failure".
3383 3. There was an off-by-one bug in the queryprogram router. An over-long
3384 return line was truncated at 256 instead of 255 characters, thereby
3385 overflowing its buffer with the terminating zero. As well as fixing this, I
3386 have increased the buffer size to 1024 (and made a note to document this).
3388 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3389 when Exim is waiting for an SMTP response from a remote server, Exim
3390 restarts its select() call on the socket, thereby resetting its timeout.
3391 This is not a problem when such interrupts are rare. Somebody set up a cron
3392 job to run exiwhat every 2 minutes, which is less than the normal select()
3393 timeout (5 or 10 minutes). This meant that the select() timeout never
3394 kicked in because it was always reset. I have fixed this by comparing the
3395 time when an interrupt arrives with the time at the start of the first call
3396 to select(). If more time than the timeout has elapsed, the interrupt is
3397 treated as a timeout.
3399 5. Some internal re-factoring in preparation for the addition of Sieve
3400 extensions (by MH). In particular, the "personal" test is moved to a
3401 separate function, and given an option for scanning Cc: and Bcc: (which is
3402 not set for Exim filters).
3404 6. When Exim created an email address using the login of the caller as the
3405 local part (e.g. when creating a From: or Sender: header line), it was not
3406 quoting the local part when it contained special characters such as @.
3408 7. Installed new OpenBSD configuration files.
3410 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3411 try to make them clearer.
3413 9. Callout options, other than the timeout value, were being ignored when
3414 verifying sender addresses in header lines. For example, when using
3416 verify = header_sender/callout=no_cache
3418 the cache was (incorrectly) being used.
3420 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3421 only the Cygwin environment.
3423 11. When return_path_on_delivery was set as a log selector, if different remote
3424 addresses in the same message used different return paths and parallel
3425 remote delivery occurred, the wrong values would sometimes be logged.
3426 (Whenever a remote delivery process finished, the return path value from
3427 the most recently started remote delivery process was logged.)
3429 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3430 lines when AUTH and/or TLS are in use. This is the "received protocol"
3431 field. Exim used to use "asmtp" for authenticated SMTP, without any
3432 indication (in the protocol name) for TLS use. Now it follows the RFC and
3433 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3434 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3435 names appear in log lines as well as in Received: header lines.
3437 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3438 extensions, and comparison tests, and to fix some bugs.
3440 14. Changes to the "personal" filter test:
3442 (1) The test was buggy in that it was just doing the equivalent of
3443 "contains" tests on header lines. For example, if a user's address was
3444 anne@some.where, the "personal" test would incorrectly be true for
3446 To: susanne@some.where
3448 This test is now done by extracting each address from the header in turn,
3449 and checking the entire address. Other tests that are part of "personal"
3450 are now done using regular expressions (for example, to check local parts
3451 of addresses in From: header lines).
3453 (2) The list of non-personal local parts in From: addresses has been
3454 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3455 taken from the Sieve specification recommendations.
3457 (3) If the message contains any header line starting with "List-" it is
3458 treated as non-personal.
3460 (4) The test for "circular" in the Subject: header line has been removed
3461 because it now seems ill-conceived.
3463 15. Minor typos in src/EDITME comments corrected.
3465 16. Installed latest exipick from John Jetmore.
3467 17. If headers_add on a router specified a text string that was too long for
3468 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3469 of string_sprintf() is now avoided.
3471 18. $message_body_size was not set (it was always zero) when running the DATA
3472 ACL and the local_scan() function.
3474 19. For the "mail" command in an Exim filter, no default was being set for
3475 the once_repeat time, causing a random time value to be used if "once" was
3476 specified. (If the value happened to be <= 0, no repeat happened.) The
3477 default is now 0s, meaning "never repeat". The "vacation" command was OK
3478 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3479 (I found it when inspecting the code).
3481 20. There is now an overall timeout for performing a callout verification. It
3482 defaults to 4 times the callout timeout, which applies to individual SMTP
3483 commands during the callout. The overall timeout applies when there is more
3484 than one host that can be tried. The timeout is checked before trying the
3485 next host. This prevents very long delays if there are a large number of
3486 hosts and all are timing out (e.g. when the network connections are timing
3487 out). The value of the overall timeout can be changed by specifying an
3488 additional sub-option for "callout", called "maxwait". For example:
3490 verify = sender/callout=5s,maxwait=20s
3492 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3493 to the end before writing, but this should make it even safer).
3495 22. Exim was forgetting that it had advertised PIPELINING for the second and
3496 subsequent messages on an SMTP connection. It was also not resetting its
3497 memory on STARTTLS and an internal HELO.
3499 23. When Exim logs an SMTP synchronization error within a session, it now
3500 records whether PIPELINING has been advertised or not.
3502 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3503 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3504 rather than long int.
3506 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3508 26. Added the never_mail option to autoreply.
3514 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3515 crash if the getsockname() call failed; this can happen if a connection is
3516 closed very soon after it is established. The problem was simply in the
3517 order in which certain operations were done, causing Exim to try to write
3518 to the SMTP stream before it had set up the file descriptor. The bug has
3519 been fixed by making things happen in the correct order.
3525 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3526 before the connection was closed, thus losing the rejection response.
3528 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3529 some early Solaris releases, but causes trouble in current releases where
3530 socklen_t is defined.
3532 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3535 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3538 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3539 that releasing the top of it at the end releases what was used for sub-
3540 expansions (unless the block got too big). However, discard this block if
3541 the first thing is a variable or header, so that we can use its block when
3542 it is dynamic (useful for very large $message_headers, for example).
3544 6. Lookups now cache *every* query, not just the most recent. A new, separate
3545 store pool is used for this. It can be recovered when all lookup caches are
3546 flushed. Lookups now release memory at the end of their result strings.
3547 This has involved some general refactoring of the lookup sources.
3549 7. Some code has been added to the store_xxx() functions to reduce the amount
3550 of flapping under certain conditions.
3552 8. log_incoming_interface used to affect only the <= reception log lines. Now
3553 it causes the local interface and port to be added to several more SMTP log
3554 lines, for example "SMTP connection from", and rejection lines.
3556 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3558 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3560 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3561 could be overwritten at the end of the current message (or the start of a
3562 new message if it was set in a HELO ACL). The value is now preserved for
3563 the duration of the SMTP connection.
3565 12. If a transport had a headers_rewrite setting, and a matching header line
3566 contained an unqualified address, that address was qualified, even if it
3567 did not match any rewriting rules. The underlying bug was that the values
3568 of the flags that permit the existence of unqualified sender and recipient
3569 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3570 non-local messages, and by -bnq for local messages) were not being
3571 preserved with the message after it was received.
3573 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3574 "next input=" as part of the text comprising the host identity instead of
3575 the correct text. The code was using the same buffer for two different
3576 strings. However, depending on which order the printing function evaluated
3577 its arguments, the bug did not always show up. Under Linux, for example, my
3578 test suite worked just fine.
3580 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3581 doesn't work with some older versions of Perl. It has been changed to "my",
3582 which in any case is probably the better facility to use.
3584 15. A really picky compiler found some instances of statements for creating
3585 error messages that either had too many or two few arguments for the format
3588 16. The size of the buffer for calls to the DNS resolver has been increased
3589 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3590 for addresses that have a lot of PTR records. This alleviates a problem; it
3591 does not fully solve it.
3593 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3594 buffer now truncates the list and logs the incident, which is the same
3595 action as happens when Exim is looking up a host name and its aliases.
3596 Previously in this situation something unpredictable would happen;
3597 sometimes it was "internal error: store_reset failed".
3599 18. If a server dropped the connection unexpectedly when an Exim client was
3600 using GnuTLS and trying to read a response, the client delivery process
3601 crashed while trying to generate an error log message.
3603 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3604 string, for example:
3606 warn message = H1: something\nH2: something
3608 the text was added as a single header line from Exim's point of view
3609 though it ended up OK in the delivered message. However, searching for the
3610 second and subsequent header lines using $h_h2: did not work. This has been
3611 fixed. Similarly, if a system filter added multiple headers in this way,
3612 the routers could not see them.
3614 20. Expanded the error message when iplsearch is called with an invalid key to
3615 suggest using net-iplsearch in a host list.
3617 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3618 ACLs are no longer actually imposed (and a message to that effect is
3621 22. If a "gecos" field in a passwd entry contained escaped characters, in
3622 particular, if it contained a \" sequence, Exim got it wrong when building
3623 a From: or a Sender: header from that name. A second bug also caused
3624 incorrect handling when an unquoted " was present following a character
3625 that needed quoting.
3627 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3628 was not being matched caselessly.
3630 24. Arranged for all hyphens in the exim.8 source to be escaped with
3633 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3634 properly. Recipient callout cache records were still being keyed to include
3635 the sender, even when use_sender was set false. This led to far more
3636 callouts that were necessary. The sender is no longer included in the key
3637 when use_sender is false.
3639 26. Added "control = submission" modifier to ACLs.
3641 27. Added the ${base62d: operator to decode base 62 numbers.
3643 28. dnsdb lookups can now access SRV records.
3645 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3646 the configuration file.
3648 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3649 (-v) mode. This makes the output for a verbose queue run more intelligible.
3651 31. Added a use_postmaster feature to recipient callouts.
3653 32. Added the $body_zerocount variable, containing the number of binary zero
3654 bytes in the message body.
3656 33. The time of last modification of the "new" subdirectory is now used as the
3657 "mailbox time last read" when there is a quota error for a maildir
3660 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3662 35. Added +ignore_unknown as a special item in host lists.
3664 36. Code for decoding IPv6 addresses in host lists is now included, even if
3665 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3666 address was recognized as an IP address, but was then not correctly decoded
3667 into binary, causing unexpected and incorrect effects when compared with
3674 1. Very minor rewording of debugging text in manualroute to say "list of
3675 hosts" instead of "hostlist".
3677 2. If verify=header_syntax was set, and a header line with an unqualified
3678 address (no domain) and a large number of spaces between the end of the
3679 name and the colon was received, the reception process suffered a buffer
3680 overflow, and (when I tested it) crashed. This was caused by some obsolete
3681 code that should have been removed. The fix is to remove it!
3683 3. When running in the test harness, delay a bit after writing a bounce
3684 message to get a bit more predictability in the log output.
3686 4. Added a call to search_tidyup() just before forking a reception process. In
3687 theory, someone could use a lookup in the expansion of smtp_accept_max_
3688 per_host which, without the tidyup, could leave open a database connection.
3690 5. Added the variables $recipient_data and $sender_data which get set from a
3691 lookup success in an ACL "recipients" or "senders" condition, or a router
3692 "senders" option, similar to $domain_data and $local_part_data.
3694 6. Moved the writing of debug_print from before to after the "senders" test
3697 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3698 problems for message scanning, either using a data ACL, or using
3699 local_scan() because the Received: header was not generated till after they
3700 were called (in order to set the time as the time of reception completion).
3701 I have revised the way this works. The header is now generated after the
3702 body is received, but before the ACL or local_scan() are called. After they
3703 are run, the timestamp in the header is updated.
3709 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3710 before starting a queue runner without re-exec. This happened only when
3711 deliver_drop_privilege was set or when the Exim user was set to root. The
3712 effect of the bug was that timeouts during subsequent deliveries caused
3713 crashes instead of being properly handled. The handler is now left at its
3714 default (and expected) setting.
3716 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3717 message, again when deliver_drop_privilege is set or Exim is run as root.
3718 The bug described in (1) was not present in this case, but the tidying up
3719 of the other signals was missing. I have made the two cases consistent.
3721 3. The ignore_target_hosts setting on a manualroute router was being ignored
3722 for hosts that were looked up using the /MX notation.
3724 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3727 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3728 operated on the sender address. After changing the $sender_address to <>
3729 for the sender address verify, Exim was re-instated it as the original
3730 (before rewriting) address, but remembering that it had rewritten it, so it
3731 wasn't rewriting it again. This bug also had the effect of breaking the
3732 sender address verification caching when the sender address was rewritten.
3734 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3735 This has been changed so that if the ip literal address matches
3736 ignore_target_hosts, the router declines.
3738 7. Added expansion conditions match_domain, match_address, and match_local_
3739 part (NOT match_host).
3741 8. The placeholder for the Received: header didn't have a length field set.
3743 9. Added code to Exim itself and to exim_lock to test for a specific race
3744 condition that could lead to file corruption when using MBX delivery. The
3745 issue is with the lockfile that is created in /tmp. If this file is removed
3746 after a process has opened it but before that process has acquired a lock,
3747 there is the potential for a second process to recreate the file and also
3748 acquire a lock. This could lead to two Exim processes writing to the file
3749 at the same time. The added code performs the same test as UW imapd; it
3750 checks after acquiring the lock that its file descriptor still refers to
3751 the same named file.
3753 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3754 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3757 11. Added the smtp_active_hostname option. If used, this will typically be made
3758 to depend on the incoming interface address. Because $interface_address is
3759 not set up until the daemon has forked a reception process, error responses
3760 that can happen earlier (such as "too many connections") no longer contain
3763 12. If an expansion in a condition on a "warn" statement fails because a lookup
3764 defers, the "warn" statement is abandoned, and the next ACL statement is
3765 processed. Previously this caused the whole ACL to be aborted.
3767 13. Added the iplsearch lookup type.
3769 14. Added ident_timeout as a log selector.
3771 15. Added tls_certificate_verified as a log selector.
3773 16. Added a global option tls_require_ciphers (compare the smtp transport
3774 option of the same name). This controls incoming TLS connections.
3776 17. I finally figured out how to make tls_require_ciphers do a similar thing
3777 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3778 before starting the TLS session.
3780 18. Tabs are now shown as \t in -bP output.
3782 19. If the log selector return_path_on_delivery was set, Exim crashed when
3783 bouncing a message because it had too many Received: header lines.
3785 20. If two routers both had headers_remove settings, and the first one included
3786 a superfluous trailing colon, the final name in the first list and the
3787 first name in the second list were incorrectly joined into one item (with a
3788 colon in the middle).
3794 1. Added -C and -D options to the exinext utility, mainly to make it easier
3795 to include in the automated testing, but these could be helpful when
3796 multiple configurations are in use.
3798 2. The exinext utility was not formatting the output nicely when there was
3799 an alternate port involved in the retry record key, nor when there was a
3800 message id as well (for retries that were specific to a specific message
3801 and a specific host). It was also confused by IPv6 addresses, because of
3802 the additional colons they contain. I have fixed the IPv4 problem, and
3803 patched it up to do a reasonable job for IPv6.
3805 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3806 delivery, the log line now contains "pipelined" if PIPELINING was used.
3808 4. An SMTP transport process used to panic and die if the bind() call to set
3809 an explicit outgoing interface failed. This has been changed; it is now
3810 treated in the same way as a connect() failure.
3812 5. A reference to $sender_host_name in the part of a conditional expansion
3813 that was being skipped was still causing a DNS lookup. This no longer
3816 6. The def: expansion condition was not recognizing references to header lines
3817 that used bh_ and bheader_.
3819 7. Added the _cache feature to named lists.
3821 8. The code for checking quota_filecount in the appendfile transport was
3822 allowing one more file than it should have been.
3824 9. For compatibility with Sendmail, the command line option
3832 and sets the incoming protocol and host name (for trusted callers). The
3833 host name and its colon can be omitted when only the protocol is to be set.
3834 Note the Exim already has two private options, -pd and -ps, that refer to
3835 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3836 "s", but I don't think that's a major issue.
3838 10. A number of refactoring changes to the code, none of which should affect
3841 (a) The number of logging options was getting close to filling up the
3842 32-bit word that was used as a bit map. I have split them into two classes:
3843 those that are passed in the argument to log_write(), and those that are
3844 only ever tested independently outside of that function. These are now in
3845 separate 32-bit words, so there is plenty of room for expansion again.
3846 There is no change in the user interface or the logging behaviour.
3848 (b) When building, for example, log lines, the code previously used a
3849 macro that called string_cat() twice, in order to add two strings. This is
3850 not really sufficiently general. Furthermore, there was one instance where
3851 it was actually wrong because one of the argument was used twice, and in
3852 one call a function was used. (As it happened, calling the function twice
3853 did not affect the overall behaviour.) The macro has been replaced by a
3854 function that can join an arbitrary number of extra strings onto a growing
3857 (c) The code for expansion conditions now uses a table and a binary chop
3858 instead of a serial search (which was left over from when there were very
3859 few conditions). Also, it now recognizes conditions like "pam" even when
3860 the relevant support is not compiled in: a suitably worded error message is
3861 given if an attempt is made to use such a condition.
3863 11. Added ${time_interval:xxxxx}.
3865 12. A bug was causing one of the ddress fields not to be passed back correctly
3866 from remote delivery subprocesses. The field in question was not being
3867 subsequently used, so this caused to problems in practice.
3869 13. Added new log selectors queue_time and deliver_time.
3871 14. Might have fixed a bug in maildirsizefile handling that threw up
3872 "unexpected character" debug warnings, and recalculated the data
3873 unnecessarily. In any case, I expanded the warning message to give more
3876 15. Added the message "Restricted characters in address" to the statements in
3877 the default ACL that block characters like @ and % in local parts.
3879 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3880 Three changes have been made:
3882 (a) There was a serious bug; a negative response to MAIL caused the whole
3883 recipient domain to be cached as invalid, thereby blocking all messages
3884 to all local parts at the same domain, from all senders. This bug has
3885 been fixed. The domain is no longer cached after a negative response to
3886 MAIL if the sender used is not empty.
3888 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3891 (c) A new callout option, "use_sender" has been added for people who want
3892 the modified behaviour.
3898 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3901 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3902 indeed breaks things for older releases.
3904 3. Added additional logging to the case where there is a problem reading data
3905 from a filter that is running in a subprocess using a pipe, in order to
3906 try to track down a specific problem.
3908 4. Testing facility fudge: when running in the test harness and attempting
3909 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3910 getting "No route to host". Convert this to a timeout.
3912 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3915 6. Some OS don't have socklen_t but use size_t instead. This affects the
3916 fifth argument of getsockopt() amongst other things. This is now
3917 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3918 can be set for individual OS. I have set it for SunOS5, OSF1, and
3919 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3920 some earlier ones do not.
3922 7. Change 4.30/15 was not doing the test caselessly.
3924 8. The standard form for an IPv6 address literal was being rejected by address
3925 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3926 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3927 this, as well as the form without the "IPv6" on the front (but only when
3928 address literals are enabled, of course).
3930 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3932 10. Exim crashed if a message with an empty sender address specified by -f
3933 encountered a router with an errors_to setting. This could be provoked only
3934 by a command such as
3938 where an empty string was supplied; "<>" did not hit this bug.
3940 11. Installed PCRE release 4.5.
3942 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3943 remained set. It is now erased.
3945 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3946 times from message ids (which are base 36 rather than the normal 62).
3948 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3949 were being counted as actual protocol errors, and logged if the log
3950 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3951 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3952 and DATA following a set of rejected RCPTs do not count as protocol errors.
3953 In other words, Exim assumes they were pipelined, though this may not
3954 actually be the case. Of course, in all cases the client gets an
3955 appropriate error code.
3957 15. If a lookup fails in an ACL condition, a message about the failure may
3958 be available; it is used if testing the ACL cannot continue, because most
3959 such messages specify what the cause of the deferral is. However, some
3960 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3961 that caused an old message to be retained and used if a later statement
3962 caused a defer, replacing the real cause of the deferral.
3964 16. If an IP address had so many PTR records that the DNS lookup buffer
3965 was not large enough to hold them, Exim could crash while trying to process
3966 the truncated data. It now detects and logs this case.
3968 17. Further to 4.21/58, another change has been made: if (and only if) the
3969 first line of a message (the first header line) ends with CRLF, a bare LF
3970 in a subsequent header line has a space inserted after it, so as not to
3971 terminate the header.
3973 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3974 unnecessarily, used atoi() instead of strtol(), and didn't check the
3975 termination when getting file sizes from file names by regex.
3977 19. Completely re-implemented the support for maildirsize files, in the light
3978 of a number of problems with the previous contributed implementation
3979 (4.30/29). In particular:
3981 . If the quota is zero, the maildirsize file is maintained, but no quota is
3984 . If the maildir directory does not exist, it is created before any attempt
3985 to write a maildirsize file.
3987 . The quota value in the file is just a cache; if the quota is changed in
3988 the transport, the new value overrides.
3990 . A regular expression is available for excluding directories from the
3993 20. The autoreply transport checks the characters in options that define the
3994 message's headers; it allows continued headers, but it was checking with
3995 isspace() after an embedded newline instead of explicitly looking for a
3998 21. If all the "regular" hosts to which an address was routed had passed their
3999 expiry times, and had not reached their retry times, the address was
4000 bounced, even if fallback hosts were defined. Now Exim should go on to try
4003 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4004 equivalent code in the SMTP transport. Some hosts send humungous responses
4005 to HELO/EHLO, more than 1024 it seems.
4007 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4008 gives compiler warnings in some environments. I've now done it "properly",
4011 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4012 (because of problems with the built-in one) was declared to return uschar *
4013 instead of char *, causing compiler failure.
4015 25. Fixed a file descriptor leak when processing alias/forward files.
4017 26. Fixed a minor format string issue in dbfn.c.
4019 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4021 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4022 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4025 29. When a sender address is verified, it is cached, to save repeating the test
4026 when there is more than one recipient in a message. However, when the
4027 verification involves a callout, it is possible for different callout
4028 options to be set for different recipients. It is too complicated to keep
4029 track of this in the cache, so now Exim always runs a verification when a
4030 callout is required, relying on the callout cache for the optimization.
4031 The overhead is duplication of the address routing, but this should not be
4034 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4035 to be verified with callout=postmaster, and the main callout worked but the
4036 postmaster check failed, the verification correctly failed. However, if a
4037 subsequent RCPT command asked for sender verification *without* the
4038 postmaster check, incorrect caching caused this verification also to fail,
4041 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4042 it was not caching the DNS options (qualify_single, search_parents) that
4043 were used when the lookup failed. A subsequent lookup with different
4044 options therefore always gave the same answer, though there were cases
4045 where it should not have. (Example: a "domains = !$mx_any" option on a
4046 dnslookup router: the "domains" option is always processed without any
4047 widening, but the router might have qualify_single set.) Now Exim uses the
4048 cached value only when the same options are set.
4050 32. Added John Jetmore's "exipick" utility to the distribution.
4052 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4053 than a timeout (e.g. a certificate is required, and is not provided), an
4054 Exim server now closes the connection immediately. Previously it waited for
4055 the client to close - but if the client is SSL, it seems that they each
4056 wait for each other, leading to a delay before one of them times out.
4058 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4059 maintained 0.8.x compatibility because I don't think many are using it, and
4060 it is clearly obsolete.
4062 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4065 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4066 client certificate was expired. A simple patch fixes this, though I don't
4067 understand the full logic of why the verify callback is called multiple
4070 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4071 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4072 which causes problems with some clients (such as the Certicom SSL Plus
4073 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4074 disables the coutermeasure allowing Eudora to connect."
4076 38. Exim was not checking that a write() to a log file succeeded. This could
4077 lead to Bad Things if a log got too big, in particular if it hit a file
4078 size limit. Exim now panics and dies if it cannot write to a log file, just
4079 as it does if it cannot open a log file.
4081 39. Modified OS/Makefile-Linux so that it now contains
4083 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4085 The two -D definitions ensure that Exim is compiled with large file
4086 support, which makes it possible to handle log files that are bigger than
4089 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4090 instance) a domain was checked against a named list that involved a lookup,
4091 causing $domain_data to be set, then another domain was checked against the
4092 same list, then the first domain was re-checked, the value of $domain_data
4093 after the final check could be wrong. In particular, if the second check
4094 failed, it could be set empty. This bug probably also applied to
4097 41. The strip_trailing_dot option was not being applied to the address given
4098 with the -f command-line option.
4100 42. The code for reading a message's header from the spool was incrementing
4101 $received_count, but never initializing it. This meant that the value was
4102 incorrect (doubled) while delivering a message in the same process in which
4103 it was received. In the most common configuration of Exim, this never
4104 happens - a fresh exec is done - but it can happen when
4105 deliver_drop_privilege is set.
4107 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4108 it now includes up to 150 characters of the unexpected data in the log
4111 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4112 and building data strings. The size of both of these buffers was 10 000
4113 bytes - far larger than anybody would *ever* want, thought I. Needless to
4114 say, somebody hit the limit. I have increased the maximum line length to
4115 20 000 and the maximum data length of concatenated lines to 100 000. I have
4116 also fixed two bugs, because there was no checking on these buffers. Tsk,
4117 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4118 buffer is too small.
4120 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4121 lsearch lookups. Now it does.
4123 46. When parsing a route_list item in a manualroute router, a fixed-length
4124 buffer was used for the list of hosts. I made this 1024 bytes long,
4125 thinking that nobody would ever have a list of hosts that long. Wrong.
4126 Somebody had a whole pile of complicated expansion conditions, and the
4127 string was silently truncated, leading to an expansion error. It turns out
4128 that it is easier to change to an unlimited length (owing to other changes
4129 that have happened since this code was originally written) than to build
4130 structure for giving a limitation error. The length of the item that
4131 expands into the list of hosts is now unlimited.
4133 47. The lsearch lookup could not handle data where the length of text line was
4134 more than 4095 characters. Such lines were truncated, leading to shortened
4135 data being returned. It should now handle lines of any length.
4137 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4138 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4141 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4142 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4143 no longer happens in comment lines. A list of replacements is now placed
4144 at the head of all of the source files, except those whose only change is
4145 to replace PERL_COMMAND in the very first #! line.
4147 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4148 messages on the queue, with a bottom-up merge sort, using code contributed
4149 by Michael Haardt. This should make operations like -bp somewhat faster on
4150 large queues. It won't affect queue runners, except when queue_run_in_order
4153 51. Installed eximstats 1.31 in the distribution.
4155 52. Added support for SRV lookups to the dnslookup router.
4157 53. If an ACL referred to $message_body or $message_body_end, the value was not
4158 reset for any messages that followed in the same SMTP session.
4160 54. The store-handling optimization for building very long strings was not
4161 differentiating between the different store pools. I don't think this
4162 actually made any difference in practice, but I've tidied it.
4164 55. While running the routers to verify a sender address, $sender_address
4165 was still set to the sender address. This is wrong, because when routing to
4166 send a bounce to the sender, it would be empty. Therefore, I have changed
4167 it so that, while verifying a sender address, $sender_address is set to <>.
4168 (There is no change to what happens when verifying a recipient address.)
4170 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4171 target A or AAAA records (if not already returned) without resetting the
4172 qualify_single or search_parents options of the DNS resolver. These are
4173 inappropriate in this case because the targets of MX and SRV records must
4174 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4175 target that, when qualified, matched something in the local domain. These
4176 two options are now turned off when doing these lookups.
4178 57. It seems that at least some releases of Reiserfs (which does not have the
4179 concept of a fixed number of inodes) returns zero and not -1 for the
4180 number of available inodes. This interacted badly with check_spool_inodes,
4181 which assumed that -1 was the "no such thing" setting. What I have done is
4182 to check that the total number of inodes is greater than zero before doing
4183 the test of how many are available.
4185 58. When a "warn" ACL statement has a log_message modifier, the message is
4186 remembered, and not repeated. This is to avoid a lot of repetition when a
4187 message has many recipients that cause the same warning to be written.
4188 Howewer, Exim was preserving the list of already written lines for an
4189 entire SMTP session, which doesn't seem right. The memory is now reset if a
4190 new message is started.
4192 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4193 debugging output unless log_rewrite was also set.
4195 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4196 of (int)(handle) when we know that handle contains (void *)(-1).
4198 61. The Exim daemon panic-logs an error return when it closes the incoming
4199 connection. However "connection reset by peer" seems to be common, and
4200 isn't really an error worthy of noting specially, so that particular error
4203 62. When Exim is trying to find all the local interfaces, it used to panic and
4204 die if the ioctl to get the interface flags failed. However, it seems that
4205 on at least one OS (Solaris 9) it is possible to have an interface that is
4206 included in the list of interfaces, but for which you get a failure error
4207 for this call. This happens when the interface is not "plumbed" into a
4208 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4209 failure of the "get flags" call assumes that the interface is down.
4211 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4212 makes life easier for people who are doing arithmetic on fields extracted
4213 from dates, where you often get leading zeros that should not be
4214 interpreted as octal.
4216 64. Added qualify_domain to the redirect router, to override the global
4219 65. If a pathologically long header line contained very many addresses (the
4220 report of this problem mentioned 10 000) and each of them was rewritten,
4221 Exim could use up a very large amount of memory. (It kept on making new
4222 copies of the header line as it rewrote, and never released the old ones.)
4223 At the expense of a bit more processing, the header rewriting function has
4224 been changed so that it no longer eats memory in this way.
4226 66. The generation of the Received: header has been moved from the time that a
4227 message starts to be received, to the time that it finishes. The timestamp
4228 in the Received: header should now be very close to that of the <= log
4229 line. There are two side-effects of this change:
4231 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4232 logged header lines no longer include the local Received: line, because
4233 it has not yet been created. The same applies to a copy of the message
4234 that is returned to a non-SMTP sender when a message is rejected.
4236 (b) When a filter file is tested using -bf, no additional Received: header
4237 is added to the test message. After some thought, I decided that this
4240 This change does not affect the value of $received_for. It is still set
4241 after address rewriting, but before local_scan() is called.
4243 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4245 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4246 gave an unhelpful panic error message, and a defer error. I have managed to
4247 change this behaviour so that it now rejects any supplied certificate,
4248 which seems right, as the list of acceptable certificates is empty.
4250 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4251 gave an unhelpful defer error. I have not managed to make this reject any
4252 supplied certificates, but the error message it gives is "no certificate
4253 supplied", which is not helpful.
4255 70. exigrep's output now also includes lines that are not associated with any
4256 message, but which match the given pattern. Implemented by a patch from
4257 Martin Sluka, which also tidied up the Perl a bit.
4259 71. Recipient callout verification, like sender verification, was using <> in
4260 the MAIL FROM command. This isn't really the right thing, since the actual
4261 sender may affect whether the remote host accepts the recipient or not. I
4262 have changed it to use the actual sender in the callout; this means that
4263 the cache record is now keyed on a recipient/sender pair, not just the
4264 recipient address. There doesn't seem to be a real danger of callout loops,
4265 since a callout by the remote host to check the sender would use <>.
4266 [SEE ABOVE: changed after hitting problems.]
4268 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4269 temporary errors. However, in the case of such a code being given after
4270 the end of a data transmission (i.e. after ".") Exim was failing to write
4271 a retry record for the message. (Yes, there was some broken host that was
4272 actually sending 8xx at this point.)
4274 73. An unknown lookup type in a host list could cause Exim to panic-die when
4275 the list was checked. (An example that provoked this was putting <; in the
4276 middle of a list instead of at the start.) If this happened during a DATA
4277 ACL check, a -D file could be left lying around. This kind of configuration
4278 error no longer causes Exim to die; instead it causes a defer errror. The
4279 incident is still logged to the main and panic logs.
4281 74. Buglet left over from Exim 3 conversion. The message "too many messages
4282 in one connection" was written to the rejectlog but not the mainlog, except
4283 when address rewriting (yes!) was being logged.
4285 75. Added write_rejectlog option.
4287 76. When a system filter was run not as root (that is, when system_filter_user
4288 was set), the values of the $n variables were not being returned to the
4289 main process; thus, they were not subsequently available in the $sn
4292 77. Added +return_path_on_delivery log selector.
4294 78. A connection timeout was being treated differently from recipients deferred
4295 when testing hosts_max_try with a message that was older than the host's
4296 retry timeout. (The host should not be counted, thus allowing all hosts to
4297 be tried at least once before bouncing.) This may have been the cause of an
4298 occasionally reported bug whereby a message would remain on the queue
4299 longer than the retry timeout, but would be bounced if a delivery was
4300 forced. I say "may" because I never totally pinned down the problem;
4301 setting up timeout/retry tests is difficult. See also the next item.
4303 79. The ultimate address timeout was not being applied to errors that involved
4304 a combination of host plus message (for example, a timeout on a MAIL
4305 command). When an address resolved to a number of possible hosts, and they
4306 were not all tried for each delivery (e.g. because of hosts_max_try), a
4307 message could remain on the queue longer than the retry timeout.
4309 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4312 81. Fixed an obscure SMTP outgoing bug which required at least the following
4313 conditions: (a) there was another message waiting for the same server;
4314 (b) the server returned 5xx to all RCPT commands in the first message so
4315 that the message was not completed; (c) the server dropped the connection
4316 or gave a negative response to the RSET that Exim sends to abort the
4317 transaction. The observed case was a dropped connection after DATA that had
4318 been sent in pipelining mode. That is, the server had advertised PIPELINING
4319 but was not implementing it correctly. The effect of the bug was incorrect
4320 behaviour, such as trying another host, and this could lead to a crash.
4326 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4327 and daemon.c were passed as pointers to ints; they should have been
4328 pointers to socklen_t variables (which are typically unsigned ints).
4330 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4333 3. Fixed a really odd bug that affected only the testing scheme; patching a
4334 certain fixed string in the binary changed the value of another string that
4335 happened to be identical to the end of the original first string.
4337 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4338 name", it returns that address as the IP address. On some operating
4339 systems (e.g. Solaris), it also passes back the IP address string as the
4340 "host name". However, on others (e.g. Linux), it passes back an empty
4341 string. Exim wasn't checking for this, and was changing the host name to an
4342 empty string, assuming it had been canonicized.
4344 5. Although rare, it is permitted to have more than one PTR record for a given
4345 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4346 all the names associated with an address, because they do in Solaris.
4347 However, it seems that they do not in Linux for data that comes from the
4348 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4349 I found this out when I moved to a new Linux workstation and tried to run
4350 the Exim test suite.
4352 To get round this problem I have changed the code so that it now does its
4353 own call to the DNS to look up PTR records when searching for a host name.
4354 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4355 addresses that are only in /etc/hosts are still found.
4357 This behaviour is, however, controlled by an option called host_lookup_
4358 order, which defaults to "bydns:byaddr". If people want to use the other
4359 order, or indeed, just use one or the other means of lookup, they can
4360 specify it in this variable.
4362 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4363 some operating systems, this comes back from gethostbyaddr() as an empty
4364 string, and this is what Exim used to test for. However, it seems that in
4365 other systems, "." is yielded. Exim now tests for this case too.
4367 7. The values of check_spool_space and check_log_space are now held internally
4368 as a number of kilobytes instead of an absolute number of bytes. If a
4369 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4370 kilobyte. This means that much larger values can be stored.
4372 8. Exim monitor: an attempt to get the action menu when not actually pointing
4373 at a message produces an empty menu entitled "No message selected". This
4374 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4375 no entries in it ("Shell widget menu has zero width and/or height"). So I
4376 have added a single, blank menu entry in this case.
4378 9. Added ${quote_local_part.
4380 10. MIME decoding is now applied to the contents of Subject: header lines when
4383 11. Now that a reference to $sender_host_address automatically causes a reverse
4384 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4385 host lookup before query-style lookups in lists that might use this
4386 variable. This has therefore been abolished, and the "net-" prefix is no
4387 longer necessary for query-style lookups.
4389 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4390 have been a typo for LFLAGS, so it has been changed.
4392 13. The install script calls Exim with "-C /dev/null" in order to find the
4393 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4394 to be output. Howeve, since Exim outputs its version number before the
4395 error, it didn't break the script. It just looked ugly. I fixed this by
4396 always allowing "-C /dev/null" if the caller is root.
4398 14. Ignore overlarge ACL variable number when reading spool file - insurance
4399 against a later release with more variables having written the file.
4401 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4402 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4403 this, as well as the form without the "IPv6" on the front.
4405 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4406 OS/Makefile-Darwin file.
4408 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4409 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4411 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4412 variable. However, some OS use ints for inodes. I've added cast to long int
4413 to get rid of the compiler warning.
4415 19. I had forgotten to lock out "/../" in configuration file names when
4416 ALT_CONFIG_PREFIX was set.
4418 20. Routers used for verification do not need to specify transports. However,
4419 if such a router generated a host list, and callout was configured, Exim
4420 crashed, because it could not find a port number from the (non-existent)
4421 transport. It now assumes port 25 in this circumstance.
4423 21. Added the -t option to exigrep.
4425 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4426 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4427 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4428 them is set, LOOKUP_LSEARCH is forced.
4430 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4432 24. Added sender and host information to the "rejected by local_scan()" log
4433 line; previously there was no indication of these.
4435 25. Added .include_if_exists.
4437 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4438 Linux. It turns out that not all file systems support this. Apparently some
4439 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4440 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4441 not supported on the file descriptor, is now ignored when Exim is trying to
4442 sync a directory. This applies only to Linux.
4444 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4446 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4447 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4450 29. Added support for maildirsize files from supplied patch (modified a bit).
4452 30. The use of :fail: followed by an empty string could lead Exim to respond to
4453 sender verification failures with (e.g.):
4455 550 Verification failed for <xxx>
4456 550 Sender verify failed
4458 where the first response line was missing the '-' that indicates it is not
4459 the final line of the response.
4461 31. The loop for finding the name of the user that called Exim had a hardwired
4462 limit of 10; it now uses the value of finduser_retries, which is used for
4463 all other user lookups.
4465 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4468 33. Exim was neglecting to zero errno before one call of strtol() when
4469 expanding a string and expecting an integer value. On some systems this
4470 resulted in spurious "integer overflow" errors. Also, it was casting the
4471 result into an int without checking.
4473 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4474 did not work. The code looks as if it has *never* worked, though it appears
4475 to have been documented since at least releast 1.62. I have made it work.
4477 35. The "timeout_DNS" error in retry rules, also documented since at least
4478 1.62, also never worked. As it isn't clear exactly what this means, and
4479 clearly it isn't a major issue, I have abolished the feature by treating it
4480 as "timeout", and writing a warning to the main and panic logs.
4482 36. The display of retry rules for -brt wasn't always showing the error code
4485 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4486 timeout_connect_A, timeout_connect_MX.
4488 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4489 to the empty sender.
4491 39. The daemon was not analysing the content of -oX till after it had closed
4492 stderr and disconnected from the controlling terminal. This meant that any
4493 syntax errors were only noted on the panic log, and the return code from
4494 the command was 0. By re-arranging the code a little, I've made the
4495 decoding happen first, so such errors now appear on stderr, and the return
4496 code is 1. However, the actual setting up of the sockets still happens in
4497 the disconnected process, so errors there are still only recorded on the
4500 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4501 connections (as happens on some IP stacks) was logged at start up time as
4502 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4503 it from "IPv6 and IPv4", which means that two separate sockets are being
4506 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4507 says whether AF_INET or AF_INET6 was passed as an argument.
4509 42. Exiwhat output was messed up when time zones were included in log
4512 43. Exiwhat now gives more information about the daemon's listening ports,
4513 and whether -tls-on-connect was used.
4515 44. The "port" option of the smtp transport is now expanded.
4517 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4518 silently ignored. Now an error message is written to the main and panic
4521 46. There's a new ACL modifier called "logwrite" which writes to a log file
4522 as soon as it is encountered.
4524 47. Added $local_user_uid and $local_user_gid at routing time.
4526 48. Exim crashed when trying to verify a sender address that was being
4529 49. Exim was recognizing only a space character after ".include". It now also
4530 recognizes a tab character.
4532 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4533 extracting the relevant information from the specification. The man page no
4534 longer contains scrambled data for the -d option, and I've added a section
4535 at the front about calling Exim under different names.
4537 51. Added "extra_headers" argument to the "mail" command in filter files.
4539 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4542 53. Installed eximstats 1.29.
4544 54. Added transport_filter_timeout as a generic transport option.
4546 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4547 Cc: header lines. This was required by RFC 822, but it not required by RFC
4550 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4551 incoming messages that did not have them. Now it does so only if the
4552 message originates locally, that is, if there is no associated remote host
4553 address. When Resent- header lines are present, this applies to the Resent-
4554 lines rather than the non-Resent- lines.
4556 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4557 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4559 58. Messages for configuration errors now include the name of the main
4560 configuration file - useful now that there may be more than one file in a
4561 list (.included file names were always shown).
4563 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4564 for those rare installations that do not start the daemon as root or run it
4565 setuid root. I've cut out the call to initgroups() if the daemon is not
4568 60. The Exim user and group can now be bound into the binary as text strings
4569 that are looked up at the start of Exim's processing.
4571 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4573 62. Added $mailstore_basename variable.
4575 63. Installed patch to sieve.c from Michael Haardt.
4577 64. When Exim failed to open the panic log after failing to open the main log,
4578 the original message it was trying to log was written to stderr and debug
4579 output, but if they were not available (the usual case in production), it
4580 was lost. Now it is written to syslog before the two lines that record the
4581 failures to open the logs.
4583 65. Users' Exim filters run in subprocesses under the user's uid. It is
4584 possible for a "deliver" command or an alias in a "personal" command to
4585 provoke an address rewrite. If logging of address rewriting is configured,
4586 this fails because the process is not running as root or exim. There may be
4587 a better way of dealing with this, but for the moment (because 4.30 needs
4588 to be released), I have disabled address rewrite logging when running a
4589 filter in a non-root, non-exim process.
4595 1. The buildconfig auxiliary program wasn't quoting the value set for
4596 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4597 not defined. This bug was masked in 4.22 by the effect that was fixed in
4600 2. Some messages that were rejected after a message id was allocated were
4601 shown as "incomplete" by exigrep. It no longer does this for messages that
4602 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4604 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4605 have allow_domain_literals set, the ID did not get logged in the <= line.
4606 Domain literals are now always recognized in Message-ID: header lines.
4608 4. The first argument for a ${extract expansion item is the key name or field
4609 number. Leading and trailing spaces in this item were not being ignored,
4610 causing some misleading effects.
4612 5. When deliver_drop_privilege was set, single queue runner processes started
4613 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4614 same command in the process it spins off) were not dropping privilege.
4616 6. When the daemon running as "exim" started a queue runner, it always
4617 re-executed Exim in the spun-off process. This is a waste of effort when
4618 deliver_drop_privilege is set. The new process now just calls the
4619 queue-runner function directly.
4625 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4628 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4629 ignored. Though the use of -oP was forcing the writing of a pid file, it
4630 was always written to the default place.
4632 3. If the message "no IP address found for host xxxx" is generated during
4633 incoming verification, it is now followed by identification of the incoming
4634 connection (so you can more easily find what provoked it).
4636 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4638 5. Added some features to "harden" Exim a bit more against certain attacks:
4640 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4641 be put in Local/Makefile. This is like the never_users runtime option,
4642 but it cannot be overridden. The default setting is "root".
4644 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4645 prefix string with which any file named in a -C command line option
4648 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4649 is retained for -C and -D only if the caller of Exim is root. Without
4650 it, the exim user may also use -C and -D and retain privilege.
4652 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4653 command line option is disabled.
4655 6. Macro names set by the -D option must start with an upper case letter, just
4656 like macro names defined in the configuration file.
4658 7. Added "dereference=" facility to LDAP.
4660 8. Two instances of the typo "uknown" in the source files are fixed.
4662 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4663 the Configure-Makefile script screwed up while processing it.
4665 10. Incorporated PCRE 4.4.
4667 11. The SMTP synchronization check was not operating right at the start of an
4668 SMTP session. For example, it could not catch a HELO sent before the client
4669 waited for the greeting. There is now a check for outstanding input at the
4670 point when the greeting is written. Because of the duplex, asynchronous
4671 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4672 way, but not yet received, when the check is performed.
4674 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4675 on TCP/IP sockets, because this apparently causes some broken clients to
4678 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4679 unchanged) from the Cygwin maintainer.
4681 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4682 was supported instead of testing for mbx. Effectively a typo.
4684 15. The spa authenticator server code was not checking that the input it
4685 received was valid base64.
4687 16. The debug output line for the "set" modifier in ACLs was not showing the
4688 name of the variable that was being set.
4690 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4692 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4693 lookup on its own. Something else was needed to trigger the lookup. For
4694 example, a match in host_lookup or the need for a host name in a host list.
4695 Now, if $sender_host_name is referenced and the host name has not yet been
4696 looked up, a lookup is performed. If the lookup fails, the variable remains
4697 empty, and $host_lookup_failed is set to "1".
4699 19. Added "eqi" as a case-independent comparison operator.
4701 20. The saslauthd authentication condition could segfault if neither service
4702 nor realm was specified.
4704 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4705 error message that was logged was misleading, and incoming SMTP
4706 connections were dropped. The message is now more accurate, and temporary
4707 errors are given to SMTP connections.
4709 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4710 (except RSET) with a 421 error, until QUIT is received. However, it was
4711 failing to send a response to QUIT.
4713 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4714 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4715 ACL is now run after the helo_try_verify_hosts code.
4717 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4718 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4719 case-independent, so other case variants are also recognized). Apparently
4720 some systems use these upper case variants.
4722 25. If more than two messages were waiting for the same host, and a transport
4723 filter was specified for the transport, Exim sent two messages over the
4724 same TCP/IP connection, and then failed with "socket operation on non-
4725 socket" when it tried to send the third.
4727 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4729 27. The extern definition of crypt16() in expand.c was not being excluded when
4730 the OS had its own crypt16() function.
4732 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4733 as a preferred synonym for return_size_limit, both as an option and as an
4736 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4738 30. Changed the default configuration ACL to relax the local part checking rule
4739 for addresses that are not in any local domains. For these addresses,
4740 slashes and pipe symbols are allowed within local parts, but the sequence
4741 /../ is explicitly forbidden.
4743 31. SPA server authentication was not clearing the challenge buffer before
4746 32. log_message in a "warn" ACL statement was writing to the reject log as
4747 well as to the main log, which contradicts the documentation and doesn't
4748 seem right (because no rejection is happening). So I have stopped it.
4750 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4751 However, I am unable to do any testing of this.
4753 34. Fixed an infelicity in the appendfile transport. When checking directories
4754 for a mailbox, to see if any needed to be created, it was accidentally
4755 using path names with one or more superfluous leading slashes; tracing
4756 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4758 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4759 contents are added to the log line that is written for every discarded
4760 recipient. (Previously a log_message setting was ignored.)
4762 36. The ${quote: operator now quotes the string if it is empty.
4764 37. The install script runs exim in order to find its version number. If for
4765 some reason other than non-existence or emptiness, which it checks, it
4766 could not run './exim', it was installing it with an empty version number,
4767 i.e. as "exim-". This error state is now caught, and the installation is
4770 38. An argument was missing from the function that creates an error message
4771 when Exim fails to connect to the socket for saslauthd authentication.
4772 This could cause Exim to crash, or give a corrupted message.
4774 39. Added isip, isip4, and isip6 to ${if conditions.
4776 40. The ACL variables $acl_xx are now saved with the message, and can be
4777 accessed later in routers, transports, and filters.
4779 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4780 strings in the file are not string-expanded.
4782 42. If a MAIL command specified a SIZE value that was too large to fit into an
4783 int variable, the check against message_size_limit failed. Such values are
4784 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4785 day this will have to be increased, but I don't think I want to be around
4786 when emails are that large.
4793 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4794 iconv() is not standard in FreeBSD.
4796 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4797 IPv6 enabled. The observed symptom was a segmentation fault on return from
4798 the function os_common_find_running_interfaces() in src/os.c.
4800 3. In the check_special_case() function in daemon.c I had used "errno" as an
4801 argument name, which causes warnings on some systems. This was basically a
4802 typo, since it was named "eno" in the comments!
4804 4. The code that waits for the clock to tick (at a resolution of some fraction
4805 of a second) so as to ensure message-id uniqueness was always waiting for
4806 at least one whole tick, when it could have waited for less. [This is
4807 almost certainly not relevant at current processor speeds, where it is
4808 unlikely to ever wait at all. But we try to future-proof.]
4810 5. The function that sleeps for a time interval that includes fractions of a
4811 second contained a race. It did not block SIGALRM between setting the
4812 timer, and suspending (a couple of lines later). If the interval was short
4813 and the sigsuspend() was delayed until after it had expired, the suspension
4814 never ended. On busy systems this could lead to processes getting stuck for
4817 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4818 process, before it forks any delivery processes. The open lookup caching
4819 mechanism meant that the open file or database connection was passed into
4820 the delivery process. The problem was that delivery processes always tidy
4821 up cached lookup data. This could cause a problem for the next delivery
4822 process started by the queue runner, because the external queue runner
4823 process does not know about the closure. So the next delivery process
4824 still has data in the lookup cache. In the case of a file lookup, there was
4825 no problem because closing a file descriptor in a subprocess doesn't affect
4826 the parent. However, if the lookup was caching a connection to a database,
4827 the connection was closed, and the second delivery process was likely to
4828 see errors such as "PGSQL: query failed: server closed the connection
4829 unexpectedly". The problem has been fixed by closing all cached lookups
4830 in a queue runner before running a delivery process.
4832 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4833 seem to have the "const" qualifier which it has on other OS. I've
4836 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4837 *of the same type* (client or server) with the same public name that an
4838 error should be diagnosed.
4840 9. When Exim looked up a host name for an IP address, but failed to find the
4841 original IP address when looking up the host name (a safety check), it
4842 output the message "<ip address> does not match any IP for NULL", which was
4843 confusing, to say the least. The bug was that the host name should have
4844 appeared instead of "NULL".
4846 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4847 user that is built into the binary, and the -C or -D options is used, root
4848 privilege is dropped before the configuration file is read. In addition,
4849 logging is switched to stderr instead of the normal log files. If the
4850 configuration then re-defines the Exim user, the unprivileged environment
4851 is probably not what is expected, so Exim logs a panic warning message (but
4854 However, if deliver_drop_privilege is set, the unprivileged state may well
4855 be exactly what is intended, so the warning has been cut out in that case,
4856 and Exim is allowed to try to write to its normal log files.
4862 1. smtp_return_error_details was not giving details for temporary sender
4863 or receiver verification errors.
4865 2. Diagnose a configuration error if two authenticators have the same public
4868 3. Exim used not to create the message log file for a message until the first
4869 delivery attempt. This could be confusing when incoming messages were held
4870 for policy or load reasons. The message log file is now created at the time
4871 the message is received, and an initial "Received" line is written to it.
4873 4. The automatically generated man page for command line options had a minor
4874 bug that caused no ill effects; however, a more serious problem was that
4875 the procedure for building the man page automatically didn't always
4876 operate. Consequently, release 4.20 contains an out-of-date version. This
4877 shouldn't happen again.
4879 5. When building Exim with embedded Perl support, the script that builds the
4880 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4881 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4883 6. The freeze_tell option was not being used for messages that were frozen on
4884 arrival, either by an ACL or by local_scan().
4886 7. Added the smtp_incomplete_transaction log selector.
4888 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4889 was accepting AUTH without a new EHLO.
4891 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4892 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4895 10. Logging of TCP/IP connections (when configured) now happens in the main
4896 daemon process instead of the child process, so that the TCP/IP connection
4897 count is more accurate (but it can never be perfect).
4899 11. The use of "drop" in a nested ACL was not being handled correctly in the
4900 outer ACL. Now, if condition failure induced by the nested "drop" causes
4901 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4902 or "require"), the connection is dropped.
4904 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4905 that yield "discard" can now be used with an "accept" or a "discard" verb,
4906 but an error is generated for any others (because I can't see a useful way
4907 to define what should happen).
4909 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4910 are now processed in the same way as lines in the Exim configuration file.
4911 In particular, continuation lines are supported.
4913 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4915 15. Added -ti meaning -t -i.
4917 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4918 domains, and warn by logging if others are found.
4920 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4921 returned from GIFCONF when Exim is trying to find the list of interfaces on
4922 a host. The code in os.c has been modified to copy each ifreq to an aligned
4923 structure in all cases.
4925 Also, in some cases, the returned ifreq's were being copied to a 'struct
4926 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4927 means the last couple of bytes of an IPv6 address could be chopped if the
4928 ifreq contained only a normal sockaddr (14 bytes storage).
4930 18. Named domain lists were not supported in the hosts_treat_as_local option.
4931 An entry such as +xxxx was not recognized, and was treated as a literal
4934 19. Ensure that header lines added by a DATA ACL are included in the reject log
4935 if the ACL subsequently rejects the message.
4937 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4938 MD5 (which is deprecated).
4940 21. When testing a filter file using -bf, Exim was writing a message when it
4941 took the sender from a "From " line in the message, but it was not doing so
4942 when it took $return_path from a Return-Path: header line. It now does.
4944 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4945 with a valid header line field name (a series of printing characters
4946 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4948 23. Changed "disc" in the source to "disk" to conform to the documentation and
4949 the book and for uniformity.
4951 24. Ignore Sendmail's -Ooption=value command line item.
4953 25. When execve() failed while trying to run a command in a pipe transport,
4954 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4955 could be confused with a return value of 69 from the command itself. This
4956 has been changed to 127, the value the shell returns if it is asked to run
4957 a non-existent command. The wording for the related log line suggests a
4958 non-existent command as the problem.
4960 26. If received_header_text expands to an empty string, do not add a Received:
4961 header line to the message. (Well, it adds a token one on the spool, but
4962 marks it "old" so that it doesn't get used or transmitted.)
4964 27. Installed eximstats 1.28 (addition of -nt option).
4966 28. There was no check for failure on the call to getsockname() in the daemon
4967 code. This can fail if there is a shortage of resources on the system, with
4968 ENOMEM, for example. A temporary error is now given on failure.
4970 29. Contrary to the C standard, it seems that in some environments, the
4971 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4972 program. Exim now does this explicitly; it affects the formatting of
4973 timestamps using strftime().
4975 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4976 complaints. I've now initialized all the variables, to avoid this.
4978 32. Header lines added by a system filter were not being "seen" during
4979 transport-time rewrites.
4981 33. The info_callback() function passed to OpenSSL is set up with type void
4982 (*)(SSL *, int, int), as described somewhere. However, when calling the
4983 function (actually a macro) that sets it up, the type void(*)() is
4984 expected. I've put in a cast to prevent warnings from picky compilers.
4986 34. If a DNS black list lookup found a CNAME record, but there were no A
4987 records associated with the domain it pointed at, Exim crashed.
4989 35. If a DNS black list lookup returned more than one A record, Exim ignored
4990 all but the first. It now scans all returned addresses if a particular IP
4991 value is being sought. In this situation, the contents of the
4992 $dnslist_value variable are a list of all the addresses, separated by a
4995 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4996 to accept a host name and all its aliases if the forward lookup for any of
4997 them yielded the IP address of the incoming connection. Now it accepts only
4998 those names whose forward lookup yields the correct IP address. Any other
4999 names are discarded. This closes a loophole whereby a rogue DNS
5000 administrator could create reverse DNS records to break through a
5001 wildcarded host restriction in an ACL.
5003 37. If a user filter or a system filter that ran in a subprocess used any of
5004 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5005 the wrong values were passed to the pipe command ($thisaddress had the
5006 value of $0, $0 had the value of $1, etc). This bug was introduced by
5007 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5009 38. Improved the line breaking for long SMTP error messages from ACLs.
5010 Previously, if there was no break point between 40 and 75 characters, Exim
5011 left the rest of the message alone. Two changes have been made: (a) I've
5012 reduced the minimum length to 35 characters; (b) if it can't find a break
5013 point between 35 and 75 characters, it looks ahead and uses the first one
5014 that it finds. This may give the occasional overlong line, but at least the
5015 remaining text gets split now.
5017 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5018 file descriptors might be low, and that setting 1000 would always raise it.
5019 It turns out that in some environments, the limit is already over 1000 and
5020 that lowering it causes trouble. So now Exim takes care not to decrease it.
5022 40. When delivering a message, the value of $return_path is set to $sender_
5023 address at the start of routing (routers may change the value). By an
5024 oversight, this default was not being set up when an address was tested by
5025 -bt or -bv, which affected the outcome if any router or filter referred to
5028 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5029 log only when "message" or "log_message" are set. However, if one of the
5030 conditions was an address verification, or a call to a nested ACL, the
5031 messages generated by the underlying test were being passed through. This
5032 no longer happens. The underlying message is available in $acl_verify_
5033 message for both "message" and "log_message" expansions, so it can be
5034 passed through if needed.
5036 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5037 new expansion $bh_ to give the encoded byte string without charset
5038 translation. Translation happens only if iconv() is available; HAVE_ICONV
5039 indicates this at build time. HEADERS_CHARSET gives the charset to
5040 translate to; headers_charset can change it in the configuration, and
5041 "headers charset" can change it in an individual filter file.
5043 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5044 that creates RFC 2047 encoded "words" labels them as that charset instead
5045 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5046 expansion operator; (ii) when Exim creates a From: line for a local
5047 message; (iii) when a header line is rewritten to include a "phrase" part.
5049 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5050 buggy, causing it to skip the first lines of messages whose message ID
5051 ended in 'D'. This would not have bitten before Exim release 4.14, because
5052 message IDs were unlikely to end in 'D' before then. The effect was to have
5053 incorrect size information for certain domains.
5055 45. #include "config.h" was missing at the start of the crypt16.c module. This
5056 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5059 46. If there was a timeout during a "random" callout check, Exim treated it as
5060 a failure of the random address, and carried on sending RSET and the real
5061 address. If the delay was just some slowness somewhere, the response to the
5062 original RCPT would be taken as a response to RSET and so on, causing
5063 mayhem of various kinds.
5065 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5066 when I implemented it. It didn't allow for the fact that some option values
5067 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5068 the right test for positive values.
5070 48. Domain names in DNS records are case-independent. Exim always looks them up
5071 in lower case. Some resolvers return domain names in exactly the case they
5072 appear in the zone file, that is, they may contain uppercase letters. Not
5073 all resolvers do this - some return always lower case. Exim was treating a
5074 change of case by a resolver as a change of domain, similar to a widening
5075 of a domain abbreviation. This triggered its re-routing code and so it was
5076 trying to route what was effectively the same domain again. This normally
5077 caused routing to fail (because the router wouldn't handle the domain
5078 twice). Now Exim checks for this case specially, and just changes the
5079 casing of the domain that it ultimately uses when it transmits the message
5082 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5085 50. If a filter generated a file delivery with a non-absolute name (possible if
5086 no home directory exists for the router), the forbid_file option was not
5089 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5090 the existing equality matching.
5092 52. Exim was using ints instead of ino_t variables in some places where it was
5093 dealing with inode numbers.
5095 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5096 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5097 called TMPDIR, and if it finds it is different, it changes its value.
5099 54. The smtp_printf() function is now made available to local_scan() so
5100 additional output lines can be written before returning. There is also an
5101 smtp_fflush() function to enable the detection of a dropped connection.
5102 The variables smtp_input and smtp_batched_input are exported to
5105 55. Changed the default runtime configuration: the message "Unknown user"
5106 has been removed from the ACL, and instead placed on the localuser router,
5107 using the cannot_route_message feature. This means that any verification
5108 failures that generate their own messages won't get overridden. Similarly,
5109 the "Unrouteable address" message that was in the ACL for unverifiable
5110 relay addresses has also been removed.
5112 56. Added hosts_avoid_esmtp to the smtp transport.
5114 57. The exicyclog script was not checking for the esoteric option
5115 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5116 will work only if exicyclog is run under the appropriate euid.
5118 58. Following a discussion on the list, the rules by which Exim recognises line
5119 endings on incoming messages have been changed. The -dropcr and drop_cr
5120 options are now no-ops, retained only for backwards compatibility. The
5121 following line terminators are recognized: LF CRLF CR. However, special
5122 processing applies to CR:
5124 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5125 nor a local message in the state where . is a terminator.
5127 (ii) If a bare CR is encountered in a header line, an extra space is added
5128 after the line terminator so as not to end the header. The reasoning
5129 behind this is that bare CRs in header lines are most likely either
5130 to be mistakes, or people trying to play silly games.
5132 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5133 was being incorrectly given as 18 bytes larger than it should have been.
5134 This is a VOB (very old bug).
5136 60. This may never have affected anything current, but just in case it has:
5137 When the local host is found other than at the start of a list of hosts,
5138 the local host, those with the same MX, and any that follow, are discarded.
5139 When the list in question was part of a longer list of hosts, the following
5140 hosts (not currently being processed) were also being discarded. This no
5141 longer happens. I'm not sure if this situation could ever has previously
5144 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5147 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5148 header. This is something that is recommended in a new Internet Draft, and
5149 is something that is documented as being done by Sendmail. There are two
5150 possible values. For messages generated by the autoreply transport, Exim
5153 Auto-Submitted: auto-replied
5155 whereas for all other generated messages (e.g. bounces) it adds
5157 Auto-Submitted: auto-generated
5159 63. The "personal" condition in filters now includes a test for the
5160 Auto-Submitted: header. If it contains the string "auto-" the message it
5161 not considered personal.
5163 64. Added rcpt_include_affixes as a generic transport option.
5165 65. Added queue_only_override (default true).
5167 66. Added the syslog_duplication option.
5169 67. If what should have been the first header line of a message consisted of
5170 a space followed by a colon, Exim was mis-interpreting it as a header line.
5171 It isn't of course - it is syntactically invalid and should therefore be
5172 treated as the start of the message body. The misbehaviour could have
5173 caused a number of strange effects, including loss of data in subsequent
5174 header lines, and spool format errors.
5176 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5177 client host had authenticated. This control can now be exercised by an ACL
5178 for more flexibility.
5180 69. By default, callouts do not happen when testing with -bh. There is now a
5181 variant, -bhc, which does actually run the callout code, including
5182 consulting and updating the callout cache.
5184 70. Added support for saslauthd authentication, courtesy of Alexander
5187 71. If statvfs() failed on the spool or log directories while checking their
5188 size for availability, Exim confusingly gave the error "space shortage".
5189 Furthermore, in debugging mode it crashed with a floating point exception.
5190 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5191 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5192 serious problem, Exim now writes to the main and panic logs when this
5193 happens, with details of the failure. It then refuses to accept the
5194 incoming message, giving the message "spool directory problem" or "log
5195 directory problem" with a 421 code for SMTP messages.
5197 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5198 0, 1, and 2 exist, because some OS complain for execs without them (see
5199 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5200 descriptors. However, the code omitted to check that the open succeeded,
5201 causing mysterious errors if for some reason the permissions on /dev/null
5202 got screwed. Now Exim writes a message to the main and panic logs, and
5203 bombs out if it can't open /dev/null.
5205 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5206 interact so that it is all more flexible. It is supposed to remain
5207 backwards compatible. Also added extra_local_interfaces.
5209 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5210 to bomb out with an assertion failure - to the client this appears as a
5211 connection drop. This problem occurs in the part of the code that was taken
5212 from the Samba project. Fortunately, the assertion is in a very simple
5213 function, so I have fixed this by reproducing the function inline in the
5214 one place where it is called, and arranging for authentication to fail
5215 instead of killing the process with assert().
5217 75. The SPA client code was not working when the server requested OEM rather
5218 than Unicode encoding.
5220 76. Added code to make require_files with a specific uid setting more usable in
5221 the case where statting the file as root fails - usually a non-root-mounted
5222 NFS file system. When this happens and the failure is EACCES, Exim now
5223 forks a subprocess and does the per-uid checking as the relevant uid.
5225 77. Added process_log_path.
5227 78. If log_file_path was not explicitly set, a setting of check_log_space or
5228 check_log_inodes was ignored.
5230 79. If a space check for the spool or log partitions fails, the incident is now
5231 logged. Of course, in the latter case the data may get lost...
5233 80. Added the %p formatting code to string_format() so that it can be used to
5234 print addresses in debug_print(). Adjusted all the address printing in the
5235 debugging in store.c to use %p rather than %d.
5237 81. There was a concern that a line of code in smtp_in.c could overflow a
5238 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5239 initially expressed, the concern was not well-founded, because trailing
5240 spaces are removed early. However, if the trailing spaces were followed by
5241 a NULL, they did not get removed, so the overflow was possible. Two fixes
5244 (a) I re-wrote the offending code in a cleaner fashion.
5245 (b) If an incoming SMTP command contains a NULL character, it is rejected
5248 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5249 runs initgroups(), so that if the Exim user is in any additional groups,
5250 they will be used during message reception.
5256 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob