1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
40 TF/01 Failure to get a lock on a hints database can have serious
41 consequences so log it to the panic log.
43 TF/02 Log LMTP confirmation messages in the same way as SMTP,
44 controlled using the smtp_confirmation log selector.
46 TF/03 Include the error message when we fail to unlink a spool file.
48 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
49 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
50 for maintaining out-of-tree patches for some time.
52 PP/01 Bugzilla 139: Documentation and portability issues.
53 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
54 Handle per-OS dynamic-module compilation flags.
56 PP/02 Let /dev/null have normal permissions.
57 The 4.73 fixes were a little too stringent and complained about the
58 permissions on /dev/null. Exempt it from some checks.
59 Reported by Andreas M. Kirchwitz.
61 PP/03 Report version information for many libraries, including
62 Exim version information for dynamically loaded libraries. Created
63 version.h, now support a version extension string for distributors
64 who patch heavily. Dynamic module ABI change.
66 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
67 privilege escalation vulnerability whereby the Exim run-time user
68 can cause root to append content of the attacker's choosing to
71 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
74 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
75 If dropping privileges for untrusted macros, we disabled normal logging
76 on the basis that it would fail; for the Exim run-time user, this is not
77 the case, and it resulted in successful deliveries going unlogged.
78 Fixed. Reported by Andreas Metzler.
84 PP/01 Date: & Message-Id: revert to normally being appended to a message,
85 only prepend for the Resent-* case. Fixes regression introduced in
86 Exim 4.70 by NM/22 for Bugzilla 607.
88 PP/02 Include check_rfc2047_length in configure.default because we're seeing
89 increasing numbers of administrators be bitten by this.
91 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
93 PP/03 Bugzilla 994: added openssl_options main configuration option.
95 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
97 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
99 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
101 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
102 without a peer certificate, leading to a segfault because of an
103 assumption that peers always have certificates. Be a little more
104 paranoid. Problem reported by Martin Tscholak.
106 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
107 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
108 NB: ClamAV planning to remove STREAM in "middle of 2010".
109 CL also introduces -bmalware, various -d+acl logging additions and
110 more caution in buffer sizes.
112 PP/09 Implemented reverse_ip expansion operator.
114 PP/10 Bugzilla 937: provide a "debug" ACL control.
116 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
118 PP/12 Bugzilla 973: Implement --version.
120 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
122 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
124 PP/15 Bugzilla 816: support multiple condition rules on Routers.
126 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
127 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
128 ignore trailing whitespace.
130 JJ/02 prevent non-panic DKIM error from being sent to paniclog
132 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
135 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
136 Notification from Dr Andrew Aitchison.
138 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
139 ExtendedDetectionInfo response format.
140 Notification from John Horne.
142 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
145 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
146 XSL and documented dependency on system catalogs, with examples of how
149 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
152 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
153 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
154 configuration file which is writeable by the Exim user or group.
156 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
157 of configuration files to cover files specified with the -C option if
158 they are going to be used with root privileges, not just the default
161 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
162 option (effectively making it always true).
164 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
165 files to be used while preserving root privileges.
167 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
168 that rogue child processes cannot use them.
170 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
171 run-time user, instead of root.
173 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
174 Exim run-time user without dropping privileges.
176 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
177 result string, instead of calling string_vformat() twice with the same
180 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
181 for other users. Others should always drop root privileges if they use
182 -C on the command line, even for a whitelisted configure file.
184 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
186 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
192 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
193 $data_path, and $header_path variables; fixed documentation bugs and
196 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
197 exipick to access non-standard spools, including the "frozen" queue
200 NM/01 Bugzilla 965: Support mysql stored procedures.
201 Patch from Alain Williams
203 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
205 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
206 Patch from Andreas Metzler
208 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
209 Patch from Kirill Miazine
211 NM/05 Bugzilla 671: Added umask to procmail example.
213 JJ/03 installed exipick 20100323.0, fixing doc bug
215 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
216 directory. Notification and patch from Dan Rosenberg.
218 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
220 TK/02 Improve log output when DKIM signing operation fails.
222 MH/01 Treat the transport option dkim_domain as a colon separated
223 list, not as a single string, and sign the message with each element,
224 omitting multiple occurences of the same signer.
226 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
227 Bugzilla 985, 986. Patch by Simon Arlott
229 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
230 Patch by Simon Arlott
232 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
233 MBX locking. Notification from Dan Rosenberg.
239 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
241 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
243 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
245 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
247 NM/04 Bugzilla 915: Flag broken perl installation during build.
253 TK/01 Added patch by Johannes Berg that expands the main option
254 "spamd_address" if it starts with a dollar sign.
256 TK/02 Write list of recipients to X-Envelope-Sender header when building
257 the mbox-format spool file for content scanning (suggested by Jakob
260 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
261 (http://www.dcc-servers.net/) support via dccifd. Activated by
262 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
264 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
265 by Mark Daniel Reidel <mr@df.eu>.
267 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
268 When building exim an external PCRE library is now needed -
269 PCRE is a system library on the majority of modern systems.
270 See entry on PCRE_LIBS in EDITME file.
272 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
273 conversation. Added nologin parameter to request.
274 Patch contributed by Kirill Miazine.
276 TF/01 Do not log submission mode rewrites if they do not change the address.
278 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
280 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
281 log files in place. Contributed by Roberto Lima.
283 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
285 TF/03 Bugzilla 615: When checking the local_parts router precondition
286 after a local_part_suffix or local_part_prefix option, Exim now
287 does not use the address's named list lookup cache, since this
288 contains cached lookups for the whole local part.
290 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
291 Robert Millan. Documentation is in experimental-spec.txt.
293 TF/04 Bugzilla 668: Fix parallel build (make -j).
295 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
297 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
298 Patch provided by Jan Srzednicki.
300 TF/05 Leading white space used to be stripped from $spam_report which
301 wrecked the formatting. Now it is preserved.
303 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
304 that they are available at delivery time.
306 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
308 TF/08 TLS error reporting now respects the incoming_interface and
309 incoming_port log selectors.
311 TF/09 Produce a more useful error message if an SMTP transport's hosts
312 setting expands to an empty string.
314 NM/06 Bugzilla 744: EXPN did not work under TLS.
315 Patch provided by Phil Pennock.
317 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
318 Patch provided by Richard Godbee.
320 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
321 acl_smtp_notquit, added index entry.
323 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
324 Patch provided by Eugene Bujak.
326 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
327 accept(). Patch provided by Maxim Dounin.
329 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
330 Patch provided by Phil Pennock.
332 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
334 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
335 Patch provided by Brad "anomie" Jorsch.
337 NM/14 Bugzilla 622: Added timeout setting to transport filter.
338 Patch provided by Dean Brooks.
340 TK/05 Add native DKIM support (does not depend on external libraries).
342 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
343 Patch provided by Graeme Fowler.
345 NM/16 Bugzilla 851: Documentation example syntax fix.
347 NM/17 Changed NOTICE file to remove references to embedded PCRE.
349 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
352 NM/19 Bugzilla 745: TLS version reporting.
353 Patch provided by Phil Pennock.
355 NM/20 Bugzilla 167: bool: condition support.
356 Patch provided by Phil Pennock.
358 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
359 clients. Patch provided by Phil Pennock.
361 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
362 Patch provided by Brad "anomie" Jorsch.
364 NM/23 Bugzilla 687: Fix misparses in eximstats.
365 Patch provided by Heiko Schlittermann.
367 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
368 Patch provided by Heiko Schlittermann.
370 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
371 plus update to original patch.
373 NM/26 Bugzilla 799: Documentation correction for ratelimit.
375 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
376 Patch provided by David Brownlee.
378 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
380 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
382 NM/30 Bugzilla 888: TLS documentation bugfixes.
384 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
386 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
387 Unlike the original bugzilla I have changed all shell scripts in src tree.
389 NM/33 Bugzilla 898: Transport filter timeout fix.
390 Patch by Todd Rinaldo.
392 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
393 Patch by Serge Demonchaux.
395 NM/35 Bugzilla 39: Base64 decode bug fixes.
396 Patch by Jakob Hirsch.
398 NM/36 Bugzilla 909: Correct connect() call in dcc code.
400 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
402 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
404 NM/39 Bugzilla 911: Fixed MakeLinks build script.
410 TK/01 Add preliminary DKIM support. Currently requires a forked version of
411 ALT-N's libdkim that I have put here:
412 http://duncanthrax.net/exim-experimental/
414 Note to Michael Haardt: I had to rename some vars in sieve.c. They
415 were called 'true' and it seems that C99 defines that as a reserved
416 keyword to be used with 'bool' variable types. That means you could
417 not include C99-style headers which use bools without triggering
418 build errors in sieve.c.
420 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
421 as mailq or other aliases. Changed the --help handling significantly
422 to do whats expected. exim_usage() emits usage/help information.
424 SC/01 Added the -bylocaldomain option to eximstats.
426 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
428 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
430 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
436 PH/01 Another patch from the Sieve maintainer.
438 PH/02 When an IPv6 address is converted to a string for single-key lookup
439 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
440 dots are used instead of colons so that keys in lsearch files need not
441 contain colons. This was done some time before quoting was made available
442 in lsearch files. However, iplsearch files do require colons in IPv6 keys
443 (notated using the quote facility) so as to distinguish them from IPv4
444 keys. This meant that lookups for IP addresses in host lists did not work
445 for iplsearch lookups.
447 This has been fixed by arranging for IPv6 addresses to be expressed with
448 colons if the lookup type is iplsearch. This is not incompatible, because
449 previously such lookups could never work.
451 The situation is now rather anomolous, since one *can* have colons in
452 ordinary lsearch keys. However, making the change in all cases is
453 incompatible and would probably break a number of configurations.
455 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
458 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
459 conversion specification without a maximum field width, thereby enabling
460 a rogue spamd server to cause a buffer overflow. While nobody in their
461 right mind would setup Exim to query an untrusted spamd server, an
462 attacker that gains access to a server running spamd could potentially
463 exploit this vulnerability to run arbitrary code as the Exim user.
465 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
466 $primary_hostname instead of what libspf2 thinks the hosts name is.
468 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
469 a directory entry by the name of the lookup key. Previously, if a
470 symlink pointed to a non-existing file or a file in a directory that
471 Exim lacked permissions to read, a lookup for a key matching that
472 symlink would fail. Now it is enough that a matching directory entry
473 exists, symlink or not. (Bugzilla 503.)
475 PH/03 The body_linecount and body_zerocount variables are now exported in the
478 PH/04 Added the $dnslist_matched variable.
480 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
481 This means they are set thereafter only if the connection becomes
484 PH/06 Added the client_condition to authenticators so that some can be skipped
485 by clients under certain conditions.
487 PH/07 The error message for a badly-placed control=no_multiline_responses left
488 "_responses" off the end of the name.
490 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
492 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
493 (without spaces) instead of just copying the configuration text.
495 PH/10 Added the /noupdate option to the ratelimit ACL condition.
497 PH/11 Added $max_received_linelength.
499 PH/12 Added +ignore_defer and +include_defer to host lists.
501 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
502 way in which PCRE > 7.0 is built.
504 PH/14 Implemented queue_only_load_latch.
506 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
507 MAIL command. The effect was to mangle the value on 64-bit systems.
509 PH/16 Another patch from the Sieve maintainer.
511 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
513 PH/18 If a system quota error occurred while trying to create the file for
514 a maildir delivery, the message "Mailbox is full" was not appended to the
515 bounce if the delivery eventually timed out. Change 4.67/27 below applied
516 only to a quota excession during the actual writing of the file.
518 PH/19 It seems that peer DN values may contain newlines (and other non-printing
519 characters?) which causes problems in log lines. The DN values are now
520 passed through string_printing() before being added to log lines.
522 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
523 and InterBase are left for another time.)
525 PH/21 Added message_body_newlines option.
527 PH/22 Guard against possible overflow in moan_check_errorcopy().
529 PH/23 POSIX allows open() to be a macro; guard against that.
531 PH/24 If the recipient of an error message contained an @ in the local part
532 (suitably quoted, of course), incorrect values were put in $domain and
533 $local_part during the evaluation of errors_copy.
539 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
540 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
543 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
544 issue a MAIL command.
546 PH/02 In an ACL statement such as
548 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
550 if a client was not listed at all, or was listed with a value other than
551 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
552 the condition was not true (as it should be), so access was not denied.
553 The bug was that the ! inversion was incorrectly passed on to the second
554 item. This has been fixed.
556 PH/03 Added additional dnslists conditions == and =& which are different from
557 = and & when the dns lookup returns more than one IP address.
559 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
560 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
562 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
563 FSYNC, which compiles an option called disable_fsync that allows for
564 bypassing fsync(). The documentation is heavily laced with warnings.
566 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
568 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
569 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
570 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
571 including adding "make clean"; (3) Added -fPIC when compiling the test
572 dynamically loaded module, to get rid of a warning.
574 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
575 message fails, move_frozen_messages = true and ignore_bounce_errors_after
576 = 0s. The bug is otherwise harmless.
578 PH/07 There was a bug in the dovecot authenticator such that the value of
579 $auth1 could be overwritten, and so not correctly preserved, after a
580 successful authentication. This usually meant that the value preserved by
581 the server_setid option was incorrect.
583 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
585 PH/09 Installed PCRE release 7.0.
587 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
588 run for batched SMTP input. It is now run at the start of every message
589 in the batch. While fixing this I discovered that the process information
590 (output by running exiwhat) was not always getting set for -bs and -bS
591 input. This is fixed, and it now also says "batched" for BSMTP.
593 PH/11 Added control=no_pipelining.
595 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
596 patch, slightly modified), and move the expansion of helo_data till after
597 the connection is made in the smtp transport (so it can use these
600 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
602 PH/14 Added log_selector = +pid.
604 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
606 PH/16 Add ${if forany and ${if forall.
608 PH/17 Added dsn_from option to vary the From: line in DSNs.
610 PH/18 Flush SMTP output before performing a callout, unless control =
611 no_callout_flush is set.
613 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
614 was true (the default) a successful delivery failed to delete the retry
615 item, thus causing premature timeout of the address. The bug is now
618 PH/20 Added hosts_avoid_pipelining to the smtp transport.
620 PH/21 Long custom messages for fakedefer and fakereject are now split up
621 into multiline reponses in the same way that messages for "deny" and
622 other ACL rejections are.
624 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
625 with slight modification.
627 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
628 draft, changing the syntax and factoring some duplicate code".
630 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
631 for deliveries of the second and subsequent messages over the same SMTP
634 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
635 ${reduce, with only minor "tidies".
637 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
639 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
640 expansion side effects.
642 PH/27 When a message times out after an over-quota error from an Exim-imposed
643 quota, the bounce message says "mailbox is full". This message was not
644 being given when it was a system quota that was exceeded. It now should
647 MH/03 Made $recipients available in local_scan(). local_scan() already has
648 better access to the recipient list through recipients_list[], but
649 $recipients can be useful in postmaster-provided expansion strings.
651 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
652 in the case of a MAIL command with additional options following the
653 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
654 were accidentally chopped off.
656 PH/29 SMTP synchronization checks are implemented when a command is read -
657 there is a check that no more input is waiting when there shouldn't be
658 any. However, for some commands, a delay in an ACL can mean that it is
659 some time before the response is written. In this time, more input might
660 arrive, invalidly. So now there are extra checks after an ACL has run for
661 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
662 pipelining has not been advertised.
664 PH/30 MH's patch to allow iscntrl() characters to be list separators.
666 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
667 returned in the SMTP response when smtp_return_error_details was false.
670 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
671 instead of the C I/O that was originally supplied, because problems were
674 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
675 Exim which did not show up earlier: it was assuming that a call to
676 SSL_CTX_set_info_callback() might give an error value. In fact, there is
677 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
678 was a macro that became an assignment, so it seemed to work. This has
679 changed to a proper function call with a void return, hence the compile
680 error. Exim's code has been fixed.
682 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
685 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
687 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
689 PH/37 If a message is not accepted after it has had an id assigned (e.g.
690 because it turns out to be too big or there is a timeout) there is no
691 "Completed" line in the log. When some messages of this type were
692 selected by exigrep, they were listed as "not completed". Others were
693 picked up by some special patterns. I have improved the selection
694 criteria to be more general.
696 PH/38 The host_find_failed option in the manualroute router can now be set
697 to "ignore", to completely ignore a host whose IP address cannot be
698 found. If all hosts are ignored, the behaviour is controlled by the new
699 host_all_ignored option.
701 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
702 homing or because of multiple MX records with /mx) generated more than
703 one IP address, and the following item turned out to be the local host,
704 all the secondary addresses of the first item were incorrectly removed
705 from the list, along with the local host and any following hosts (which
706 is what is supposed to happen).
708 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
709 whoever called Exim into the -H file. In the case of the daemon it was
710 behaving confusingly. When first started, it used values for whoever
711 started the daemon, but after a SIGHUP it used the Exim user (because it
712 calls itself on a restart). I have changed the code so that it now always
715 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
716 message are rejected with the same error (e.g. no authentication or bad
717 sender address), and a DATA command is nevertheless sent (as can happen
718 with PIPELINING or a stupid MUA), the error message that was given to the
719 RCPT commands is included in the rejection of the DATA command. This is
720 intended to be helpful for MUAs that show only the final error to their
723 PH/42 Another patch from the Sieve maintainer.
725 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
726 Eximstats - Fixed some broken HTML links and added missing column headers
728 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
729 columns for Rejects, Temp Rejects, Ham, and Spam rows.
731 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
733 PH/43 Yet another patch from the Sieve maintainer.
735 PH/44 I found a way to check for a TCP/IP connection going away before sending
736 the response to the final '.' that terminates a message, but only in the
737 case where the client has not sent further data following the '.'
738 (unfortunately, this is allowed). However, in many cases there won't be
739 any further data because there won't be any more messages to send. A call
740 to select() can be used: if it shows that the input is "ready", there is
741 either input waiting, or the socket has been closed. An attempt to read
742 the next input character can distinguish the two cases. Previously, Exim
743 would have sent an OK response which the client would never have see.
744 This could lead to message repetition. This fix should cure that, at
745 least in a lot of common cases.
747 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
748 advertised in response to EHLO.
754 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
755 fixed by 4.65/MH/01 (is this a record?) are fixed:
757 (i) An empty string was always treated as zero by the numeric comparison
758 operators. This behaviour has been restored.
760 (ii) It is documented that the numeric comparison operators always treat
761 their arguments as decimal numbers. This was broken in that numbers
762 starting with 0 were being interpreted as octal.
764 While fixing these problems I realized that there was another issue that
765 hadn't been noticed. Values of message_size_limit (both the global option
766 and the transport option) were treated as octal if they started with 0.
767 The documentation was vague. These values are now always treated as
768 decimal, and I will make that clear in the documentation.
774 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
775 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
778 MH/01 Don't check that the operands of numeric comparison operators are
779 integers when their expansion is in "skipping" mode (fixes bug
780 introduced by 4.64-PH/07).
782 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
783 child addresses, Exim now panics and dies. Previously, because the count
784 is held in a short int, deliveries were likely to be lost. As such a
785 large number of recipients for a single message is ridiculous
786 (performance will be very, very poor), I have chosen to impose a limit
787 rather than extend the field.
793 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
794 leftover -K file (the existence of which was triggered by #402).
795 While we were at it, introduced process PID as part of the -K
796 filename. This should rule out race conditions when creating
799 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
800 processing considerably. Previous code took too long for large mails,
801 triggering a timeout which in turn triggers #401.
803 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
804 in the DK code in transports.c. sendfile() is not really portable,
805 hence the _LINUX specificness.
807 TF/01 In the add_headers option to the mail command in an Exim filter,
808 there was a bug that Exim would claim a syntax error in any
809 header after the first one which had an odd number of characters
812 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
813 callout verification, Exim cached a "reject" for the entire domain. This
814 is correct for most verifications, but it is not correct for a recipient
815 verification with use_sender or use_postmaster set, because in that case
816 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
817 case of MAIL FROM:<> rejection from other early rejections (e.g.
818 rejection of HELO). When verifying a recipient using a non-null MAIL
819 address, the cache is ignored if it shows MAIL FROM:<> rejection.
820 Whatever the result of the callout, the value of the domain cache is
821 left unchanged (for any other kind of callout, getting as far as trying
822 RCPT means that the domain itself is ok).
824 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
827 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
828 manifest itself as EPIPE rather than ECONNECT. When tidying away a
829 session, the daemon ignores ECONNECT errors and logs others; it now
830 ignores EPIPE as well.
832 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
833 (quoted-printable decoding).
835 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
836 later the small subsequent patch to fix an introduced bug.
838 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
840 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
842 PH/08 An error is now given if message_size_limit is specified negative.
844 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
845 to be given (somewhat) arbitrary names.
847 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
850 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
851 miscellaneous code fixes
853 PH/10 Added the log_reject_target ACL modifier to specify where to log
856 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
857 hostname. This is wrong, because it relates to the incoming message (and
858 probably the interface on which it is arriving) and not to the outgoing
859 callout (which could be using a different interface). This has been
860 changed to use the value of the helo_data option from the smtp transport
861 instead - this is what is used when a message is actually being sent. If
862 there is no remote transport (possible with a router that sets up host
863 addresses), $smtp_active_hostname is used.
865 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
866 tweaks were necessary in order to get it to work (see also 21 below):
867 (a) The code assumed that strncpy() returns a negative number on buffer
868 overflow, which isn't the case. Replaced with Exim's string_format()
870 (b) There were several signed/unsigned issues. I just did the minimum
871 hacking in of casts. There is scope for a larger refactoring.
872 (c) The code used strcasecmp() which is not a standard C function.
873 Replaced with Exim's strcmpic() function.
874 (d) The code set only $1; it now sets $auth1 as well.
875 (e) A simple test gave the error "authentication client didn't specify
876 service in request". It would seem that Dovecot has changed its
877 interface. Fortunately there's a specification; I followed it and
878 changed what the client sends and it appears to be working now.
880 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
883 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
884 address A is aliased to B and C, where B exists and C does not. Without
885 -v the output is "A verified" because verification stops after a
886 successful redirection if more than one address is generated. However,
887 with -v the child addresses are also verified. Exim was outputting "A
888 failed to verify" and then showing the successful verification for C,
889 with its parentage. It now outputs "B failed to verify", showing B's
890 parentage before showing the successful verification of C.
892 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
893 look up a TXT record in a specific list after matching in a combined
896 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
897 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
898 they consult the DNS. I had assumed they would set it the way they
899 wanted; and indeed my experiments on Linux seem to show that in some
900 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
901 To be on the safe side, however, I have now made the interface to
902 host_find_byname() similar to host_find_bydns(), with an argument
903 containing the DNS resolver options. The host_find_byname() function now
904 sets these options at its start, just as host_find_bydns() does. The smtp
905 transport options dns_qualify_single and dns_search_parents are passed to
906 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
907 of host_find_byname() use the default settings of RES_DEFNAMES
908 (qualify_single) but not RES_DNSRCH (search_parents).
910 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
911 spool_read_header() do less string testing, by means of a preliminary
912 switch on the second character of optional "-foo" lines. (This is
913 overdue, caused by the large number of possibilities that now exist.
914 Originally there were few.) While I was there, I also converted the
915 str(n)cmp tests so they don't re-test the leading "-" and the first
916 character, in the hope this might squeeze out yet more improvement.
918 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
919 flag allowing group syntax was set by the header_syntax check but not
920 turned off, possible causing trouble later; (2) The flag was not being
921 set at all for the header_verify test, causing "group"-style headers to
922 be rejected. I have now set it in this case, and also caused header_
923 verify to ignore an empty address taken from a group. While doing this, I
924 came across some other cases where the code for allowing group syntax
925 while scanning a header line wasn't quite right (mostly, not resetting
926 the flag correctly in the right place). These bugs could have caused
927 trouble for malformed header lines. I hope it is now all correct.
929 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
930 with the "reply" argument non-NULL. The code, however (which originally
931 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
932 but it didn't always do it. This confused somebody who was copying the
933 code for some other use. I have removed all the tests.
935 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
936 feature that was used to support insecure browsers during the U.S. crypto
937 embargo. It requires special client support, and Exim is probably the
938 only MTA that supported it -- and would never use it because real RSA is
939 always available. This code has been removed, because it had the bad
940 effect of slowing Exim down by computing (never used) parameters for the
941 RSA_EXPORT functionality.
943 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
944 authenticator to fail if there's a tab character in the incoming data
945 (there should never be unless someone is messing about, as it's supposed
946 to be base64-encoded). Also added, on Timo's advice, the "secured" option
947 if the connection is using TLS or if the remote IP is the same as the
948 local IP, and the "valid-client-cert option" if a client certificate has
951 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
952 authenticators. This can be used for authorization after authentication
953 succeeds. (In the case of plaintext, it servers for both authentication
956 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
957 if any retry times were supplied.
959 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
960 connection, where there is no client IP address to check. In this
961 situation, the verify now always succeeds.
963 PH/25 Applied John Jetmore's -Mset patch.
965 PH/26 Added -bem to be like -Mset, but loading a message from a file.
967 PH/27 In a string expansion for a processed (not raw) header when multiple
968 headers of the same name were present, leading whitespace was being
969 removed from all of them, but trailing whitespace was being removed only
970 from the last one. Now trailing whitespace is removed from each header
971 before concatenation. Completely empty headers in a concatenation (as
974 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
975 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
977 PH/29 [Removed. This was a change that I later backed out, and forgot to
978 correct the ChangeLog entry (that I had efficiently created) before
979 committing the later change.]
981 PH/30 Exim was sometimes attempting to deliver messages that had suffered
982 address errors (4xx response to RCPT) over the same connection as other
983 messages routed to the same hosts. Such deliveries are always "forced",
984 so retry times are not inspected. This resulted in far too many retries
985 for the affected addresses. The effect occurred only when there were more
986 hosts than the hosts_max_try setting in the smtp transport when it had
987 the 4xx errors. Those hosts that it had tried were not added to the list
988 of hosts for which the message was waiting, so if all were tried, there
989 was no problem. Two fixes have been applied:
991 (i) If there are any address or message errors in an SMTP delivery, none
992 of the hosts (tried or untried) are now added to the list of hosts
993 for which the message is waiting, so the message should not be a
994 candidate for sending over the same connection that was used for a
995 successful delivery of some other message. This seems entirely
996 reasonable: after all the message is NOT "waiting for some host".
997 This is so "obvious" that I'm not sure why it wasn't done
998 previously. Hope I haven't missed anything, but it can't do any
999 harm, as the worst effect is to miss an optimization.
1001 (ii) If, despite (i), such a delivery is accidentally attempted, the
1002 routing retry time is respected, so at least it doesn't keep
1003 hammering the server.
1005 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1006 in ${readsocket because some servers need this prod.
1008 PH/32 Added some extra debug output when updating a wait-xxx database.
1010 PH/33 The hint "could be header name not terminated by colon", which has been
1011 given for certain expansion errors for a long time, was not being given
1012 for the ${if def:h_colon_omitted{... case.
1014 PH/34 The spec says: "With one important exception, whenever a domain list is
1015 being scanned, $domain contains the subject domain." There was at least
1016 one case where this was not true.
1018 PH/35 The error "getsockname() failed: connection reset by peer" was being
1019 written to the panic log as well as the main log, but it isn't really
1020 panic-worthy as it just means the connection died rather early on. I have
1021 removed the panic log writing for the ECONNRESET error when getsockname()
1024 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1025 runs only) independently of the message's sender address. This meant
1026 that, if the 4xx error was in fact related to the sender, a different
1027 message to the same recipient with a different sender could confuse
1028 things. In particualar, this can happen when sending to a greylisting
1029 server, but other circumstances could also provoke similar problems.
1030 I have changed the default so that the retry time for these errors is now
1031 based a combination of the sender and recipient addresses. This change
1032 can be overridden by setting address_retry_include_sender=false in the
1035 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1036 remote server are returned as part of bounce messages. This was not
1037 happening for LMTP over a pipe (the lmtp transport), but now it is the
1038 same for both kinds of LMTP.
1040 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1041 in header lines that were in fact CNAMEs. This is no longer the case.
1043 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1044 and queue runs started by the daemon processed all messages. This has
1045 been fixed so that -R and -S can now usefully be given with -q<time>.
1047 PH/40 Import PCRE release 6.7 (fixes some bugs).
1049 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1051 PH/42 Give an error if -q is specified more than once.
1053 PH/43 Renamed the variables $interface_address and $interface_port as
1054 $received_ip_address and $received_port, to make it clear that these
1055 values apply to message reception, and not to the outgoing interface when
1056 a message is delivered. (The old names remain recognized, of course.)
1058 PH/44 There was no timeout on the connect() call when using a Unix domain
1059 socket in the ${readsocket expansion. There now is.
1061 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1062 be meaningful with "accept".
1064 SC/01 Eximstats V1.43
1065 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1067 SC/02 Eximstats V1.44
1068 Use a glob alias rather than an array ref in the generated
1069 parser. This improves both readability and performance.
1071 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1072 Collect SpamAssassin and rejection statistics.
1073 Don't display local sender or destination tables unless
1074 there is data to show.
1075 Added average volumes into the top table text output.
1077 SC/04 Eximstats V1.46
1078 Collect data on the number of addresses (recipients)
1079 as well as the number of messages.
1081 SC/05 Eximstats V1.47
1082 Added 'Message too big' to the list of mail rejection
1083 reasons (thanks to Marco Gaiarin).
1085 SC/06 Eximstats V1.48
1086 Mainlog lines which have GMT offsets and are too short to
1087 have a flag are now skipped.
1089 SC/07 Eximstats V1.49 (Alain Williams)
1090 Added the -emptyok flag.
1092 SC/08 Eximstats V1.50
1093 Fixes for obtaining the IP address from reject messages.
1095 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1096 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1097 whitesspace changes from 4.64-PH/27
1099 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1102 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1103 are found, allow negative numbers in numeric criteria)
1105 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1107 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1108 to match changes made in 4.64-PH/43
1110 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1112 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1113 "domain" is actually an IP address, and force a failure. This locks out
1114 those revolvers/nameservers that support "A-for-A" lookups, in
1115 contravention of the specifications.
1117 PH/48 When a host name was looked up from an IP address, and the subsequent
1118 forward lookup of the name timed out, the host name was left in
1119 $sender_host_name, contrary to the specification.
1121 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1122 restricted to single-key lookups, Exim was not diagnosing an error if
1123 * or *@ was used with a query-style lookup.
1125 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1127 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1128 long ago, but noone interested enough thought of it. Let's just say that
1129 the "1.1" means that there are some new functions that weren't there at
1130 some point in the past.
1132 PH/51 Error processing for expansion failure of helo_data from an smtp
1133 transport during callout processing was broken.
1135 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1136 tested/used via the -bh/-bhc/-bs options.
1138 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1139 bug, fixed in subsequent PCRE releases).
1141 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1142 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1144 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1150 SC/01 Use a glob alias rather than an array ref in eximstats generated
1151 parser. This improves both readability and performance.
1153 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1154 Don't display local sender or destination tables in eximstats unless
1155 there is data to show.
1156 Added average volumes into the eximstats top table text output.
1158 SC/03 Collect data on the number of addresses (recipients) as well
1159 as the number of messages in eximstats.
1161 TF/01 Correct an error in the documentation for the redirect router. Exim
1162 does (usually) call initgroups() when daemonizing.
1164 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1165 with consistent privilege compared to when running as a daemon.
1167 TF/03 Note in the spec that $authenticated_id is not set for local
1168 submissions from trusted users.
1170 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1171 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1173 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1174 by adding some example configuration directives to the default
1175 configuration file. A little bit of work is required to uncomment the
1176 directives and define how usernames and passwords are checked, but
1177 there is now a framework to start from.
1179 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1180 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1181 without this. I don't know how relevant this is to other LDAP libraries.
1183 PH/02 Add the verb name to the "unknown ACL verb" error.
1185 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1187 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1189 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1190 directory not expanded when it should be if an expanded home directory
1191 was set for the address (which is overridden by the transport).
1193 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1196 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1197 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1198 because it is too late at that time, and has no effect.
1200 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1201 security issue with \' (bugzilla #107). I could not use the
1202 PQescapeStringConn() function, because it needs a PGconn value as one of
1205 PH/08 When testing addresses using -bt, indicate those final addresses that
1206 are duplicates that would not cause an additional delivery. At least one
1207 person was confused, thinking that -bt output corresponded to deliveries.
1208 (Suppressing duplicates isn't a good idea as you lose the information
1209 about possibly different redirections that led to the duplicates.)
1211 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1212 systems where poll() doesn't work, in particular OS X.
1214 PH/10 Added more information to debugging output for retry time not reached.
1216 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1217 operations in malware.c.
1219 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1222 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1223 syslog_duplication set false, log lines that would normally be written
1224 both the the main log and to the reject log were not written to syslog at
1227 PH/14 In the default configuration, change the use of "message" in ACL warn
1228 statements to "add_header".
1230 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1231 not followed by a command (e.g. "seen endif").
1233 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1234 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1237 PH/17 Added extra conditions to the default value of delay_warning_condition
1241 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1242 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1243 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1246 The Auto-Submitted: and various List- headers are standardised, whereas I
1247 don't think Precedence: ever was.
1249 PH/18 Refactored debugging code in route_finduser() to show more information,
1250 in particular, the error code if getpwnam() issues one.
1252 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1253 This is apparently needed in addition to the PH/07 change above to avoid
1254 any possible encoding problems.
1256 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1257 but not after initializing Perl.
1259 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1260 output them only if debugging. By default they are written stderr,
1261 apparently, which is not desirable.
1263 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1266 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1269 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1271 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1272 authenticated or an ident call has been made. Suppress the default
1273 values for $authenticated_id and $authenticated_sender (but permit -oMai
1274 and -oMas) when testing with -bh.
1276 PH/24 Re-jigged the order of the tests in the default configuration so that the
1277 tests for valid domains and recipients precede the DNS black list and CSA
1278 tests, on the grounds that those ones are more expensive.
1280 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1281 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1282 This bug exists in every version of Exim that I still have, right back to
1285 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1286 However, an attempt to turn on case-sensitivity in a regex key by
1287 including (?-i) didn't work because the subject string was already
1288 lowercased, and the effects were non-intuitive. It turns out that a
1289 one-line patch can be used to allow (?-i) to work as expected.
1295 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1296 other effects) broke the use of negated acl sub-conditions.
1298 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1301 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1302 "Deny" causes Exim to reject the incoming connection with a 554 error.
1303 Unfortunately, if there is a major crisis, such as a disk failure,
1304 tcp-wrappers gives "deny", whereas what one would like would be some
1305 kind of temporary error. A kludge has been added to help with this.
1306 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1307 554 error is used if errno is still zero or contains ENOENT (which occurs
1308 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1311 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1313 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1314 errors. Otherwise a message that provokes a temporary error (when other
1315 messages do not) can cause a whole host to time out.
1317 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1318 addresses were routed directly to files or pipes from a redirect router.
1319 File deliveries just didn't batch; pipe deliveries might have suffered
1322 PH/06 A failure to get a lock for a hints database would erroneously always say
1323 "Failed to get write lock", even when it was really a read lock.
1325 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1326 of 0600. This has been changed to use the value of the lockfile_mode
1327 option (which defaults to 0600).
1329 PH/08 Applied small patch from the Sieve maintainer.
1331 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1332 folder from quota calculations, a direct delivery into this folder messed
1333 up the contents of the maildirsize file. This was because the regex was
1334 used only to exclude .Trash (or whatever) when the size of the mailbox
1335 was calculated. There was no check that a delivery was happening into an
1336 excluded directory. This bug has been fixed by ignoring all quota
1337 processing for deliveries into excluded directories.
1339 PH/10 Added the maildirfolder_create_regex option to appendfile.
1345 PH/01 The code for finding all the local interface addresses on a FreeBSD
1346 system running IPv6 was broken. This may well have applied to all BSD
1347 systems, as well as to others that have similar system calls. The broken
1348 code found IPv4 interfaces correctly, but gave incorrect values for the
1349 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1350 that it would not match correctly against @[] and not recognize the IPv6
1353 PH/02 The ipliteral router was not recognizing addresses of the form user@
1354 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1356 PH/03 Added disable_ipv6.
1358 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1359 decoded form, because it is most often used to construct To: headers
1360 lines in autoreplies, and the decoded form may well be syntactically
1361 invalid. However, $reply_address has leading white space removed, and all
1362 newlines turned into spaces so that the autoreply transport does not
1365 PH/05 If group was specified without a user on a router, and no group or user
1366 was specified on a transport, the group from the router was ignored.
1368 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1369 for visible compile-time settings that can be used to change these
1370 numbers, for those that want even more. Backwards compatibility with old
1371 spool files has been maintained. However, going back to a previous Exim
1372 release will lost any variables that are in spool files.
1374 PH/07 Two small changes when running in the test harness: increase delay when
1375 passing a TCP/IP connection to a new process, in case the original
1376 process has to generate a bounce, and remove special handling of
1377 127.0.0.2 (sic), which is no longer necessary.
1379 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1380 be the same on different OS.
1382 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1385 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1386 whether --show-vars was specified or not
1388 JJ/02 exipick: Added support for new ACL variable spool format introduced
1391 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1392 syntactically invalid From: or Reply-to: line, and a filter used this to
1393 generate an autoreply, and therefore failed to obtain an address for the
1394 autoreply, Exim could try to deliver to a non-existent relative file
1395 name, causing unrelated and misleading errors. What now happens is that
1396 it logs this as a hard delivery error, but does not attempt to create a
1399 PH/11 The exinext utility has a -C option for testing purposes, but although
1400 the given file was scanned by exinext itself; it wasn't being passed on
1401 when Exim was called.
1403 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1404 an end-of-file indication when reading a command response.
1406 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1407 compiled. In many other places in Exim, IPv6 addresses are always
1408 recognized, so I have changed this. It also means that IPv4 domain
1409 literals of the form [IPV4:n.n.n.n] are now always recognized.
1411 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1412 used if the router is not running as root, for example, when verifying at
1413 ACL time, or when using -bh. The debugging output from this situation was
1414 non-existent - all you got was a failure to exec. I have made two
1417 (a) Failures to set uid/gid, the current directory, or a process leader
1418 in a subprocess such as that created by queryprogram now generate
1419 suitable debugging ouput when -d is set.
1421 (b) The queryprogram router detects when it is not running as root,
1422 outputs suitable debugging information if -d is set, and then runs
1423 the subprocess without attempting to change uid/gid.
1425 PH/15 Minor change to Makefile for building test_host (undocumented testing
1428 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1429 additional section of a DNS packet that returns MX or SRV records.
1430 Instead, it always explicitly searches for A/AAAA records. This avoids
1431 major problems that occur when a DNS server includes only records of one
1432 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1433 fixed another bug: if SRV records were looked up and the corresponding
1434 address records were *not* found in the additional section, the port
1435 values from the SRV records were lost.
1437 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1438 using the correct key (the original address) when searching the retry
1439 rules in order to find which one to use for generating the retry hint.
1441 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1442 adding the default one. Similarly, if it contains a Reply-To: header, the
1443 errors_reply_to option, if set, is not used.
1445 PH/19 When calculating a retry time, Exim used to measure the "time since
1446 failure" by looking at the "first failed" field in the retry record. Now
1447 it does not use this if it is later than than the arrival time of the
1448 message. Instead it uses the arrival time. This makes for better
1449 behaviour in cases where some deliveries succeed, thus re-setting the
1450 "first failed" field. An example is a quota failure for a huge message
1451 when small messages continue to be delivered. Without this change, the
1452 "time since failure" will always be short, possible causing more frequent
1453 delivery attempts for the huge message than are intended.
1454 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1456 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1457 $1, $2, $3) because the numerical variables can be reset during some
1458 expansion items (e.g. "match"), thereby losing the authentication data.
1460 PH/21 Make -bV show the size of off_t variables so that the test suite can
1461 decide whether to run tests for quotas > 2G.
1463 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1464 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1465 filecount value is greater than 2G or if a quota value is greater than 2G
1466 on a system where the size of off_t is not greater than 4, a panic error
1469 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1470 never match. The debug and -bh output now contains an explicit error
1471 message indicating a malformed IPv4 address or mask.
1473 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1474 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1475 PH/23 above applies.
1477 PH/25 Do not write to syslog when running in the test harness. The only
1478 occasion when this arises is a failure to open the main or panic logs
1479 (for which there is an explicit test).
1481 PH/26 Added the /no_tell option to "control=freeze".
1483 PH/27 If a host name lookup failed very early in a connection, for example, if
1484 the IP address matched host_lookup and the reverse lookup yielded a name
1485 that did not have a forward lookup, an error message of the form "no IP
1486 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1487 could be logged. Now it outputs the IP address instead of "NULL".
1489 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1490 allows the sender and the authenticated sender to be set when
1491 submitting a message from within Exim. Since child_open_exim() is
1492 documented for local_scan(), the new function should be too.
1494 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1495 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1496 results in an empty string is now treated as unset.
1498 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1500 PH/31 Added sender_verify_fail logging option.
1502 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1503 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1504 tidied the source and removed it altogether.
1506 PH/33 When a queue run was abandoned because the load average was too high, a
1507 log line was always written; now it is written only if the queue_run log
1508 selector is set. In addition, the log line for abandonment now contains
1509 information about the queue run such as the pid. This is always present
1510 in "start" and "stop" lines but was omitted from the "abandon" line.
1512 PH/34 Omit spaces between a header name and the colon in the error message that
1513 is given when verify = headers_syntax fails (if there are lots of them,
1514 the message gets confusing).
1516 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1517 names, as there are now some PTR records that contain slashes. This check
1518 is only to protect against broken name servers that fall over on strange
1519 characters, so the fact that it applies to all lookups doesn't matter.
1521 PH/36 Now that the new test suite is complete, we can remove some of the
1522 special code in Exim that was needed for the old test suite. For example,
1523 sorting DNS records because real resolvers return them in an arbitrary
1524 order. The new test suite's fake resolver always returns records in the
1527 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1528 bounces) except when queue_only is set, to avoid logging races between
1529 the different processes.
1531 PH/38 Panic-die if .include specifies a non-absolute path.
1533 PH/39 A tweak to the "H" retry rule from its user.
1535 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1536 a label. They prevented compilation on older perls.
1538 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1539 a warning to be raised on newish perls.
1541 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1542 on queue. Changes to match documented behaviour of showing count of
1543 messages matching specified criteria.
1545 PH/40 Changed the default ident timeout from 30s to 5s.
1547 PH/41 Added support for the use of login_cap features, on those BSD systems
1548 that have them, for controlling the resources used by pipe deliveries.
1550 PH/42 The content-scanning code uses fopen() to create files in which to put
1551 message data. Previously it was not paying any attention to the mode of
1552 the files. Exim runs with umask(0) because the rest of the code creates
1553 files with open(), and sets the required mode explicitly. Thus, these
1554 files were ending up world-writeable. This was not a big issue, because,
1555 being within the spool directory, they were not world-accessible. I have
1556 created a function called modefopen, which takes an additional mode
1557 argument. It sets umask(777), creates the file, chmods it to the required
1558 mode, then resets the umask. All the relevant calls to fopen() in the
1559 content scanning code have been changed to use this function.
1561 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1562 to 24 hours. This avoids potential overflow problems when processing G
1563 and H retry rules. I suspect nobody ever tinkers with this value.
1565 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1567 PH/45 When the plaintext authenticator is running as a client, the server's
1568 challenges are checked to ensure they are valid base64 strings. By
1569 default, the authentication attempt is cancelled if an invalid string is
1570 received. Setting client_ignore_invalid_base64 true ignores these errors.
1571 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1572 they are received. Thus, the responses can be made to depend on the
1573 challenges. If an invalid string is ignored, an empty string is placed in
1576 PH/46 Messages that are created by the autoreply transport now contains a
1577 References: header, in accordance with RFCs 2822 and 3834.
1579 PH/47 Added authenticated_sender_force to the smtp transport.
1581 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1583 PH/49 Installed latest patch from the Sieve maintainer.
1585 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1586 was also set, the appendfile transport was unnecessarily scanning a
1587 directory of message files (e.g. for maildir delivery) to find the count
1588 of files (along with the size), even though it did not need this
1589 information. It now does the scan only if it needs to find either the
1590 size of the count of files.
1592 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1594 PH/52 Two bugs concerned with error handling when the smtp transport is
1597 (i) Exim was not creating retry information for temporary errors given
1598 for individual recipients after the DATA command when the smtp transport
1599 was used in LMTP mode. This meant that they could be retried too
1600 frequently, and not timed out correctly.
1602 (ii) Exim was setting the flag that allows error details to be returned
1603 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1604 recipients that were returned after the DATA command.
1606 PH/53 This is related to PH/52, but is more general: for any failing address,
1607 when detailed error information was permitted to be returned to the
1608 sender, but the error was temporary, then after the final timeout, only
1609 "retry timeout exceeded" was returned. Now it returns the full error as
1610 well as "retry timeout exceeded".
1612 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1613 do this, and (what is worse) MTAs that accept it.
1615 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1616 will now be deprecated.
1618 PH/56 New os.c-cygwin from the Cygwin maintainer.
1620 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1621 formats (previously only available in exim formats via -bpr, -bpru,
1622 and -bpra. Now also available in native and exiqgrep formats)
1624 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1625 with very large, slow to parse queues
1627 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1629 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1631 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1632 responses to authentication challenges, though it was showing the
1633 challenges; (ii) I've removed the CR characters from the debug output for
1636 PH/58 Allow for the insertion of a newline as well as a space when a string
1637 is turned into more than one encoded-word during RFC 2047 encoding. The
1638 Sieve code now uses this.
1640 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1641 data_4xx, lost_connection, tls_required.
1643 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1644 message was being sent as an SMTP response.
1646 PH/61 Add -l and -k options to exicyclog.
1648 PH/62 When verifying, if an address was redirected to one new address, so that
1649 verification continued, and the new address failed or deferred after
1650 having set something in $address_data, the value of $address_data was not
1651 passed back to the ACL. This was different to the case when no
1652 redirection occurred. The value is now passed back in both cases.
1654 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1655 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1656 use login_cap.h, so on its own it isn't the distinguishing feature. The
1657 new name refers directly to the setclassresources() function.
1659 PH/65 Added configuration files for NetBSD3.
1661 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1663 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1664 is preferred over IPv4.
1666 PH/68 The bounce_return_message and bounce_return_body options were not being
1667 honoured for bounces generated during the reception of non-SMTP messages.
1668 In particular, this applied to messages rejected by the ACL. This bug has
1669 been fixed. However, if bounce_return_message is true and bounce_return_
1670 body is false, the headers that are returned for a non-SMTP message
1671 include only those that have been read before the error was detected.
1672 (In the case of an ACL rejection, they have all been read.)
1674 PH/69 The HTML version of the specification is now built in a directory called
1675 spec_html instead of spec.html, because the latter looks like a path with
1676 a MIME-type, and this confuses some software.
1678 PH/70 Catch two compiler warnings in sieve.c.
1680 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1681 function verify_get_ident() calls ip_connect() to connect a socket, but
1682 if the "connect()" function timed out, ip_connect() used to close the
1683 socket. However, verify_get_ident() also closes the socket later, and in
1684 between Exim writes to the log, which may get opened at this point. When
1685 the socket was closed in ip_connect(), the log could get the same file
1686 descriptor number as the socket. This naturally causes chaos. The fix is
1687 not to close the socket in ip_connect(); the socket should be closed by
1688 the function that creates it. There was only one place in the code where
1689 this was missing, in the iplookup router, which I don't think anybody now
1690 uses, but I've fixed it anyway.
1692 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1693 well as to direct DNS lookups. Otherwise the handling of names in host
1694 lists is inconsistent and therefore confusing.
1700 PH/01 Two changes to the default runtime configuration:
1702 (1) Move the checks for relay_from_hosts and authenticated clients from
1703 after to before the (commented out) DNS black list checks.
1705 (2) Add control=submission to the relay_from_hosts and authenticated
1706 clients checks, on the grounds that messages accepted by these
1707 statements are most likely to be submissions.
1709 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1711 (1) Generate an error if the third argument for the ${prvs expansion is
1714 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1717 (3) Reset the variables that are obtained from the first argument of
1718 ${prvscheck and used in the second argument before leaving the code,
1719 because their memory is reclaimed, so using them afterwards may do
1722 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1723 one (it's much easier than Tom thought :-).
1725 (5) Because of (4), we can now allow for the use of $prvscheck_result
1726 inside the third argument.
1728 PH/03 For some reason, the default setting of PATH when running a command from
1729 a pipe transport was just "/usr/bin". I have changed it to
1732 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1733 anything to be listed in the output from -bV.
1735 PH/05 When a filter generated an autoreply, the entire To: header line was
1736 quoted in the delivery log line, like this:
1738 => >A.N.Other <ano@some.domain> <original@ddress> ...
1740 This has been changed so that it extracts the operative address. There
1741 may be more than one such address. If so, they are comma-separated, like
1744 => >ano@some.domain,ona@other.domain <original@ddress> ...
1746 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1747 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1748 not being looked up in the rDNS to get a host name, Exim was showing the
1749 IP address twice in Received: lines, even though the IP addresses were
1750 identical. For example:
1752 Received: from [1.2.3.4] (helo=[1.2.3.4])
1754 However, if the real host name was known, it was omitting the HELO data
1755 if it matched the actual IP address. This has been tidied up so that it
1756 doesn't show the same IP address twice.
1758 PH/07 When both +timestamp and +memory debugging was on, the value given by
1759 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1760 called by the debug printing, thereby overwriting the timestamp buffer.
1761 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1763 PH/08 When the original message was included in an autoreply transport, it
1764 always said "this is a copy of the message, including all the headers",
1765 even if body_only or headers_only was set. It now gives an appropriate
1768 PH/09 Applied a patch from the Sieve maintainer which:
1770 o fixes some comments
1771 o adds the (disabled) notify extension core
1772 o adds some debug output for the result of if/elsif tests
1773 o points to the current vacation draft in the documentation
1774 and documents the missing references header update
1778 o fixes a bug in processing the envelope test (when testing
1779 multiple envelope elements, the last element determinted the
1782 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1783 Electronic Mail") by including:
1785 Auto-submitted: auto-generated
1787 in the messages that it generates (bounce messages and others, such as
1788 warnings). In the case of bounce messages for non-SMTP mesages, there was
1789 also a typo: it was using "Auto_submitted" (underscore instead of
1790 hyphen). Since every message generated by Exim is necessarily in response
1791 to another message, thes have all been changed to:
1793 Auto-Submitted: auto-replied
1795 in accordance with these statements in the RFC:
1797 The auto-replied keyword:
1799 - SHOULD be used on messages sent in direct response to another
1800 message by an automatic process,
1802 - MUST NOT be used on manually-generated messages,
1804 - MAY be used on Delivery Status Notifications (DSNs) and Message
1805 Disposition Notifications (MDNs),
1807 - MUST NOT be used on messages generated by automatic or periodic
1808 processes, except for messages which are automatic responses to
1811 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1812 to the default Received: header definition.
1814 PH/12 Added log selector acl_warn_skipped (default on).
1816 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1817 variables because (a) they are in the wrong storage pool and (b) even if
1818 they were copied, it wouldn't work properly because of the caching.
1820 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1821 checking when decoding. Apparently there are clients that generate
1822 overlong encoded strings. Why am I not surprised?
1824 PH/15 If the first argument of "${if match_address" was not empty, but did not
1825 contain an "@" character, Exim crashed. Now it writes a panic log message
1826 and treats the condition as false.
1828 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1830 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1831 extension "envelope-auth". The code is finished and in agreement with
1832 other implementations, but there is no documentation so far and in fact,
1833 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1834 not changing the active code.
1836 Print executed "if" and "elsif" statements when debugging is used. This
1837 helps a great deal to understand what a filter does.
1839 Document more things not specified clearly in RFC3028. I had all this
1840 sorted out, when out of a sudden new issues came to my mind. Oops."
1842 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1845 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1846 canonical form (as documented). However, after a host name lookup from
1847 the IP address, check_host() was doing a simple string comparison with
1848 addresses acquired from the DNS when checking that the found name did
1849 have the original IP as one of its addresses. Since any found IPv6
1850 addresses are likely to be in abbreviated form, the comparison could
1851 fail. Luckily, there already exists a function for doing the comparison
1852 by converting both addresses to binary, so now that is used instead of
1853 the text comparison.
1855 PH/20 There was another similar case to PH/19, when a complete host name was
1856 given in a host list; looking up its IP address could give an abbreviated
1857 form, whereas the current host's name might or might not be abbreviated.
1858 The same fix has been applied.
1864 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1865 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1868 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1869 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1871 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1873 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1874 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1875 Also, for "Auto-Submitted", treat anything other than "no" as
1876 non-personal, in accordance with RFC 3834. (Previously it treated
1877 anything starting "auto-" as non-personal.)
1879 TF/01 The control=submission/name=... option had a problem with syntax
1880 errors if the name included a slash character. The /name= option
1881 now slurps the rest of the string, so it can include any characters
1882 but it must come last in the list of options (after /sender_retain
1885 PH/05 Some modifications to the interface to the fake nameserver for the new
1893 TK/01 Added the "success_on_redirect" address verification option. See
1894 NewStuff for rationale and an example.
1896 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1898 PH/02 Patch to exigrep to allow it to work on syslog lines.
1900 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1901 fread() to skip over the body file's header line, because in Cygwin the
1902 header line is locked and is inaccessible.
1904 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1905 co-exist for some time) to make it clear that it is the Exim ID that is
1906 referenced, not the Message-ID: header line.
1908 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1909 string_format() function, because snprintf() does not exist on all
1912 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1913 ${stat: expansion item.
1915 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1916 protocol synchronization error", to keep the pedants happy.
1918 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1919 well as for IRIX systems, when gcc is being used. See the host.c source
1922 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1924 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1927 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1928 transport and to the smtp transport in LMTP mode.
1930 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1932 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1933 run a filter in a subprocess. This could lead to confusion in subsequent
1934 lookups in the parent process. There should also be a search_tidyup() at
1935 the end of the subprocess.
1937 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1938 only if the host matched helo_try_verify_hosts, which caused the
1939 verification to occur when the EHLO/HELO command was issued. The ACL just
1940 tested the remembered result. Now, if a previous verification attempt has
1941 not happened, "verify = helo" does it there and then.
1943 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1945 TK/03 Fix log output including CR from clamd.
1947 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1948 exist provoked a memory error which could cause a segfault.
1950 PH/15 Installed PCRE 6.2
1952 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1954 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1955 of the problem. Specifically, suggested +O2 rather than +O1 for the
1958 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1960 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1961 record, the port was not correctly logged when the outgoing_port log
1962 selector was set (it logged the transort's default port).
1964 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1965 fallback_hosts, and "hosts" in the smtp transport.
1967 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1968 host errors such as "Connection refused".
1970 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1971 authentication with radiusclient 0.4.9:
1973 - Error returned from rc_read_config was caught wrongly
1974 - Username/password not passed on to radius server due to wrong length.
1976 The presumption is that some radiusclient API changes for 4.51/PH/17
1977 were not taken care of correctly. The code is still untested by me (my
1978 Linux distribution still has 0.3.2 of radiusclient), but it was
1979 contributed by a Radius user.
1981 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1982 expanding the "port" option of the smtp transport.
1984 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1985 while reading a MIME header. Thanks to Tom Hughes for a patch.
1987 PH/24 Include config.h inside local_scan.h so that configuration settings are
1990 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1991 that in an ACL for RCPT (for example), you can examine exactly what was
1994 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1995 commands, but it was not correctly comparing the address with the actual
1996 client host address. Thus, it would show the EHLO address in Received:
1997 header lines when this was not necessary.
1999 PH/27 Added the % operator to ${eval:}.
2001 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2002 it should be ignoring failures (because with -C, for example, it has lost
2003 privilege). It wasn't ignoring creation failures other than "already
2006 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2009 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2010 because an input line was too long, either on its own, or by virtue of
2011 too many continuations, the temporary file was not being removed, and the
2012 return code was incorrect.
2014 PH/31 Missing "BOOL" in function definition in filtertest.c.
2016 PH/32 Applied Sieve patches from the maintainer.
2018 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2020 PH/33 Added "verify = not_blind".
2022 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2023 Local/Makefile (with some defaults set). These are used in built scripts
2024 such as exicyclog, but they have never been used in the exim_install
2025 script (though there are many overriding facilities there). I have
2026 arranged that the exim_install script now takes note of these two
2029 PH/35 Installed configuration files for Dragonfly.
2031 PH/36 When a locally submitted message by a trusted user did not contain a
2032 From: header, and the sender address was obtained from -f or from an SMTP
2033 MAIL command, and the trusted user did not use -F to supply a sender
2034 name, $originator_name was incorrectly used when constructing a From:
2035 header. Furthermore, $originator_name was used for submission mode
2036 messages from external hosts without From: headers in a similar way,
2037 which is clearly wrong.
2039 PH/37 Added control=suppress_local_fixups.
2041 PH/38 When log_selector = +received_sender was set, and the addition of the
2042 sender made the log line's construction buffer exactly full, or one byte
2043 less than full, an overflow happened when the terminating "\n" was
2046 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2047 when the result of a list match is failure because a DNS lookup failed.
2049 PH/40 RM_COMMAND is now used in the building process.
2051 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2052 the "build-* directories that it finds.
2054 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2055 address in a domain literal was a prefix of an interface address.
2057 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2058 when verifying a sender address, unless rewrite_headers is false.
2060 PH/44 Wrote a long comment about why errors_to addresses are verified as
2061 recipients, not senders.
2063 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2064 the ratelimit ACL was added.
2066 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2068 PH/46 Added extra information about PostgreSQL errors to the error string.
2070 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2071 avoiding the need to install special zones in a real server. This is
2072 backwards compatible; if it can't find the fake resolver, it drops back.
2073 Thus, both old and new test suites can be run.
2075 TF/02 Added util/ratelimit.pl
2077 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2078 clock is set back in time.
2080 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2081 Candler <B.Candler@pobox.com>.
2083 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2084 OK for addresses that are the result of redirections.
2086 PH/48 A number of further additions for the benefit of the new test suite,
2087 including a fake gethostbyname() that interfaces to the fake DNS resolver
2090 TF/06 The fix for widen_domains has also been applied to qualify_single and
2091 search_parents which are the other dnslookup options that can cause
2094 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2097 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2099 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2100 service" if it encountered an MX record with an empty target hostname.
2101 The message is now "an MX or SRV record indicated no SMTP service".
2103 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2104 if the DNS of the sending site is misconfigured. This is quite a
2105 common situation. This change restores the behaviour of treating a
2106 helo verification defer as a failure.
2108 PH/51 If self=fail was set on a router, the bounce message did not include the
2109 actual error message.
2115 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2117 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2118 command itself ended in error, the underlying message about the transport
2119 filter timeout was being overwritten with the pipe command error. Now the
2120 underlying error message should be appended to the second error message.
2122 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2124 PH/02 Reduce the amount of output that "make" produces by default. Full output
2125 can still be requested.
2127 PH/03 The warning log line about a condition test deferring for a "warn" verb
2128 was being output only once per connection, rather than after each
2129 occurrence (because it was using the same function as for successful
2130 "warn" verbs). This seems wrong, so I have changed it.
2132 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2133 it should not have, which might have caused a crash in the right
2134 circumstances, but probably never did.
2136 PH/04 Installed a modified version of Tony Finch's patch to make submission
2137 mode fix the return path as well as the Sender: header line, and to
2138 add a /name= option so that you can make the user's friendly name appear
2141 TF/03 Added the control = fakedefer ACL modifier.
2143 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2144 Mark Lowes for thorough testing.
2146 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2148 TK/03 Merged latest SRS patch from Miles Wilton.
2150 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2151 with the definition in sysexits.h (which is #included earlier).
2152 Fortunately, Exim does not actually use EX_OK. The code used to try to
2153 preserve the sysexits.h value, by assumimg that macro definitions were
2154 scanned for macro replacements. I have been disabused of this notion,
2155 so now the code just undefines EX_OK before #including unistd.h.
2157 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2158 in the smtp transport. When a block could not be written in a single
2159 write() function, the timeout was being re-applied to each part-write.
2160 This seems wrong - if the receiver was accepting one byte at a time it
2161 would take for ever. The timeout is now adjusted when this happens. It
2162 doesn't have to be particularly precise.
2164 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2165 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2167 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2168 without a domain if the check to <postmaster@domain> fails.
2170 SC/01 Eximstats: added -xls and the ability to specify output files
2171 (patch written by Frank Heydlauf).
2173 SC/02 Eximstats: use FileHandles for outputing results.
2175 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2177 SC/04 Eximstats: fixed display of large numbers with -nvr option
2179 SC/05 Eximstats: fixed merging of reports with empty tables.
2181 SC/06 Eximstats: added the -include_original_destination flag
2183 SC/07 Eximstats: removed tabs and trailing whitespace.
2185 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2187 TK/06 MBOX spool code: Add real "From " MBOX separator line
2188 so the .eml file is really in mbox format (even though
2189 most programs do not really care). Patch from Alex Miller.
2191 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2192 The latter is generated from $received_to and is only set if the
2193 message has one envelope recipient. SA can use these headers,
2194 obviously out-of-the-box. Patch from Alex Miller.
2196 PH/08 The ${def test on a variable was returning false if the variable's
2197 value was "0", contrary to what the specification has always said!
2198 The result should be true unless the variable is empty.
2200 PH/09 The syntax error of a character other than { following "${if
2201 def:variable_name" (after optional whitespace) was not being diagnosed.
2202 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2203 accidental colon was present, for example, could give incorrect results.
2205 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2206 result is used (not including appendfile, where other changes are about
2209 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2210 This involved changing a lot of size variables from int to off_t. It
2211 should work with maildirs and everything.
2213 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2214 spamd dying while we are connected to it.
2216 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2219 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2220 function for BDB 4.3.
2222 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2224 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2225 the sources for PCRE have been re-arranged and more files are now
2228 PH/15 The code I had for printing potentially long long variables in PH/11
2229 above was not the best (it lost precision). The length of off_t variables
2230 is now inspected at build time, and an appropriate printing format (%ld
2231 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2232 to be "long long int" or "long int". This is needed for the internal
2233 formatting function string_vformat().
2235 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2236 the configuration file to be ":syslog", then the script "guesses" where
2237 the logs files are, rather than using the compiled in default. In our
2238 case the guess is not the same as the compiled default, so the script
2239 suddenly stopped working when I started to use syslog. The patch checks
2240 to see if log_file_path is "". If so, it attempts to read it from exim
2241 with no configuration file to get the compiled in version, before it
2242 falls back to the previous guessing code."
2244 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2245 implementing BATV in an Exim configuration. See NewStuff for the gory
2248 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2249 Makefile that are specific to HP-UX.
2251 PH/18 If the "use_postmaster" option was set for a recipient callout together
2252 with the "random" option, the postmaster address was used as the MAIL
2253 FROM address for the random test, but not for the subsequent recipient
2254 test. It is now used for both.
2256 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2257 patch removes a few documentation additions to RFC 3028, because the
2258 latest draft now contains them. It adds the new en;ascii-case comparator
2259 and a new error check for 8bit text in MIME parts. Comparator and
2260 require names are now matched exactly. I enabled the subaddress
2261 extension, but it is not well tested yet (read: it works for me)."
2263 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2264 rework some of the code of TK/09 above to avoid the hardwired use of
2265 "%lld" and "long long". Replaced the call to snprintf() with a call to
2268 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2269 records point to non-existent hosts", "retry timeout exceeded", and
2270 "retry time not reached for any host after a long failure period".
2272 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2273 experimental DomainKeys support:
2275 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2276 (2) On an error such as an illegally used "control", the wrong name for
2277 the control was given.
2279 These problems did NOT occur unless DomainKeys support was compiled.
2281 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2283 PH/24 Added ${if match_ip condition.
2285 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2286 databases so that it will be absolutely obvious if a crash occurs in the
2287 DB library. This is a regular occurrence (often caused by mis-matched
2290 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2291 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2292 picked up on a user's system that detects such things. There doesn't seem
2293 to be a gcc warning option for this - only an attribute that has to be
2294 put on the function's prototype. It seems that in Fedora Core 4 they have
2295 set this on a number of new functions. No doubt there will be more in due
2298 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2299 specify a transport. However, if an address that was verified by such a
2300 router was the subject of a callout, Exim crashed because it tried to
2301 read the rcpt_include_affixes from the non-existent transport. Now it
2302 just assumes that the setting of that option is false. This bug was
2303 introduced by 4.51/PH/31.
2305 PH/28 Changed -d+all to exclude +memory, because that information is very
2306 rarely of interest, but it makes the output a lot bigger. People tend to
2307 do -d+all out of habit.
2309 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2310 code in os-type was giving problems when libc.so lives in lib64, like on
2313 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2314 aren't the modern standard, and it seems that some systems' include files
2315 don't always have them. Exim was already checking for some of the newer
2316 ones like T_AAAA, and defining it itself. I've added checks for all the
2317 record types that Exim uses.
2319 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2320 not automatically generating a new one, as it is supposed to. This
2321 prevented TLS from working. If the file did exist, but contained invalid
2322 data, a new version was generated, as expected. It was only the case of a
2323 non-existent file that was broken.
2325 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2326 with a change in libdomainkeys > 0.64.
2328 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2329 from DNS. If the selector record carries the flag, it now has
2330 precedence over the domain-wide flag.
2332 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2334 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2335 the use of a port name that isn't defined in /etc/services) occurred, the
2336 message was deferred as in a normal delivery, and thus remained on the
2337 spool, instead of being failed because of the mua_wrapper setting. This
2338 is now fixed, and I tidied up some of the mua_wrapper messages at the
2341 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2342 the messages in a hash of arrays rather than using individual hashes.
2343 This is a bit cleaner and results in dramatic memory savings, albeit
2344 at a slight CPU cost.
2346 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2347 as requested by Marc Sherman.
2349 SC/10 Eximstats: added histograms for user specified patterns as requested
2352 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2354 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2355 fopen() in the content-scanning modules that did not already have it.
2361 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2362 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2364 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2366 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2368 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2369 to test Sieve filters that use "vacation".
2371 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2372 that changes the way the GnuTLS parameters are stored in the cache file.
2373 The new format can be generated externally. For backward compatibility,
2374 if the data in the cache doesn't make sense, Exim assumes it has read an
2375 old-format file, and it generates new data and writes a new file. This
2376 means that you can't go back to an older release without removing the
2379 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2380 work if there are any delivery delays because "one_time" forces the
2381 parent to be marked "delivered", so its unseen clone is never tried
2382 again. For this reason, Exim now forbids the simultaneous setting of
2385 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2386 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2387 entry if you want to know the details. The fix, however, neglected to
2388 consider the case when local delivery batching is involved. The test for
2389 "previously delivered" was not happening when checking to see if an
2390 address could be batched with a previous (undelivered) one; under
2391 certain circumstances this could lead to multiple deliveries to the same
2394 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2395 in its include files, and this causes problems building Exim.
2397 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2398 header_syntax) but Exim was just ignoring anything given after a slash.
2399 In particular, this caused confusion with an attempt to use "verify =
2400 reverse_host_lookup/defer_ok". An error is now given when options are
2401 supplied for verify items that do not have them. (Maybe reverse_host_
2402 lookup should have a defer_ok option, but that's a different point.)
2404 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2405 defined by RFC 821) to 2048, because there were problems with some AUTH
2406 commands, and RFC 1869 says the size should be increased for extended
2407 SMTP commands that take arguments.
2409 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2412 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2413 "unknown" error; now it says that the functionality isn't in the binary.
2415 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2416 an address' error message when a string expansion fails (syntax or
2417 whatever). Otherwise the password may appear in the log. Following change
2418 PH/42 below, there is no longer a chance of it appearing in a bounce
2421 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2423 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2424 of its addresses was ever tried. (Bugzilla bug #2.)
2426 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2427 the result incorrectly in the debug output. (It correctly added a newline
2428 to what was transported.)
2430 TF/01 Added $received_time.
2432 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2433 commented out examples of how to interface to a virus scanner and to
2434 SpamAssassin. Also added commented examples of av_scanner and
2435 spamd_address settings.
2437 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2438 and controls are allowed in which ACLs. There were a couple of minor
2439 errors. Some of the entries in the conditions table (which is a table of
2440 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2441 negation of where the condition IS allowed.
2443 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2445 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2446 header file does not have a version number, so I've had to invent a new
2447 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2448 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2449 radiusclient), but it was contributed by a Radius user.
2451 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2452 files or directories, for OpenSSL.
2454 PH/19 When an Exim process that is running as root has to create an Exim log
2455 file, it does so in a subprocess that runs as exim:exim so as to get the
2456 ownership right at creation (otherwise, other Exim processes might see
2457 the file with the wrong ownership). There was no test for failure of this
2458 fork() call, which would lead to the process getting stuck as it waited
2459 for a non-existent subprocess. Forks do occasionally fail when resources
2460 run out. I reviewed all the other calls to fork(); they all seem to check
2463 PH/20 When checking for unexpected SMTP input at connect time (before writing
2464 the banner), Exim was not dealing correctly with a non-positive return
2465 from the read() function. If the client had disconnected by this time,
2466 the result was a log entry for a synchronization error with an empty
2467 string after "input=" when read() returned zero. If read() returned -1
2468 (an event I could not check), uninitialized data bytes were printed.
2469 There were reports of junk text (parts of files, etc) appearing after
2472 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2474 PH/22 Added support for macro redefinition, and (re)definition in between
2475 driver and ACL definitions.
2477 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2478 forgetting to use the resulting value; it was using the unexpanded value.
2480 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2481 hadn't been configured. The fix is from Juergen Kreileder, who
2482 understands it better than I do:
2484 "Here's what I see happening with three configured cyrus_sasl
2485 authenticators configured (plain, login, cram-md5):
2487 On startup auth_cyrus_sasl_init() gets called for each of these.
2488 This means three calls to sasl_listmech() without a specified mech_list.
2489 => SASL tests which mechs of all available mechs actually work
2490 => three warnings about OTP not working
2491 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2493 With the patch, sasl_listmech() also gets called three times. But now
2494 SASL's mech_list option is set to the server_mech specified in the the
2495 authenticator. Or in other words, the answer from sasl_listmech()
2496 gets limited to just the mech you're testing for (which is different
2498 => the return list contains just 'plain' or 'login', 'cram-md5' or
2499 nothing depending on the value of ob->server_mech.
2501 I've just tested the patch: Authentication still works fine,
2502 unavailable mechs specified in the exim configuration are still
2503 caught, and the auth.log warnings about OTP are gone."
2505 PH/25 When debugging is enabled, the contents of the command line are added
2506 to the debugging output, even when log_selector=+arguments is not
2509 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2510 answer is "GNU", and only if the return is "GNU/something" is the answer
2513 PH/27 $acl_verify_message is now set immediately after the failure of a
2514 verification in an ACL, and so is available in subsequent modifiers. In
2515 particular, the message can be preserved by coding like this:
2517 warn !verify = sender
2518 set acl_m0 = $acl_verify_message
2520 Previously, $acl_verify_message was set only while expanding "message"
2521 and "log_message" when a very denied access.
2523 PH/28 Modified OS/os.c-Linux with
2525 -#ifndef OS_LOAD_AVERAGE
2526 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2528 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2529 nomenclature these days.)
2531 PH/29 Installed patch from the Sieve maintainer that adds the options
2532 sieve_useraddress and sieve_subaddress to the redirect router.
2534 PH/30 In these circumstances:
2535 . Two addresses routed to the same list of hosts;
2536 . First host does not offer TLS;
2537 . First host accepts first address;
2538 . First host gives temporary error to second address;
2539 . Second host offers TLS and a TLS session is established;
2540 . Second host accepts second address.
2541 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2542 and peerdn, if requested) that were in fact used only for the second
2545 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2546 attention to any local part prefix or suffix that was matched by the
2547 router that accepted the address. It now behaves in the same way as it
2548 does for delivery: the affixes are removed from the local part unless
2549 rcpt_include_affixes is set on the transport.
2551 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2552 timeout during the DATA phase of an incoming message.
2554 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2555 applied a patch sanctioned by the Sieve maintainer.
2557 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2558 the uid or gid is negative. A case of a negative gid caused this to be
2559 noticed. The fix allows for either to be negative.
2561 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2562 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2565 PH/36 The change PH/12 above was broken. Fixed it.
2567 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2568 the grounds that routing the same address twice would always produce the
2569 same answer. This might have been true once, but it is certainly no
2570 longer true now. Routing a child address may depend on the previous
2571 routing that produced that child. Some complicated redirection strategies
2572 went wrong when messages had multiple recipients, and made Exim's
2573 behaviour dependent on the order in which the addresses were given.
2575 I have moved the duplicate checking until after the routing is complete.
2576 Exim scans the addresses that are assigned to local and remote
2577 transports, and removes any duplicates. This means that more work will be
2578 done, as duplicates will always all be routed, but duplicates are
2579 presumably rare, so I don't expect this is of any significance.
2581 For deliveries to pipes, files, and autoreplies, the duplicate checking
2582 still happens during the routing process, since they are not going to be
2585 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2586 It corrects a timeout issue with spamd. This is Ian's comment: "The
2587 background is that sometimes spamd either never reads data from a
2588 connection it has accepted, or it never writes response data. The exiscan
2589 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2590 blindly assumes that writes won't block so it may never time out."
2592 PH/39 Allow G after quota size as well as K and M.
2594 PH/40 The value set for $authenticated_id in an authenticator may not contain
2595 binary zeroes or newlines because the value is written to log lines and
2596 to spool files. There was no check on this. Now the value is run through
2597 the string_printing() function so that such characters are converted to
2598 printable escape sequences.
2600 PH/41 $message_linecount is a new variable that contains the total number of
2601 lines in the message. Compare $body_linecount, which is the count for the
2604 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2605 bounce and delay warning messages, except in certain special cases, which
2608 (a) An SMTP error message from a remote host;
2609 (b) A message specified in a :fail: redirection;
2610 (c) A message specified in a "fail" command in a system filter;
2611 (d) A message specified in a FAIL return from the queryprogram router;
2612 (e) A message specified by the cannot_route_message router option.
2614 In these cases only, Exim does include the error details in bounce and
2615 warning messages. There are also a few cases where bland messages such
2616 as "unrouteable address" or "local delivery error" are given.
2618 PH/43 $value is now also set for the "else" part of a ${run expansion.
2620 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2621 being worked on, but at least Exim now implements the latest version to
2624 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2625 process to complete was treated as a delivery failure, a timeout while
2626 writing the message to the pipe was logged, but erroneously treated as a
2627 successful delivery. Such timeouts include transport filter timeouts. For
2628 consistency with the overall process timeout, these timeouts are now
2629 treated as errors, giving rise to delivery failures by default. However,
2630 there is now a new Boolean option for the pipe transport called
2631 timeout_defer, which, if set TRUE, converts the failures into defers for
2632 both kinds of timeout. A transport filter timeout is now identified in
2635 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2636 systems where "make" and "gmake" are different, calling "gmake" at top
2637 level broke things. I've arranged for the value of $(MAKE) to be passed
2638 from the Makefile to this script so that it can call the same version of
2642 A note about Exim versions 4.44 and 4.50
2643 ----------------------------------------
2645 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2646 changes of various kinds. As a consequence, a big documentation update was
2647 needed. This delayed the release for rather longer than seemed good, especially
2648 in the light of a couple of (minor) security issues. Therefore, the changes
2649 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2650 release. So 4.44 and 4.50 are in effect two different branches that both start
2653 I have left the 4.50 change log unchanged; it contains all the changes since
2654 4.43. The change log for 4.44 is below; many of its items are identical to
2655 those for 4.50. This seems to be the most sensible way to preserve the
2656 historical information.
2662 1. Minor wording change to the doc/README.SIEVE file.
2664 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2665 computation of the current number of files was incorrect.
2667 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2668 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2671 4. Give more explanation in the error message when the command for a transport
2672 filter fails to execute.
2674 5. There are several places where Exim runs a non-Exim command in a
2675 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2676 was being done only for the command run by the queryprogram router. It is
2677 now done for all such subprocesses. The other cases are: ${run, transport
2678 filters, and the commands run by the lmtp and pipe transports.
2680 6. Added CONFIGURE_GROUP build-time option.
2682 7. Some older OS have a limit of 256 on the maximum number of file
2683 descriptors. Exim was using setrlimit() to set 1000 as a large value
2684 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2685 systems. I've change it so that if it can't get 1000, it tries for 256.
2687 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2688 was an oversight, and furthermore, ever since the addition of extra
2689 controls (e.g. 4.43/32), the checks on when to allow different forms of
2690 "control" were broken. There should now be diagnostics for all cases when a
2691 control that does not make sense is encountered.
2693 9. Added the /retain_sender option to "control=submission".
2695 10. $recipients is now available in the predata ACL (oversight).
2697 11. Tidy the search cache before the fork to do a delivery from a message
2698 received from the command line. Otherwise the child will trigger a lookup
2699 failure and thereby defer the delivery if it tries to use (for example) a
2700 cached ldap connection that the parent has called unbind on.
2702 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2703 of $address_data from the recipient verification was clobbered by the
2704 sender verification.
2706 13. The value of address_data from a sender verification is now available in
2707 $sender_address_data in subsequent conditions in the ACL statement.
2709 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2711 15. Added a new option "connect=<time>" to callout options, to set a different
2714 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2715 was its contents. (It was OK if the option was not defined at all.)
2717 17. A "Completed" log line is now written for messages that are removed from
2718 the spool by the -Mrm option.
2720 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2721 information about exactly what failed.
2723 19. Added -dd to debug only the daemon process.
2725 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2726 handles timeouts, both on the server side and network timeouts. Renamed the
2727 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2729 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2730 It is now set to "smtps".
2732 22. $host_address is now set to the target address during the checking of
2733 ignore_target_hosts.
2735 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2736 being passed; this would have caused $sender_host_name to have been used if
2737 matching the list had actually called for a host name (not very likely,
2738 since this list is usually IP addresses). A host name is now passed as
2741 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2742 code that specifies a non-restarting handler (typically sigaction() in
2743 modern systems) in an attempt to fix a rare and obscure crash bug.
2745 25. Narrowed the window for a race in the daemon that could cause it to ignore
2746 SIGCHLD signals. This is not a major problem, because they are used only to
2747 wake it up if nothing else does.
2749 26. A malformed maildirsize file could cause Exim to calculate negative values
2750 for the mailbox size or file count. Odd effects could occur as a result.
2751 The maildirsize information is now recalculated if the size or filecount
2754 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2755 support for a long time. Removed HAVE_SYS_VFS_H.
2757 28. Installed the latest version of exipick from John Jetmore.
2759 29. In an address list, if the pattern was not a regular expression, an empty
2760 subject address (from a bounce message) matched only if the pattern was an
2761 empty string. Non-empty patterns were not even tested. This was the wrong
2762 because it is perfectly reasonable to use an empty address as part of a
2763 database query. An empty address is now tested by patterns that are
2764 lookups. However, all the other forms of pattern expect the subject to
2765 contain a local part and a domain, and therefore, for them, an empty
2766 address still always fails if the pattern is not itself empty.
2768 30. Exim went into a mad DNS loop when attempting to do a callout where the
2769 host was specified on an smtp transport, and looking it up yielded more
2770 than one IP address.
2772 31. Re-factored the code for checking spool and log partition space into a
2773 function that finds that data and another that does the check. The former
2774 is then used to implement four new variables: $spool_space, $log_space,
2775 $spool_inodes, and $log_inodes.
2777 32. The RFC2047 encoding function was originally intended for short strings
2778 such as real names; it was not keeping to the 75-character limit for
2779 encoded words that the RFC imposes. It now respects the limit, and
2780 generates multiple encoded words if necessary. To be on the safe side, I
2781 have increased the buffer size for the ${rfc2047: expansion operator from
2784 33. It is now permitted to omit both strings after an "if" condition; if the
2785 condition is true, the result is "true". As before, when the second string
2786 is omitted, a false condition yields an empty string. This makes it less
2787 cumbersome to write custom ACL and router conditions.
2789 34. Failure to deliver a bounce message always caused it to be frozen, even if
2790 there was an errors_to setting on the router. The errors_to setting is now
2793 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2794 canonical form (fully expanded) before being placed in
2795 $sender_host_address.
2797 36. The table in the code that translates DNS record types into text (T_A to
2798 "A" for instance) was missing entries for NS and CNAME. It is just possible
2799 that this could have caused confusion if both these types were looked up
2800 for the same domain, because the text type is used as part of Exim's
2801 per-process caching. But the chance of anyone hitting this buglet seems
2804 37. The dnsdb lookup has been extended in a number of ways.
2806 (1) There is a new type, "zns", which walks up the domain tree until it
2807 finds some nameserver records. It should be used with care.
2809 (2) There is a new type, "mxh", which is like "mx" except that it returns
2810 just the host names, not the priorities.
2812 (3) It is now possible to give a list of domains (or IP addresses) to be
2813 looked up. The behaviour when one of the lookups defers can be
2814 controlled by a keyword.
2816 (4) It is now possible to specify the separator character for use when
2817 multiple records are returned.
2819 38. The dnslists ACL condition has been extended: it is now possible to supply
2820 a list of IP addresses and/or domains to be looked up in a particular DNS
2823 39. Added log_selector=+queue_time_overall.
2825 40. When running the queue in the test harness, wait just a tad after forking a
2826 delivery process, to get repeatability of debugging output.
2828 41. Include certificate and key file names in error message when GnuTLS fails
2829 to set them up, because the GnuTLS error message doesn't include the name
2830 of the failing file when there is a problem reading it.
2832 42. Allow both -bf and -bF in the same test run.
2834 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2836 44. The "Exiscan patch" is now merged into the mainline Exim source.
2838 45. Sometimes the final signoff response after QUIT could fail to get
2839 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2840 < 0 before doing a fflush(). This bug looks as though it goes back to the
2841 introduction of TLS in release 3.20, but "sometimes" must have been rare
2842 because the tests only now provoked it.
2844 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2845 (this can affect the format of dates).
2847 47. exim_tidydb, when checking for the continued existence of a message for
2848 which it has found a message-specific retry record, was not finding
2849 messages that were in split spool directories. Consequently, it was
2850 deleting retry records that should have stayed in existence.
2852 48. Steve fixed some bugs in eximstats.
2854 49. The SPA authentication driver was not abandoning authentication and moving
2855 on to the next authenticator when an expansion was forced to fail,
2856 contradicting the general specification for all authenticators. Instead it
2857 was generating a temporary error. It now behaves as specified.
2859 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2860 (the order specifies the preference for clients). The order is now AES256,
2861 AES128, 3DES, ARCFOUR128.
2863 51. Small patch to Sieve code - explicitly set From: when generating an
2866 52. Exim crashed if a remote delivery caused a very long error message to be
2867 recorded - for instance if somebody sent an entire SpamAssassin report back
2868 as a large number of 550 error lines. This bug was coincidentally fixed by
2869 increasing the size of one of Exim's internal buffers (big_buffer) that
2870 happened as part of the Exiscan merge. However, to be on the safe side, I
2871 have made the code more robust (and fixed the comments that describe what
2874 53. Now that there can be additional text after "Completed" in log lines (if
2875 the queue_time_overall log selector is set), a one-byte patch to exigrep
2876 was needed to allow it to recognize "Completed" as not the last thing in
2879 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2880 patch that reportedly fixes this has been added. I am not expert enough to
2881 create a test for it. This is what the patch creator wrote:
2883 "I found a little strange behaviour of ldap code when working with
2884 Windows 2003 AD Domain, where users was placed in more than one
2885 Organization Units. When I tried to give exim partial DN, the exit code
2886 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2887 But simultaneously result of request was absolutely normal ldap result,
2888 so I produce this patch..."
2890 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2891 REFERENCE, so I have modified the code to exclude the patch when that macro
2894 55. Some experimental protocols are using DNS PTR records for new purposes. The
2895 keys for these records are domain names, not reversed IP addresses. The
2896 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2897 leaves it alone. Component reversal etc. now happens only for IP addresses.
2900 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2902 57. Double the size of the debug message buffer (to 2048) so that more of very
2903 long debug lines gets shown.
2905 58. The exicyclog utility now does better if the number of log files to keep
2906 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2908 59. Two changes related to the smtp_active_hostname option:
2910 (1) $smtp_active_hostname is now available as a variable.
2911 (2) The default for smtp_banner uses $smtp_active_hostname instead
2912 of $primary_hostname.
2914 60. The host_aton() function is supposed to be passed a string that is known
2915 to be a valid IP address. However, in the case of IPv6 addresses, it was
2916 not checking this. This is a hostage to fortune. Exim now panics and dies
2917 if the condition is not met. A case was found where this could be provoked
2918 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2919 components; fortuitously, this particular loophole had already been fixed
2920 by change 4.50/55 above.
2922 If there are any other similar loopholes, the new check in host_aton()
2923 itself should stop them being exploited. The report I received stated that
2924 data on the command line could provoke the exploit when Exim was running as
2925 exim, but did not say which command line option was involved. All I could
2926 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2927 running as the user.
2930 61. There was a buffer overflow vulnerability in the SPA authentication code
2931 (which came originally from the Samba project). I have added a test to the
2932 spa_base64_to_bits() function which I hope fixes it.
2935 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2936 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2938 63. The daemon start-up calls getloadavg() while still root for those OS that
2939 need the first call to be done as root, but it missed one case: when
2940 deliver_queue_load_max is set with deliver_drop_privilege. This is
2941 necessary for the benefit of the queue runner, because there is no re-exec
2942 when deliver_drop_privilege is set.
2944 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2945 This has been fixed.
2947 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2948 was in use, was not putting the data itself into the right store pool;
2949 consequently, it could be overwritten for a subsequent message in the same
2950 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2953 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2955 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2956 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2957 the function were treating the return as a boolean value, which happened to
2958 work because 0=false and not-0=true, but is not correct code.
2960 68. The host_aton() function was not handling scoped IPv6 addresses (those
2961 with, for example, "%eth0" on the end) correctly.
2963 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2964 negated items (that is, ~something) in unsigned ints. Some compilers
2965 apparently mutter when there is no cast.
2967 70. If an address verification called from an ACL failed, and did not produce a
2968 user-specific message (i.e. there was only a "system" message), nothing was
2969 put in $acl_verify_message. In this situation, it now puts the system
2972 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2973 session; change 4.31/43 added the unwanted input to the log line - except
2974 that it did not do this in the start of session case. It now does.
2976 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2977 This is wrong and can cause the other end to generate a synchronization
2978 error if it is another Exim or anything else that does the synchronization
2979 check. A QUIT command is no longer sent after a timeout.
2981 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2982 during host lookups.
2984 74. The defer_ok option of callout verification was not working if it was used
2985 when verifying addresses in header lines, that is, for this case:
2987 verify = header_sender/callout=defer_ok
2989 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2990 those file descriptors could be used for SMTP connections. If anything
2991 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2992 could be sent to the SMTP client, causing chaos. The daemon now opens
2993 stdin, stdout, and stderr to /dev/null when it puts itself into the
2996 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2997 log by default. The user can override this with suitable Perl magic.
2999 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3000 the log message when discard triggers, was not working for the DATA ACL or
3001 for the non-SMTP ACL.
3003 78. Error message wording change in sieve.c.
3005 79. If smtp_accept_max_per_host was set, the number of connections could be
3006 restricted to fewer than expected, because the daemon was trying to set up
3007 a new connection before checking whether the processes handling previous
3008 connections had finished. The check for completed processes is now done
3009 earlier. On busy systems, this bug wouldn't be noticed because something
3010 else would have woken the daemon, and it would have reaped the completed
3013 80. If a message was submitted locally by a user whose login name contained one
3014 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3015 It caused a spool format error. I have fixed the spool reading code. A
3016 related problem was that the "from" clause in the Received: line became
3017 illegal because of the space(s). It is now covered by ${quote_local_part.
3019 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3022 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3024 83. Patch from Sieve maintainer to fix unterminated string problem in
3025 "vacation" handling.
3027 84. Some minor changes to the Linux configuration files to help with other
3028 OS variants using glibc.
3030 85. One more patch for Sieve to update vacation handling to latest spec.
3033 ----------------------------------------------------
3034 See the note above about the 4.44 and 4.50 releases.
3035 ----------------------------------------------------
3041 1. Change 4.43/35 introduced a bug that caused file counts to be
3042 incorrectly computed when quota_filecount was set in an appendfile
3045 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3046 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3049 3. Give more explanation in the error message when the command for a transport
3050 filter fails to execute.
3052 4. There are several places where Exim runs a non-Exim command in a
3053 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3054 was being done only for the command run by the queryprogram router. It is
3055 now done for all such subprocesses. The other cases are: ${run, transport
3056 filters, and the commands run by the lmtp and pipe transports.
3058 5. Some older OS have a limit of 256 on the maximum number of file
3059 descriptors. Exim was using setrlimit() to set 1000 as a large value
3060 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3061 systems. I've change it so that if it can't get 1000, it tries for 256.
3063 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3064 was an oversight, and furthermore, ever since the addition of extra
3065 controls (e.g. 4.43/32), the checks on when to allow different forms of
3066 "control" were broken. There should now be diagnostics for all cases when a
3067 control that does not make sense is encountered.
3069 7. $recipients is now available in the predata ACL (oversight).
3071 8. Tidy the search cache before the fork to do a delivery from a message
3072 received from the command line. Otherwise the child will trigger a lookup
3073 failure and thereby defer the delivery if it tries to use (for example) a
3074 cached ldap connection that the parent has called unbind on.
3076 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3077 of $address_data from the recipient verification was clobbered by the
3078 sender verification.
3080 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3081 was its contents. (It was OK if the option was not defined at all.)
3083 11. A "Completed" log line is now written for messages that are removed from
3084 the spool by the -Mrm option.
3086 12. $host_address is now set to the target address during the checking of
3087 ignore_target_hosts.
3089 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3090 being passed; this would have caused $sender_host_name to have been used if
3091 matching the list had actually called for a host name (not very likely,
3092 since this list is usually IP addresses). A host name is now passed as
3095 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3096 code that specifies a non-restarting handler (typically sigaction() in
3097 modern systems) in an attempt to fix a rare and obscure crash bug.
3099 15. Narrowed the window for a race in the daemon that could cause it to ignore
3100 SIGCHLD signals. This is not a major problem, because they are used only to
3101 wake it up if nothing else does.
3103 16. A malformed maildirsize file could cause Exim to calculate negative values
3104 for the mailbox size or file count. Odd effects could occur as a result.
3105 The maildirsize information is now recalculated if the size or filecount
3108 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3109 support for a long time. Removed HAVE_SYS_VFS_H.
3111 18. Updated exipick to current release from John Jetmore.
3113 19. Allow an empty sender to be matched against a lookup in an address list.
3114 Previously the only cases considered were a regular expression, or an
3117 20. Exim went into a mad DNS lookup loop when doing a callout where the
3118 host was specified on the transport, if the DNS lookup yielded more than
3121 21. The RFC2047 encoding function was originally intended for short strings
3122 such as real names; it was not keeping to the 75-character limit for
3123 encoded words that the RFC imposes. It now respects the limit, and
3124 generates multiple encoded words if necessary. To be on the safe side, I
3125 have increased the buffer size for the ${rfc2047: expansion operator from
3128 22. Failure to deliver a bounce message always caused it to be frozen, even if
3129 there was an errors_to setting on the router. The errors_to setting is now
3132 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3133 canonical form (fully expanded) before being placed in
3134 $sender_host_address.
3136 24. Updated eximstats to version 1.33
3138 25. Include certificate and key file names in error message when GnuTLS fails
3139 to set them up, because the GnuTLS error message doesn't include the name
3140 of the failing file when there is a problem reading it.
3142 26. Expand error message when OpenSSL has problems setting up cert/key files.
3145 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3146 (this can affect the format of dates).
3148 28. exim_tidydb, when checking for the continued existence of a message for
3149 which it has found a message-specific retry record, was not finding
3150 messages that were in split spool directories. Consequently, it was
3151 deleting retry records that should have stayed in existence.
3153 29. eximstats updated to version 1.35
3154 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3155 1.35 - bugfix such that pie charts by volume are generated correctly
3157 30. The SPA authentication driver was not abandoning authentication and moving
3158 on to the next authenticator when an expansion was forced to fail,
3159 contradicting the general specification for all authenticators. Instead it
3160 was generating a temporary error. It now behaves as specified.
3162 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3163 (the order specifies the preference for clients). The order is now AES256,
3164 AES128, 3DES, ARCFOUR128.
3166 31. Small patch to Sieve code - explicitly set From: when generating an
3169 32. Exim crashed if a remote delivery caused a very long error message to be
3170 recorded - for instance if somebody sent an entire SpamAssassin report back
3171 as a large number of 550 error lines. This bug was coincidentally fixed by
3172 increasing the size of one of Exim's internal buffers (big_buffer) that
3173 happened as part of the Exiscan merge. However, to be on the safe side, I
3174 have made the code more robust (and fixed the comments that describe what
3177 33. Some experimental protocols are using DNS PTR records for new purposes. The
3178 keys for these records are domain names, not reversed IP addresses. The
3179 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3180 leaves it alone. Component reversal etc. now happens only for IP addresses.
3183 34. The host_aton() function is supposed to be passed a string that is known
3184 to be a valid IP address. However, in the case of IPv6 addresses, it was
3185 not checking this. This is a hostage to fortune. Exim now panics and dies
3186 if the condition is not met. A case was found where this could be provoked
3187 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3188 components; fortuitously, this particular loophole had already been fixed
3189 by change 4.50/55 or 4.44/33 above.
3191 If there are any other similar loopholes, the new check in host_aton()
3192 itself should stop them being exploited. The report I received stated that
3193 data on the command line could provoke the exploit when Exim was running as
3194 exim, but did not say which command line option was involved. All I could
3195 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3196 running as the user.
3199 35. There was a buffer overflow vulnerability in the SPA authentication code
3200 (which came originally from the Samba project). I have added a test to the
3201 spa_base64_to_bits() function which I hope fixes it.
3204 36. The daemon start-up calls getloadavg() while still root for those OS that
3205 need the first call to be done as root, but it missed one case: when
3206 deliver_queue_load_max is set with deliver_drop_privilege. This is
3207 necessary for the benefit of the queue runner, because there is no re-exec
3208 when deliver_drop_privilege is set.
3210 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3211 was in use, was not putting the data itself into the right store pool;
3212 consequently, it could be overwritten for a subsequent message in the same
3213 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3216 38. Sometimes the final signoff response after QUIT could fail to get
3217 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3218 < 0 before doing a fflush(). This bug looks as though it goes back to the
3219 introduction of TLS in release 3.20, but "sometimes" must have been rare
3220 because the tests only now provoked it.
3226 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3227 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3228 Now it returns an integer. A number of calls were still expecting a T/F
3229 return. Fortuitously, in all cases, the tests worked in OK situations,
3230 which is the norm. However, things would have gone wrong on any write
3231 failures on the smtp file descriptor. This function is used when sending
3232 messages over SMTP and also when doing verify callouts.
3234 2. When Exim is called to do synchronous delivery of a locally submitted
3235 message (the -odf or -odi options), it no longer closes stderr before doing
3238 3. Implemented the mua_wrapper option.
3240 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3242 5. Implemented the functions header_remove(), header_testname(),
3243 header_add_at_position(), and receive_remove_recipient(), and exported them
3246 6. If an ACL "warn" statement specified the addition of headers, Exim already
3247 inserted X-ACL-Warn: at the start if there was no header name. However, it
3248 was not making this test for the second and subsequent header lines if
3249 there were newlines in the string. This meant that an invalid header could
3250 be inserted if Exim was badly configured.
3252 7. Allow an ACL "warn" statement to add header lines at the start or after all
3253 the Received: headers, as well as at the end.
3255 8. Added the rcpt_4xx retry error code.
3257 9. Added postmaster_mailfrom=xxx to callout verification option.
3259 10. Added mailfrom=xxxx to the callout verification option, for verify=
3262 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3263 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3265 12. Inserted some casts to stop certain compilers warning when using pointer
3266 differences as field lengths or precisions in printf-type calls (mostly
3267 affecting debugging statements).
3269 13. Added optional readline() support for -be (dynamically loaded).
3271 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3272 same clock tick as a message's arrival, so that its received time was the
3273 same as the "first fail" time on the retry record, and that message
3274 remained on the queue past the ultimate address timeout, every queue runner
3275 would try a delivery (because it was past the ultimate address timeout) but
3276 after another failure, the ultimate address timeout, which should have then
3277 bounced the address, did not kick in. This was a "< instead of <=" error;
3278 in most cases the first failure would have been in the next clock tick
3279 after the received time, and all would be well.
3281 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3282 being recognized when the domain list was tested by the match_domain
3283 condition in an expansion string.
3285 16. Added the ${str2b64: operator.
3287 17. Exim was always calling setrlimit() to set a large limit for the number of
3288 processes, without checking whether the existing limit was already
3289 adequate. (It did check for the limit on file descriptors.) Furthermore,
3290 errors from getrlimit() and setrlimit() were being ignored. Now they are
3291 logged to the main and panic logs, but Exim does carry on, to try to do its
3292 job under whatever limits there are.
3294 18. Imported PCRE 5.0.
3296 19. Trivial typo in log message " temporarily refused connection" (the leading
3299 20. If the log selector return_path_on_delivery was set and an address was
3300 redirected to /dev/null, the delivery process crashed because it assumed
3301 that a return path would always be set for a "successful" delivery. In this
3302 case, the whole delivery is bypassed as an optimization, and therefore no
3305 21. Internal re-arrangement: the function for sending a challenge and reading
3306 a response while authentication was assuming a zero-terminated challenge
3307 string. It's now changed to take a pointer and a length, to allow for
3308 binary data in such strings.
3310 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3312 23. Exim was not respecting finduser_retries when seeking the login of the
3313 uid under which it was called; it was always trying 10 times. (The default
3314 setting of finduser_retries is zero.) Also, it was sleeping after the final
3315 failure, which is pointless.
3317 24. Implemented tls_on_connect_ports.
3319 25. Implemented acl_smtp_predata.
3321 26. If the domain in control=submission is set empty, Exim assumes that the
3322 authenticated id is a complete email address when it generates From: or
3323 Sender: header lines.
3325 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3326 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3327 chown and chgrp in /bin and hostname in /usr/bin.
3329 28. Exim was keeping the "process log" file open after each use, just as it
3330 does for the main log. This opens the possibility of it remaining open for
3331 long periods when the USR1 signal hits a daemon. Occasional processlog
3332 errors were reported, that could have been caused by this. Anyway, it seems
3333 much more sensible not to leave this file open at all, so that is what now
3336 29. The long-running daemon process does not normally write to the log once it
3337 has entered its main loop, and it closes the log before doing so. This is
3338 so that log files can straightforwardly be renamed and moved. However,
3339 there are a couple of unusual error situations where the daemon does write
3340 log entries, and I had neglected to close the log afterwards.
3342 30. The text of an SMTP error response that was received during a remote
3343 delivery was being truncated at 512 bytes. This is too short for some of
3344 the long messages that one sometimes sees. I've increased the limit to
3347 31. It is now possible to make retry rules that apply only when a message has a
3348 specific sender, in particular, an empty sender.
3350 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3351 it possible to be selective about when SMTP synchronization is enforced.
3353 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3355 32. Implemented hosts_connection_nolog.
3357 33. Added an ACL for QUIT.
3359 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3362 35. Added mailbox_size and mailbox_filecount to appendfile.
3364 36. Added control = no_multiline_responses to ACLs.
3366 37. There was a bug in the logic of the code that waits for the clock to tick
3367 in the case where the clock went backwards by a substantial amount such
3368 that the microsecond fraction of "now" was more than the microsecond
3369 fraction of "then" (but the whole seconds number was less).
3371 38. Added support for the libradius Radius client library this is found on
3372 FreeBSD (previously only the radiusclient library was supported).
3378 1. When certain lookups returned multiple values in the form name=value, the
3379 quoting of the values was not always being done properly. Specifically:
3380 (a) If the value started with a double quote, but contained no whitespace,
3382 (b) If the value contained whitespace other than a space character (i.e.
3383 tabs or newlines or carriage returns) it was not quoted.
3384 This fix has been applied to the mysql and pgsql lookups by writing a
3385 separate quoting function and calling it from the lookup code. The fix
3386 should probably also be applied to nisplus, ibase and oracle lookups, but
3387 since I cannot test any of those, I have not disturbed their existing code.
3389 2. A hit in the callout cache for a specific address caused a log line with no
3390 reason for rejecting RCPT. Now it says "Previous (cached) callout
3391 verification failure".
3393 3. There was an off-by-one bug in the queryprogram router. An over-long
3394 return line was truncated at 256 instead of 255 characters, thereby
3395 overflowing its buffer with the terminating zero. As well as fixing this, I
3396 have increased the buffer size to 1024 (and made a note to document this).
3398 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3399 when Exim is waiting for an SMTP response from a remote server, Exim
3400 restarts its select() call on the socket, thereby resetting its timeout.
3401 This is not a problem when such interrupts are rare. Somebody set up a cron
3402 job to run exiwhat every 2 minutes, which is less than the normal select()
3403 timeout (5 or 10 minutes). This meant that the select() timeout never
3404 kicked in because it was always reset. I have fixed this by comparing the
3405 time when an interrupt arrives with the time at the start of the first call
3406 to select(). If more time than the timeout has elapsed, the interrupt is
3407 treated as a timeout.
3409 5. Some internal re-factoring in preparation for the addition of Sieve
3410 extensions (by MH). In particular, the "personal" test is moved to a
3411 separate function, and given an option for scanning Cc: and Bcc: (which is
3412 not set for Exim filters).
3414 6. When Exim created an email address using the login of the caller as the
3415 local part (e.g. when creating a From: or Sender: header line), it was not
3416 quoting the local part when it contained special characters such as @.
3418 7. Installed new OpenBSD configuration files.
3420 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3421 try to make them clearer.
3423 9. Callout options, other than the timeout value, were being ignored when
3424 verifying sender addresses in header lines. For example, when using
3426 verify = header_sender/callout=no_cache
3428 the cache was (incorrectly) being used.
3430 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3431 only the Cygwin environment.
3433 11. When return_path_on_delivery was set as a log selector, if different remote
3434 addresses in the same message used different return paths and parallel
3435 remote delivery occurred, the wrong values would sometimes be logged.
3436 (Whenever a remote delivery process finished, the return path value from
3437 the most recently started remote delivery process was logged.)
3439 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3440 lines when AUTH and/or TLS are in use. This is the "received protocol"
3441 field. Exim used to use "asmtp" for authenticated SMTP, without any
3442 indication (in the protocol name) for TLS use. Now it follows the RFC and
3443 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3444 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3445 names appear in log lines as well as in Received: header lines.
3447 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3448 extensions, and comparison tests, and to fix some bugs.
3450 14. Changes to the "personal" filter test:
3452 (1) The test was buggy in that it was just doing the equivalent of
3453 "contains" tests on header lines. For example, if a user's address was
3454 anne@some.where, the "personal" test would incorrectly be true for
3456 To: susanne@some.where
3458 This test is now done by extracting each address from the header in turn,
3459 and checking the entire address. Other tests that are part of "personal"
3460 are now done using regular expressions (for example, to check local parts
3461 of addresses in From: header lines).
3463 (2) The list of non-personal local parts in From: addresses has been
3464 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3465 taken from the Sieve specification recommendations.
3467 (3) If the message contains any header line starting with "List-" it is
3468 treated as non-personal.
3470 (4) The test for "circular" in the Subject: header line has been removed
3471 because it now seems ill-conceived.
3473 15. Minor typos in src/EDITME comments corrected.
3475 16. Installed latest exipick from John Jetmore.
3477 17. If headers_add on a router specified a text string that was too long for
3478 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3479 of string_sprintf() is now avoided.
3481 18. $message_body_size was not set (it was always zero) when running the DATA
3482 ACL and the local_scan() function.
3484 19. For the "mail" command in an Exim filter, no default was being set for
3485 the once_repeat time, causing a random time value to be used if "once" was
3486 specified. (If the value happened to be <= 0, no repeat happened.) The
3487 default is now 0s, meaning "never repeat". The "vacation" command was OK
3488 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3489 (I found it when inspecting the code).
3491 20. There is now an overall timeout for performing a callout verification. It
3492 defaults to 4 times the callout timeout, which applies to individual SMTP
3493 commands during the callout. The overall timeout applies when there is more
3494 than one host that can be tried. The timeout is checked before trying the
3495 next host. This prevents very long delays if there are a large number of
3496 hosts and all are timing out (e.g. when the network connections are timing
3497 out). The value of the overall timeout can be changed by specifying an
3498 additional sub-option for "callout", called "maxwait". For example:
3500 verify = sender/callout=5s,maxwait=20s
3502 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3503 to the end before writing, but this should make it even safer).
3505 22. Exim was forgetting that it had advertised PIPELINING for the second and
3506 subsequent messages on an SMTP connection. It was also not resetting its
3507 memory on STARTTLS and an internal HELO.
3509 23. When Exim logs an SMTP synchronization error within a session, it now
3510 records whether PIPELINING has been advertised or not.
3512 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3513 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3514 rather than long int.
3516 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3518 26. Added the never_mail option to autoreply.
3524 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3525 crash if the getsockname() call failed; this can happen if a connection is
3526 closed very soon after it is established. The problem was simply in the
3527 order in which certain operations were done, causing Exim to try to write
3528 to the SMTP stream before it had set up the file descriptor. The bug has
3529 been fixed by making things happen in the correct order.
3535 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3536 before the connection was closed, thus losing the rejection response.
3538 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3539 some early Solaris releases, but causes trouble in current releases where
3540 socklen_t is defined.
3542 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3545 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3548 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3549 that releasing the top of it at the end releases what was used for sub-
3550 expansions (unless the block got too big). However, discard this block if
3551 the first thing is a variable or header, so that we can use its block when
3552 it is dynamic (useful for very large $message_headers, for example).
3554 6. Lookups now cache *every* query, not just the most recent. A new, separate
3555 store pool is used for this. It can be recovered when all lookup caches are
3556 flushed. Lookups now release memory at the end of their result strings.
3557 This has involved some general refactoring of the lookup sources.
3559 7. Some code has been added to the store_xxx() functions to reduce the amount
3560 of flapping under certain conditions.
3562 8. log_incoming_interface used to affect only the <= reception log lines. Now
3563 it causes the local interface and port to be added to several more SMTP log
3564 lines, for example "SMTP connection from", and rejection lines.
3566 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3568 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3570 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3571 could be overwritten at the end of the current message (or the start of a
3572 new message if it was set in a HELO ACL). The value is now preserved for
3573 the duration of the SMTP connection.
3575 12. If a transport had a headers_rewrite setting, and a matching header line
3576 contained an unqualified address, that address was qualified, even if it
3577 did not match any rewriting rules. The underlying bug was that the values
3578 of the flags that permit the existence of unqualified sender and recipient
3579 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3580 non-local messages, and by -bnq for local messages) were not being
3581 preserved with the message after it was received.
3583 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3584 "next input=" as part of the text comprising the host identity instead of
3585 the correct text. The code was using the same buffer for two different
3586 strings. However, depending on which order the printing function evaluated
3587 its arguments, the bug did not always show up. Under Linux, for example, my
3588 test suite worked just fine.
3590 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3591 doesn't work with some older versions of Perl. It has been changed to "my",
3592 which in any case is probably the better facility to use.
3594 15. A really picky compiler found some instances of statements for creating
3595 error messages that either had too many or two few arguments for the format
3598 16. The size of the buffer for calls to the DNS resolver has been increased
3599 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3600 for addresses that have a lot of PTR records. This alleviates a problem; it
3601 does not fully solve it.
3603 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3604 buffer now truncates the list and logs the incident, which is the same
3605 action as happens when Exim is looking up a host name and its aliases.
3606 Previously in this situation something unpredictable would happen;
3607 sometimes it was "internal error: store_reset failed".
3609 18. If a server dropped the connection unexpectedly when an Exim client was
3610 using GnuTLS and trying to read a response, the client delivery process
3611 crashed while trying to generate an error log message.
3613 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3614 string, for example:
3616 warn message = H1: something\nH2: something
3618 the text was added as a single header line from Exim's point of view
3619 though it ended up OK in the delivered message. However, searching for the
3620 second and subsequent header lines using $h_h2: did not work. This has been
3621 fixed. Similarly, if a system filter added multiple headers in this way,
3622 the routers could not see them.
3624 20. Expanded the error message when iplsearch is called with an invalid key to
3625 suggest using net-iplsearch in a host list.
3627 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3628 ACLs are no longer actually imposed (and a message to that effect is
3631 22. If a "gecos" field in a passwd entry contained escaped characters, in
3632 particular, if it contained a \" sequence, Exim got it wrong when building
3633 a From: or a Sender: header from that name. A second bug also caused
3634 incorrect handling when an unquoted " was present following a character
3635 that needed quoting.
3637 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3638 was not being matched caselessly.
3640 24. Arranged for all hyphens in the exim.8 source to be escaped with
3643 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3644 properly. Recipient callout cache records were still being keyed to include
3645 the sender, even when use_sender was set false. This led to far more
3646 callouts that were necessary. The sender is no longer included in the key
3647 when use_sender is false.
3649 26. Added "control = submission" modifier to ACLs.
3651 27. Added the ${base62d: operator to decode base 62 numbers.
3653 28. dnsdb lookups can now access SRV records.
3655 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3656 the configuration file.
3658 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3659 (-v) mode. This makes the output for a verbose queue run more intelligible.
3661 31. Added a use_postmaster feature to recipient callouts.
3663 32. Added the $body_zerocount variable, containing the number of binary zero
3664 bytes in the message body.
3666 33. The time of last modification of the "new" subdirectory is now used as the
3667 "mailbox time last read" when there is a quota error for a maildir
3670 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3672 35. Added +ignore_unknown as a special item in host lists.
3674 36. Code for decoding IPv6 addresses in host lists is now included, even if
3675 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3676 address was recognized as an IP address, but was then not correctly decoded
3677 into binary, causing unexpected and incorrect effects when compared with
3684 1. Very minor rewording of debugging text in manualroute to say "list of
3685 hosts" instead of "hostlist".
3687 2. If verify=header_syntax was set, and a header line with an unqualified
3688 address (no domain) and a large number of spaces between the end of the
3689 name and the colon was received, the reception process suffered a buffer
3690 overflow, and (when I tested it) crashed. This was caused by some obsolete
3691 code that should have been removed. The fix is to remove it!
3693 3. When running in the test harness, delay a bit after writing a bounce
3694 message to get a bit more predictability in the log output.
3696 4. Added a call to search_tidyup() just before forking a reception process. In
3697 theory, someone could use a lookup in the expansion of smtp_accept_max_
3698 per_host which, without the tidyup, could leave open a database connection.
3700 5. Added the variables $recipient_data and $sender_data which get set from a
3701 lookup success in an ACL "recipients" or "senders" condition, or a router
3702 "senders" option, similar to $domain_data and $local_part_data.
3704 6. Moved the writing of debug_print from before to after the "senders" test
3707 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3708 problems for message scanning, either using a data ACL, or using
3709 local_scan() because the Received: header was not generated till after they
3710 were called (in order to set the time as the time of reception completion).
3711 I have revised the way this works. The header is now generated after the
3712 body is received, but before the ACL or local_scan() are called. After they
3713 are run, the timestamp in the header is updated.
3719 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3720 before starting a queue runner without re-exec. This happened only when
3721 deliver_drop_privilege was set or when the Exim user was set to root. The
3722 effect of the bug was that timeouts during subsequent deliveries caused
3723 crashes instead of being properly handled. The handler is now left at its
3724 default (and expected) setting.
3726 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3727 message, again when deliver_drop_privilege is set or Exim is run as root.
3728 The bug described in (1) was not present in this case, but the tidying up
3729 of the other signals was missing. I have made the two cases consistent.
3731 3. The ignore_target_hosts setting on a manualroute router was being ignored
3732 for hosts that were looked up using the /MX notation.
3734 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3737 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3738 operated on the sender address. After changing the $sender_address to <>
3739 for the sender address verify, Exim was re-instated it as the original
3740 (before rewriting) address, but remembering that it had rewritten it, so it
3741 wasn't rewriting it again. This bug also had the effect of breaking the
3742 sender address verification caching when the sender address was rewritten.
3744 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3745 This has been changed so that if the ip literal address matches
3746 ignore_target_hosts, the router declines.
3748 7. Added expansion conditions match_domain, match_address, and match_local_
3749 part (NOT match_host).
3751 8. The placeholder for the Received: header didn't have a length field set.
3753 9. Added code to Exim itself and to exim_lock to test for a specific race
3754 condition that could lead to file corruption when using MBX delivery. The
3755 issue is with the lockfile that is created in /tmp. If this file is removed
3756 after a process has opened it but before that process has acquired a lock,
3757 there is the potential for a second process to recreate the file and also
3758 acquire a lock. This could lead to two Exim processes writing to the file
3759 at the same time. The added code performs the same test as UW imapd; it
3760 checks after acquiring the lock that its file descriptor still refers to
3761 the same named file.
3763 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3764 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3767 11. Added the smtp_active_hostname option. If used, this will typically be made
3768 to depend on the incoming interface address. Because $interface_address is
3769 not set up until the daemon has forked a reception process, error responses
3770 that can happen earlier (such as "too many connections") no longer contain
3773 12. If an expansion in a condition on a "warn" statement fails because a lookup
3774 defers, the "warn" statement is abandoned, and the next ACL statement is
3775 processed. Previously this caused the whole ACL to be aborted.
3777 13. Added the iplsearch lookup type.
3779 14. Added ident_timeout as a log selector.
3781 15. Added tls_certificate_verified as a log selector.
3783 16. Added a global option tls_require_ciphers (compare the smtp transport
3784 option of the same name). This controls incoming TLS connections.
3786 17. I finally figured out how to make tls_require_ciphers do a similar thing
3787 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3788 before starting the TLS session.
3790 18. Tabs are now shown as \t in -bP output.
3792 19. If the log selector return_path_on_delivery was set, Exim crashed when
3793 bouncing a message because it had too many Received: header lines.
3795 20. If two routers both had headers_remove settings, and the first one included
3796 a superfluous trailing colon, the final name in the first list and the
3797 first name in the second list were incorrectly joined into one item (with a
3798 colon in the middle).
3804 1. Added -C and -D options to the exinext utility, mainly to make it easier
3805 to include in the automated testing, but these could be helpful when
3806 multiple configurations are in use.
3808 2. The exinext utility was not formatting the output nicely when there was
3809 an alternate port involved in the retry record key, nor when there was a
3810 message id as well (for retries that were specific to a specific message
3811 and a specific host). It was also confused by IPv6 addresses, because of
3812 the additional colons they contain. I have fixed the IPv4 problem, and
3813 patched it up to do a reasonable job for IPv6.
3815 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3816 delivery, the log line now contains "pipelined" if PIPELINING was used.
3818 4. An SMTP transport process used to panic and die if the bind() call to set
3819 an explicit outgoing interface failed. This has been changed; it is now
3820 treated in the same way as a connect() failure.
3822 5. A reference to $sender_host_name in the part of a conditional expansion
3823 that was being skipped was still causing a DNS lookup. This no longer
3826 6. The def: expansion condition was not recognizing references to header lines
3827 that used bh_ and bheader_.
3829 7. Added the _cache feature to named lists.
3831 8. The code for checking quota_filecount in the appendfile transport was
3832 allowing one more file than it should have been.
3834 9. For compatibility with Sendmail, the command line option
3842 and sets the incoming protocol and host name (for trusted callers). The
3843 host name and its colon can be omitted when only the protocol is to be set.
3844 Note the Exim already has two private options, -pd and -ps, that refer to
3845 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3846 "s", but I don't think that's a major issue.
3848 10. A number of refactoring changes to the code, none of which should affect
3851 (a) The number of logging options was getting close to filling up the
3852 32-bit word that was used as a bit map. I have split them into two classes:
3853 those that are passed in the argument to log_write(), and those that are
3854 only ever tested independently outside of that function. These are now in
3855 separate 32-bit words, so there is plenty of room for expansion again.
3856 There is no change in the user interface or the logging behaviour.
3858 (b) When building, for example, log lines, the code previously used a
3859 macro that called string_cat() twice, in order to add two strings. This is
3860 not really sufficiently general. Furthermore, there was one instance where
3861 it was actually wrong because one of the argument was used twice, and in
3862 one call a function was used. (As it happened, calling the function twice
3863 did not affect the overall behaviour.) The macro has been replaced by a
3864 function that can join an arbitrary number of extra strings onto a growing
3867 (c) The code for expansion conditions now uses a table and a binary chop
3868 instead of a serial search (which was left over from when there were very
3869 few conditions). Also, it now recognizes conditions like "pam" even when
3870 the relevant support is not compiled in: a suitably worded error message is
3871 given if an attempt is made to use such a condition.
3873 11. Added ${time_interval:xxxxx}.
3875 12. A bug was causing one of the ddress fields not to be passed back correctly
3876 from remote delivery subprocesses. The field in question was not being
3877 subsequently used, so this caused to problems in practice.
3879 13. Added new log selectors queue_time and deliver_time.
3881 14. Might have fixed a bug in maildirsizefile handling that threw up
3882 "unexpected character" debug warnings, and recalculated the data
3883 unnecessarily. In any case, I expanded the warning message to give more
3886 15. Added the message "Restricted characters in address" to the statements in
3887 the default ACL that block characters like @ and % in local parts.
3889 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3890 Three changes have been made:
3892 (a) There was a serious bug; a negative response to MAIL caused the whole
3893 recipient domain to be cached as invalid, thereby blocking all messages
3894 to all local parts at the same domain, from all senders. This bug has
3895 been fixed. The domain is no longer cached after a negative response to
3896 MAIL if the sender used is not empty.
3898 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3901 (c) A new callout option, "use_sender" has been added for people who want
3902 the modified behaviour.
3908 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3911 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3912 indeed breaks things for older releases.
3914 3. Added additional logging to the case where there is a problem reading data
3915 from a filter that is running in a subprocess using a pipe, in order to
3916 try to track down a specific problem.
3918 4. Testing facility fudge: when running in the test harness and attempting
3919 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3920 getting "No route to host". Convert this to a timeout.
3922 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3925 6. Some OS don't have socklen_t but use size_t instead. This affects the
3926 fifth argument of getsockopt() amongst other things. This is now
3927 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3928 can be set for individual OS. I have set it for SunOS5, OSF1, and
3929 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3930 some earlier ones do not.
3932 7. Change 4.30/15 was not doing the test caselessly.
3934 8. The standard form for an IPv6 address literal was being rejected by address
3935 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3936 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3937 this, as well as the form without the "IPv6" on the front (but only when
3938 address literals are enabled, of course).
3940 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3942 10. Exim crashed if a message with an empty sender address specified by -f
3943 encountered a router with an errors_to setting. This could be provoked only
3944 by a command such as
3948 where an empty string was supplied; "<>" did not hit this bug.
3950 11. Installed PCRE release 4.5.
3952 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3953 remained set. It is now erased.
3955 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3956 times from message ids (which are base 36 rather than the normal 62).
3958 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3959 were being counted as actual protocol errors, and logged if the log
3960 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3961 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3962 and DATA following a set of rejected RCPTs do not count as protocol errors.
3963 In other words, Exim assumes they were pipelined, though this may not
3964 actually be the case. Of course, in all cases the client gets an
3965 appropriate error code.
3967 15. If a lookup fails in an ACL condition, a message about the failure may
3968 be available; it is used if testing the ACL cannot continue, because most
3969 such messages specify what the cause of the deferral is. However, some
3970 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3971 that caused an old message to be retained and used if a later statement
3972 caused a defer, replacing the real cause of the deferral.
3974 16. If an IP address had so many PTR records that the DNS lookup buffer
3975 was not large enough to hold them, Exim could crash while trying to process
3976 the truncated data. It now detects and logs this case.
3978 17. Further to 4.21/58, another change has been made: if (and only if) the
3979 first line of a message (the first header line) ends with CRLF, a bare LF
3980 in a subsequent header line has a space inserted after it, so as not to
3981 terminate the header.
3983 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3984 unnecessarily, used atoi() instead of strtol(), and didn't check the
3985 termination when getting file sizes from file names by regex.
3987 19. Completely re-implemented the support for maildirsize files, in the light
3988 of a number of problems with the previous contributed implementation
3989 (4.30/29). In particular:
3991 . If the quota is zero, the maildirsize file is maintained, but no quota is
3994 . If the maildir directory does not exist, it is created before any attempt
3995 to write a maildirsize file.
3997 . The quota value in the file is just a cache; if the quota is changed in
3998 the transport, the new value overrides.
4000 . A regular expression is available for excluding directories from the
4003 20. The autoreply transport checks the characters in options that define the
4004 message's headers; it allows continued headers, but it was checking with
4005 isspace() after an embedded newline instead of explicitly looking for a
4008 21. If all the "regular" hosts to which an address was routed had passed their
4009 expiry times, and had not reached their retry times, the address was
4010 bounced, even if fallback hosts were defined. Now Exim should go on to try
4013 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4014 equivalent code in the SMTP transport. Some hosts send humungous responses
4015 to HELO/EHLO, more than 1024 it seems.
4017 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4018 gives compiler warnings in some environments. I've now done it "properly",
4021 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4022 (because of problems with the built-in one) was declared to return uschar *
4023 instead of char *, causing compiler failure.
4025 25. Fixed a file descriptor leak when processing alias/forward files.
4027 26. Fixed a minor format string issue in dbfn.c.
4029 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4031 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4032 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4035 29. When a sender address is verified, it is cached, to save repeating the test
4036 when there is more than one recipient in a message. However, when the
4037 verification involves a callout, it is possible for different callout
4038 options to be set for different recipients. It is too complicated to keep
4039 track of this in the cache, so now Exim always runs a verification when a
4040 callout is required, relying on the callout cache for the optimization.
4041 The overhead is duplication of the address routing, but this should not be
4044 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4045 to be verified with callout=postmaster, and the main callout worked but the
4046 postmaster check failed, the verification correctly failed. However, if a
4047 subsequent RCPT command asked for sender verification *without* the
4048 postmaster check, incorrect caching caused this verification also to fail,
4051 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4052 it was not caching the DNS options (qualify_single, search_parents) that
4053 were used when the lookup failed. A subsequent lookup with different
4054 options therefore always gave the same answer, though there were cases
4055 where it should not have. (Example: a "domains = !$mx_any" option on a
4056 dnslookup router: the "domains" option is always processed without any
4057 widening, but the router might have qualify_single set.) Now Exim uses the
4058 cached value only when the same options are set.
4060 32. Added John Jetmore's "exipick" utility to the distribution.
4062 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4063 than a timeout (e.g. a certificate is required, and is not provided), an
4064 Exim server now closes the connection immediately. Previously it waited for
4065 the client to close - but if the client is SSL, it seems that they each
4066 wait for each other, leading to a delay before one of them times out.
4068 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4069 maintained 0.8.x compatibility because I don't think many are using it, and
4070 it is clearly obsolete.
4072 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4075 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4076 client certificate was expired. A simple patch fixes this, though I don't
4077 understand the full logic of why the verify callback is called multiple
4080 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4081 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4082 which causes problems with some clients (such as the Certicom SSL Plus
4083 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4084 disables the coutermeasure allowing Eudora to connect."
4086 38. Exim was not checking that a write() to a log file succeeded. This could
4087 lead to Bad Things if a log got too big, in particular if it hit a file
4088 size limit. Exim now panics and dies if it cannot write to a log file, just
4089 as it does if it cannot open a log file.
4091 39. Modified OS/Makefile-Linux so that it now contains
4093 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4095 The two -D definitions ensure that Exim is compiled with large file
4096 support, which makes it possible to handle log files that are bigger than
4099 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4100 instance) a domain was checked against a named list that involved a lookup,
4101 causing $domain_data to be set, then another domain was checked against the
4102 same list, then the first domain was re-checked, the value of $domain_data
4103 after the final check could be wrong. In particular, if the second check
4104 failed, it could be set empty. This bug probably also applied to
4107 41. The strip_trailing_dot option was not being applied to the address given
4108 with the -f command-line option.
4110 42. The code for reading a message's header from the spool was incrementing
4111 $received_count, but never initializing it. This meant that the value was
4112 incorrect (doubled) while delivering a message in the same process in which
4113 it was received. In the most common configuration of Exim, this never
4114 happens - a fresh exec is done - but it can happen when
4115 deliver_drop_privilege is set.
4117 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4118 it now includes up to 150 characters of the unexpected data in the log
4121 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4122 and building data strings. The size of both of these buffers was 10 000
4123 bytes - far larger than anybody would *ever* want, thought I. Needless to
4124 say, somebody hit the limit. I have increased the maximum line length to
4125 20 000 and the maximum data length of concatenated lines to 100 000. I have
4126 also fixed two bugs, because there was no checking on these buffers. Tsk,
4127 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4128 buffer is too small.
4130 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4131 lsearch lookups. Now it does.
4133 46. When parsing a route_list item in a manualroute router, a fixed-length
4134 buffer was used for the list of hosts. I made this 1024 bytes long,
4135 thinking that nobody would ever have a list of hosts that long. Wrong.
4136 Somebody had a whole pile of complicated expansion conditions, and the
4137 string was silently truncated, leading to an expansion error. It turns out
4138 that it is easier to change to an unlimited length (owing to other changes
4139 that have happened since this code was originally written) than to build
4140 structure for giving a limitation error. The length of the item that
4141 expands into the list of hosts is now unlimited.
4143 47. The lsearch lookup could not handle data where the length of text line was
4144 more than 4095 characters. Such lines were truncated, leading to shortened
4145 data being returned. It should now handle lines of any length.
4147 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4148 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4151 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4152 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4153 no longer happens in comment lines. A list of replacements is now placed
4154 at the head of all of the source files, except those whose only change is
4155 to replace PERL_COMMAND in the very first #! line.
4157 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4158 messages on the queue, with a bottom-up merge sort, using code contributed
4159 by Michael Haardt. This should make operations like -bp somewhat faster on
4160 large queues. It won't affect queue runners, except when queue_run_in_order
4163 51. Installed eximstats 1.31 in the distribution.
4165 52. Added support for SRV lookups to the dnslookup router.
4167 53. If an ACL referred to $message_body or $message_body_end, the value was not
4168 reset for any messages that followed in the same SMTP session.
4170 54. The store-handling optimization for building very long strings was not
4171 differentiating between the different store pools. I don't think this
4172 actually made any difference in practice, but I've tidied it.
4174 55. While running the routers to verify a sender address, $sender_address
4175 was still set to the sender address. This is wrong, because when routing to
4176 send a bounce to the sender, it would be empty. Therefore, I have changed
4177 it so that, while verifying a sender address, $sender_address is set to <>.
4178 (There is no change to what happens when verifying a recipient address.)
4180 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4181 target A or AAAA records (if not already returned) without resetting the
4182 qualify_single or search_parents options of the DNS resolver. These are
4183 inappropriate in this case because the targets of MX and SRV records must
4184 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4185 target that, when qualified, matched something in the local domain. These
4186 two options are now turned off when doing these lookups.
4188 57. It seems that at least some releases of Reiserfs (which does not have the
4189 concept of a fixed number of inodes) returns zero and not -1 for the
4190 number of available inodes. This interacted badly with check_spool_inodes,
4191 which assumed that -1 was the "no such thing" setting. What I have done is
4192 to check that the total number of inodes is greater than zero before doing
4193 the test of how many are available.
4195 58. When a "warn" ACL statement has a log_message modifier, the message is
4196 remembered, and not repeated. This is to avoid a lot of repetition when a
4197 message has many recipients that cause the same warning to be written.
4198 Howewer, Exim was preserving the list of already written lines for an
4199 entire SMTP session, which doesn't seem right. The memory is now reset if a
4200 new message is started.
4202 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4203 debugging output unless log_rewrite was also set.
4205 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4206 of (int)(handle) when we know that handle contains (void *)(-1).
4208 61. The Exim daemon panic-logs an error return when it closes the incoming
4209 connection. However "connection reset by peer" seems to be common, and
4210 isn't really an error worthy of noting specially, so that particular error
4213 62. When Exim is trying to find all the local interfaces, it used to panic and
4214 die if the ioctl to get the interface flags failed. However, it seems that
4215 on at least one OS (Solaris 9) it is possible to have an interface that is
4216 included in the list of interfaces, but for which you get a failure error
4217 for this call. This happens when the interface is not "plumbed" into a
4218 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4219 failure of the "get flags" call assumes that the interface is down.
4221 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4222 makes life easier for people who are doing arithmetic on fields extracted
4223 from dates, where you often get leading zeros that should not be
4224 interpreted as octal.
4226 64. Added qualify_domain to the redirect router, to override the global
4229 65. If a pathologically long header line contained very many addresses (the
4230 report of this problem mentioned 10 000) and each of them was rewritten,
4231 Exim could use up a very large amount of memory. (It kept on making new
4232 copies of the header line as it rewrote, and never released the old ones.)
4233 At the expense of a bit more processing, the header rewriting function has
4234 been changed so that it no longer eats memory in this way.
4236 66. The generation of the Received: header has been moved from the time that a
4237 message starts to be received, to the time that it finishes. The timestamp
4238 in the Received: header should now be very close to that of the <= log
4239 line. There are two side-effects of this change:
4241 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4242 logged header lines no longer include the local Received: line, because
4243 it has not yet been created. The same applies to a copy of the message
4244 that is returned to a non-SMTP sender when a message is rejected.
4246 (b) When a filter file is tested using -bf, no additional Received: header
4247 is added to the test message. After some thought, I decided that this
4250 This change does not affect the value of $received_for. It is still set
4251 after address rewriting, but before local_scan() is called.
4253 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4255 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4256 gave an unhelpful panic error message, and a defer error. I have managed to
4257 change this behaviour so that it now rejects any supplied certificate,
4258 which seems right, as the list of acceptable certificates is empty.
4260 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4261 gave an unhelpful defer error. I have not managed to make this reject any
4262 supplied certificates, but the error message it gives is "no certificate
4263 supplied", which is not helpful.
4265 70. exigrep's output now also includes lines that are not associated with any
4266 message, but which match the given pattern. Implemented by a patch from
4267 Martin Sluka, which also tidied up the Perl a bit.
4269 71. Recipient callout verification, like sender verification, was using <> in
4270 the MAIL FROM command. This isn't really the right thing, since the actual
4271 sender may affect whether the remote host accepts the recipient or not. I
4272 have changed it to use the actual sender in the callout; this means that
4273 the cache record is now keyed on a recipient/sender pair, not just the
4274 recipient address. There doesn't seem to be a real danger of callout loops,
4275 since a callout by the remote host to check the sender would use <>.
4276 [SEE ABOVE: changed after hitting problems.]
4278 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4279 temporary errors. However, in the case of such a code being given after
4280 the end of a data transmission (i.e. after ".") Exim was failing to write
4281 a retry record for the message. (Yes, there was some broken host that was
4282 actually sending 8xx at this point.)
4284 73. An unknown lookup type in a host list could cause Exim to panic-die when
4285 the list was checked. (An example that provoked this was putting <; in the
4286 middle of a list instead of at the start.) If this happened during a DATA
4287 ACL check, a -D file could be left lying around. This kind of configuration
4288 error no longer causes Exim to die; instead it causes a defer errror. The
4289 incident is still logged to the main and panic logs.
4291 74. Buglet left over from Exim 3 conversion. The message "too many messages
4292 in one connection" was written to the rejectlog but not the mainlog, except
4293 when address rewriting (yes!) was being logged.
4295 75. Added write_rejectlog option.
4297 76. When a system filter was run not as root (that is, when system_filter_user
4298 was set), the values of the $n variables were not being returned to the
4299 main process; thus, they were not subsequently available in the $sn
4302 77. Added +return_path_on_delivery log selector.
4304 78. A connection timeout was being treated differently from recipients deferred
4305 when testing hosts_max_try with a message that was older than the host's
4306 retry timeout. (The host should not be counted, thus allowing all hosts to
4307 be tried at least once before bouncing.) This may have been the cause of an
4308 occasionally reported bug whereby a message would remain on the queue
4309 longer than the retry timeout, but would be bounced if a delivery was
4310 forced. I say "may" because I never totally pinned down the problem;
4311 setting up timeout/retry tests is difficult. See also the next item.
4313 79. The ultimate address timeout was not being applied to errors that involved
4314 a combination of host plus message (for example, a timeout on a MAIL
4315 command). When an address resolved to a number of possible hosts, and they
4316 were not all tried for each delivery (e.g. because of hosts_max_try), a
4317 message could remain on the queue longer than the retry timeout.
4319 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4322 81. Fixed an obscure SMTP outgoing bug which required at least the following
4323 conditions: (a) there was another message waiting for the same server;
4324 (b) the server returned 5xx to all RCPT commands in the first message so
4325 that the message was not completed; (c) the server dropped the connection
4326 or gave a negative response to the RSET that Exim sends to abort the
4327 transaction. The observed case was a dropped connection after DATA that had
4328 been sent in pipelining mode. That is, the server had advertised PIPELINING
4329 but was not implementing it correctly. The effect of the bug was incorrect
4330 behaviour, such as trying another host, and this could lead to a crash.
4336 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4337 and daemon.c were passed as pointers to ints; they should have been
4338 pointers to socklen_t variables (which are typically unsigned ints).
4340 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4343 3. Fixed a really odd bug that affected only the testing scheme; patching a
4344 certain fixed string in the binary changed the value of another string that
4345 happened to be identical to the end of the original first string.
4347 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4348 name", it returns that address as the IP address. On some operating
4349 systems (e.g. Solaris), it also passes back the IP address string as the
4350 "host name". However, on others (e.g. Linux), it passes back an empty
4351 string. Exim wasn't checking for this, and was changing the host name to an
4352 empty string, assuming it had been canonicized.
4354 5. Although rare, it is permitted to have more than one PTR record for a given
4355 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4356 all the names associated with an address, because they do in Solaris.
4357 However, it seems that they do not in Linux for data that comes from the
4358 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4359 I found this out when I moved to a new Linux workstation and tried to run
4360 the Exim test suite.
4362 To get round this problem I have changed the code so that it now does its
4363 own call to the DNS to look up PTR records when searching for a host name.
4364 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4365 addresses that are only in /etc/hosts are still found.
4367 This behaviour is, however, controlled by an option called host_lookup_
4368 order, which defaults to "bydns:byaddr". If people want to use the other
4369 order, or indeed, just use one or the other means of lookup, they can
4370 specify it in this variable.
4372 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4373 some operating systems, this comes back from gethostbyaddr() as an empty
4374 string, and this is what Exim used to test for. However, it seems that in
4375 other systems, "." is yielded. Exim now tests for this case too.
4377 7. The values of check_spool_space and check_log_space are now held internally
4378 as a number of kilobytes instead of an absolute number of bytes. If a
4379 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4380 kilobyte. This means that much larger values can be stored.
4382 8. Exim monitor: an attempt to get the action menu when not actually pointing
4383 at a message produces an empty menu entitled "No message selected". This
4384 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4385 no entries in it ("Shell widget menu has zero width and/or height"). So I
4386 have added a single, blank menu entry in this case.
4388 9. Added ${quote_local_part.
4390 10. MIME decoding is now applied to the contents of Subject: header lines when
4393 11. Now that a reference to $sender_host_address automatically causes a reverse
4394 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4395 host lookup before query-style lookups in lists that might use this
4396 variable. This has therefore been abolished, and the "net-" prefix is no
4397 longer necessary for query-style lookups.
4399 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4400 have been a typo for LFLAGS, so it has been changed.
4402 13. The install script calls Exim with "-C /dev/null" in order to find the
4403 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4404 to be output. Howeve, since Exim outputs its version number before the
4405 error, it didn't break the script. It just looked ugly. I fixed this by
4406 always allowing "-C /dev/null" if the caller is root.
4408 14. Ignore overlarge ACL variable number when reading spool file - insurance
4409 against a later release with more variables having written the file.
4411 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4412 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4413 this, as well as the form without the "IPv6" on the front.
4415 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4416 OS/Makefile-Darwin file.
4418 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4419 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4421 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4422 variable. However, some OS use ints for inodes. I've added cast to long int
4423 to get rid of the compiler warning.
4425 19. I had forgotten to lock out "/../" in configuration file names when
4426 ALT_CONFIG_PREFIX was set.
4428 20. Routers used for verification do not need to specify transports. However,
4429 if such a router generated a host list, and callout was configured, Exim
4430 crashed, because it could not find a port number from the (non-existent)
4431 transport. It now assumes port 25 in this circumstance.
4433 21. Added the -t option to exigrep.
4435 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4436 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4437 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4438 them is set, LOOKUP_LSEARCH is forced.
4440 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4442 24. Added sender and host information to the "rejected by local_scan()" log
4443 line; previously there was no indication of these.
4445 25. Added .include_if_exists.
4447 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4448 Linux. It turns out that not all file systems support this. Apparently some
4449 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4450 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4451 not supported on the file descriptor, is now ignored when Exim is trying to
4452 sync a directory. This applies only to Linux.
4454 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4456 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4457 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4460 29. Added support for maildirsize files from supplied patch (modified a bit).
4462 30. The use of :fail: followed by an empty string could lead Exim to respond to
4463 sender verification failures with (e.g.):
4465 550 Verification failed for <xxx>
4466 550 Sender verify failed
4468 where the first response line was missing the '-' that indicates it is not
4469 the final line of the response.
4471 31. The loop for finding the name of the user that called Exim had a hardwired
4472 limit of 10; it now uses the value of finduser_retries, which is used for
4473 all other user lookups.
4475 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4478 33. Exim was neglecting to zero errno before one call of strtol() when
4479 expanding a string and expecting an integer value. On some systems this
4480 resulted in spurious "integer overflow" errors. Also, it was casting the
4481 result into an int without checking.
4483 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4484 did not work. The code looks as if it has *never* worked, though it appears
4485 to have been documented since at least releast 1.62. I have made it work.
4487 35. The "timeout_DNS" error in retry rules, also documented since at least
4488 1.62, also never worked. As it isn't clear exactly what this means, and
4489 clearly it isn't a major issue, I have abolished the feature by treating it
4490 as "timeout", and writing a warning to the main and panic logs.
4492 36. The display of retry rules for -brt wasn't always showing the error code
4495 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4496 timeout_connect_A, timeout_connect_MX.
4498 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4499 to the empty sender.
4501 39. The daemon was not analysing the content of -oX till after it had closed
4502 stderr and disconnected from the controlling terminal. This meant that any
4503 syntax errors were only noted on the panic log, and the return code from
4504 the command was 0. By re-arranging the code a little, I've made the
4505 decoding happen first, so such errors now appear on stderr, and the return
4506 code is 1. However, the actual setting up of the sockets still happens in
4507 the disconnected process, so errors there are still only recorded on the
4510 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4511 connections (as happens on some IP stacks) was logged at start up time as
4512 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4513 it from "IPv6 and IPv4", which means that two separate sockets are being
4516 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4517 says whether AF_INET or AF_INET6 was passed as an argument.
4519 42. Exiwhat output was messed up when time zones were included in log
4522 43. Exiwhat now gives more information about the daemon's listening ports,
4523 and whether -tls-on-connect was used.
4525 44. The "port" option of the smtp transport is now expanded.
4527 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4528 silently ignored. Now an error message is written to the main and panic
4531 46. There's a new ACL modifier called "logwrite" which writes to a log file
4532 as soon as it is encountered.
4534 47. Added $local_user_uid and $local_user_gid at routing time.
4536 48. Exim crashed when trying to verify a sender address that was being
4539 49. Exim was recognizing only a space character after ".include". It now also
4540 recognizes a tab character.
4542 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4543 extracting the relevant information from the specification. The man page no
4544 longer contains scrambled data for the -d option, and I've added a section
4545 at the front about calling Exim under different names.
4547 51. Added "extra_headers" argument to the "mail" command in filter files.
4549 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4552 53. Installed eximstats 1.29.
4554 54. Added transport_filter_timeout as a generic transport option.
4556 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4557 Cc: header lines. This was required by RFC 822, but it not required by RFC
4560 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4561 incoming messages that did not have them. Now it does so only if the
4562 message originates locally, that is, if there is no associated remote host
4563 address. When Resent- header lines are present, this applies to the Resent-
4564 lines rather than the non-Resent- lines.
4566 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4567 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4569 58. Messages for configuration errors now include the name of the main
4570 configuration file - useful now that there may be more than one file in a
4571 list (.included file names were always shown).
4573 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4574 for those rare installations that do not start the daemon as root or run it
4575 setuid root. I've cut out the call to initgroups() if the daemon is not
4578 60. The Exim user and group can now be bound into the binary as text strings
4579 that are looked up at the start of Exim's processing.
4581 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4583 62. Added $mailstore_basename variable.
4585 63. Installed patch to sieve.c from Michael Haardt.
4587 64. When Exim failed to open the panic log after failing to open the main log,
4588 the original message it was trying to log was written to stderr and debug
4589 output, but if they were not available (the usual case in production), it
4590 was lost. Now it is written to syslog before the two lines that record the
4591 failures to open the logs.
4593 65. Users' Exim filters run in subprocesses under the user's uid. It is
4594 possible for a "deliver" command or an alias in a "personal" command to
4595 provoke an address rewrite. If logging of address rewriting is configured,
4596 this fails because the process is not running as root or exim. There may be
4597 a better way of dealing with this, but for the moment (because 4.30 needs
4598 to be released), I have disabled address rewrite logging when running a
4599 filter in a non-root, non-exim process.
4605 1. The buildconfig auxiliary program wasn't quoting the value set for
4606 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4607 not defined. This bug was masked in 4.22 by the effect that was fixed in
4610 2. Some messages that were rejected after a message id was allocated were
4611 shown as "incomplete" by exigrep. It no longer does this for messages that
4612 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4614 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4615 have allow_domain_literals set, the ID did not get logged in the <= line.
4616 Domain literals are now always recognized in Message-ID: header lines.
4618 4. The first argument for a ${extract expansion item is the key name or field
4619 number. Leading and trailing spaces in this item were not being ignored,
4620 causing some misleading effects.
4622 5. When deliver_drop_privilege was set, single queue runner processes started
4623 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4624 same command in the process it spins off) were not dropping privilege.
4626 6. When the daemon running as "exim" started a queue runner, it always
4627 re-executed Exim in the spun-off process. This is a waste of effort when
4628 deliver_drop_privilege is set. The new process now just calls the
4629 queue-runner function directly.
4635 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4638 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4639 ignored. Though the use of -oP was forcing the writing of a pid file, it
4640 was always written to the default place.
4642 3. If the message "no IP address found for host xxxx" is generated during
4643 incoming verification, it is now followed by identification of the incoming
4644 connection (so you can more easily find what provoked it).
4646 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4648 5. Added some features to "harden" Exim a bit more against certain attacks:
4650 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4651 be put in Local/Makefile. This is like the never_users runtime option,
4652 but it cannot be overridden. The default setting is "root".
4654 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4655 prefix string with which any file named in a -C command line option
4658 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4659 is retained for -C and -D only if the caller of Exim is root. Without
4660 it, the exim user may also use -C and -D and retain privilege.
4662 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4663 command line option is disabled.
4665 6. Macro names set by the -D option must start with an upper case letter, just
4666 like macro names defined in the configuration file.
4668 7. Added "dereference=" facility to LDAP.
4670 8. Two instances of the typo "uknown" in the source files are fixed.
4672 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4673 the Configure-Makefile script screwed up while processing it.
4675 10. Incorporated PCRE 4.4.
4677 11. The SMTP synchronization check was not operating right at the start of an
4678 SMTP session. For example, it could not catch a HELO sent before the client
4679 waited for the greeting. There is now a check for outstanding input at the
4680 point when the greeting is written. Because of the duplex, asynchronous
4681 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4682 way, but not yet received, when the check is performed.
4684 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4685 on TCP/IP sockets, because this apparently causes some broken clients to
4688 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4689 unchanged) from the Cygwin maintainer.
4691 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4692 was supported instead of testing for mbx. Effectively a typo.
4694 15. The spa authenticator server code was not checking that the input it
4695 received was valid base64.
4697 16. The debug output line for the "set" modifier in ACLs was not showing the
4698 name of the variable that was being set.
4700 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4702 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4703 lookup on its own. Something else was needed to trigger the lookup. For
4704 example, a match in host_lookup or the need for a host name in a host list.
4705 Now, if $sender_host_name is referenced and the host name has not yet been
4706 looked up, a lookup is performed. If the lookup fails, the variable remains
4707 empty, and $host_lookup_failed is set to "1".
4709 19. Added "eqi" as a case-independent comparison operator.
4711 20. The saslauthd authentication condition could segfault if neither service
4712 nor realm was specified.
4714 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4715 error message that was logged was misleading, and incoming SMTP
4716 connections were dropped. The message is now more accurate, and temporary
4717 errors are given to SMTP connections.
4719 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4720 (except RSET) with a 421 error, until QUIT is received. However, it was
4721 failing to send a response to QUIT.
4723 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4724 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4725 ACL is now run after the helo_try_verify_hosts code.
4727 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4728 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4729 case-independent, so other case variants are also recognized). Apparently
4730 some systems use these upper case variants.
4732 25. If more than two messages were waiting for the same host, and a transport
4733 filter was specified for the transport, Exim sent two messages over the
4734 same TCP/IP connection, and then failed with "socket operation on non-
4735 socket" when it tried to send the third.
4737 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4739 27. The extern definition of crypt16() in expand.c was not being excluded when
4740 the OS had its own crypt16() function.
4742 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4743 as a preferred synonym for return_size_limit, both as an option and as an
4746 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4748 30. Changed the default configuration ACL to relax the local part checking rule
4749 for addresses that are not in any local domains. For these addresses,
4750 slashes and pipe symbols are allowed within local parts, but the sequence
4751 /../ is explicitly forbidden.
4753 31. SPA server authentication was not clearing the challenge buffer before
4756 32. log_message in a "warn" ACL statement was writing to the reject log as
4757 well as to the main log, which contradicts the documentation and doesn't
4758 seem right (because no rejection is happening). So I have stopped it.
4760 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4761 However, I am unable to do any testing of this.
4763 34. Fixed an infelicity in the appendfile transport. When checking directories
4764 for a mailbox, to see if any needed to be created, it was accidentally
4765 using path names with one or more superfluous leading slashes; tracing
4766 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4768 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4769 contents are added to the log line that is written for every discarded
4770 recipient. (Previously a log_message setting was ignored.)
4772 36. The ${quote: operator now quotes the string if it is empty.
4774 37. The install script runs exim in order to find its version number. If for
4775 some reason other than non-existence or emptiness, which it checks, it
4776 could not run './exim', it was installing it with an empty version number,
4777 i.e. as "exim-". This error state is now caught, and the installation is
4780 38. An argument was missing from the function that creates an error message
4781 when Exim fails to connect to the socket for saslauthd authentication.
4782 This could cause Exim to crash, or give a corrupted message.
4784 39. Added isip, isip4, and isip6 to ${if conditions.
4786 40. The ACL variables $acl_xx are now saved with the message, and can be
4787 accessed later in routers, transports, and filters.
4789 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4790 strings in the file are not string-expanded.
4792 42. If a MAIL command specified a SIZE value that was too large to fit into an
4793 int variable, the check against message_size_limit failed. Such values are
4794 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4795 day this will have to be increased, but I don't think I want to be around
4796 when emails are that large.
4803 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4804 iconv() is not standard in FreeBSD.
4806 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4807 IPv6 enabled. The observed symptom was a segmentation fault on return from
4808 the function os_common_find_running_interfaces() in src/os.c.
4810 3. In the check_special_case() function in daemon.c I had used "errno" as an
4811 argument name, which causes warnings on some systems. This was basically a
4812 typo, since it was named "eno" in the comments!
4814 4. The code that waits for the clock to tick (at a resolution of some fraction
4815 of a second) so as to ensure message-id uniqueness was always waiting for
4816 at least one whole tick, when it could have waited for less. [This is
4817 almost certainly not relevant at current processor speeds, where it is
4818 unlikely to ever wait at all. But we try to future-proof.]
4820 5. The function that sleeps for a time interval that includes fractions of a
4821 second contained a race. It did not block SIGALRM between setting the
4822 timer, and suspending (a couple of lines later). If the interval was short
4823 and the sigsuspend() was delayed until after it had expired, the suspension
4824 never ended. On busy systems this could lead to processes getting stuck for
4827 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4828 process, before it forks any delivery processes. The open lookup caching
4829 mechanism meant that the open file or database connection was passed into
4830 the delivery process. The problem was that delivery processes always tidy
4831 up cached lookup data. This could cause a problem for the next delivery
4832 process started by the queue runner, because the external queue runner
4833 process does not know about the closure. So the next delivery process
4834 still has data in the lookup cache. In the case of a file lookup, there was
4835 no problem because closing a file descriptor in a subprocess doesn't affect
4836 the parent. However, if the lookup was caching a connection to a database,
4837 the connection was closed, and the second delivery process was likely to
4838 see errors such as "PGSQL: query failed: server closed the connection
4839 unexpectedly". The problem has been fixed by closing all cached lookups
4840 in a queue runner before running a delivery process.
4842 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4843 seem to have the "const" qualifier which it has on other OS. I've
4846 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4847 *of the same type* (client or server) with the same public name that an
4848 error should be diagnosed.
4850 9. When Exim looked up a host name for an IP address, but failed to find the
4851 original IP address when looking up the host name (a safety check), it
4852 output the message "<ip address> does not match any IP for NULL", which was
4853 confusing, to say the least. The bug was that the host name should have
4854 appeared instead of "NULL".
4856 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4857 user that is built into the binary, and the -C or -D options is used, root
4858 privilege is dropped before the configuration file is read. In addition,
4859 logging is switched to stderr instead of the normal log files. If the
4860 configuration then re-defines the Exim user, the unprivileged environment
4861 is probably not what is expected, so Exim logs a panic warning message (but
4864 However, if deliver_drop_privilege is set, the unprivileged state may well
4865 be exactly what is intended, so the warning has been cut out in that case,
4866 and Exim is allowed to try to write to its normal log files.
4872 1. smtp_return_error_details was not giving details for temporary sender
4873 or receiver verification errors.
4875 2. Diagnose a configuration error if two authenticators have the same public
4878 3. Exim used not to create the message log file for a message until the first
4879 delivery attempt. This could be confusing when incoming messages were held
4880 for policy or load reasons. The message log file is now created at the time
4881 the message is received, and an initial "Received" line is written to it.
4883 4. The automatically generated man page for command line options had a minor
4884 bug that caused no ill effects; however, a more serious problem was that
4885 the procedure for building the man page automatically didn't always
4886 operate. Consequently, release 4.20 contains an out-of-date version. This
4887 shouldn't happen again.
4889 5. When building Exim with embedded Perl support, the script that builds the
4890 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4891 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4893 6. The freeze_tell option was not being used for messages that were frozen on
4894 arrival, either by an ACL or by local_scan().
4896 7. Added the smtp_incomplete_transaction log selector.
4898 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4899 was accepting AUTH without a new EHLO.
4901 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4902 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4905 10. Logging of TCP/IP connections (when configured) now happens in the main
4906 daemon process instead of the child process, so that the TCP/IP connection
4907 count is more accurate (but it can never be perfect).
4909 11. The use of "drop" in a nested ACL was not being handled correctly in the
4910 outer ACL. Now, if condition failure induced by the nested "drop" causes
4911 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4912 or "require"), the connection is dropped.
4914 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4915 that yield "discard" can now be used with an "accept" or a "discard" verb,
4916 but an error is generated for any others (because I can't see a useful way
4917 to define what should happen).
4919 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4920 are now processed in the same way as lines in the Exim configuration file.
4921 In particular, continuation lines are supported.
4923 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4925 15. Added -ti meaning -t -i.
4927 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4928 domains, and warn by logging if others are found.
4930 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4931 returned from GIFCONF when Exim is trying to find the list of interfaces on
4932 a host. The code in os.c has been modified to copy each ifreq to an aligned
4933 structure in all cases.
4935 Also, in some cases, the returned ifreq's were being copied to a 'struct
4936 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4937 means the last couple of bytes of an IPv6 address could be chopped if the
4938 ifreq contained only a normal sockaddr (14 bytes storage).
4940 18. Named domain lists were not supported in the hosts_treat_as_local option.
4941 An entry such as +xxxx was not recognized, and was treated as a literal
4944 19. Ensure that header lines added by a DATA ACL are included in the reject log
4945 if the ACL subsequently rejects the message.
4947 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4948 MD5 (which is deprecated).
4950 21. When testing a filter file using -bf, Exim was writing a message when it
4951 took the sender from a "From " line in the message, but it was not doing so
4952 when it took $return_path from a Return-Path: header line. It now does.
4954 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4955 with a valid header line field name (a series of printing characters
4956 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4958 23. Changed "disc" in the source to "disk" to conform to the documentation and
4959 the book and for uniformity.
4961 24. Ignore Sendmail's -Ooption=value command line item.
4963 25. When execve() failed while trying to run a command in a pipe transport,
4964 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4965 could be confused with a return value of 69 from the command itself. This
4966 has been changed to 127, the value the shell returns if it is asked to run
4967 a non-existent command. The wording for the related log line suggests a
4968 non-existent command as the problem.
4970 26. If received_header_text expands to an empty string, do not add a Received:
4971 header line to the message. (Well, it adds a token one on the spool, but
4972 marks it "old" so that it doesn't get used or transmitted.)
4974 27. Installed eximstats 1.28 (addition of -nt option).
4976 28. There was no check for failure on the call to getsockname() in the daemon
4977 code. This can fail if there is a shortage of resources on the system, with
4978 ENOMEM, for example. A temporary error is now given on failure.
4980 29. Contrary to the C standard, it seems that in some environments, the
4981 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4982 program. Exim now does this explicitly; it affects the formatting of
4983 timestamps using strftime().
4985 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4986 complaints. I've now initialized all the variables, to avoid this.
4988 32. Header lines added by a system filter were not being "seen" during
4989 transport-time rewrites.
4991 33. The info_callback() function passed to OpenSSL is set up with type void
4992 (*)(SSL *, int, int), as described somewhere. However, when calling the
4993 function (actually a macro) that sets it up, the type void(*)() is
4994 expected. I've put in a cast to prevent warnings from picky compilers.
4996 34. If a DNS black list lookup found a CNAME record, but there were no A
4997 records associated with the domain it pointed at, Exim crashed.
4999 35. If a DNS black list lookup returned more than one A record, Exim ignored
5000 all but the first. It now scans all returned addresses if a particular IP
5001 value is being sought. In this situation, the contents of the
5002 $dnslist_value variable are a list of all the addresses, separated by a
5005 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5006 to accept a host name and all its aliases if the forward lookup for any of
5007 them yielded the IP address of the incoming connection. Now it accepts only
5008 those names whose forward lookup yields the correct IP address. Any other
5009 names are discarded. This closes a loophole whereby a rogue DNS
5010 administrator could create reverse DNS records to break through a
5011 wildcarded host restriction in an ACL.
5013 37. If a user filter or a system filter that ran in a subprocess used any of
5014 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5015 the wrong values were passed to the pipe command ($thisaddress had the
5016 value of $0, $0 had the value of $1, etc). This bug was introduced by
5017 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5019 38. Improved the line breaking for long SMTP error messages from ACLs.
5020 Previously, if there was no break point between 40 and 75 characters, Exim
5021 left the rest of the message alone. Two changes have been made: (a) I've
5022 reduced the minimum length to 35 characters; (b) if it can't find a break
5023 point between 35 and 75 characters, it looks ahead and uses the first one
5024 that it finds. This may give the occasional overlong line, but at least the
5025 remaining text gets split now.
5027 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5028 file descriptors might be low, and that setting 1000 would always raise it.
5029 It turns out that in some environments, the limit is already over 1000 and
5030 that lowering it causes trouble. So now Exim takes care not to decrease it.
5032 40. When delivering a message, the value of $return_path is set to $sender_
5033 address at the start of routing (routers may change the value). By an
5034 oversight, this default was not being set up when an address was tested by
5035 -bt or -bv, which affected the outcome if any router or filter referred to
5038 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5039 log only when "message" or "log_message" are set. However, if one of the
5040 conditions was an address verification, or a call to a nested ACL, the
5041 messages generated by the underlying test were being passed through. This
5042 no longer happens. The underlying message is available in $acl_verify_
5043 message for both "message" and "log_message" expansions, so it can be
5044 passed through if needed.
5046 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5047 new expansion $bh_ to give the encoded byte string without charset
5048 translation. Translation happens only if iconv() is available; HAVE_ICONV
5049 indicates this at build time. HEADERS_CHARSET gives the charset to
5050 translate to; headers_charset can change it in the configuration, and
5051 "headers charset" can change it in an individual filter file.
5053 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5054 that creates RFC 2047 encoded "words" labels them as that charset instead
5055 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5056 expansion operator; (ii) when Exim creates a From: line for a local
5057 message; (iii) when a header line is rewritten to include a "phrase" part.
5059 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5060 buggy, causing it to skip the first lines of messages whose message ID
5061 ended in 'D'. This would not have bitten before Exim release 4.14, because
5062 message IDs were unlikely to end in 'D' before then. The effect was to have
5063 incorrect size information for certain domains.
5065 45. #include "config.h" was missing at the start of the crypt16.c module. This
5066 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5069 46. If there was a timeout during a "random" callout check, Exim treated it as
5070 a failure of the random address, and carried on sending RSET and the real
5071 address. If the delay was just some slowness somewhere, the response to the
5072 original RCPT would be taken as a response to RSET and so on, causing
5073 mayhem of various kinds.
5075 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5076 when I implemented it. It didn't allow for the fact that some option values
5077 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5078 the right test for positive values.
5080 48. Domain names in DNS records are case-independent. Exim always looks them up
5081 in lower case. Some resolvers return domain names in exactly the case they
5082 appear in the zone file, that is, they may contain uppercase letters. Not
5083 all resolvers do this - some return always lower case. Exim was treating a
5084 change of case by a resolver as a change of domain, similar to a widening
5085 of a domain abbreviation. This triggered its re-routing code and so it was
5086 trying to route what was effectively the same domain again. This normally
5087 caused routing to fail (because the router wouldn't handle the domain
5088 twice). Now Exim checks for this case specially, and just changes the
5089 casing of the domain that it ultimately uses when it transmits the message
5092 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5095 50. If a filter generated a file delivery with a non-absolute name (possible if
5096 no home directory exists for the router), the forbid_file option was not
5099 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5100 the existing equality matching.
5102 52. Exim was using ints instead of ino_t variables in some places where it was
5103 dealing with inode numbers.
5105 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5106 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5107 called TMPDIR, and if it finds it is different, it changes its value.
5109 54. The smtp_printf() function is now made available to local_scan() so
5110 additional output lines can be written before returning. There is also an
5111 smtp_fflush() function to enable the detection of a dropped connection.
5112 The variables smtp_input and smtp_batched_input are exported to
5115 55. Changed the default runtime configuration: the message "Unknown user"
5116 has been removed from the ACL, and instead placed on the localuser router,
5117 using the cannot_route_message feature. This means that any verification
5118 failures that generate their own messages won't get overridden. Similarly,
5119 the "Unrouteable address" message that was in the ACL for unverifiable
5120 relay addresses has also been removed.
5122 56. Added hosts_avoid_esmtp to the smtp transport.
5124 57. The exicyclog script was not checking for the esoteric option
5125 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5126 will work only if exicyclog is run under the appropriate euid.
5128 58. Following a discussion on the list, the rules by which Exim recognises line
5129 endings on incoming messages have been changed. The -dropcr and drop_cr
5130 options are now no-ops, retained only for backwards compatibility. The
5131 following line terminators are recognized: LF CRLF CR. However, special
5132 processing applies to CR:
5134 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5135 nor a local message in the state where . is a terminator.
5137 (ii) If a bare CR is encountered in a header line, an extra space is added
5138 after the line terminator so as not to end the header. The reasoning
5139 behind this is that bare CRs in header lines are most likely either
5140 to be mistakes, or people trying to play silly games.
5142 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5143 was being incorrectly given as 18 bytes larger than it should have been.
5144 This is a VOB (very old bug).
5146 60. This may never have affected anything current, but just in case it has:
5147 When the local host is found other than at the start of a list of hosts,
5148 the local host, those with the same MX, and any that follow, are discarded.
5149 When the list in question was part of a longer list of hosts, the following
5150 hosts (not currently being processed) were also being discarded. This no
5151 longer happens. I'm not sure if this situation could ever has previously
5154 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5157 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5158 header. This is something that is recommended in a new Internet Draft, and
5159 is something that is documented as being done by Sendmail. There are two
5160 possible values. For messages generated by the autoreply transport, Exim
5163 Auto-Submitted: auto-replied
5165 whereas for all other generated messages (e.g. bounces) it adds
5167 Auto-Submitted: auto-generated
5169 63. The "personal" condition in filters now includes a test for the
5170 Auto-Submitted: header. If it contains the string "auto-" the message it
5171 not considered personal.
5173 64. Added rcpt_include_affixes as a generic transport option.
5175 65. Added queue_only_override (default true).
5177 66. Added the syslog_duplication option.
5179 67. If what should have been the first header line of a message consisted of
5180 a space followed by a colon, Exim was mis-interpreting it as a header line.
5181 It isn't of course - it is syntactically invalid and should therefore be
5182 treated as the start of the message body. The misbehaviour could have
5183 caused a number of strange effects, including loss of data in subsequent
5184 header lines, and spool format errors.
5186 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5187 client host had authenticated. This control can now be exercised by an ACL
5188 for more flexibility.
5190 69. By default, callouts do not happen when testing with -bh. There is now a
5191 variant, -bhc, which does actually run the callout code, including
5192 consulting and updating the callout cache.
5194 70. Added support for saslauthd authentication, courtesy of Alexander
5197 71. If statvfs() failed on the spool or log directories while checking their
5198 size for availability, Exim confusingly gave the error "space shortage".
5199 Furthermore, in debugging mode it crashed with a floating point exception.
5200 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5201 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5202 serious problem, Exim now writes to the main and panic logs when this
5203 happens, with details of the failure. It then refuses to accept the
5204 incoming message, giving the message "spool directory problem" or "log
5205 directory problem" with a 421 code for SMTP messages.
5207 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5208 0, 1, and 2 exist, because some OS complain for execs without them (see
5209 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5210 descriptors. However, the code omitted to check that the open succeeded,
5211 causing mysterious errors if for some reason the permissions on /dev/null
5212 got screwed. Now Exim writes a message to the main and panic logs, and
5213 bombs out if it can't open /dev/null.
5215 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5216 interact so that it is all more flexible. It is supposed to remain
5217 backwards compatible. Also added extra_local_interfaces.
5219 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5220 to bomb out with an assertion failure - to the client this appears as a
5221 connection drop. This problem occurs in the part of the code that was taken
5222 from the Samba project. Fortunately, the assertion is in a very simple
5223 function, so I have fixed this by reproducing the function inline in the
5224 one place where it is called, and arranging for authentication to fail
5225 instead of killing the process with assert().
5227 75. The SPA client code was not working when the server requested OEM rather
5228 than Unicode encoding.
5230 76. Added code to make require_files with a specific uid setting more usable in
5231 the case where statting the file as root fails - usually a non-root-mounted
5232 NFS file system. When this happens and the failure is EACCES, Exim now
5233 forks a subprocess and does the per-uid checking as the relevant uid.
5235 77. Added process_log_path.
5237 78. If log_file_path was not explicitly set, a setting of check_log_space or
5238 check_log_inodes was ignored.
5240 79. If a space check for the spool or log partitions fails, the incident is now
5241 logged. Of course, in the latter case the data may get lost...
5243 80. Added the %p formatting code to string_format() so that it can be used to
5244 print addresses in debug_print(). Adjusted all the address printing in the
5245 debugging in store.c to use %p rather than %d.
5247 81. There was a concern that a line of code in smtp_in.c could overflow a
5248 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5249 initially expressed, the concern was not well-founded, because trailing
5250 spaces are removed early. However, if the trailing spaces were followed by
5251 a NULL, they did not get removed, so the overflow was possible. Two fixes
5254 (a) I re-wrote the offending code in a cleaner fashion.
5255 (b) If an incoming SMTP command contains a NULL character, it is rejected
5258 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5259 runs initgroups(), so that if the Exim user is in any additional groups,
5260 they will be used during message reception.
5266 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob