1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
48 TF/01 Failure to get a lock on a hints database can have serious
49 consequences so log it to the panic log.
51 TF/02 Log LMTP confirmation messages in the same way as SMTP,
52 controlled using the smtp_confirmation log selector.
54 TF/03 Include the error message when we fail to unlink a spool file.
56 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
57 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
58 for maintaining out-of-tree patches for some time.
60 PP/01 Bugzilla 139: Documentation and portability issues.
61 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
62 Handle per-OS dynamic-module compilation flags.
64 PP/02 Let /dev/null have normal permissions.
65 The 4.73 fixes were a little too stringent and complained about the
66 permissions on /dev/null. Exempt it from some checks.
67 Reported by Andreas M. Kirchwitz.
69 PP/03 Report version information for many libraries, including
70 Exim version information for dynamically loaded libraries. Created
71 version.h, now support a version extension string for distributors
72 who patch heavily. Dynamic module ABI change.
74 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
75 privilege escalation vulnerability whereby the Exim run-time user
76 can cause root to append content of the attacker's choosing to
79 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
82 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
83 If dropping privileges for untrusted macros, we disabled normal logging
84 on the basis that it would fail; for the Exim run-time user, this is not
85 the case, and it resulted in successful deliveries going unlogged.
86 Fixed. Reported by Andreas Metzler.
92 PP/01 Date: & Message-Id: revert to normally being appended to a message,
93 only prepend for the Resent-* case. Fixes regression introduced in
94 Exim 4.70 by NM/22 for Bugzilla 607.
96 PP/02 Include check_rfc2047_length in configure.default because we're seeing
97 increasing numbers of administrators be bitten by this.
99 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
101 PP/03 Bugzilla 994: added openssl_options main configuration option.
103 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
105 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
107 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
109 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
110 without a peer certificate, leading to a segfault because of an
111 assumption that peers always have certificates. Be a little more
112 paranoid. Problem reported by Martin Tscholak.
114 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
115 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
116 NB: ClamAV planning to remove STREAM in "middle of 2010".
117 CL also introduces -bmalware, various -d+acl logging additions and
118 more caution in buffer sizes.
120 PP/09 Implemented reverse_ip expansion operator.
122 PP/10 Bugzilla 937: provide a "debug" ACL control.
124 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
126 PP/12 Bugzilla 973: Implement --version.
128 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
130 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
132 PP/15 Bugzilla 816: support multiple condition rules on Routers.
134 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
135 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
136 ignore trailing whitespace.
138 JJ/02 prevent non-panic DKIM error from being sent to paniclog
140 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
143 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
144 Notification from Dr Andrew Aitchison.
146 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
147 ExtendedDetectionInfo response format.
148 Notification from John Horne.
150 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
153 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
154 XSL and documented dependency on system catalogs, with examples of how
157 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
160 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
161 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
162 configuration file which is writeable by the Exim user or group.
164 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
165 of configuration files to cover files specified with the -C option if
166 they are going to be used with root privileges, not just the default
169 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
170 option (effectively making it always true).
172 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
173 files to be used while preserving root privileges.
175 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
176 that rogue child processes cannot use them.
178 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
179 run-time user, instead of root.
181 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
182 Exim run-time user without dropping privileges.
184 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
185 result string, instead of calling string_vformat() twice with the same
188 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
189 for other users. Others should always drop root privileges if they use
190 -C on the command line, even for a whitelisted configure file.
192 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
194 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
200 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
201 $data_path, and $header_path variables; fixed documentation bugs and
204 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
205 exipick to access non-standard spools, including the "frozen" queue
208 NM/01 Bugzilla 965: Support mysql stored procedures.
209 Patch from Alain Williams
211 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
213 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
214 Patch from Andreas Metzler
216 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
217 Patch from Kirill Miazine
219 NM/05 Bugzilla 671: Added umask to procmail example.
221 JJ/03 installed exipick 20100323.0, fixing doc bug
223 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
224 directory. Notification and patch from Dan Rosenberg.
226 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
228 TK/02 Improve log output when DKIM signing operation fails.
230 MH/01 Treat the transport option dkim_domain as a colon separated
231 list, not as a single string, and sign the message with each element,
232 omitting multiple occurences of the same signer.
234 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
235 Bugzilla 985, 986. Patch by Simon Arlott
237 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
238 Patch by Simon Arlott
240 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
241 MBX locking. Notification from Dan Rosenberg.
247 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
249 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
251 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
253 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
255 NM/04 Bugzilla 915: Flag broken perl installation during build.
261 TK/01 Added patch by Johannes Berg that expands the main option
262 "spamd_address" if it starts with a dollar sign.
264 TK/02 Write list of recipients to X-Envelope-Sender header when building
265 the mbox-format spool file for content scanning (suggested by Jakob
268 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
269 (http://www.dcc-servers.net/) support via dccifd. Activated by
270 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
272 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
273 by Mark Daniel Reidel <mr@df.eu>.
275 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
276 When building exim an external PCRE library is now needed -
277 PCRE is a system library on the majority of modern systems.
278 See entry on PCRE_LIBS in EDITME file.
280 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
281 conversation. Added nologin parameter to request.
282 Patch contributed by Kirill Miazine.
284 TF/01 Do not log submission mode rewrites if they do not change the address.
286 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
288 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
289 log files in place. Contributed by Roberto Lima.
291 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
293 TF/03 Bugzilla 615: When checking the local_parts router precondition
294 after a local_part_suffix or local_part_prefix option, Exim now
295 does not use the address's named list lookup cache, since this
296 contains cached lookups for the whole local part.
298 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
299 Robert Millan. Documentation is in experimental-spec.txt.
301 TF/04 Bugzilla 668: Fix parallel build (make -j).
303 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
305 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
306 Patch provided by Jan Srzednicki.
308 TF/05 Leading white space used to be stripped from $spam_report which
309 wrecked the formatting. Now it is preserved.
311 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
312 that they are available at delivery time.
314 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
316 TF/08 TLS error reporting now respects the incoming_interface and
317 incoming_port log selectors.
319 TF/09 Produce a more useful error message if an SMTP transport's hosts
320 setting expands to an empty string.
322 NM/06 Bugzilla 744: EXPN did not work under TLS.
323 Patch provided by Phil Pennock.
325 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
326 Patch provided by Richard Godbee.
328 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
329 acl_smtp_notquit, added index entry.
331 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
332 Patch provided by Eugene Bujak.
334 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
335 accept(). Patch provided by Maxim Dounin.
337 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
338 Patch provided by Phil Pennock.
340 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
342 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
343 Patch provided by Brad "anomie" Jorsch.
345 NM/14 Bugzilla 622: Added timeout setting to transport filter.
346 Patch provided by Dean Brooks.
348 TK/05 Add native DKIM support (does not depend on external libraries).
350 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
351 Patch provided by Graeme Fowler.
353 NM/16 Bugzilla 851: Documentation example syntax fix.
355 NM/17 Changed NOTICE file to remove references to embedded PCRE.
357 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
360 NM/19 Bugzilla 745: TLS version reporting.
361 Patch provided by Phil Pennock.
363 NM/20 Bugzilla 167: bool: condition support.
364 Patch provided by Phil Pennock.
366 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
367 clients. Patch provided by Phil Pennock.
369 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
370 Patch provided by Brad "anomie" Jorsch.
372 NM/23 Bugzilla 687: Fix misparses in eximstats.
373 Patch provided by Heiko Schlittermann.
375 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
376 Patch provided by Heiko Schlittermann.
378 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
379 plus update to original patch.
381 NM/26 Bugzilla 799: Documentation correction for ratelimit.
383 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
384 Patch provided by David Brownlee.
386 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
388 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
390 NM/30 Bugzilla 888: TLS documentation bugfixes.
392 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
394 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
395 Unlike the original bugzilla I have changed all shell scripts in src tree.
397 NM/33 Bugzilla 898: Transport filter timeout fix.
398 Patch by Todd Rinaldo.
400 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
401 Patch by Serge Demonchaux.
403 NM/35 Bugzilla 39: Base64 decode bug fixes.
404 Patch by Jakob Hirsch.
406 NM/36 Bugzilla 909: Correct connect() call in dcc code.
408 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
410 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
412 NM/39 Bugzilla 911: Fixed MakeLinks build script.
418 TK/01 Add preliminary DKIM support. Currently requires a forked version of
419 ALT-N's libdkim that I have put here:
420 http://duncanthrax.net/exim-experimental/
422 Note to Michael Haardt: I had to rename some vars in sieve.c. They
423 were called 'true' and it seems that C99 defines that as a reserved
424 keyword to be used with 'bool' variable types. That means you could
425 not include C99-style headers which use bools without triggering
426 build errors in sieve.c.
428 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
429 as mailq or other aliases. Changed the --help handling significantly
430 to do whats expected. exim_usage() emits usage/help information.
432 SC/01 Added the -bylocaldomain option to eximstats.
434 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
436 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
438 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
444 PH/01 Another patch from the Sieve maintainer.
446 PH/02 When an IPv6 address is converted to a string for single-key lookup
447 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
448 dots are used instead of colons so that keys in lsearch files need not
449 contain colons. This was done some time before quoting was made available
450 in lsearch files. However, iplsearch files do require colons in IPv6 keys
451 (notated using the quote facility) so as to distinguish them from IPv4
452 keys. This meant that lookups for IP addresses in host lists did not work
453 for iplsearch lookups.
455 This has been fixed by arranging for IPv6 addresses to be expressed with
456 colons if the lookup type is iplsearch. This is not incompatible, because
457 previously such lookups could never work.
459 The situation is now rather anomolous, since one *can* have colons in
460 ordinary lsearch keys. However, making the change in all cases is
461 incompatible and would probably break a number of configurations.
463 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
466 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
467 conversion specification without a maximum field width, thereby enabling
468 a rogue spamd server to cause a buffer overflow. While nobody in their
469 right mind would setup Exim to query an untrusted spamd server, an
470 attacker that gains access to a server running spamd could potentially
471 exploit this vulnerability to run arbitrary code as the Exim user.
473 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
474 $primary_hostname instead of what libspf2 thinks the hosts name is.
476 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
477 a directory entry by the name of the lookup key. Previously, if a
478 symlink pointed to a non-existing file or a file in a directory that
479 Exim lacked permissions to read, a lookup for a key matching that
480 symlink would fail. Now it is enough that a matching directory entry
481 exists, symlink or not. (Bugzilla 503.)
483 PH/03 The body_linecount and body_zerocount variables are now exported in the
486 PH/04 Added the $dnslist_matched variable.
488 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
489 This means they are set thereafter only if the connection becomes
492 PH/06 Added the client_condition to authenticators so that some can be skipped
493 by clients under certain conditions.
495 PH/07 The error message for a badly-placed control=no_multiline_responses left
496 "_responses" off the end of the name.
498 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
500 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
501 (without spaces) instead of just copying the configuration text.
503 PH/10 Added the /noupdate option to the ratelimit ACL condition.
505 PH/11 Added $max_received_linelength.
507 PH/12 Added +ignore_defer and +include_defer to host lists.
509 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
510 way in which PCRE > 7.0 is built.
512 PH/14 Implemented queue_only_load_latch.
514 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
515 MAIL command. The effect was to mangle the value on 64-bit systems.
517 PH/16 Another patch from the Sieve maintainer.
519 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
521 PH/18 If a system quota error occurred while trying to create the file for
522 a maildir delivery, the message "Mailbox is full" was not appended to the
523 bounce if the delivery eventually timed out. Change 4.67/27 below applied
524 only to a quota excession during the actual writing of the file.
526 PH/19 It seems that peer DN values may contain newlines (and other non-printing
527 characters?) which causes problems in log lines. The DN values are now
528 passed through string_printing() before being added to log lines.
530 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
531 and InterBase are left for another time.)
533 PH/21 Added message_body_newlines option.
535 PH/22 Guard against possible overflow in moan_check_errorcopy().
537 PH/23 POSIX allows open() to be a macro; guard against that.
539 PH/24 If the recipient of an error message contained an @ in the local part
540 (suitably quoted, of course), incorrect values were put in $domain and
541 $local_part during the evaluation of errors_copy.
547 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
548 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
551 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
552 issue a MAIL command.
554 PH/02 In an ACL statement such as
556 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
558 if a client was not listed at all, or was listed with a value other than
559 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
560 the condition was not true (as it should be), so access was not denied.
561 The bug was that the ! inversion was incorrectly passed on to the second
562 item. This has been fixed.
564 PH/03 Added additional dnslists conditions == and =& which are different from
565 = and & when the dns lookup returns more than one IP address.
567 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
568 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
570 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
571 FSYNC, which compiles an option called disable_fsync that allows for
572 bypassing fsync(). The documentation is heavily laced with warnings.
574 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
576 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
577 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
578 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
579 including adding "make clean"; (3) Added -fPIC when compiling the test
580 dynamically loaded module, to get rid of a warning.
582 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
583 message fails, move_frozen_messages = true and ignore_bounce_errors_after
584 = 0s. The bug is otherwise harmless.
586 PH/07 There was a bug in the dovecot authenticator such that the value of
587 $auth1 could be overwritten, and so not correctly preserved, after a
588 successful authentication. This usually meant that the value preserved by
589 the server_setid option was incorrect.
591 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
593 PH/09 Installed PCRE release 7.0.
595 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
596 run for batched SMTP input. It is now run at the start of every message
597 in the batch. While fixing this I discovered that the process information
598 (output by running exiwhat) was not always getting set for -bs and -bS
599 input. This is fixed, and it now also says "batched" for BSMTP.
601 PH/11 Added control=no_pipelining.
603 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
604 patch, slightly modified), and move the expansion of helo_data till after
605 the connection is made in the smtp transport (so it can use these
608 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
610 PH/14 Added log_selector = +pid.
612 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
614 PH/16 Add ${if forany and ${if forall.
616 PH/17 Added dsn_from option to vary the From: line in DSNs.
618 PH/18 Flush SMTP output before performing a callout, unless control =
619 no_callout_flush is set.
621 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
622 was true (the default) a successful delivery failed to delete the retry
623 item, thus causing premature timeout of the address. The bug is now
626 PH/20 Added hosts_avoid_pipelining to the smtp transport.
628 PH/21 Long custom messages for fakedefer and fakereject are now split up
629 into multiline reponses in the same way that messages for "deny" and
630 other ACL rejections are.
632 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
633 with slight modification.
635 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
636 draft, changing the syntax and factoring some duplicate code".
638 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
639 for deliveries of the second and subsequent messages over the same SMTP
642 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
643 ${reduce, with only minor "tidies".
645 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
647 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
648 expansion side effects.
650 PH/27 When a message times out after an over-quota error from an Exim-imposed
651 quota, the bounce message says "mailbox is full". This message was not
652 being given when it was a system quota that was exceeded. It now should
655 MH/03 Made $recipients available in local_scan(). local_scan() already has
656 better access to the recipient list through recipients_list[], but
657 $recipients can be useful in postmaster-provided expansion strings.
659 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
660 in the case of a MAIL command with additional options following the
661 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
662 were accidentally chopped off.
664 PH/29 SMTP synchronization checks are implemented when a command is read -
665 there is a check that no more input is waiting when there shouldn't be
666 any. However, for some commands, a delay in an ACL can mean that it is
667 some time before the response is written. In this time, more input might
668 arrive, invalidly. So now there are extra checks after an ACL has run for
669 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
670 pipelining has not been advertised.
672 PH/30 MH's patch to allow iscntrl() characters to be list separators.
674 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
675 returned in the SMTP response when smtp_return_error_details was false.
678 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
679 instead of the C I/O that was originally supplied, because problems were
682 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
683 Exim which did not show up earlier: it was assuming that a call to
684 SSL_CTX_set_info_callback() might give an error value. In fact, there is
685 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
686 was a macro that became an assignment, so it seemed to work. This has
687 changed to a proper function call with a void return, hence the compile
688 error. Exim's code has been fixed.
690 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
693 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
695 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
697 PH/37 If a message is not accepted after it has had an id assigned (e.g.
698 because it turns out to be too big or there is a timeout) there is no
699 "Completed" line in the log. When some messages of this type were
700 selected by exigrep, they were listed as "not completed". Others were
701 picked up by some special patterns. I have improved the selection
702 criteria to be more general.
704 PH/38 The host_find_failed option in the manualroute router can now be set
705 to "ignore", to completely ignore a host whose IP address cannot be
706 found. If all hosts are ignored, the behaviour is controlled by the new
707 host_all_ignored option.
709 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
710 homing or because of multiple MX records with /mx) generated more than
711 one IP address, and the following item turned out to be the local host,
712 all the secondary addresses of the first item were incorrectly removed
713 from the list, along with the local host and any following hosts (which
714 is what is supposed to happen).
716 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
717 whoever called Exim into the -H file. In the case of the daemon it was
718 behaving confusingly. When first started, it used values for whoever
719 started the daemon, but after a SIGHUP it used the Exim user (because it
720 calls itself on a restart). I have changed the code so that it now always
723 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
724 message are rejected with the same error (e.g. no authentication or bad
725 sender address), and a DATA command is nevertheless sent (as can happen
726 with PIPELINING or a stupid MUA), the error message that was given to the
727 RCPT commands is included in the rejection of the DATA command. This is
728 intended to be helpful for MUAs that show only the final error to their
731 PH/42 Another patch from the Sieve maintainer.
733 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
734 Eximstats - Fixed some broken HTML links and added missing column headers
736 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
737 columns for Rejects, Temp Rejects, Ham, and Spam rows.
739 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
741 PH/43 Yet another patch from the Sieve maintainer.
743 PH/44 I found a way to check for a TCP/IP connection going away before sending
744 the response to the final '.' that terminates a message, but only in the
745 case where the client has not sent further data following the '.'
746 (unfortunately, this is allowed). However, in many cases there won't be
747 any further data because there won't be any more messages to send. A call
748 to select() can be used: if it shows that the input is "ready", there is
749 either input waiting, or the socket has been closed. An attempt to read
750 the next input character can distinguish the two cases. Previously, Exim
751 would have sent an OK response which the client would never have see.
752 This could lead to message repetition. This fix should cure that, at
753 least in a lot of common cases.
755 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
756 advertised in response to EHLO.
762 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
763 fixed by 4.65/MH/01 (is this a record?) are fixed:
765 (i) An empty string was always treated as zero by the numeric comparison
766 operators. This behaviour has been restored.
768 (ii) It is documented that the numeric comparison operators always treat
769 their arguments as decimal numbers. This was broken in that numbers
770 starting with 0 were being interpreted as octal.
772 While fixing these problems I realized that there was another issue that
773 hadn't been noticed. Values of message_size_limit (both the global option
774 and the transport option) were treated as octal if they started with 0.
775 The documentation was vague. These values are now always treated as
776 decimal, and I will make that clear in the documentation.
782 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
783 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
786 MH/01 Don't check that the operands of numeric comparison operators are
787 integers when their expansion is in "skipping" mode (fixes bug
788 introduced by 4.64-PH/07).
790 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
791 child addresses, Exim now panics and dies. Previously, because the count
792 is held in a short int, deliveries were likely to be lost. As such a
793 large number of recipients for a single message is ridiculous
794 (performance will be very, very poor), I have chosen to impose a limit
795 rather than extend the field.
801 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
802 leftover -K file (the existence of which was triggered by #402).
803 While we were at it, introduced process PID as part of the -K
804 filename. This should rule out race conditions when creating
807 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
808 processing considerably. Previous code took too long for large mails,
809 triggering a timeout which in turn triggers #401.
811 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
812 in the DK code in transports.c. sendfile() is not really portable,
813 hence the _LINUX specificness.
815 TF/01 In the add_headers option to the mail command in an Exim filter,
816 there was a bug that Exim would claim a syntax error in any
817 header after the first one which had an odd number of characters
820 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
821 callout verification, Exim cached a "reject" for the entire domain. This
822 is correct for most verifications, but it is not correct for a recipient
823 verification with use_sender or use_postmaster set, because in that case
824 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
825 case of MAIL FROM:<> rejection from other early rejections (e.g.
826 rejection of HELO). When verifying a recipient using a non-null MAIL
827 address, the cache is ignored if it shows MAIL FROM:<> rejection.
828 Whatever the result of the callout, the value of the domain cache is
829 left unchanged (for any other kind of callout, getting as far as trying
830 RCPT means that the domain itself is ok).
832 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
835 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
836 manifest itself as EPIPE rather than ECONNECT. When tidying away a
837 session, the daemon ignores ECONNECT errors and logs others; it now
838 ignores EPIPE as well.
840 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
841 (quoted-printable decoding).
843 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
844 later the small subsequent patch to fix an introduced bug.
846 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
848 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
850 PH/08 An error is now given if message_size_limit is specified negative.
852 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
853 to be given (somewhat) arbitrary names.
855 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
858 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
859 miscellaneous code fixes
861 PH/10 Added the log_reject_target ACL modifier to specify where to log
864 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
865 hostname. This is wrong, because it relates to the incoming message (and
866 probably the interface on which it is arriving) and not to the outgoing
867 callout (which could be using a different interface). This has been
868 changed to use the value of the helo_data option from the smtp transport
869 instead - this is what is used when a message is actually being sent. If
870 there is no remote transport (possible with a router that sets up host
871 addresses), $smtp_active_hostname is used.
873 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
874 tweaks were necessary in order to get it to work (see also 21 below):
875 (a) The code assumed that strncpy() returns a negative number on buffer
876 overflow, which isn't the case. Replaced with Exim's string_format()
878 (b) There were several signed/unsigned issues. I just did the minimum
879 hacking in of casts. There is scope for a larger refactoring.
880 (c) The code used strcasecmp() which is not a standard C function.
881 Replaced with Exim's strcmpic() function.
882 (d) The code set only $1; it now sets $auth1 as well.
883 (e) A simple test gave the error "authentication client didn't specify
884 service in request". It would seem that Dovecot has changed its
885 interface. Fortunately there's a specification; I followed it and
886 changed what the client sends and it appears to be working now.
888 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
891 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
892 address A is aliased to B and C, where B exists and C does not. Without
893 -v the output is "A verified" because verification stops after a
894 successful redirection if more than one address is generated. However,
895 with -v the child addresses are also verified. Exim was outputting "A
896 failed to verify" and then showing the successful verification for C,
897 with its parentage. It now outputs "B failed to verify", showing B's
898 parentage before showing the successful verification of C.
900 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
901 look up a TXT record in a specific list after matching in a combined
904 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
905 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
906 they consult the DNS. I had assumed they would set it the way they
907 wanted; and indeed my experiments on Linux seem to show that in some
908 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
909 To be on the safe side, however, I have now made the interface to
910 host_find_byname() similar to host_find_bydns(), with an argument
911 containing the DNS resolver options. The host_find_byname() function now
912 sets these options at its start, just as host_find_bydns() does. The smtp
913 transport options dns_qualify_single and dns_search_parents are passed to
914 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
915 of host_find_byname() use the default settings of RES_DEFNAMES
916 (qualify_single) but not RES_DNSRCH (search_parents).
918 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
919 spool_read_header() do less string testing, by means of a preliminary
920 switch on the second character of optional "-foo" lines. (This is
921 overdue, caused by the large number of possibilities that now exist.
922 Originally there were few.) While I was there, I also converted the
923 str(n)cmp tests so they don't re-test the leading "-" and the first
924 character, in the hope this might squeeze out yet more improvement.
926 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
927 flag allowing group syntax was set by the header_syntax check but not
928 turned off, possible causing trouble later; (2) The flag was not being
929 set at all for the header_verify test, causing "group"-style headers to
930 be rejected. I have now set it in this case, and also caused header_
931 verify to ignore an empty address taken from a group. While doing this, I
932 came across some other cases where the code for allowing group syntax
933 while scanning a header line wasn't quite right (mostly, not resetting
934 the flag correctly in the right place). These bugs could have caused
935 trouble for malformed header lines. I hope it is now all correct.
937 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
938 with the "reply" argument non-NULL. The code, however (which originally
939 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
940 but it didn't always do it. This confused somebody who was copying the
941 code for some other use. I have removed all the tests.
943 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
944 feature that was used to support insecure browsers during the U.S. crypto
945 embargo. It requires special client support, and Exim is probably the
946 only MTA that supported it -- and would never use it because real RSA is
947 always available. This code has been removed, because it had the bad
948 effect of slowing Exim down by computing (never used) parameters for the
949 RSA_EXPORT functionality.
951 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
952 authenticator to fail if there's a tab character in the incoming data
953 (there should never be unless someone is messing about, as it's supposed
954 to be base64-encoded). Also added, on Timo's advice, the "secured" option
955 if the connection is using TLS or if the remote IP is the same as the
956 local IP, and the "valid-client-cert option" if a client certificate has
959 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
960 authenticators. This can be used for authorization after authentication
961 succeeds. (In the case of plaintext, it servers for both authentication
964 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
965 if any retry times were supplied.
967 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
968 connection, where there is no client IP address to check. In this
969 situation, the verify now always succeeds.
971 PH/25 Applied John Jetmore's -Mset patch.
973 PH/26 Added -bem to be like -Mset, but loading a message from a file.
975 PH/27 In a string expansion for a processed (not raw) header when multiple
976 headers of the same name were present, leading whitespace was being
977 removed from all of them, but trailing whitespace was being removed only
978 from the last one. Now trailing whitespace is removed from each header
979 before concatenation. Completely empty headers in a concatenation (as
982 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
983 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
985 PH/29 [Removed. This was a change that I later backed out, and forgot to
986 correct the ChangeLog entry (that I had efficiently created) before
987 committing the later change.]
989 PH/30 Exim was sometimes attempting to deliver messages that had suffered
990 address errors (4xx response to RCPT) over the same connection as other
991 messages routed to the same hosts. Such deliveries are always "forced",
992 so retry times are not inspected. This resulted in far too many retries
993 for the affected addresses. The effect occurred only when there were more
994 hosts than the hosts_max_try setting in the smtp transport when it had
995 the 4xx errors. Those hosts that it had tried were not added to the list
996 of hosts for which the message was waiting, so if all were tried, there
997 was no problem. Two fixes have been applied:
999 (i) If there are any address or message errors in an SMTP delivery, none
1000 of the hosts (tried or untried) are now added to the list of hosts
1001 for which the message is waiting, so the message should not be a
1002 candidate for sending over the same connection that was used for a
1003 successful delivery of some other message. This seems entirely
1004 reasonable: after all the message is NOT "waiting for some host".
1005 This is so "obvious" that I'm not sure why it wasn't done
1006 previously. Hope I haven't missed anything, but it can't do any
1007 harm, as the worst effect is to miss an optimization.
1009 (ii) If, despite (i), such a delivery is accidentally attempted, the
1010 routing retry time is respected, so at least it doesn't keep
1011 hammering the server.
1013 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1014 in ${readsocket because some servers need this prod.
1016 PH/32 Added some extra debug output when updating a wait-xxx database.
1018 PH/33 The hint "could be header name not terminated by colon", which has been
1019 given for certain expansion errors for a long time, was not being given
1020 for the ${if def:h_colon_omitted{... case.
1022 PH/34 The spec says: "With one important exception, whenever a domain list is
1023 being scanned, $domain contains the subject domain." There was at least
1024 one case where this was not true.
1026 PH/35 The error "getsockname() failed: connection reset by peer" was being
1027 written to the panic log as well as the main log, but it isn't really
1028 panic-worthy as it just means the connection died rather early on. I have
1029 removed the panic log writing for the ECONNRESET error when getsockname()
1032 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1033 runs only) independently of the message's sender address. This meant
1034 that, if the 4xx error was in fact related to the sender, a different
1035 message to the same recipient with a different sender could confuse
1036 things. In particualar, this can happen when sending to a greylisting
1037 server, but other circumstances could also provoke similar problems.
1038 I have changed the default so that the retry time for these errors is now
1039 based a combination of the sender and recipient addresses. This change
1040 can be overridden by setting address_retry_include_sender=false in the
1043 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1044 remote server are returned as part of bounce messages. This was not
1045 happening for LMTP over a pipe (the lmtp transport), but now it is the
1046 same for both kinds of LMTP.
1048 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1049 in header lines that were in fact CNAMEs. This is no longer the case.
1051 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1052 and queue runs started by the daemon processed all messages. This has
1053 been fixed so that -R and -S can now usefully be given with -q<time>.
1055 PH/40 Import PCRE release 6.7 (fixes some bugs).
1057 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1059 PH/42 Give an error if -q is specified more than once.
1061 PH/43 Renamed the variables $interface_address and $interface_port as
1062 $received_ip_address and $received_port, to make it clear that these
1063 values apply to message reception, and not to the outgoing interface when
1064 a message is delivered. (The old names remain recognized, of course.)
1066 PH/44 There was no timeout on the connect() call when using a Unix domain
1067 socket in the ${readsocket expansion. There now is.
1069 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1070 be meaningful with "accept".
1072 SC/01 Eximstats V1.43
1073 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1075 SC/02 Eximstats V1.44
1076 Use a glob alias rather than an array ref in the generated
1077 parser. This improves both readability and performance.
1079 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1080 Collect SpamAssassin and rejection statistics.
1081 Don't display local sender or destination tables unless
1082 there is data to show.
1083 Added average volumes into the top table text output.
1085 SC/04 Eximstats V1.46
1086 Collect data on the number of addresses (recipients)
1087 as well as the number of messages.
1089 SC/05 Eximstats V1.47
1090 Added 'Message too big' to the list of mail rejection
1091 reasons (thanks to Marco Gaiarin).
1093 SC/06 Eximstats V1.48
1094 Mainlog lines which have GMT offsets and are too short to
1095 have a flag are now skipped.
1097 SC/07 Eximstats V1.49 (Alain Williams)
1098 Added the -emptyok flag.
1100 SC/08 Eximstats V1.50
1101 Fixes for obtaining the IP address from reject messages.
1103 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1104 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1105 whitesspace changes from 4.64-PH/27
1107 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1110 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1111 are found, allow negative numbers in numeric criteria)
1113 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1115 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1116 to match changes made in 4.64-PH/43
1118 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1120 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1121 "domain" is actually an IP address, and force a failure. This locks out
1122 those revolvers/nameservers that support "A-for-A" lookups, in
1123 contravention of the specifications.
1125 PH/48 When a host name was looked up from an IP address, and the subsequent
1126 forward lookup of the name timed out, the host name was left in
1127 $sender_host_name, contrary to the specification.
1129 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1130 restricted to single-key lookups, Exim was not diagnosing an error if
1131 * or *@ was used with a query-style lookup.
1133 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1135 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1136 long ago, but noone interested enough thought of it. Let's just say that
1137 the "1.1" means that there are some new functions that weren't there at
1138 some point in the past.
1140 PH/51 Error processing for expansion failure of helo_data from an smtp
1141 transport during callout processing was broken.
1143 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1144 tested/used via the -bh/-bhc/-bs options.
1146 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1147 bug, fixed in subsequent PCRE releases).
1149 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1150 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1152 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1158 SC/01 Use a glob alias rather than an array ref in eximstats generated
1159 parser. This improves both readability and performance.
1161 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1162 Don't display local sender or destination tables in eximstats unless
1163 there is data to show.
1164 Added average volumes into the eximstats top table text output.
1166 SC/03 Collect data on the number of addresses (recipients) as well
1167 as the number of messages in eximstats.
1169 TF/01 Correct an error in the documentation for the redirect router. Exim
1170 does (usually) call initgroups() when daemonizing.
1172 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1173 with consistent privilege compared to when running as a daemon.
1175 TF/03 Note in the spec that $authenticated_id is not set for local
1176 submissions from trusted users.
1178 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1179 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1181 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1182 by adding some example configuration directives to the default
1183 configuration file. A little bit of work is required to uncomment the
1184 directives and define how usernames and passwords are checked, but
1185 there is now a framework to start from.
1187 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1188 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1189 without this. I don't know how relevant this is to other LDAP libraries.
1191 PH/02 Add the verb name to the "unknown ACL verb" error.
1193 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1195 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1197 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1198 directory not expanded when it should be if an expanded home directory
1199 was set for the address (which is overridden by the transport).
1201 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1204 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1205 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1206 because it is too late at that time, and has no effect.
1208 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1209 security issue with \' (bugzilla #107). I could not use the
1210 PQescapeStringConn() function, because it needs a PGconn value as one of
1213 PH/08 When testing addresses using -bt, indicate those final addresses that
1214 are duplicates that would not cause an additional delivery. At least one
1215 person was confused, thinking that -bt output corresponded to deliveries.
1216 (Suppressing duplicates isn't a good idea as you lose the information
1217 about possibly different redirections that led to the duplicates.)
1219 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1220 systems where poll() doesn't work, in particular OS X.
1222 PH/10 Added more information to debugging output for retry time not reached.
1224 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1225 operations in malware.c.
1227 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1230 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1231 syslog_duplication set false, log lines that would normally be written
1232 both the the main log and to the reject log were not written to syslog at
1235 PH/14 In the default configuration, change the use of "message" in ACL warn
1236 statements to "add_header".
1238 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1239 not followed by a command (e.g. "seen endif").
1241 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1242 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1245 PH/17 Added extra conditions to the default value of delay_warning_condition
1249 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1250 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1251 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1254 The Auto-Submitted: and various List- headers are standardised, whereas I
1255 don't think Precedence: ever was.
1257 PH/18 Refactored debugging code in route_finduser() to show more information,
1258 in particular, the error code if getpwnam() issues one.
1260 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1261 This is apparently needed in addition to the PH/07 change above to avoid
1262 any possible encoding problems.
1264 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1265 but not after initializing Perl.
1267 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1268 output them only if debugging. By default they are written stderr,
1269 apparently, which is not desirable.
1271 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1274 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1277 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1279 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1280 authenticated or an ident call has been made. Suppress the default
1281 values for $authenticated_id and $authenticated_sender (but permit -oMai
1282 and -oMas) when testing with -bh.
1284 PH/24 Re-jigged the order of the tests in the default configuration so that the
1285 tests for valid domains and recipients precede the DNS black list and CSA
1286 tests, on the grounds that those ones are more expensive.
1288 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1289 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1290 This bug exists in every version of Exim that I still have, right back to
1293 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1294 However, an attempt to turn on case-sensitivity in a regex key by
1295 including (?-i) didn't work because the subject string was already
1296 lowercased, and the effects were non-intuitive. It turns out that a
1297 one-line patch can be used to allow (?-i) to work as expected.
1303 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1304 other effects) broke the use of negated acl sub-conditions.
1306 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1309 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1310 "Deny" causes Exim to reject the incoming connection with a 554 error.
1311 Unfortunately, if there is a major crisis, such as a disk failure,
1312 tcp-wrappers gives "deny", whereas what one would like would be some
1313 kind of temporary error. A kludge has been added to help with this.
1314 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1315 554 error is used if errno is still zero or contains ENOENT (which occurs
1316 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1319 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1321 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1322 errors. Otherwise a message that provokes a temporary error (when other
1323 messages do not) can cause a whole host to time out.
1325 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1326 addresses were routed directly to files or pipes from a redirect router.
1327 File deliveries just didn't batch; pipe deliveries might have suffered
1330 PH/06 A failure to get a lock for a hints database would erroneously always say
1331 "Failed to get write lock", even when it was really a read lock.
1333 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1334 of 0600. This has been changed to use the value of the lockfile_mode
1335 option (which defaults to 0600).
1337 PH/08 Applied small patch from the Sieve maintainer.
1339 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1340 folder from quota calculations, a direct delivery into this folder messed
1341 up the contents of the maildirsize file. This was because the regex was
1342 used only to exclude .Trash (or whatever) when the size of the mailbox
1343 was calculated. There was no check that a delivery was happening into an
1344 excluded directory. This bug has been fixed by ignoring all quota
1345 processing for deliveries into excluded directories.
1347 PH/10 Added the maildirfolder_create_regex option to appendfile.
1353 PH/01 The code for finding all the local interface addresses on a FreeBSD
1354 system running IPv6 was broken. This may well have applied to all BSD
1355 systems, as well as to others that have similar system calls. The broken
1356 code found IPv4 interfaces correctly, but gave incorrect values for the
1357 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1358 that it would not match correctly against @[] and not recognize the IPv6
1361 PH/02 The ipliteral router was not recognizing addresses of the form user@
1362 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1364 PH/03 Added disable_ipv6.
1366 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1367 decoded form, because it is most often used to construct To: headers
1368 lines in autoreplies, and the decoded form may well be syntactically
1369 invalid. However, $reply_address has leading white space removed, and all
1370 newlines turned into spaces so that the autoreply transport does not
1373 PH/05 If group was specified without a user on a router, and no group or user
1374 was specified on a transport, the group from the router was ignored.
1376 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1377 for visible compile-time settings that can be used to change these
1378 numbers, for those that want even more. Backwards compatibility with old
1379 spool files has been maintained. However, going back to a previous Exim
1380 release will lost any variables that are in spool files.
1382 PH/07 Two small changes when running in the test harness: increase delay when
1383 passing a TCP/IP connection to a new process, in case the original
1384 process has to generate a bounce, and remove special handling of
1385 127.0.0.2 (sic), which is no longer necessary.
1387 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1388 be the same on different OS.
1390 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1393 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1394 whether --show-vars was specified or not
1396 JJ/02 exipick: Added support for new ACL variable spool format introduced
1399 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1400 syntactically invalid From: or Reply-to: line, and a filter used this to
1401 generate an autoreply, and therefore failed to obtain an address for the
1402 autoreply, Exim could try to deliver to a non-existent relative file
1403 name, causing unrelated and misleading errors. What now happens is that
1404 it logs this as a hard delivery error, but does not attempt to create a
1407 PH/11 The exinext utility has a -C option for testing purposes, but although
1408 the given file was scanned by exinext itself; it wasn't being passed on
1409 when Exim was called.
1411 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1412 an end-of-file indication when reading a command response.
1414 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1415 compiled. In many other places in Exim, IPv6 addresses are always
1416 recognized, so I have changed this. It also means that IPv4 domain
1417 literals of the form [IPV4:n.n.n.n] are now always recognized.
1419 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1420 used if the router is not running as root, for example, when verifying at
1421 ACL time, or when using -bh. The debugging output from this situation was
1422 non-existent - all you got was a failure to exec. I have made two
1425 (a) Failures to set uid/gid, the current directory, or a process leader
1426 in a subprocess such as that created by queryprogram now generate
1427 suitable debugging ouput when -d is set.
1429 (b) The queryprogram router detects when it is not running as root,
1430 outputs suitable debugging information if -d is set, and then runs
1431 the subprocess without attempting to change uid/gid.
1433 PH/15 Minor change to Makefile for building test_host (undocumented testing
1436 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1437 additional section of a DNS packet that returns MX or SRV records.
1438 Instead, it always explicitly searches for A/AAAA records. This avoids
1439 major problems that occur when a DNS server includes only records of one
1440 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1441 fixed another bug: if SRV records were looked up and the corresponding
1442 address records were *not* found in the additional section, the port
1443 values from the SRV records were lost.
1445 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1446 using the correct key (the original address) when searching the retry
1447 rules in order to find which one to use for generating the retry hint.
1449 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1450 adding the default one. Similarly, if it contains a Reply-To: header, the
1451 errors_reply_to option, if set, is not used.
1453 PH/19 When calculating a retry time, Exim used to measure the "time since
1454 failure" by looking at the "first failed" field in the retry record. Now
1455 it does not use this if it is later than than the arrival time of the
1456 message. Instead it uses the arrival time. This makes for better
1457 behaviour in cases where some deliveries succeed, thus re-setting the
1458 "first failed" field. An example is a quota failure for a huge message
1459 when small messages continue to be delivered. Without this change, the
1460 "time since failure" will always be short, possible causing more frequent
1461 delivery attempts for the huge message than are intended.
1462 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1464 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1465 $1, $2, $3) because the numerical variables can be reset during some
1466 expansion items (e.g. "match"), thereby losing the authentication data.
1468 PH/21 Make -bV show the size of off_t variables so that the test suite can
1469 decide whether to run tests for quotas > 2G.
1471 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1472 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1473 filecount value is greater than 2G or if a quota value is greater than 2G
1474 on a system where the size of off_t is not greater than 4, a panic error
1477 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1478 never match. The debug and -bh output now contains an explicit error
1479 message indicating a malformed IPv4 address or mask.
1481 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1482 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1483 PH/23 above applies.
1485 PH/25 Do not write to syslog when running in the test harness. The only
1486 occasion when this arises is a failure to open the main or panic logs
1487 (for which there is an explicit test).
1489 PH/26 Added the /no_tell option to "control=freeze".
1491 PH/27 If a host name lookup failed very early in a connection, for example, if
1492 the IP address matched host_lookup and the reverse lookup yielded a name
1493 that did not have a forward lookup, an error message of the form "no IP
1494 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1495 could be logged. Now it outputs the IP address instead of "NULL".
1497 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1498 allows the sender and the authenticated sender to be set when
1499 submitting a message from within Exim. Since child_open_exim() is
1500 documented for local_scan(), the new function should be too.
1502 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1503 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1504 results in an empty string is now treated as unset.
1506 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1508 PH/31 Added sender_verify_fail logging option.
1510 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1511 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1512 tidied the source and removed it altogether.
1514 PH/33 When a queue run was abandoned because the load average was too high, a
1515 log line was always written; now it is written only if the queue_run log
1516 selector is set. In addition, the log line for abandonment now contains
1517 information about the queue run such as the pid. This is always present
1518 in "start" and "stop" lines but was omitted from the "abandon" line.
1520 PH/34 Omit spaces between a header name and the colon in the error message that
1521 is given when verify = headers_syntax fails (if there are lots of them,
1522 the message gets confusing).
1524 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1525 names, as there are now some PTR records that contain slashes. This check
1526 is only to protect against broken name servers that fall over on strange
1527 characters, so the fact that it applies to all lookups doesn't matter.
1529 PH/36 Now that the new test suite is complete, we can remove some of the
1530 special code in Exim that was needed for the old test suite. For example,
1531 sorting DNS records because real resolvers return them in an arbitrary
1532 order. The new test suite's fake resolver always returns records in the
1535 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1536 bounces) except when queue_only is set, to avoid logging races between
1537 the different processes.
1539 PH/38 Panic-die if .include specifies a non-absolute path.
1541 PH/39 A tweak to the "H" retry rule from its user.
1543 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1544 a label. They prevented compilation on older perls.
1546 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1547 a warning to be raised on newish perls.
1549 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1550 on queue. Changes to match documented behaviour of showing count of
1551 messages matching specified criteria.
1553 PH/40 Changed the default ident timeout from 30s to 5s.
1555 PH/41 Added support for the use of login_cap features, on those BSD systems
1556 that have them, for controlling the resources used by pipe deliveries.
1558 PH/42 The content-scanning code uses fopen() to create files in which to put
1559 message data. Previously it was not paying any attention to the mode of
1560 the files. Exim runs with umask(0) because the rest of the code creates
1561 files with open(), and sets the required mode explicitly. Thus, these
1562 files were ending up world-writeable. This was not a big issue, because,
1563 being within the spool directory, they were not world-accessible. I have
1564 created a function called modefopen, which takes an additional mode
1565 argument. It sets umask(777), creates the file, chmods it to the required
1566 mode, then resets the umask. All the relevant calls to fopen() in the
1567 content scanning code have been changed to use this function.
1569 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1570 to 24 hours. This avoids potential overflow problems when processing G
1571 and H retry rules. I suspect nobody ever tinkers with this value.
1573 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1575 PH/45 When the plaintext authenticator is running as a client, the server's
1576 challenges are checked to ensure they are valid base64 strings. By
1577 default, the authentication attempt is cancelled if an invalid string is
1578 received. Setting client_ignore_invalid_base64 true ignores these errors.
1579 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1580 they are received. Thus, the responses can be made to depend on the
1581 challenges. If an invalid string is ignored, an empty string is placed in
1584 PH/46 Messages that are created by the autoreply transport now contains a
1585 References: header, in accordance with RFCs 2822 and 3834.
1587 PH/47 Added authenticated_sender_force to the smtp transport.
1589 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1591 PH/49 Installed latest patch from the Sieve maintainer.
1593 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1594 was also set, the appendfile transport was unnecessarily scanning a
1595 directory of message files (e.g. for maildir delivery) to find the count
1596 of files (along with the size), even though it did not need this
1597 information. It now does the scan only if it needs to find either the
1598 size of the count of files.
1600 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1602 PH/52 Two bugs concerned with error handling when the smtp transport is
1605 (i) Exim was not creating retry information for temporary errors given
1606 for individual recipients after the DATA command when the smtp transport
1607 was used in LMTP mode. This meant that they could be retried too
1608 frequently, and not timed out correctly.
1610 (ii) Exim was setting the flag that allows error details to be returned
1611 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1612 recipients that were returned after the DATA command.
1614 PH/53 This is related to PH/52, but is more general: for any failing address,
1615 when detailed error information was permitted to be returned to the
1616 sender, but the error was temporary, then after the final timeout, only
1617 "retry timeout exceeded" was returned. Now it returns the full error as
1618 well as "retry timeout exceeded".
1620 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1621 do this, and (what is worse) MTAs that accept it.
1623 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1624 will now be deprecated.
1626 PH/56 New os.c-cygwin from the Cygwin maintainer.
1628 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1629 formats (previously only available in exim formats via -bpr, -bpru,
1630 and -bpra. Now also available in native and exiqgrep formats)
1632 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1633 with very large, slow to parse queues
1635 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1637 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1639 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1640 responses to authentication challenges, though it was showing the
1641 challenges; (ii) I've removed the CR characters from the debug output for
1644 PH/58 Allow for the insertion of a newline as well as a space when a string
1645 is turned into more than one encoded-word during RFC 2047 encoding. The
1646 Sieve code now uses this.
1648 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1649 data_4xx, lost_connection, tls_required.
1651 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1652 message was being sent as an SMTP response.
1654 PH/61 Add -l and -k options to exicyclog.
1656 PH/62 When verifying, if an address was redirected to one new address, so that
1657 verification continued, and the new address failed or deferred after
1658 having set something in $address_data, the value of $address_data was not
1659 passed back to the ACL. This was different to the case when no
1660 redirection occurred. The value is now passed back in both cases.
1662 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1663 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1664 use login_cap.h, so on its own it isn't the distinguishing feature. The
1665 new name refers directly to the setclassresources() function.
1667 PH/65 Added configuration files for NetBSD3.
1669 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1671 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1672 is preferred over IPv4.
1674 PH/68 The bounce_return_message and bounce_return_body options were not being
1675 honoured for bounces generated during the reception of non-SMTP messages.
1676 In particular, this applied to messages rejected by the ACL. This bug has
1677 been fixed. However, if bounce_return_message is true and bounce_return_
1678 body is false, the headers that are returned for a non-SMTP message
1679 include only those that have been read before the error was detected.
1680 (In the case of an ACL rejection, they have all been read.)
1682 PH/69 The HTML version of the specification is now built in a directory called
1683 spec_html instead of spec.html, because the latter looks like a path with
1684 a MIME-type, and this confuses some software.
1686 PH/70 Catch two compiler warnings in sieve.c.
1688 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1689 function verify_get_ident() calls ip_connect() to connect a socket, but
1690 if the "connect()" function timed out, ip_connect() used to close the
1691 socket. However, verify_get_ident() also closes the socket later, and in
1692 between Exim writes to the log, which may get opened at this point. When
1693 the socket was closed in ip_connect(), the log could get the same file
1694 descriptor number as the socket. This naturally causes chaos. The fix is
1695 not to close the socket in ip_connect(); the socket should be closed by
1696 the function that creates it. There was only one place in the code where
1697 this was missing, in the iplookup router, which I don't think anybody now
1698 uses, but I've fixed it anyway.
1700 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1701 well as to direct DNS lookups. Otherwise the handling of names in host
1702 lists is inconsistent and therefore confusing.
1708 PH/01 Two changes to the default runtime configuration:
1710 (1) Move the checks for relay_from_hosts and authenticated clients from
1711 after to before the (commented out) DNS black list checks.
1713 (2) Add control=submission to the relay_from_hosts and authenticated
1714 clients checks, on the grounds that messages accepted by these
1715 statements are most likely to be submissions.
1717 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1719 (1) Generate an error if the third argument for the ${prvs expansion is
1722 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1725 (3) Reset the variables that are obtained from the first argument of
1726 ${prvscheck and used in the second argument before leaving the code,
1727 because their memory is reclaimed, so using them afterwards may do
1730 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1731 one (it's much easier than Tom thought :-).
1733 (5) Because of (4), we can now allow for the use of $prvscheck_result
1734 inside the third argument.
1736 PH/03 For some reason, the default setting of PATH when running a command from
1737 a pipe transport was just "/usr/bin". I have changed it to
1740 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1741 anything to be listed in the output from -bV.
1743 PH/05 When a filter generated an autoreply, the entire To: header line was
1744 quoted in the delivery log line, like this:
1746 => >A.N.Other <ano@some.domain> <original@ddress> ...
1748 This has been changed so that it extracts the operative address. There
1749 may be more than one such address. If so, they are comma-separated, like
1752 => >ano@some.domain,ona@other.domain <original@ddress> ...
1754 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1755 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1756 not being looked up in the rDNS to get a host name, Exim was showing the
1757 IP address twice in Received: lines, even though the IP addresses were
1758 identical. For example:
1760 Received: from [1.2.3.4] (helo=[1.2.3.4])
1762 However, if the real host name was known, it was omitting the HELO data
1763 if it matched the actual IP address. This has been tidied up so that it
1764 doesn't show the same IP address twice.
1766 PH/07 When both +timestamp and +memory debugging was on, the value given by
1767 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1768 called by the debug printing, thereby overwriting the timestamp buffer.
1769 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1771 PH/08 When the original message was included in an autoreply transport, it
1772 always said "this is a copy of the message, including all the headers",
1773 even if body_only or headers_only was set. It now gives an appropriate
1776 PH/09 Applied a patch from the Sieve maintainer which:
1778 o fixes some comments
1779 o adds the (disabled) notify extension core
1780 o adds some debug output for the result of if/elsif tests
1781 o points to the current vacation draft in the documentation
1782 and documents the missing references header update
1786 o fixes a bug in processing the envelope test (when testing
1787 multiple envelope elements, the last element determinted the
1790 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1791 Electronic Mail") by including:
1793 Auto-submitted: auto-generated
1795 in the messages that it generates (bounce messages and others, such as
1796 warnings). In the case of bounce messages for non-SMTP mesages, there was
1797 also a typo: it was using "Auto_submitted" (underscore instead of
1798 hyphen). Since every message generated by Exim is necessarily in response
1799 to another message, thes have all been changed to:
1801 Auto-Submitted: auto-replied
1803 in accordance with these statements in the RFC:
1805 The auto-replied keyword:
1807 - SHOULD be used on messages sent in direct response to another
1808 message by an automatic process,
1810 - MUST NOT be used on manually-generated messages,
1812 - MAY be used on Delivery Status Notifications (DSNs) and Message
1813 Disposition Notifications (MDNs),
1815 - MUST NOT be used on messages generated by automatic or periodic
1816 processes, except for messages which are automatic responses to
1819 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1820 to the default Received: header definition.
1822 PH/12 Added log selector acl_warn_skipped (default on).
1824 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1825 variables because (a) they are in the wrong storage pool and (b) even if
1826 they were copied, it wouldn't work properly because of the caching.
1828 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1829 checking when decoding. Apparently there are clients that generate
1830 overlong encoded strings. Why am I not surprised?
1832 PH/15 If the first argument of "${if match_address" was not empty, but did not
1833 contain an "@" character, Exim crashed. Now it writes a panic log message
1834 and treats the condition as false.
1836 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1838 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1839 extension "envelope-auth". The code is finished and in agreement with
1840 other implementations, but there is no documentation so far and in fact,
1841 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1842 not changing the active code.
1844 Print executed "if" and "elsif" statements when debugging is used. This
1845 helps a great deal to understand what a filter does.
1847 Document more things not specified clearly in RFC3028. I had all this
1848 sorted out, when out of a sudden new issues came to my mind. Oops."
1850 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1853 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1854 canonical form (as documented). However, after a host name lookup from
1855 the IP address, check_host() was doing a simple string comparison with
1856 addresses acquired from the DNS when checking that the found name did
1857 have the original IP as one of its addresses. Since any found IPv6
1858 addresses are likely to be in abbreviated form, the comparison could
1859 fail. Luckily, there already exists a function for doing the comparison
1860 by converting both addresses to binary, so now that is used instead of
1861 the text comparison.
1863 PH/20 There was another similar case to PH/19, when a complete host name was
1864 given in a host list; looking up its IP address could give an abbreviated
1865 form, whereas the current host's name might or might not be abbreviated.
1866 The same fix has been applied.
1872 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1873 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1876 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1877 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1879 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1881 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1882 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1883 Also, for "Auto-Submitted", treat anything other than "no" as
1884 non-personal, in accordance with RFC 3834. (Previously it treated
1885 anything starting "auto-" as non-personal.)
1887 TF/01 The control=submission/name=... option had a problem with syntax
1888 errors if the name included a slash character. The /name= option
1889 now slurps the rest of the string, so it can include any characters
1890 but it must come last in the list of options (after /sender_retain
1893 PH/05 Some modifications to the interface to the fake nameserver for the new
1901 TK/01 Added the "success_on_redirect" address verification option. See
1902 NewStuff for rationale and an example.
1904 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1906 PH/02 Patch to exigrep to allow it to work on syslog lines.
1908 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1909 fread() to skip over the body file's header line, because in Cygwin the
1910 header line is locked and is inaccessible.
1912 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1913 co-exist for some time) to make it clear that it is the Exim ID that is
1914 referenced, not the Message-ID: header line.
1916 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1917 string_format() function, because snprintf() does not exist on all
1920 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1921 ${stat: expansion item.
1923 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1924 protocol synchronization error", to keep the pedants happy.
1926 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1927 well as for IRIX systems, when gcc is being used. See the host.c source
1930 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1932 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1935 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1936 transport and to the smtp transport in LMTP mode.
1938 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1940 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1941 run a filter in a subprocess. This could lead to confusion in subsequent
1942 lookups in the parent process. There should also be a search_tidyup() at
1943 the end of the subprocess.
1945 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1946 only if the host matched helo_try_verify_hosts, which caused the
1947 verification to occur when the EHLO/HELO command was issued. The ACL just
1948 tested the remembered result. Now, if a previous verification attempt has
1949 not happened, "verify = helo" does it there and then.
1951 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1953 TK/03 Fix log output including CR from clamd.
1955 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1956 exist provoked a memory error which could cause a segfault.
1958 PH/15 Installed PCRE 6.2
1960 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1962 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1963 of the problem. Specifically, suggested +O2 rather than +O1 for the
1966 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1968 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1969 record, the port was not correctly logged when the outgoing_port log
1970 selector was set (it logged the transort's default port).
1972 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1973 fallback_hosts, and "hosts" in the smtp transport.
1975 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1976 host errors such as "Connection refused".
1978 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1979 authentication with radiusclient 0.4.9:
1981 - Error returned from rc_read_config was caught wrongly
1982 - Username/password not passed on to radius server due to wrong length.
1984 The presumption is that some radiusclient API changes for 4.51/PH/17
1985 were not taken care of correctly. The code is still untested by me (my
1986 Linux distribution still has 0.3.2 of radiusclient), but it was
1987 contributed by a Radius user.
1989 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1990 expanding the "port" option of the smtp transport.
1992 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1993 while reading a MIME header. Thanks to Tom Hughes for a patch.
1995 PH/24 Include config.h inside local_scan.h so that configuration settings are
1998 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1999 that in an ACL for RCPT (for example), you can examine exactly what was
2002 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2003 commands, but it was not correctly comparing the address with the actual
2004 client host address. Thus, it would show the EHLO address in Received:
2005 header lines when this was not necessary.
2007 PH/27 Added the % operator to ${eval:}.
2009 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2010 it should be ignoring failures (because with -C, for example, it has lost
2011 privilege). It wasn't ignoring creation failures other than "already
2014 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2017 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2018 because an input line was too long, either on its own, or by virtue of
2019 too many continuations, the temporary file was not being removed, and the
2020 return code was incorrect.
2022 PH/31 Missing "BOOL" in function definition in filtertest.c.
2024 PH/32 Applied Sieve patches from the maintainer.
2026 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2028 PH/33 Added "verify = not_blind".
2030 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2031 Local/Makefile (with some defaults set). These are used in built scripts
2032 such as exicyclog, but they have never been used in the exim_install
2033 script (though there are many overriding facilities there). I have
2034 arranged that the exim_install script now takes note of these two
2037 PH/35 Installed configuration files for Dragonfly.
2039 PH/36 When a locally submitted message by a trusted user did not contain a
2040 From: header, and the sender address was obtained from -f or from an SMTP
2041 MAIL command, and the trusted user did not use -F to supply a sender
2042 name, $originator_name was incorrectly used when constructing a From:
2043 header. Furthermore, $originator_name was used for submission mode
2044 messages from external hosts without From: headers in a similar way,
2045 which is clearly wrong.
2047 PH/37 Added control=suppress_local_fixups.
2049 PH/38 When log_selector = +received_sender was set, and the addition of the
2050 sender made the log line's construction buffer exactly full, or one byte
2051 less than full, an overflow happened when the terminating "\n" was
2054 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2055 when the result of a list match is failure because a DNS lookup failed.
2057 PH/40 RM_COMMAND is now used in the building process.
2059 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2060 the "build-* directories that it finds.
2062 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2063 address in a domain literal was a prefix of an interface address.
2065 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2066 when verifying a sender address, unless rewrite_headers is false.
2068 PH/44 Wrote a long comment about why errors_to addresses are verified as
2069 recipients, not senders.
2071 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2072 the ratelimit ACL was added.
2074 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2076 PH/46 Added extra information about PostgreSQL errors to the error string.
2078 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2079 avoiding the need to install special zones in a real server. This is
2080 backwards compatible; if it can't find the fake resolver, it drops back.
2081 Thus, both old and new test suites can be run.
2083 TF/02 Added util/ratelimit.pl
2085 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2086 clock is set back in time.
2088 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2089 Candler <B.Candler@pobox.com>.
2091 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2092 OK for addresses that are the result of redirections.
2094 PH/48 A number of further additions for the benefit of the new test suite,
2095 including a fake gethostbyname() that interfaces to the fake DNS resolver
2098 TF/06 The fix for widen_domains has also been applied to qualify_single and
2099 search_parents which are the other dnslookup options that can cause
2102 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2105 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2107 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2108 service" if it encountered an MX record with an empty target hostname.
2109 The message is now "an MX or SRV record indicated no SMTP service".
2111 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2112 if the DNS of the sending site is misconfigured. This is quite a
2113 common situation. This change restores the behaviour of treating a
2114 helo verification defer as a failure.
2116 PH/51 If self=fail was set on a router, the bounce message did not include the
2117 actual error message.
2123 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2125 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2126 command itself ended in error, the underlying message about the transport
2127 filter timeout was being overwritten with the pipe command error. Now the
2128 underlying error message should be appended to the second error message.
2130 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2132 PH/02 Reduce the amount of output that "make" produces by default. Full output
2133 can still be requested.
2135 PH/03 The warning log line about a condition test deferring for a "warn" verb
2136 was being output only once per connection, rather than after each
2137 occurrence (because it was using the same function as for successful
2138 "warn" verbs). This seems wrong, so I have changed it.
2140 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2141 it should not have, which might have caused a crash in the right
2142 circumstances, but probably never did.
2144 PH/04 Installed a modified version of Tony Finch's patch to make submission
2145 mode fix the return path as well as the Sender: header line, and to
2146 add a /name= option so that you can make the user's friendly name appear
2149 TF/03 Added the control = fakedefer ACL modifier.
2151 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2152 Mark Lowes for thorough testing.
2154 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2156 TK/03 Merged latest SRS patch from Miles Wilton.
2158 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2159 with the definition in sysexits.h (which is #included earlier).
2160 Fortunately, Exim does not actually use EX_OK. The code used to try to
2161 preserve the sysexits.h value, by assumimg that macro definitions were
2162 scanned for macro replacements. I have been disabused of this notion,
2163 so now the code just undefines EX_OK before #including unistd.h.
2165 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2166 in the smtp transport. When a block could not be written in a single
2167 write() function, the timeout was being re-applied to each part-write.
2168 This seems wrong - if the receiver was accepting one byte at a time it
2169 would take for ever. The timeout is now adjusted when this happens. It
2170 doesn't have to be particularly precise.
2172 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2173 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2175 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2176 without a domain if the check to <postmaster@domain> fails.
2178 SC/01 Eximstats: added -xls and the ability to specify output files
2179 (patch written by Frank Heydlauf).
2181 SC/02 Eximstats: use FileHandles for outputing results.
2183 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2185 SC/04 Eximstats: fixed display of large numbers with -nvr option
2187 SC/05 Eximstats: fixed merging of reports with empty tables.
2189 SC/06 Eximstats: added the -include_original_destination flag
2191 SC/07 Eximstats: removed tabs and trailing whitespace.
2193 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2195 TK/06 MBOX spool code: Add real "From " MBOX separator line
2196 so the .eml file is really in mbox format (even though
2197 most programs do not really care). Patch from Alex Miller.
2199 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2200 The latter is generated from $received_to and is only set if the
2201 message has one envelope recipient. SA can use these headers,
2202 obviously out-of-the-box. Patch from Alex Miller.
2204 PH/08 The ${def test on a variable was returning false if the variable's
2205 value was "0", contrary to what the specification has always said!
2206 The result should be true unless the variable is empty.
2208 PH/09 The syntax error of a character other than { following "${if
2209 def:variable_name" (after optional whitespace) was not being diagnosed.
2210 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2211 accidental colon was present, for example, could give incorrect results.
2213 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2214 result is used (not including appendfile, where other changes are about
2217 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2218 This involved changing a lot of size variables from int to off_t. It
2219 should work with maildirs and everything.
2221 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2222 spamd dying while we are connected to it.
2224 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2227 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2228 function for BDB 4.3.
2230 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2232 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2233 the sources for PCRE have been re-arranged and more files are now
2236 PH/15 The code I had for printing potentially long long variables in PH/11
2237 above was not the best (it lost precision). The length of off_t variables
2238 is now inspected at build time, and an appropriate printing format (%ld
2239 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2240 to be "long long int" or "long int". This is needed for the internal
2241 formatting function string_vformat().
2243 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2244 the configuration file to be ":syslog", then the script "guesses" where
2245 the logs files are, rather than using the compiled in default. In our
2246 case the guess is not the same as the compiled default, so the script
2247 suddenly stopped working when I started to use syslog. The patch checks
2248 to see if log_file_path is "". If so, it attempts to read it from exim
2249 with no configuration file to get the compiled in version, before it
2250 falls back to the previous guessing code."
2252 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2253 implementing BATV in an Exim configuration. See NewStuff for the gory
2256 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2257 Makefile that are specific to HP-UX.
2259 PH/18 If the "use_postmaster" option was set for a recipient callout together
2260 with the "random" option, the postmaster address was used as the MAIL
2261 FROM address for the random test, but not for the subsequent recipient
2262 test. It is now used for both.
2264 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2265 patch removes a few documentation additions to RFC 3028, because the
2266 latest draft now contains them. It adds the new en;ascii-case comparator
2267 and a new error check for 8bit text in MIME parts. Comparator and
2268 require names are now matched exactly. I enabled the subaddress
2269 extension, but it is not well tested yet (read: it works for me)."
2271 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2272 rework some of the code of TK/09 above to avoid the hardwired use of
2273 "%lld" and "long long". Replaced the call to snprintf() with a call to
2276 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2277 records point to non-existent hosts", "retry timeout exceeded", and
2278 "retry time not reached for any host after a long failure period".
2280 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2281 experimental DomainKeys support:
2283 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2284 (2) On an error such as an illegally used "control", the wrong name for
2285 the control was given.
2287 These problems did NOT occur unless DomainKeys support was compiled.
2289 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2291 PH/24 Added ${if match_ip condition.
2293 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2294 databases so that it will be absolutely obvious if a crash occurs in the
2295 DB library. This is a regular occurrence (often caused by mis-matched
2298 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2299 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2300 picked up on a user's system that detects such things. There doesn't seem
2301 to be a gcc warning option for this - only an attribute that has to be
2302 put on the function's prototype. It seems that in Fedora Core 4 they have
2303 set this on a number of new functions. No doubt there will be more in due
2306 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2307 specify a transport. However, if an address that was verified by such a
2308 router was the subject of a callout, Exim crashed because it tried to
2309 read the rcpt_include_affixes from the non-existent transport. Now it
2310 just assumes that the setting of that option is false. This bug was
2311 introduced by 4.51/PH/31.
2313 PH/28 Changed -d+all to exclude +memory, because that information is very
2314 rarely of interest, but it makes the output a lot bigger. People tend to
2315 do -d+all out of habit.
2317 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2318 code in os-type was giving problems when libc.so lives in lib64, like on
2321 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2322 aren't the modern standard, and it seems that some systems' include files
2323 don't always have them. Exim was already checking for some of the newer
2324 ones like T_AAAA, and defining it itself. I've added checks for all the
2325 record types that Exim uses.
2327 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2328 not automatically generating a new one, as it is supposed to. This
2329 prevented TLS from working. If the file did exist, but contained invalid
2330 data, a new version was generated, as expected. It was only the case of a
2331 non-existent file that was broken.
2333 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2334 with a change in libdomainkeys > 0.64.
2336 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2337 from DNS. If the selector record carries the flag, it now has
2338 precedence over the domain-wide flag.
2340 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2342 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2343 the use of a port name that isn't defined in /etc/services) occurred, the
2344 message was deferred as in a normal delivery, and thus remained on the
2345 spool, instead of being failed because of the mua_wrapper setting. This
2346 is now fixed, and I tidied up some of the mua_wrapper messages at the
2349 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2350 the messages in a hash of arrays rather than using individual hashes.
2351 This is a bit cleaner and results in dramatic memory savings, albeit
2352 at a slight CPU cost.
2354 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2355 as requested by Marc Sherman.
2357 SC/10 Eximstats: added histograms for user specified patterns as requested
2360 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2362 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2363 fopen() in the content-scanning modules that did not already have it.
2369 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2370 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2372 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2374 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2376 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2377 to test Sieve filters that use "vacation".
2379 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2380 that changes the way the GnuTLS parameters are stored in the cache file.
2381 The new format can be generated externally. For backward compatibility,
2382 if the data in the cache doesn't make sense, Exim assumes it has read an
2383 old-format file, and it generates new data and writes a new file. This
2384 means that you can't go back to an older release without removing the
2387 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2388 work if there are any delivery delays because "one_time" forces the
2389 parent to be marked "delivered", so its unseen clone is never tried
2390 again. For this reason, Exim now forbids the simultaneous setting of
2393 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2394 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2395 entry if you want to know the details. The fix, however, neglected to
2396 consider the case when local delivery batching is involved. The test for
2397 "previously delivered" was not happening when checking to see if an
2398 address could be batched with a previous (undelivered) one; under
2399 certain circumstances this could lead to multiple deliveries to the same
2402 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2403 in its include files, and this causes problems building Exim.
2405 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2406 header_syntax) but Exim was just ignoring anything given after a slash.
2407 In particular, this caused confusion with an attempt to use "verify =
2408 reverse_host_lookup/defer_ok". An error is now given when options are
2409 supplied for verify items that do not have them. (Maybe reverse_host_
2410 lookup should have a defer_ok option, but that's a different point.)
2412 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2413 defined by RFC 821) to 2048, because there were problems with some AUTH
2414 commands, and RFC 1869 says the size should be increased for extended
2415 SMTP commands that take arguments.
2417 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2420 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2421 "unknown" error; now it says that the functionality isn't in the binary.
2423 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2424 an address' error message when a string expansion fails (syntax or
2425 whatever). Otherwise the password may appear in the log. Following change
2426 PH/42 below, there is no longer a chance of it appearing in a bounce
2429 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2431 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2432 of its addresses was ever tried. (Bugzilla bug #2.)
2434 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2435 the result incorrectly in the debug output. (It correctly added a newline
2436 to what was transported.)
2438 TF/01 Added $received_time.
2440 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2441 commented out examples of how to interface to a virus scanner and to
2442 SpamAssassin. Also added commented examples of av_scanner and
2443 spamd_address settings.
2445 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2446 and controls are allowed in which ACLs. There were a couple of minor
2447 errors. Some of the entries in the conditions table (which is a table of
2448 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2449 negation of where the condition IS allowed.
2451 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2453 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2454 header file does not have a version number, so I've had to invent a new
2455 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2456 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2457 radiusclient), but it was contributed by a Radius user.
2459 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2460 files or directories, for OpenSSL.
2462 PH/19 When an Exim process that is running as root has to create an Exim log
2463 file, it does so in a subprocess that runs as exim:exim so as to get the
2464 ownership right at creation (otherwise, other Exim processes might see
2465 the file with the wrong ownership). There was no test for failure of this
2466 fork() call, which would lead to the process getting stuck as it waited
2467 for a non-existent subprocess. Forks do occasionally fail when resources
2468 run out. I reviewed all the other calls to fork(); they all seem to check
2471 PH/20 When checking for unexpected SMTP input at connect time (before writing
2472 the banner), Exim was not dealing correctly with a non-positive return
2473 from the read() function. If the client had disconnected by this time,
2474 the result was a log entry for a synchronization error with an empty
2475 string after "input=" when read() returned zero. If read() returned -1
2476 (an event I could not check), uninitialized data bytes were printed.
2477 There were reports of junk text (parts of files, etc) appearing after
2480 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2482 PH/22 Added support for macro redefinition, and (re)definition in between
2483 driver and ACL definitions.
2485 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2486 forgetting to use the resulting value; it was using the unexpanded value.
2488 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2489 hadn't been configured. The fix is from Juergen Kreileder, who
2490 understands it better than I do:
2492 "Here's what I see happening with three configured cyrus_sasl
2493 authenticators configured (plain, login, cram-md5):
2495 On startup auth_cyrus_sasl_init() gets called for each of these.
2496 This means three calls to sasl_listmech() without a specified mech_list.
2497 => SASL tests which mechs of all available mechs actually work
2498 => three warnings about OTP not working
2499 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2501 With the patch, sasl_listmech() also gets called three times. But now
2502 SASL's mech_list option is set to the server_mech specified in the the
2503 authenticator. Or in other words, the answer from sasl_listmech()
2504 gets limited to just the mech you're testing for (which is different
2506 => the return list contains just 'plain' or 'login', 'cram-md5' or
2507 nothing depending on the value of ob->server_mech.
2509 I've just tested the patch: Authentication still works fine,
2510 unavailable mechs specified in the exim configuration are still
2511 caught, and the auth.log warnings about OTP are gone."
2513 PH/25 When debugging is enabled, the contents of the command line are added
2514 to the debugging output, even when log_selector=+arguments is not
2517 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2518 answer is "GNU", and only if the return is "GNU/something" is the answer
2521 PH/27 $acl_verify_message is now set immediately after the failure of a
2522 verification in an ACL, and so is available in subsequent modifiers. In
2523 particular, the message can be preserved by coding like this:
2525 warn !verify = sender
2526 set acl_m0 = $acl_verify_message
2528 Previously, $acl_verify_message was set only while expanding "message"
2529 and "log_message" when a very denied access.
2531 PH/28 Modified OS/os.c-Linux with
2533 -#ifndef OS_LOAD_AVERAGE
2534 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2536 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2537 nomenclature these days.)
2539 PH/29 Installed patch from the Sieve maintainer that adds the options
2540 sieve_useraddress and sieve_subaddress to the redirect router.
2542 PH/30 In these circumstances:
2543 . Two addresses routed to the same list of hosts;
2544 . First host does not offer TLS;
2545 . First host accepts first address;
2546 . First host gives temporary error to second address;
2547 . Second host offers TLS and a TLS session is established;
2548 . Second host accepts second address.
2549 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2550 and peerdn, if requested) that were in fact used only for the second
2553 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2554 attention to any local part prefix or suffix that was matched by the
2555 router that accepted the address. It now behaves in the same way as it
2556 does for delivery: the affixes are removed from the local part unless
2557 rcpt_include_affixes is set on the transport.
2559 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2560 timeout during the DATA phase of an incoming message.
2562 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2563 applied a patch sanctioned by the Sieve maintainer.
2565 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2566 the uid or gid is negative. A case of a negative gid caused this to be
2567 noticed. The fix allows for either to be negative.
2569 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2570 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2573 PH/36 The change PH/12 above was broken. Fixed it.
2575 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2576 the grounds that routing the same address twice would always produce the
2577 same answer. This might have been true once, but it is certainly no
2578 longer true now. Routing a child address may depend on the previous
2579 routing that produced that child. Some complicated redirection strategies
2580 went wrong when messages had multiple recipients, and made Exim's
2581 behaviour dependent on the order in which the addresses were given.
2583 I have moved the duplicate checking until after the routing is complete.
2584 Exim scans the addresses that are assigned to local and remote
2585 transports, and removes any duplicates. This means that more work will be
2586 done, as duplicates will always all be routed, but duplicates are
2587 presumably rare, so I don't expect this is of any significance.
2589 For deliveries to pipes, files, and autoreplies, the duplicate checking
2590 still happens during the routing process, since they are not going to be
2593 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2594 It corrects a timeout issue with spamd. This is Ian's comment: "The
2595 background is that sometimes spamd either never reads data from a
2596 connection it has accepted, or it never writes response data. The exiscan
2597 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2598 blindly assumes that writes won't block so it may never time out."
2600 PH/39 Allow G after quota size as well as K and M.
2602 PH/40 The value set for $authenticated_id in an authenticator may not contain
2603 binary zeroes or newlines because the value is written to log lines and
2604 to spool files. There was no check on this. Now the value is run through
2605 the string_printing() function so that such characters are converted to
2606 printable escape sequences.
2608 PH/41 $message_linecount is a new variable that contains the total number of
2609 lines in the message. Compare $body_linecount, which is the count for the
2612 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2613 bounce and delay warning messages, except in certain special cases, which
2616 (a) An SMTP error message from a remote host;
2617 (b) A message specified in a :fail: redirection;
2618 (c) A message specified in a "fail" command in a system filter;
2619 (d) A message specified in a FAIL return from the queryprogram router;
2620 (e) A message specified by the cannot_route_message router option.
2622 In these cases only, Exim does include the error details in bounce and
2623 warning messages. There are also a few cases where bland messages such
2624 as "unrouteable address" or "local delivery error" are given.
2626 PH/43 $value is now also set for the "else" part of a ${run expansion.
2628 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2629 being worked on, but at least Exim now implements the latest version to
2632 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2633 process to complete was treated as a delivery failure, a timeout while
2634 writing the message to the pipe was logged, but erroneously treated as a
2635 successful delivery. Such timeouts include transport filter timeouts. For
2636 consistency with the overall process timeout, these timeouts are now
2637 treated as errors, giving rise to delivery failures by default. However,
2638 there is now a new Boolean option for the pipe transport called
2639 timeout_defer, which, if set TRUE, converts the failures into defers for
2640 both kinds of timeout. A transport filter timeout is now identified in
2643 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2644 systems where "make" and "gmake" are different, calling "gmake" at top
2645 level broke things. I've arranged for the value of $(MAKE) to be passed
2646 from the Makefile to this script so that it can call the same version of
2650 A note about Exim versions 4.44 and 4.50
2651 ----------------------------------------
2653 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2654 changes of various kinds. As a consequence, a big documentation update was
2655 needed. This delayed the release for rather longer than seemed good, especially
2656 in the light of a couple of (minor) security issues. Therefore, the changes
2657 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2658 release. So 4.44 and 4.50 are in effect two different branches that both start
2661 I have left the 4.50 change log unchanged; it contains all the changes since
2662 4.43. The change log for 4.44 is below; many of its items are identical to
2663 those for 4.50. This seems to be the most sensible way to preserve the
2664 historical information.
2670 1. Minor wording change to the doc/README.SIEVE file.
2672 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2673 computation of the current number of files was incorrect.
2675 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2676 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2679 4. Give more explanation in the error message when the command for a transport
2680 filter fails to execute.
2682 5. There are several places where Exim runs a non-Exim command in a
2683 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2684 was being done only for the command run by the queryprogram router. It is
2685 now done for all such subprocesses. The other cases are: ${run, transport
2686 filters, and the commands run by the lmtp and pipe transports.
2688 6. Added CONFIGURE_GROUP build-time option.
2690 7. Some older OS have a limit of 256 on the maximum number of file
2691 descriptors. Exim was using setrlimit() to set 1000 as a large value
2692 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2693 systems. I've change it so that if it can't get 1000, it tries for 256.
2695 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2696 was an oversight, and furthermore, ever since the addition of extra
2697 controls (e.g. 4.43/32), the checks on when to allow different forms of
2698 "control" were broken. There should now be diagnostics for all cases when a
2699 control that does not make sense is encountered.
2701 9. Added the /retain_sender option to "control=submission".
2703 10. $recipients is now available in the predata ACL (oversight).
2705 11. Tidy the search cache before the fork to do a delivery from a message
2706 received from the command line. Otherwise the child will trigger a lookup
2707 failure and thereby defer the delivery if it tries to use (for example) a
2708 cached ldap connection that the parent has called unbind on.
2710 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2711 of $address_data from the recipient verification was clobbered by the
2712 sender verification.
2714 13. The value of address_data from a sender verification is now available in
2715 $sender_address_data in subsequent conditions in the ACL statement.
2717 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2719 15. Added a new option "connect=<time>" to callout options, to set a different
2722 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2723 was its contents. (It was OK if the option was not defined at all.)
2725 17. A "Completed" log line is now written for messages that are removed from
2726 the spool by the -Mrm option.
2728 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2729 information about exactly what failed.
2731 19. Added -dd to debug only the daemon process.
2733 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2734 handles timeouts, both on the server side and network timeouts. Renamed the
2735 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2737 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2738 It is now set to "smtps".
2740 22. $host_address is now set to the target address during the checking of
2741 ignore_target_hosts.
2743 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2744 being passed; this would have caused $sender_host_name to have been used if
2745 matching the list had actually called for a host name (not very likely,
2746 since this list is usually IP addresses). A host name is now passed as
2749 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2750 code that specifies a non-restarting handler (typically sigaction() in
2751 modern systems) in an attempt to fix a rare and obscure crash bug.
2753 25. Narrowed the window for a race in the daemon that could cause it to ignore
2754 SIGCHLD signals. This is not a major problem, because they are used only to
2755 wake it up if nothing else does.
2757 26. A malformed maildirsize file could cause Exim to calculate negative values
2758 for the mailbox size or file count. Odd effects could occur as a result.
2759 The maildirsize information is now recalculated if the size or filecount
2762 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2763 support for a long time. Removed HAVE_SYS_VFS_H.
2765 28. Installed the latest version of exipick from John Jetmore.
2767 29. In an address list, if the pattern was not a regular expression, an empty
2768 subject address (from a bounce message) matched only if the pattern was an
2769 empty string. Non-empty patterns were not even tested. This was the wrong
2770 because it is perfectly reasonable to use an empty address as part of a
2771 database query. An empty address is now tested by patterns that are
2772 lookups. However, all the other forms of pattern expect the subject to
2773 contain a local part and a domain, and therefore, for them, an empty
2774 address still always fails if the pattern is not itself empty.
2776 30. Exim went into a mad DNS loop when attempting to do a callout where the
2777 host was specified on an smtp transport, and looking it up yielded more
2778 than one IP address.
2780 31. Re-factored the code for checking spool and log partition space into a
2781 function that finds that data and another that does the check. The former
2782 is then used to implement four new variables: $spool_space, $log_space,
2783 $spool_inodes, and $log_inodes.
2785 32. The RFC2047 encoding function was originally intended for short strings
2786 such as real names; it was not keeping to the 75-character limit for
2787 encoded words that the RFC imposes. It now respects the limit, and
2788 generates multiple encoded words if necessary. To be on the safe side, I
2789 have increased the buffer size for the ${rfc2047: expansion operator from
2792 33. It is now permitted to omit both strings after an "if" condition; if the
2793 condition is true, the result is "true". As before, when the second string
2794 is omitted, a false condition yields an empty string. This makes it less
2795 cumbersome to write custom ACL and router conditions.
2797 34. Failure to deliver a bounce message always caused it to be frozen, even if
2798 there was an errors_to setting on the router. The errors_to setting is now
2801 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2802 canonical form (fully expanded) before being placed in
2803 $sender_host_address.
2805 36. The table in the code that translates DNS record types into text (T_A to
2806 "A" for instance) was missing entries for NS and CNAME. It is just possible
2807 that this could have caused confusion if both these types were looked up
2808 for the same domain, because the text type is used as part of Exim's
2809 per-process caching. But the chance of anyone hitting this buglet seems
2812 37. The dnsdb lookup has been extended in a number of ways.
2814 (1) There is a new type, "zns", which walks up the domain tree until it
2815 finds some nameserver records. It should be used with care.
2817 (2) There is a new type, "mxh", which is like "mx" except that it returns
2818 just the host names, not the priorities.
2820 (3) It is now possible to give a list of domains (or IP addresses) to be
2821 looked up. The behaviour when one of the lookups defers can be
2822 controlled by a keyword.
2824 (4) It is now possible to specify the separator character for use when
2825 multiple records are returned.
2827 38. The dnslists ACL condition has been extended: it is now possible to supply
2828 a list of IP addresses and/or domains to be looked up in a particular DNS
2831 39. Added log_selector=+queue_time_overall.
2833 40. When running the queue in the test harness, wait just a tad after forking a
2834 delivery process, to get repeatability of debugging output.
2836 41. Include certificate and key file names in error message when GnuTLS fails
2837 to set them up, because the GnuTLS error message doesn't include the name
2838 of the failing file when there is a problem reading it.
2840 42. Allow both -bf and -bF in the same test run.
2842 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2844 44. The "Exiscan patch" is now merged into the mainline Exim source.
2846 45. Sometimes the final signoff response after QUIT could fail to get
2847 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2848 < 0 before doing a fflush(). This bug looks as though it goes back to the
2849 introduction of TLS in release 3.20, but "sometimes" must have been rare
2850 because the tests only now provoked it.
2852 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2853 (this can affect the format of dates).
2855 47. exim_tidydb, when checking for the continued existence of a message for
2856 which it has found a message-specific retry record, was not finding
2857 messages that were in split spool directories. Consequently, it was
2858 deleting retry records that should have stayed in existence.
2860 48. Steve fixed some bugs in eximstats.
2862 49. The SPA authentication driver was not abandoning authentication and moving
2863 on to the next authenticator when an expansion was forced to fail,
2864 contradicting the general specification for all authenticators. Instead it
2865 was generating a temporary error. It now behaves as specified.
2867 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2868 (the order specifies the preference for clients). The order is now AES256,
2869 AES128, 3DES, ARCFOUR128.
2871 51. Small patch to Sieve code - explicitly set From: when generating an
2874 52. Exim crashed if a remote delivery caused a very long error message to be
2875 recorded - for instance if somebody sent an entire SpamAssassin report back
2876 as a large number of 550 error lines. This bug was coincidentally fixed by
2877 increasing the size of one of Exim's internal buffers (big_buffer) that
2878 happened as part of the Exiscan merge. However, to be on the safe side, I
2879 have made the code more robust (and fixed the comments that describe what
2882 53. Now that there can be additional text after "Completed" in log lines (if
2883 the queue_time_overall log selector is set), a one-byte patch to exigrep
2884 was needed to allow it to recognize "Completed" as not the last thing in
2887 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2888 patch that reportedly fixes this has been added. I am not expert enough to
2889 create a test for it. This is what the patch creator wrote:
2891 "I found a little strange behaviour of ldap code when working with
2892 Windows 2003 AD Domain, where users was placed in more than one
2893 Organization Units. When I tried to give exim partial DN, the exit code
2894 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2895 But simultaneously result of request was absolutely normal ldap result,
2896 so I produce this patch..."
2898 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2899 REFERENCE, so I have modified the code to exclude the patch when that macro
2902 55. Some experimental protocols are using DNS PTR records for new purposes. The
2903 keys for these records are domain names, not reversed IP addresses. The
2904 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2905 leaves it alone. Component reversal etc. now happens only for IP addresses.
2908 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2910 57. Double the size of the debug message buffer (to 2048) so that more of very
2911 long debug lines gets shown.
2913 58. The exicyclog utility now does better if the number of log files to keep
2914 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2916 59. Two changes related to the smtp_active_hostname option:
2918 (1) $smtp_active_hostname is now available as a variable.
2919 (2) The default for smtp_banner uses $smtp_active_hostname instead
2920 of $primary_hostname.
2922 60. The host_aton() function is supposed to be passed a string that is known
2923 to be a valid IP address. However, in the case of IPv6 addresses, it was
2924 not checking this. This is a hostage to fortune. Exim now panics and dies
2925 if the condition is not met. A case was found where this could be provoked
2926 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2927 components; fortuitously, this particular loophole had already been fixed
2928 by change 4.50/55 above.
2930 If there are any other similar loopholes, the new check in host_aton()
2931 itself should stop them being exploited. The report I received stated that
2932 data on the command line could provoke the exploit when Exim was running as
2933 exim, but did not say which command line option was involved. All I could
2934 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2935 running as the user.
2938 61. There was a buffer overflow vulnerability in the SPA authentication code
2939 (which came originally from the Samba project). I have added a test to the
2940 spa_base64_to_bits() function which I hope fixes it.
2943 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2944 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2946 63. The daemon start-up calls getloadavg() while still root for those OS that
2947 need the first call to be done as root, but it missed one case: when
2948 deliver_queue_load_max is set with deliver_drop_privilege. This is
2949 necessary for the benefit of the queue runner, because there is no re-exec
2950 when deliver_drop_privilege is set.
2952 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2953 This has been fixed.
2955 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2956 was in use, was not putting the data itself into the right store pool;
2957 consequently, it could be overwritten for a subsequent message in the same
2958 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2961 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2963 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2964 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2965 the function were treating the return as a boolean value, which happened to
2966 work because 0=false and not-0=true, but is not correct code.
2968 68. The host_aton() function was not handling scoped IPv6 addresses (those
2969 with, for example, "%eth0" on the end) correctly.
2971 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2972 negated items (that is, ~something) in unsigned ints. Some compilers
2973 apparently mutter when there is no cast.
2975 70. If an address verification called from an ACL failed, and did not produce a
2976 user-specific message (i.e. there was only a "system" message), nothing was
2977 put in $acl_verify_message. In this situation, it now puts the system
2980 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2981 session; change 4.31/43 added the unwanted input to the log line - except
2982 that it did not do this in the start of session case. It now does.
2984 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2985 This is wrong and can cause the other end to generate a synchronization
2986 error if it is another Exim or anything else that does the synchronization
2987 check. A QUIT command is no longer sent after a timeout.
2989 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2990 during host lookups.
2992 74. The defer_ok option of callout verification was not working if it was used
2993 when verifying addresses in header lines, that is, for this case:
2995 verify = header_sender/callout=defer_ok
2997 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2998 those file descriptors could be used for SMTP connections. If anything
2999 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3000 could be sent to the SMTP client, causing chaos. The daemon now opens
3001 stdin, stdout, and stderr to /dev/null when it puts itself into the
3004 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3005 log by default. The user can override this with suitable Perl magic.
3007 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3008 the log message when discard triggers, was not working for the DATA ACL or
3009 for the non-SMTP ACL.
3011 78. Error message wording change in sieve.c.
3013 79. If smtp_accept_max_per_host was set, the number of connections could be
3014 restricted to fewer than expected, because the daemon was trying to set up
3015 a new connection before checking whether the processes handling previous
3016 connections had finished. The check for completed processes is now done
3017 earlier. On busy systems, this bug wouldn't be noticed because something
3018 else would have woken the daemon, and it would have reaped the completed
3021 80. If a message was submitted locally by a user whose login name contained one
3022 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3023 It caused a spool format error. I have fixed the spool reading code. A
3024 related problem was that the "from" clause in the Received: line became
3025 illegal because of the space(s). It is now covered by ${quote_local_part.
3027 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3030 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3032 83. Patch from Sieve maintainer to fix unterminated string problem in
3033 "vacation" handling.
3035 84. Some minor changes to the Linux configuration files to help with other
3036 OS variants using glibc.
3038 85. One more patch for Sieve to update vacation handling to latest spec.
3041 ----------------------------------------------------
3042 See the note above about the 4.44 and 4.50 releases.
3043 ----------------------------------------------------
3049 1. Change 4.43/35 introduced a bug that caused file counts to be
3050 incorrectly computed when quota_filecount was set in an appendfile
3053 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3054 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3057 3. Give more explanation in the error message when the command for a transport
3058 filter fails to execute.
3060 4. There are several places where Exim runs a non-Exim command in a
3061 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3062 was being done only for the command run by the queryprogram router. It is
3063 now done for all such subprocesses. The other cases are: ${run, transport
3064 filters, and the commands run by the lmtp and pipe transports.
3066 5. Some older OS have a limit of 256 on the maximum number of file
3067 descriptors. Exim was using setrlimit() to set 1000 as a large value
3068 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3069 systems. I've change it so that if it can't get 1000, it tries for 256.
3071 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3072 was an oversight, and furthermore, ever since the addition of extra
3073 controls (e.g. 4.43/32), the checks on when to allow different forms of
3074 "control" were broken. There should now be diagnostics for all cases when a
3075 control that does not make sense is encountered.
3077 7. $recipients is now available in the predata ACL (oversight).
3079 8. Tidy the search cache before the fork to do a delivery from a message
3080 received from the command line. Otherwise the child will trigger a lookup
3081 failure and thereby defer the delivery if it tries to use (for example) a
3082 cached ldap connection that the parent has called unbind on.
3084 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3085 of $address_data from the recipient verification was clobbered by the
3086 sender verification.
3088 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3089 was its contents. (It was OK if the option was not defined at all.)
3091 11. A "Completed" log line is now written for messages that are removed from
3092 the spool by the -Mrm option.
3094 12. $host_address is now set to the target address during the checking of
3095 ignore_target_hosts.
3097 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3098 being passed; this would have caused $sender_host_name to have been used if
3099 matching the list had actually called for a host name (not very likely,
3100 since this list is usually IP addresses). A host name is now passed as
3103 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3104 code that specifies a non-restarting handler (typically sigaction() in
3105 modern systems) in an attempt to fix a rare and obscure crash bug.
3107 15. Narrowed the window for a race in the daemon that could cause it to ignore
3108 SIGCHLD signals. This is not a major problem, because they are used only to
3109 wake it up if nothing else does.
3111 16. A malformed maildirsize file could cause Exim to calculate negative values
3112 for the mailbox size or file count. Odd effects could occur as a result.
3113 The maildirsize information is now recalculated if the size or filecount
3116 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3117 support for a long time. Removed HAVE_SYS_VFS_H.
3119 18. Updated exipick to current release from John Jetmore.
3121 19. Allow an empty sender to be matched against a lookup in an address list.
3122 Previously the only cases considered were a regular expression, or an
3125 20. Exim went into a mad DNS lookup loop when doing a callout where the
3126 host was specified on the transport, if the DNS lookup yielded more than
3129 21. The RFC2047 encoding function was originally intended for short strings
3130 such as real names; it was not keeping to the 75-character limit for
3131 encoded words that the RFC imposes. It now respects the limit, and
3132 generates multiple encoded words if necessary. To be on the safe side, I
3133 have increased the buffer size for the ${rfc2047: expansion operator from
3136 22. Failure to deliver a bounce message always caused it to be frozen, even if
3137 there was an errors_to setting on the router. The errors_to setting is now
3140 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3141 canonical form (fully expanded) before being placed in
3142 $sender_host_address.
3144 24. Updated eximstats to version 1.33
3146 25. Include certificate and key file names in error message when GnuTLS fails
3147 to set them up, because the GnuTLS error message doesn't include the name
3148 of the failing file when there is a problem reading it.
3150 26. Expand error message when OpenSSL has problems setting up cert/key files.
3153 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3154 (this can affect the format of dates).
3156 28. exim_tidydb, when checking for the continued existence of a message for
3157 which it has found a message-specific retry record, was not finding
3158 messages that were in split spool directories. Consequently, it was
3159 deleting retry records that should have stayed in existence.
3161 29. eximstats updated to version 1.35
3162 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3163 1.35 - bugfix such that pie charts by volume are generated correctly
3165 30. The SPA authentication driver was not abandoning authentication and moving
3166 on to the next authenticator when an expansion was forced to fail,
3167 contradicting the general specification for all authenticators. Instead it
3168 was generating a temporary error. It now behaves as specified.
3170 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3171 (the order specifies the preference for clients). The order is now AES256,
3172 AES128, 3DES, ARCFOUR128.
3174 31. Small patch to Sieve code - explicitly set From: when generating an
3177 32. Exim crashed if a remote delivery caused a very long error message to be
3178 recorded - for instance if somebody sent an entire SpamAssassin report back
3179 as a large number of 550 error lines. This bug was coincidentally fixed by
3180 increasing the size of one of Exim's internal buffers (big_buffer) that
3181 happened as part of the Exiscan merge. However, to be on the safe side, I
3182 have made the code more robust (and fixed the comments that describe what
3185 33. Some experimental protocols are using DNS PTR records for new purposes. The
3186 keys for these records are domain names, not reversed IP addresses. The
3187 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3188 leaves it alone. Component reversal etc. now happens only for IP addresses.
3191 34. The host_aton() function is supposed to be passed a string that is known
3192 to be a valid IP address. However, in the case of IPv6 addresses, it was
3193 not checking this. This is a hostage to fortune. Exim now panics and dies
3194 if the condition is not met. A case was found where this could be provoked
3195 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3196 components; fortuitously, this particular loophole had already been fixed
3197 by change 4.50/55 or 4.44/33 above.
3199 If there are any other similar loopholes, the new check in host_aton()
3200 itself should stop them being exploited. The report I received stated that
3201 data on the command line could provoke the exploit when Exim was running as
3202 exim, but did not say which command line option was involved. All I could
3203 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3204 running as the user.
3207 35. There was a buffer overflow vulnerability in the SPA authentication code
3208 (which came originally from the Samba project). I have added a test to the
3209 spa_base64_to_bits() function which I hope fixes it.
3212 36. The daemon start-up calls getloadavg() while still root for those OS that
3213 need the first call to be done as root, but it missed one case: when
3214 deliver_queue_load_max is set with deliver_drop_privilege. This is
3215 necessary for the benefit of the queue runner, because there is no re-exec
3216 when deliver_drop_privilege is set.
3218 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3219 was in use, was not putting the data itself into the right store pool;
3220 consequently, it could be overwritten for a subsequent message in the same
3221 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3224 38. Sometimes the final signoff response after QUIT could fail to get
3225 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3226 < 0 before doing a fflush(). This bug looks as though it goes back to the
3227 introduction of TLS in release 3.20, but "sometimes" must have been rare
3228 because the tests only now provoked it.
3234 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3235 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3236 Now it returns an integer. A number of calls were still expecting a T/F
3237 return. Fortuitously, in all cases, the tests worked in OK situations,
3238 which is the norm. However, things would have gone wrong on any write
3239 failures on the smtp file descriptor. This function is used when sending
3240 messages over SMTP and also when doing verify callouts.
3242 2. When Exim is called to do synchronous delivery of a locally submitted
3243 message (the -odf or -odi options), it no longer closes stderr before doing
3246 3. Implemented the mua_wrapper option.
3248 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3250 5. Implemented the functions header_remove(), header_testname(),
3251 header_add_at_position(), and receive_remove_recipient(), and exported them
3254 6. If an ACL "warn" statement specified the addition of headers, Exim already
3255 inserted X-ACL-Warn: at the start if there was no header name. However, it
3256 was not making this test for the second and subsequent header lines if
3257 there were newlines in the string. This meant that an invalid header could
3258 be inserted if Exim was badly configured.
3260 7. Allow an ACL "warn" statement to add header lines at the start or after all
3261 the Received: headers, as well as at the end.
3263 8. Added the rcpt_4xx retry error code.
3265 9. Added postmaster_mailfrom=xxx to callout verification option.
3267 10. Added mailfrom=xxxx to the callout verification option, for verify=
3270 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3271 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3273 12. Inserted some casts to stop certain compilers warning when using pointer
3274 differences as field lengths or precisions in printf-type calls (mostly
3275 affecting debugging statements).
3277 13. Added optional readline() support for -be (dynamically loaded).
3279 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3280 same clock tick as a message's arrival, so that its received time was the
3281 same as the "first fail" time on the retry record, and that message
3282 remained on the queue past the ultimate address timeout, every queue runner
3283 would try a delivery (because it was past the ultimate address timeout) but
3284 after another failure, the ultimate address timeout, which should have then
3285 bounced the address, did not kick in. This was a "< instead of <=" error;
3286 in most cases the first failure would have been in the next clock tick
3287 after the received time, and all would be well.
3289 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3290 being recognized when the domain list was tested by the match_domain
3291 condition in an expansion string.
3293 16. Added the ${str2b64: operator.
3295 17. Exim was always calling setrlimit() to set a large limit for the number of
3296 processes, without checking whether the existing limit was already
3297 adequate. (It did check for the limit on file descriptors.) Furthermore,
3298 errors from getrlimit() and setrlimit() were being ignored. Now they are
3299 logged to the main and panic logs, but Exim does carry on, to try to do its
3300 job under whatever limits there are.
3302 18. Imported PCRE 5.0.
3304 19. Trivial typo in log message " temporarily refused connection" (the leading
3307 20. If the log selector return_path_on_delivery was set and an address was
3308 redirected to /dev/null, the delivery process crashed because it assumed
3309 that a return path would always be set for a "successful" delivery. In this
3310 case, the whole delivery is bypassed as an optimization, and therefore no
3313 21. Internal re-arrangement: the function for sending a challenge and reading
3314 a response while authentication was assuming a zero-terminated challenge
3315 string. It's now changed to take a pointer and a length, to allow for
3316 binary data in such strings.
3318 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3320 23. Exim was not respecting finduser_retries when seeking the login of the
3321 uid under which it was called; it was always trying 10 times. (The default
3322 setting of finduser_retries is zero.) Also, it was sleeping after the final
3323 failure, which is pointless.
3325 24. Implemented tls_on_connect_ports.
3327 25. Implemented acl_smtp_predata.
3329 26. If the domain in control=submission is set empty, Exim assumes that the
3330 authenticated id is a complete email address when it generates From: or
3331 Sender: header lines.
3333 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3334 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3335 chown and chgrp in /bin and hostname in /usr/bin.
3337 28. Exim was keeping the "process log" file open after each use, just as it
3338 does for the main log. This opens the possibility of it remaining open for
3339 long periods when the USR1 signal hits a daemon. Occasional processlog
3340 errors were reported, that could have been caused by this. Anyway, it seems
3341 much more sensible not to leave this file open at all, so that is what now
3344 29. The long-running daemon process does not normally write to the log once it
3345 has entered its main loop, and it closes the log before doing so. This is
3346 so that log files can straightforwardly be renamed and moved. However,
3347 there are a couple of unusual error situations where the daemon does write
3348 log entries, and I had neglected to close the log afterwards.
3350 30. The text of an SMTP error response that was received during a remote
3351 delivery was being truncated at 512 bytes. This is too short for some of
3352 the long messages that one sometimes sees. I've increased the limit to
3355 31. It is now possible to make retry rules that apply only when a message has a
3356 specific sender, in particular, an empty sender.
3358 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3359 it possible to be selective about when SMTP synchronization is enforced.
3361 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3363 32. Implemented hosts_connection_nolog.
3365 33. Added an ACL for QUIT.
3367 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3370 35. Added mailbox_size and mailbox_filecount to appendfile.
3372 36. Added control = no_multiline_responses to ACLs.
3374 37. There was a bug in the logic of the code that waits for the clock to tick
3375 in the case where the clock went backwards by a substantial amount such
3376 that the microsecond fraction of "now" was more than the microsecond
3377 fraction of "then" (but the whole seconds number was less).
3379 38. Added support for the libradius Radius client library this is found on
3380 FreeBSD (previously only the radiusclient library was supported).
3386 1. When certain lookups returned multiple values in the form name=value, the
3387 quoting of the values was not always being done properly. Specifically:
3388 (a) If the value started with a double quote, but contained no whitespace,
3390 (b) If the value contained whitespace other than a space character (i.e.
3391 tabs or newlines or carriage returns) it was not quoted.
3392 This fix has been applied to the mysql and pgsql lookups by writing a
3393 separate quoting function and calling it from the lookup code. The fix
3394 should probably also be applied to nisplus, ibase and oracle lookups, but
3395 since I cannot test any of those, I have not disturbed their existing code.
3397 2. A hit in the callout cache for a specific address caused a log line with no
3398 reason for rejecting RCPT. Now it says "Previous (cached) callout
3399 verification failure".
3401 3. There was an off-by-one bug in the queryprogram router. An over-long
3402 return line was truncated at 256 instead of 255 characters, thereby
3403 overflowing its buffer with the terminating zero. As well as fixing this, I
3404 have increased the buffer size to 1024 (and made a note to document this).
3406 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3407 when Exim is waiting for an SMTP response from a remote server, Exim
3408 restarts its select() call on the socket, thereby resetting its timeout.
3409 This is not a problem when such interrupts are rare. Somebody set up a cron
3410 job to run exiwhat every 2 minutes, which is less than the normal select()
3411 timeout (5 or 10 minutes). This meant that the select() timeout never
3412 kicked in because it was always reset. I have fixed this by comparing the
3413 time when an interrupt arrives with the time at the start of the first call
3414 to select(). If more time than the timeout has elapsed, the interrupt is
3415 treated as a timeout.
3417 5. Some internal re-factoring in preparation for the addition of Sieve
3418 extensions (by MH). In particular, the "personal" test is moved to a
3419 separate function, and given an option for scanning Cc: and Bcc: (which is
3420 not set for Exim filters).
3422 6. When Exim created an email address using the login of the caller as the
3423 local part (e.g. when creating a From: or Sender: header line), it was not
3424 quoting the local part when it contained special characters such as @.
3426 7. Installed new OpenBSD configuration files.
3428 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3429 try to make them clearer.
3431 9. Callout options, other than the timeout value, were being ignored when
3432 verifying sender addresses in header lines. For example, when using
3434 verify = header_sender/callout=no_cache
3436 the cache was (incorrectly) being used.
3438 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3439 only the Cygwin environment.
3441 11. When return_path_on_delivery was set as a log selector, if different remote
3442 addresses in the same message used different return paths and parallel
3443 remote delivery occurred, the wrong values would sometimes be logged.
3444 (Whenever a remote delivery process finished, the return path value from
3445 the most recently started remote delivery process was logged.)
3447 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3448 lines when AUTH and/or TLS are in use. This is the "received protocol"
3449 field. Exim used to use "asmtp" for authenticated SMTP, without any
3450 indication (in the protocol name) for TLS use. Now it follows the RFC and
3451 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3452 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3453 names appear in log lines as well as in Received: header lines.
3455 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3456 extensions, and comparison tests, and to fix some bugs.
3458 14. Changes to the "personal" filter test:
3460 (1) The test was buggy in that it was just doing the equivalent of
3461 "contains" tests on header lines. For example, if a user's address was
3462 anne@some.where, the "personal" test would incorrectly be true for
3464 To: susanne@some.where
3466 This test is now done by extracting each address from the header in turn,
3467 and checking the entire address. Other tests that are part of "personal"
3468 are now done using regular expressions (for example, to check local parts
3469 of addresses in From: header lines).
3471 (2) The list of non-personal local parts in From: addresses has been
3472 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3473 taken from the Sieve specification recommendations.
3475 (3) If the message contains any header line starting with "List-" it is
3476 treated as non-personal.
3478 (4) The test for "circular" in the Subject: header line has been removed
3479 because it now seems ill-conceived.
3481 15. Minor typos in src/EDITME comments corrected.
3483 16. Installed latest exipick from John Jetmore.
3485 17. If headers_add on a router specified a text string that was too long for
3486 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3487 of string_sprintf() is now avoided.
3489 18. $message_body_size was not set (it was always zero) when running the DATA
3490 ACL and the local_scan() function.
3492 19. For the "mail" command in an Exim filter, no default was being set for
3493 the once_repeat time, causing a random time value to be used if "once" was
3494 specified. (If the value happened to be <= 0, no repeat happened.) The
3495 default is now 0s, meaning "never repeat". The "vacation" command was OK
3496 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3497 (I found it when inspecting the code).
3499 20. There is now an overall timeout for performing a callout verification. It
3500 defaults to 4 times the callout timeout, which applies to individual SMTP
3501 commands during the callout. The overall timeout applies when there is more
3502 than one host that can be tried. The timeout is checked before trying the
3503 next host. This prevents very long delays if there are a large number of
3504 hosts and all are timing out (e.g. when the network connections are timing
3505 out). The value of the overall timeout can be changed by specifying an
3506 additional sub-option for "callout", called "maxwait". For example:
3508 verify = sender/callout=5s,maxwait=20s
3510 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3511 to the end before writing, but this should make it even safer).
3513 22. Exim was forgetting that it had advertised PIPELINING for the second and
3514 subsequent messages on an SMTP connection. It was also not resetting its
3515 memory on STARTTLS and an internal HELO.
3517 23. When Exim logs an SMTP synchronization error within a session, it now
3518 records whether PIPELINING has been advertised or not.
3520 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3521 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3522 rather than long int.
3524 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3526 26. Added the never_mail option to autoreply.
3532 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3533 crash if the getsockname() call failed; this can happen if a connection is
3534 closed very soon after it is established. The problem was simply in the
3535 order in which certain operations were done, causing Exim to try to write
3536 to the SMTP stream before it had set up the file descriptor. The bug has
3537 been fixed by making things happen in the correct order.
3543 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3544 before the connection was closed, thus losing the rejection response.
3546 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3547 some early Solaris releases, but causes trouble in current releases where
3548 socklen_t is defined.
3550 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3553 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3556 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3557 that releasing the top of it at the end releases what was used for sub-
3558 expansions (unless the block got too big). However, discard this block if
3559 the first thing is a variable or header, so that we can use its block when
3560 it is dynamic (useful for very large $message_headers, for example).
3562 6. Lookups now cache *every* query, not just the most recent. A new, separate
3563 store pool is used for this. It can be recovered when all lookup caches are
3564 flushed. Lookups now release memory at the end of their result strings.
3565 This has involved some general refactoring of the lookup sources.
3567 7. Some code has been added to the store_xxx() functions to reduce the amount
3568 of flapping under certain conditions.
3570 8. log_incoming_interface used to affect only the <= reception log lines. Now
3571 it causes the local interface and port to be added to several more SMTP log
3572 lines, for example "SMTP connection from", and rejection lines.
3574 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3576 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3578 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3579 could be overwritten at the end of the current message (or the start of a
3580 new message if it was set in a HELO ACL). The value is now preserved for
3581 the duration of the SMTP connection.
3583 12. If a transport had a headers_rewrite setting, and a matching header line
3584 contained an unqualified address, that address was qualified, even if it
3585 did not match any rewriting rules. The underlying bug was that the values
3586 of the flags that permit the existence of unqualified sender and recipient
3587 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3588 non-local messages, and by -bnq for local messages) were not being
3589 preserved with the message after it was received.
3591 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3592 "next input=" as part of the text comprising the host identity instead of
3593 the correct text. The code was using the same buffer for two different
3594 strings. However, depending on which order the printing function evaluated
3595 its arguments, the bug did not always show up. Under Linux, for example, my
3596 test suite worked just fine.
3598 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3599 doesn't work with some older versions of Perl. It has been changed to "my",
3600 which in any case is probably the better facility to use.
3602 15. A really picky compiler found some instances of statements for creating
3603 error messages that either had too many or two few arguments for the format
3606 16. The size of the buffer for calls to the DNS resolver has been increased
3607 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3608 for addresses that have a lot of PTR records. This alleviates a problem; it
3609 does not fully solve it.
3611 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3612 buffer now truncates the list and logs the incident, which is the same
3613 action as happens when Exim is looking up a host name and its aliases.
3614 Previously in this situation something unpredictable would happen;
3615 sometimes it was "internal error: store_reset failed".
3617 18. If a server dropped the connection unexpectedly when an Exim client was
3618 using GnuTLS and trying to read a response, the client delivery process
3619 crashed while trying to generate an error log message.
3621 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3622 string, for example:
3624 warn message = H1: something\nH2: something
3626 the text was added as a single header line from Exim's point of view
3627 though it ended up OK in the delivered message. However, searching for the
3628 second and subsequent header lines using $h_h2: did not work. This has been
3629 fixed. Similarly, if a system filter added multiple headers in this way,
3630 the routers could not see them.
3632 20. Expanded the error message when iplsearch is called with an invalid key to
3633 suggest using net-iplsearch in a host list.
3635 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3636 ACLs are no longer actually imposed (and a message to that effect is
3639 22. If a "gecos" field in a passwd entry contained escaped characters, in
3640 particular, if it contained a \" sequence, Exim got it wrong when building
3641 a From: or a Sender: header from that name. A second bug also caused
3642 incorrect handling when an unquoted " was present following a character
3643 that needed quoting.
3645 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3646 was not being matched caselessly.
3648 24. Arranged for all hyphens in the exim.8 source to be escaped with
3651 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3652 properly. Recipient callout cache records were still being keyed to include
3653 the sender, even when use_sender was set false. This led to far more
3654 callouts that were necessary. The sender is no longer included in the key
3655 when use_sender is false.
3657 26. Added "control = submission" modifier to ACLs.
3659 27. Added the ${base62d: operator to decode base 62 numbers.
3661 28. dnsdb lookups can now access SRV records.
3663 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3664 the configuration file.
3666 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3667 (-v) mode. This makes the output for a verbose queue run more intelligible.
3669 31. Added a use_postmaster feature to recipient callouts.
3671 32. Added the $body_zerocount variable, containing the number of binary zero
3672 bytes in the message body.
3674 33. The time of last modification of the "new" subdirectory is now used as the
3675 "mailbox time last read" when there is a quota error for a maildir
3678 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3680 35. Added +ignore_unknown as a special item in host lists.
3682 36. Code for decoding IPv6 addresses in host lists is now included, even if
3683 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3684 address was recognized as an IP address, but was then not correctly decoded
3685 into binary, causing unexpected and incorrect effects when compared with
3692 1. Very minor rewording of debugging text in manualroute to say "list of
3693 hosts" instead of "hostlist".
3695 2. If verify=header_syntax was set, and a header line with an unqualified
3696 address (no domain) and a large number of spaces between the end of the
3697 name and the colon was received, the reception process suffered a buffer
3698 overflow, and (when I tested it) crashed. This was caused by some obsolete
3699 code that should have been removed. The fix is to remove it!
3701 3. When running in the test harness, delay a bit after writing a bounce
3702 message to get a bit more predictability in the log output.
3704 4. Added a call to search_tidyup() just before forking a reception process. In
3705 theory, someone could use a lookup in the expansion of smtp_accept_max_
3706 per_host which, without the tidyup, could leave open a database connection.
3708 5. Added the variables $recipient_data and $sender_data which get set from a
3709 lookup success in an ACL "recipients" or "senders" condition, or a router
3710 "senders" option, similar to $domain_data and $local_part_data.
3712 6. Moved the writing of debug_print from before to after the "senders" test
3715 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3716 problems for message scanning, either using a data ACL, or using
3717 local_scan() because the Received: header was not generated till after they
3718 were called (in order to set the time as the time of reception completion).
3719 I have revised the way this works. The header is now generated after the
3720 body is received, but before the ACL or local_scan() are called. After they
3721 are run, the timestamp in the header is updated.
3727 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3728 before starting a queue runner without re-exec. This happened only when
3729 deliver_drop_privilege was set or when the Exim user was set to root. The
3730 effect of the bug was that timeouts during subsequent deliveries caused
3731 crashes instead of being properly handled. The handler is now left at its
3732 default (and expected) setting.
3734 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3735 message, again when deliver_drop_privilege is set or Exim is run as root.
3736 The bug described in (1) was not present in this case, but the tidying up
3737 of the other signals was missing. I have made the two cases consistent.
3739 3. The ignore_target_hosts setting on a manualroute router was being ignored
3740 for hosts that were looked up using the /MX notation.
3742 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3745 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3746 operated on the sender address. After changing the $sender_address to <>
3747 for the sender address verify, Exim was re-instated it as the original
3748 (before rewriting) address, but remembering that it had rewritten it, so it
3749 wasn't rewriting it again. This bug also had the effect of breaking the
3750 sender address verification caching when the sender address was rewritten.
3752 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3753 This has been changed so that if the ip literal address matches
3754 ignore_target_hosts, the router declines.
3756 7. Added expansion conditions match_domain, match_address, and match_local_
3757 part (NOT match_host).
3759 8. The placeholder for the Received: header didn't have a length field set.
3761 9. Added code to Exim itself and to exim_lock to test for a specific race
3762 condition that could lead to file corruption when using MBX delivery. The
3763 issue is with the lockfile that is created in /tmp. If this file is removed
3764 after a process has opened it but before that process has acquired a lock,
3765 there is the potential for a second process to recreate the file and also
3766 acquire a lock. This could lead to two Exim processes writing to the file
3767 at the same time. The added code performs the same test as UW imapd; it
3768 checks after acquiring the lock that its file descriptor still refers to
3769 the same named file.
3771 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3772 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3775 11. Added the smtp_active_hostname option. If used, this will typically be made
3776 to depend on the incoming interface address. Because $interface_address is
3777 not set up until the daemon has forked a reception process, error responses
3778 that can happen earlier (such as "too many connections") no longer contain
3781 12. If an expansion in a condition on a "warn" statement fails because a lookup
3782 defers, the "warn" statement is abandoned, and the next ACL statement is
3783 processed. Previously this caused the whole ACL to be aborted.
3785 13. Added the iplsearch lookup type.
3787 14. Added ident_timeout as a log selector.
3789 15. Added tls_certificate_verified as a log selector.
3791 16. Added a global option tls_require_ciphers (compare the smtp transport
3792 option of the same name). This controls incoming TLS connections.
3794 17. I finally figured out how to make tls_require_ciphers do a similar thing
3795 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3796 before starting the TLS session.
3798 18. Tabs are now shown as \t in -bP output.
3800 19. If the log selector return_path_on_delivery was set, Exim crashed when
3801 bouncing a message because it had too many Received: header lines.
3803 20. If two routers both had headers_remove settings, and the first one included
3804 a superfluous trailing colon, the final name in the first list and the
3805 first name in the second list were incorrectly joined into one item (with a
3806 colon in the middle).
3812 1. Added -C and -D options to the exinext utility, mainly to make it easier
3813 to include in the automated testing, but these could be helpful when
3814 multiple configurations are in use.
3816 2. The exinext utility was not formatting the output nicely when there was
3817 an alternate port involved in the retry record key, nor when there was a
3818 message id as well (for retries that were specific to a specific message
3819 and a specific host). It was also confused by IPv6 addresses, because of
3820 the additional colons they contain. I have fixed the IPv4 problem, and
3821 patched it up to do a reasonable job for IPv6.
3823 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3824 delivery, the log line now contains "pipelined" if PIPELINING was used.
3826 4. An SMTP transport process used to panic and die if the bind() call to set
3827 an explicit outgoing interface failed. This has been changed; it is now
3828 treated in the same way as a connect() failure.
3830 5. A reference to $sender_host_name in the part of a conditional expansion
3831 that was being skipped was still causing a DNS lookup. This no longer
3834 6. The def: expansion condition was not recognizing references to header lines
3835 that used bh_ and bheader_.
3837 7. Added the _cache feature to named lists.
3839 8. The code for checking quota_filecount in the appendfile transport was
3840 allowing one more file than it should have been.
3842 9. For compatibility with Sendmail, the command line option
3850 and sets the incoming protocol and host name (for trusted callers). The
3851 host name and its colon can be omitted when only the protocol is to be set.
3852 Note the Exim already has two private options, -pd and -ps, that refer to
3853 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3854 "s", but I don't think that's a major issue.
3856 10. A number of refactoring changes to the code, none of which should affect
3859 (a) The number of logging options was getting close to filling up the
3860 32-bit word that was used as a bit map. I have split them into two classes:
3861 those that are passed in the argument to log_write(), and those that are
3862 only ever tested independently outside of that function. These are now in
3863 separate 32-bit words, so there is plenty of room for expansion again.
3864 There is no change in the user interface or the logging behaviour.
3866 (b) When building, for example, log lines, the code previously used a
3867 macro that called string_cat() twice, in order to add two strings. This is
3868 not really sufficiently general. Furthermore, there was one instance where
3869 it was actually wrong because one of the argument was used twice, and in
3870 one call a function was used. (As it happened, calling the function twice
3871 did not affect the overall behaviour.) The macro has been replaced by a
3872 function that can join an arbitrary number of extra strings onto a growing
3875 (c) The code for expansion conditions now uses a table and a binary chop
3876 instead of a serial search (which was left over from when there were very
3877 few conditions). Also, it now recognizes conditions like "pam" even when
3878 the relevant support is not compiled in: a suitably worded error message is
3879 given if an attempt is made to use such a condition.
3881 11. Added ${time_interval:xxxxx}.
3883 12. A bug was causing one of the ddress fields not to be passed back correctly
3884 from remote delivery subprocesses. The field in question was not being
3885 subsequently used, so this caused to problems in practice.
3887 13. Added new log selectors queue_time and deliver_time.
3889 14. Might have fixed a bug in maildirsizefile handling that threw up
3890 "unexpected character" debug warnings, and recalculated the data
3891 unnecessarily. In any case, I expanded the warning message to give more
3894 15. Added the message "Restricted characters in address" to the statements in
3895 the default ACL that block characters like @ and % in local parts.
3897 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3898 Three changes have been made:
3900 (a) There was a serious bug; a negative response to MAIL caused the whole
3901 recipient domain to be cached as invalid, thereby blocking all messages
3902 to all local parts at the same domain, from all senders. This bug has
3903 been fixed. The domain is no longer cached after a negative response to
3904 MAIL if the sender used is not empty.
3906 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3909 (c) A new callout option, "use_sender" has been added for people who want
3910 the modified behaviour.
3916 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3919 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3920 indeed breaks things for older releases.
3922 3. Added additional logging to the case where there is a problem reading data
3923 from a filter that is running in a subprocess using a pipe, in order to
3924 try to track down a specific problem.
3926 4. Testing facility fudge: when running in the test harness and attempting
3927 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3928 getting "No route to host". Convert this to a timeout.
3930 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3933 6. Some OS don't have socklen_t but use size_t instead. This affects the
3934 fifth argument of getsockopt() amongst other things. This is now
3935 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3936 can be set for individual OS. I have set it for SunOS5, OSF1, and
3937 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3938 some earlier ones do not.
3940 7. Change 4.30/15 was not doing the test caselessly.
3942 8. The standard form for an IPv6 address literal was being rejected by address
3943 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3944 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3945 this, as well as the form without the "IPv6" on the front (but only when
3946 address literals are enabled, of course).
3948 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3950 10. Exim crashed if a message with an empty sender address specified by -f
3951 encountered a router with an errors_to setting. This could be provoked only
3952 by a command such as
3956 where an empty string was supplied; "<>" did not hit this bug.
3958 11. Installed PCRE release 4.5.
3960 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3961 remained set. It is now erased.
3963 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3964 times from message ids (which are base 36 rather than the normal 62).
3966 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3967 were being counted as actual protocol errors, and logged if the log
3968 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3969 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3970 and DATA following a set of rejected RCPTs do not count as protocol errors.
3971 In other words, Exim assumes they were pipelined, though this may not
3972 actually be the case. Of course, in all cases the client gets an
3973 appropriate error code.
3975 15. If a lookup fails in an ACL condition, a message about the failure may
3976 be available; it is used if testing the ACL cannot continue, because most
3977 such messages specify what the cause of the deferral is. However, some
3978 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3979 that caused an old message to be retained and used if a later statement
3980 caused a defer, replacing the real cause of the deferral.
3982 16. If an IP address had so many PTR records that the DNS lookup buffer
3983 was not large enough to hold them, Exim could crash while trying to process
3984 the truncated data. It now detects and logs this case.
3986 17. Further to 4.21/58, another change has been made: if (and only if) the
3987 first line of a message (the first header line) ends with CRLF, a bare LF
3988 in a subsequent header line has a space inserted after it, so as not to
3989 terminate the header.
3991 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3992 unnecessarily, used atoi() instead of strtol(), and didn't check the
3993 termination when getting file sizes from file names by regex.
3995 19. Completely re-implemented the support for maildirsize files, in the light
3996 of a number of problems with the previous contributed implementation
3997 (4.30/29). In particular:
3999 . If the quota is zero, the maildirsize file is maintained, but no quota is
4002 . If the maildir directory does not exist, it is created before any attempt
4003 to write a maildirsize file.
4005 . The quota value in the file is just a cache; if the quota is changed in
4006 the transport, the new value overrides.
4008 . A regular expression is available for excluding directories from the
4011 20. The autoreply transport checks the characters in options that define the
4012 message's headers; it allows continued headers, but it was checking with
4013 isspace() after an embedded newline instead of explicitly looking for a
4016 21. If all the "regular" hosts to which an address was routed had passed their
4017 expiry times, and had not reached their retry times, the address was
4018 bounced, even if fallback hosts were defined. Now Exim should go on to try
4021 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4022 equivalent code in the SMTP transport. Some hosts send humungous responses
4023 to HELO/EHLO, more than 1024 it seems.
4025 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4026 gives compiler warnings in some environments. I've now done it "properly",
4029 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4030 (because of problems with the built-in one) was declared to return uschar *
4031 instead of char *, causing compiler failure.
4033 25. Fixed a file descriptor leak when processing alias/forward files.
4035 26. Fixed a minor format string issue in dbfn.c.
4037 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4039 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4040 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4043 29. When a sender address is verified, it is cached, to save repeating the test
4044 when there is more than one recipient in a message. However, when the
4045 verification involves a callout, it is possible for different callout
4046 options to be set for different recipients. It is too complicated to keep
4047 track of this in the cache, so now Exim always runs a verification when a
4048 callout is required, relying on the callout cache for the optimization.
4049 The overhead is duplication of the address routing, but this should not be
4052 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4053 to be verified with callout=postmaster, and the main callout worked but the
4054 postmaster check failed, the verification correctly failed. However, if a
4055 subsequent RCPT command asked for sender verification *without* the
4056 postmaster check, incorrect caching caused this verification also to fail,
4059 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4060 it was not caching the DNS options (qualify_single, search_parents) that
4061 were used when the lookup failed. A subsequent lookup with different
4062 options therefore always gave the same answer, though there were cases
4063 where it should not have. (Example: a "domains = !$mx_any" option on a
4064 dnslookup router: the "domains" option is always processed without any
4065 widening, but the router might have qualify_single set.) Now Exim uses the
4066 cached value only when the same options are set.
4068 32. Added John Jetmore's "exipick" utility to the distribution.
4070 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4071 than a timeout (e.g. a certificate is required, and is not provided), an
4072 Exim server now closes the connection immediately. Previously it waited for
4073 the client to close - but if the client is SSL, it seems that they each
4074 wait for each other, leading to a delay before one of them times out.
4076 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4077 maintained 0.8.x compatibility because I don't think many are using it, and
4078 it is clearly obsolete.
4080 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4083 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4084 client certificate was expired. A simple patch fixes this, though I don't
4085 understand the full logic of why the verify callback is called multiple
4088 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4089 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4090 which causes problems with some clients (such as the Certicom SSL Plus
4091 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4092 disables the coutermeasure allowing Eudora to connect."
4094 38. Exim was not checking that a write() to a log file succeeded. This could
4095 lead to Bad Things if a log got too big, in particular if it hit a file
4096 size limit. Exim now panics and dies if it cannot write to a log file, just
4097 as it does if it cannot open a log file.
4099 39. Modified OS/Makefile-Linux so that it now contains
4101 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4103 The two -D definitions ensure that Exim is compiled with large file
4104 support, which makes it possible to handle log files that are bigger than
4107 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4108 instance) a domain was checked against a named list that involved a lookup,
4109 causing $domain_data to be set, then another domain was checked against the
4110 same list, then the first domain was re-checked, the value of $domain_data
4111 after the final check could be wrong. In particular, if the second check
4112 failed, it could be set empty. This bug probably also applied to
4115 41. The strip_trailing_dot option was not being applied to the address given
4116 with the -f command-line option.
4118 42. The code for reading a message's header from the spool was incrementing
4119 $received_count, but never initializing it. This meant that the value was
4120 incorrect (doubled) while delivering a message in the same process in which
4121 it was received. In the most common configuration of Exim, this never
4122 happens - a fresh exec is done - but it can happen when
4123 deliver_drop_privilege is set.
4125 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4126 it now includes up to 150 characters of the unexpected data in the log
4129 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4130 and building data strings. The size of both of these buffers was 10 000
4131 bytes - far larger than anybody would *ever* want, thought I. Needless to
4132 say, somebody hit the limit. I have increased the maximum line length to
4133 20 000 and the maximum data length of concatenated lines to 100 000. I have
4134 also fixed two bugs, because there was no checking on these buffers. Tsk,
4135 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4136 buffer is too small.
4138 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4139 lsearch lookups. Now it does.
4141 46. When parsing a route_list item in a manualroute router, a fixed-length
4142 buffer was used for the list of hosts. I made this 1024 bytes long,
4143 thinking that nobody would ever have a list of hosts that long. Wrong.
4144 Somebody had a whole pile of complicated expansion conditions, and the
4145 string was silently truncated, leading to an expansion error. It turns out
4146 that it is easier to change to an unlimited length (owing to other changes
4147 that have happened since this code was originally written) than to build
4148 structure for giving a limitation error. The length of the item that
4149 expands into the list of hosts is now unlimited.
4151 47. The lsearch lookup could not handle data where the length of text line was
4152 more than 4095 characters. Such lines were truncated, leading to shortened
4153 data being returned. It should now handle lines of any length.
4155 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4156 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4159 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4160 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4161 no longer happens in comment lines. A list of replacements is now placed
4162 at the head of all of the source files, except those whose only change is
4163 to replace PERL_COMMAND in the very first #! line.
4165 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4166 messages on the queue, with a bottom-up merge sort, using code contributed
4167 by Michael Haardt. This should make operations like -bp somewhat faster on
4168 large queues. It won't affect queue runners, except when queue_run_in_order
4171 51. Installed eximstats 1.31 in the distribution.
4173 52. Added support for SRV lookups to the dnslookup router.
4175 53. If an ACL referred to $message_body or $message_body_end, the value was not
4176 reset for any messages that followed in the same SMTP session.
4178 54. The store-handling optimization for building very long strings was not
4179 differentiating between the different store pools. I don't think this
4180 actually made any difference in practice, but I've tidied it.
4182 55. While running the routers to verify a sender address, $sender_address
4183 was still set to the sender address. This is wrong, because when routing to
4184 send a bounce to the sender, it would be empty. Therefore, I have changed
4185 it so that, while verifying a sender address, $sender_address is set to <>.
4186 (There is no change to what happens when verifying a recipient address.)
4188 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4189 target A or AAAA records (if not already returned) without resetting the
4190 qualify_single or search_parents options of the DNS resolver. These are
4191 inappropriate in this case because the targets of MX and SRV records must
4192 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4193 target that, when qualified, matched something in the local domain. These
4194 two options are now turned off when doing these lookups.
4196 57. It seems that at least some releases of Reiserfs (which does not have the
4197 concept of a fixed number of inodes) returns zero and not -1 for the
4198 number of available inodes. This interacted badly with check_spool_inodes,
4199 which assumed that -1 was the "no such thing" setting. What I have done is
4200 to check that the total number of inodes is greater than zero before doing
4201 the test of how many are available.
4203 58. When a "warn" ACL statement has a log_message modifier, the message is
4204 remembered, and not repeated. This is to avoid a lot of repetition when a
4205 message has many recipients that cause the same warning to be written.
4206 Howewer, Exim was preserving the list of already written lines for an
4207 entire SMTP session, which doesn't seem right. The memory is now reset if a
4208 new message is started.
4210 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4211 debugging output unless log_rewrite was also set.
4213 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4214 of (int)(handle) when we know that handle contains (void *)(-1).
4216 61. The Exim daemon panic-logs an error return when it closes the incoming
4217 connection. However "connection reset by peer" seems to be common, and
4218 isn't really an error worthy of noting specially, so that particular error
4221 62. When Exim is trying to find all the local interfaces, it used to panic and
4222 die if the ioctl to get the interface flags failed. However, it seems that
4223 on at least one OS (Solaris 9) it is possible to have an interface that is
4224 included in the list of interfaces, but for which you get a failure error
4225 for this call. This happens when the interface is not "plumbed" into a
4226 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4227 failure of the "get flags" call assumes that the interface is down.
4229 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4230 makes life easier for people who are doing arithmetic on fields extracted
4231 from dates, where you often get leading zeros that should not be
4232 interpreted as octal.
4234 64. Added qualify_domain to the redirect router, to override the global
4237 65. If a pathologically long header line contained very many addresses (the
4238 report of this problem mentioned 10 000) and each of them was rewritten,
4239 Exim could use up a very large amount of memory. (It kept on making new
4240 copies of the header line as it rewrote, and never released the old ones.)
4241 At the expense of a bit more processing, the header rewriting function has
4242 been changed so that it no longer eats memory in this way.
4244 66. The generation of the Received: header has been moved from the time that a
4245 message starts to be received, to the time that it finishes. The timestamp
4246 in the Received: header should now be very close to that of the <= log
4247 line. There are two side-effects of this change:
4249 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4250 logged header lines no longer include the local Received: line, because
4251 it has not yet been created. The same applies to a copy of the message
4252 that is returned to a non-SMTP sender when a message is rejected.
4254 (b) When a filter file is tested using -bf, no additional Received: header
4255 is added to the test message. After some thought, I decided that this
4258 This change does not affect the value of $received_for. It is still set
4259 after address rewriting, but before local_scan() is called.
4261 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4263 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4264 gave an unhelpful panic error message, and a defer error. I have managed to
4265 change this behaviour so that it now rejects any supplied certificate,
4266 which seems right, as the list of acceptable certificates is empty.
4268 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4269 gave an unhelpful defer error. I have not managed to make this reject any
4270 supplied certificates, but the error message it gives is "no certificate
4271 supplied", which is not helpful.
4273 70. exigrep's output now also includes lines that are not associated with any
4274 message, but which match the given pattern. Implemented by a patch from
4275 Martin Sluka, which also tidied up the Perl a bit.
4277 71. Recipient callout verification, like sender verification, was using <> in
4278 the MAIL FROM command. This isn't really the right thing, since the actual
4279 sender may affect whether the remote host accepts the recipient or not. I
4280 have changed it to use the actual sender in the callout; this means that
4281 the cache record is now keyed on a recipient/sender pair, not just the
4282 recipient address. There doesn't seem to be a real danger of callout loops,
4283 since a callout by the remote host to check the sender would use <>.
4284 [SEE ABOVE: changed after hitting problems.]
4286 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4287 temporary errors. However, in the case of such a code being given after
4288 the end of a data transmission (i.e. after ".") Exim was failing to write
4289 a retry record for the message. (Yes, there was some broken host that was
4290 actually sending 8xx at this point.)
4292 73. An unknown lookup type in a host list could cause Exim to panic-die when
4293 the list was checked. (An example that provoked this was putting <; in the
4294 middle of a list instead of at the start.) If this happened during a DATA
4295 ACL check, a -D file could be left lying around. This kind of configuration
4296 error no longer causes Exim to die; instead it causes a defer errror. The
4297 incident is still logged to the main and panic logs.
4299 74. Buglet left over from Exim 3 conversion. The message "too many messages
4300 in one connection" was written to the rejectlog but not the mainlog, except
4301 when address rewriting (yes!) was being logged.
4303 75. Added write_rejectlog option.
4305 76. When a system filter was run not as root (that is, when system_filter_user
4306 was set), the values of the $n variables were not being returned to the
4307 main process; thus, they were not subsequently available in the $sn
4310 77. Added +return_path_on_delivery log selector.
4312 78. A connection timeout was being treated differently from recipients deferred
4313 when testing hosts_max_try with a message that was older than the host's
4314 retry timeout. (The host should not be counted, thus allowing all hosts to
4315 be tried at least once before bouncing.) This may have been the cause of an
4316 occasionally reported bug whereby a message would remain on the queue
4317 longer than the retry timeout, but would be bounced if a delivery was
4318 forced. I say "may" because I never totally pinned down the problem;
4319 setting up timeout/retry tests is difficult. See also the next item.
4321 79. The ultimate address timeout was not being applied to errors that involved
4322 a combination of host plus message (for example, a timeout on a MAIL
4323 command). When an address resolved to a number of possible hosts, and they
4324 were not all tried for each delivery (e.g. because of hosts_max_try), a
4325 message could remain on the queue longer than the retry timeout.
4327 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4330 81. Fixed an obscure SMTP outgoing bug which required at least the following
4331 conditions: (a) there was another message waiting for the same server;
4332 (b) the server returned 5xx to all RCPT commands in the first message so
4333 that the message was not completed; (c) the server dropped the connection
4334 or gave a negative response to the RSET that Exim sends to abort the
4335 transaction. The observed case was a dropped connection after DATA that had
4336 been sent in pipelining mode. That is, the server had advertised PIPELINING
4337 but was not implementing it correctly. The effect of the bug was incorrect
4338 behaviour, such as trying another host, and this could lead to a crash.
4344 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4345 and daemon.c were passed as pointers to ints; they should have been
4346 pointers to socklen_t variables (which are typically unsigned ints).
4348 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4351 3. Fixed a really odd bug that affected only the testing scheme; patching a
4352 certain fixed string in the binary changed the value of another string that
4353 happened to be identical to the end of the original first string.
4355 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4356 name", it returns that address as the IP address. On some operating
4357 systems (e.g. Solaris), it also passes back the IP address string as the
4358 "host name". However, on others (e.g. Linux), it passes back an empty
4359 string. Exim wasn't checking for this, and was changing the host name to an
4360 empty string, assuming it had been canonicized.
4362 5. Although rare, it is permitted to have more than one PTR record for a given
4363 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4364 all the names associated with an address, because they do in Solaris.
4365 However, it seems that they do not in Linux for data that comes from the
4366 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4367 I found this out when I moved to a new Linux workstation and tried to run
4368 the Exim test suite.
4370 To get round this problem I have changed the code so that it now does its
4371 own call to the DNS to look up PTR records when searching for a host name.
4372 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4373 addresses that are only in /etc/hosts are still found.
4375 This behaviour is, however, controlled by an option called host_lookup_
4376 order, which defaults to "bydns:byaddr". If people want to use the other
4377 order, or indeed, just use one or the other means of lookup, they can
4378 specify it in this variable.
4380 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4381 some operating systems, this comes back from gethostbyaddr() as an empty
4382 string, and this is what Exim used to test for. However, it seems that in
4383 other systems, "." is yielded. Exim now tests for this case too.
4385 7. The values of check_spool_space and check_log_space are now held internally
4386 as a number of kilobytes instead of an absolute number of bytes. If a
4387 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4388 kilobyte. This means that much larger values can be stored.
4390 8. Exim monitor: an attempt to get the action menu when not actually pointing
4391 at a message produces an empty menu entitled "No message selected". This
4392 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4393 no entries in it ("Shell widget menu has zero width and/or height"). So I
4394 have added a single, blank menu entry in this case.
4396 9. Added ${quote_local_part.
4398 10. MIME decoding is now applied to the contents of Subject: header lines when
4401 11. Now that a reference to $sender_host_address automatically causes a reverse
4402 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4403 host lookup before query-style lookups in lists that might use this
4404 variable. This has therefore been abolished, and the "net-" prefix is no
4405 longer necessary for query-style lookups.
4407 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4408 have been a typo for LFLAGS, so it has been changed.
4410 13. The install script calls Exim with "-C /dev/null" in order to find the
4411 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4412 to be output. Howeve, since Exim outputs its version number before the
4413 error, it didn't break the script. It just looked ugly. I fixed this by
4414 always allowing "-C /dev/null" if the caller is root.
4416 14. Ignore overlarge ACL variable number when reading spool file - insurance
4417 against a later release with more variables having written the file.
4419 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4420 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4421 this, as well as the form without the "IPv6" on the front.
4423 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4424 OS/Makefile-Darwin file.
4426 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4427 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4429 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4430 variable. However, some OS use ints for inodes. I've added cast to long int
4431 to get rid of the compiler warning.
4433 19. I had forgotten to lock out "/../" in configuration file names when
4434 ALT_CONFIG_PREFIX was set.
4436 20. Routers used for verification do not need to specify transports. However,
4437 if such a router generated a host list, and callout was configured, Exim
4438 crashed, because it could not find a port number from the (non-existent)
4439 transport. It now assumes port 25 in this circumstance.
4441 21. Added the -t option to exigrep.
4443 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4444 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4445 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4446 them is set, LOOKUP_LSEARCH is forced.
4448 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4450 24. Added sender and host information to the "rejected by local_scan()" log
4451 line; previously there was no indication of these.
4453 25. Added .include_if_exists.
4455 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4456 Linux. It turns out that not all file systems support this. Apparently some
4457 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4458 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4459 not supported on the file descriptor, is now ignored when Exim is trying to
4460 sync a directory. This applies only to Linux.
4462 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4464 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4465 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4468 29. Added support for maildirsize files from supplied patch (modified a bit).
4470 30. The use of :fail: followed by an empty string could lead Exim to respond to
4471 sender verification failures with (e.g.):
4473 550 Verification failed for <xxx>
4474 550 Sender verify failed
4476 where the first response line was missing the '-' that indicates it is not
4477 the final line of the response.
4479 31. The loop for finding the name of the user that called Exim had a hardwired
4480 limit of 10; it now uses the value of finduser_retries, which is used for
4481 all other user lookups.
4483 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4486 33. Exim was neglecting to zero errno before one call of strtol() when
4487 expanding a string and expecting an integer value. On some systems this
4488 resulted in spurious "integer overflow" errors. Also, it was casting the
4489 result into an int without checking.
4491 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4492 did not work. The code looks as if it has *never* worked, though it appears
4493 to have been documented since at least releast 1.62. I have made it work.
4495 35. The "timeout_DNS" error in retry rules, also documented since at least
4496 1.62, also never worked. As it isn't clear exactly what this means, and
4497 clearly it isn't a major issue, I have abolished the feature by treating it
4498 as "timeout", and writing a warning to the main and panic logs.
4500 36. The display of retry rules for -brt wasn't always showing the error code
4503 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4504 timeout_connect_A, timeout_connect_MX.
4506 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4507 to the empty sender.
4509 39. The daemon was not analysing the content of -oX till after it had closed
4510 stderr and disconnected from the controlling terminal. This meant that any
4511 syntax errors were only noted on the panic log, and the return code from
4512 the command was 0. By re-arranging the code a little, I've made the
4513 decoding happen first, so such errors now appear on stderr, and the return
4514 code is 1. However, the actual setting up of the sockets still happens in
4515 the disconnected process, so errors there are still only recorded on the
4518 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4519 connections (as happens on some IP stacks) was logged at start up time as
4520 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4521 it from "IPv6 and IPv4", which means that two separate sockets are being
4524 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4525 says whether AF_INET or AF_INET6 was passed as an argument.
4527 42. Exiwhat output was messed up when time zones were included in log
4530 43. Exiwhat now gives more information about the daemon's listening ports,
4531 and whether -tls-on-connect was used.
4533 44. The "port" option of the smtp transport is now expanded.
4535 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4536 silently ignored. Now an error message is written to the main and panic
4539 46. There's a new ACL modifier called "logwrite" which writes to a log file
4540 as soon as it is encountered.
4542 47. Added $local_user_uid and $local_user_gid at routing time.
4544 48. Exim crashed when trying to verify a sender address that was being
4547 49. Exim was recognizing only a space character after ".include". It now also
4548 recognizes a tab character.
4550 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4551 extracting the relevant information from the specification. The man page no
4552 longer contains scrambled data for the -d option, and I've added a section
4553 at the front about calling Exim under different names.
4555 51. Added "extra_headers" argument to the "mail" command in filter files.
4557 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4560 53. Installed eximstats 1.29.
4562 54. Added transport_filter_timeout as a generic transport option.
4564 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4565 Cc: header lines. This was required by RFC 822, but it not required by RFC
4568 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4569 incoming messages that did not have them. Now it does so only if the
4570 message originates locally, that is, if there is no associated remote host
4571 address. When Resent- header lines are present, this applies to the Resent-
4572 lines rather than the non-Resent- lines.
4574 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4575 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4577 58. Messages for configuration errors now include the name of the main
4578 configuration file - useful now that there may be more than one file in a
4579 list (.included file names were always shown).
4581 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4582 for those rare installations that do not start the daemon as root or run it
4583 setuid root. I've cut out the call to initgroups() if the daemon is not
4586 60. The Exim user and group can now be bound into the binary as text strings
4587 that are looked up at the start of Exim's processing.
4589 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4591 62. Added $mailstore_basename variable.
4593 63. Installed patch to sieve.c from Michael Haardt.
4595 64. When Exim failed to open the panic log after failing to open the main log,
4596 the original message it was trying to log was written to stderr and debug
4597 output, but if they were not available (the usual case in production), it
4598 was lost. Now it is written to syslog before the two lines that record the
4599 failures to open the logs.
4601 65. Users' Exim filters run in subprocesses under the user's uid. It is
4602 possible for a "deliver" command or an alias in a "personal" command to
4603 provoke an address rewrite. If logging of address rewriting is configured,
4604 this fails because the process is not running as root or exim. There may be
4605 a better way of dealing with this, but for the moment (because 4.30 needs
4606 to be released), I have disabled address rewrite logging when running a
4607 filter in a non-root, non-exim process.
4613 1. The buildconfig auxiliary program wasn't quoting the value set for
4614 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4615 not defined. This bug was masked in 4.22 by the effect that was fixed in
4618 2. Some messages that were rejected after a message id was allocated were
4619 shown as "incomplete" by exigrep. It no longer does this for messages that
4620 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4622 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4623 have allow_domain_literals set, the ID did not get logged in the <= line.
4624 Domain literals are now always recognized in Message-ID: header lines.
4626 4. The first argument for a ${extract expansion item is the key name or field
4627 number. Leading and trailing spaces in this item were not being ignored,
4628 causing some misleading effects.
4630 5. When deliver_drop_privilege was set, single queue runner processes started
4631 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4632 same command in the process it spins off) were not dropping privilege.
4634 6. When the daemon running as "exim" started a queue runner, it always
4635 re-executed Exim in the spun-off process. This is a waste of effort when
4636 deliver_drop_privilege is set. The new process now just calls the
4637 queue-runner function directly.
4643 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4646 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4647 ignored. Though the use of -oP was forcing the writing of a pid file, it
4648 was always written to the default place.
4650 3. If the message "no IP address found for host xxxx" is generated during
4651 incoming verification, it is now followed by identification of the incoming
4652 connection (so you can more easily find what provoked it).
4654 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4656 5. Added some features to "harden" Exim a bit more against certain attacks:
4658 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4659 be put in Local/Makefile. This is like the never_users runtime option,
4660 but it cannot be overridden. The default setting is "root".
4662 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4663 prefix string with which any file named in a -C command line option
4666 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4667 is retained for -C and -D only if the caller of Exim is root. Without
4668 it, the exim user may also use -C and -D and retain privilege.
4670 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4671 command line option is disabled.
4673 6. Macro names set by the -D option must start with an upper case letter, just
4674 like macro names defined in the configuration file.
4676 7. Added "dereference=" facility to LDAP.
4678 8. Two instances of the typo "uknown" in the source files are fixed.
4680 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4681 the Configure-Makefile script screwed up while processing it.
4683 10. Incorporated PCRE 4.4.
4685 11. The SMTP synchronization check was not operating right at the start of an
4686 SMTP session. For example, it could not catch a HELO sent before the client
4687 waited for the greeting. There is now a check for outstanding input at the
4688 point when the greeting is written. Because of the duplex, asynchronous
4689 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4690 way, but not yet received, when the check is performed.
4692 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4693 on TCP/IP sockets, because this apparently causes some broken clients to
4696 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4697 unchanged) from the Cygwin maintainer.
4699 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4700 was supported instead of testing for mbx. Effectively a typo.
4702 15. The spa authenticator server code was not checking that the input it
4703 received was valid base64.
4705 16. The debug output line for the "set" modifier in ACLs was not showing the
4706 name of the variable that was being set.
4708 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4710 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4711 lookup on its own. Something else was needed to trigger the lookup. For
4712 example, a match in host_lookup or the need for a host name in a host list.
4713 Now, if $sender_host_name is referenced and the host name has not yet been
4714 looked up, a lookup is performed. If the lookup fails, the variable remains
4715 empty, and $host_lookup_failed is set to "1".
4717 19. Added "eqi" as a case-independent comparison operator.
4719 20. The saslauthd authentication condition could segfault if neither service
4720 nor realm was specified.
4722 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4723 error message that was logged was misleading, and incoming SMTP
4724 connections were dropped. The message is now more accurate, and temporary
4725 errors are given to SMTP connections.
4727 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4728 (except RSET) with a 421 error, until QUIT is received. However, it was
4729 failing to send a response to QUIT.
4731 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4732 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4733 ACL is now run after the helo_try_verify_hosts code.
4735 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4736 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4737 case-independent, so other case variants are also recognized). Apparently
4738 some systems use these upper case variants.
4740 25. If more than two messages were waiting for the same host, and a transport
4741 filter was specified for the transport, Exim sent two messages over the
4742 same TCP/IP connection, and then failed with "socket operation on non-
4743 socket" when it tried to send the third.
4745 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4747 27. The extern definition of crypt16() in expand.c was not being excluded when
4748 the OS had its own crypt16() function.
4750 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4751 as a preferred synonym for return_size_limit, both as an option and as an
4754 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4756 30. Changed the default configuration ACL to relax the local part checking rule
4757 for addresses that are not in any local domains. For these addresses,
4758 slashes and pipe symbols are allowed within local parts, but the sequence
4759 /../ is explicitly forbidden.
4761 31. SPA server authentication was not clearing the challenge buffer before
4764 32. log_message in a "warn" ACL statement was writing to the reject log as
4765 well as to the main log, which contradicts the documentation and doesn't
4766 seem right (because no rejection is happening). So I have stopped it.
4768 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4769 However, I am unable to do any testing of this.
4771 34. Fixed an infelicity in the appendfile transport. When checking directories
4772 for a mailbox, to see if any needed to be created, it was accidentally
4773 using path names with one or more superfluous leading slashes; tracing
4774 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4776 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4777 contents are added to the log line that is written for every discarded
4778 recipient. (Previously a log_message setting was ignored.)
4780 36. The ${quote: operator now quotes the string if it is empty.
4782 37. The install script runs exim in order to find its version number. If for
4783 some reason other than non-existence or emptiness, which it checks, it
4784 could not run './exim', it was installing it with an empty version number,
4785 i.e. as "exim-". This error state is now caught, and the installation is
4788 38. An argument was missing from the function that creates an error message
4789 when Exim fails to connect to the socket for saslauthd authentication.
4790 This could cause Exim to crash, or give a corrupted message.
4792 39. Added isip, isip4, and isip6 to ${if conditions.
4794 40. The ACL variables $acl_xx are now saved with the message, and can be
4795 accessed later in routers, transports, and filters.
4797 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4798 strings in the file are not string-expanded.
4800 42. If a MAIL command specified a SIZE value that was too large to fit into an
4801 int variable, the check against message_size_limit failed. Such values are
4802 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4803 day this will have to be increased, but I don't think I want to be around
4804 when emails are that large.
4811 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4812 iconv() is not standard in FreeBSD.
4814 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4815 IPv6 enabled. The observed symptom was a segmentation fault on return from
4816 the function os_common_find_running_interfaces() in src/os.c.
4818 3. In the check_special_case() function in daemon.c I had used "errno" as an
4819 argument name, which causes warnings on some systems. This was basically a
4820 typo, since it was named "eno" in the comments!
4822 4. The code that waits for the clock to tick (at a resolution of some fraction
4823 of a second) so as to ensure message-id uniqueness was always waiting for
4824 at least one whole tick, when it could have waited for less. [This is
4825 almost certainly not relevant at current processor speeds, where it is
4826 unlikely to ever wait at all. But we try to future-proof.]
4828 5. The function that sleeps for a time interval that includes fractions of a
4829 second contained a race. It did not block SIGALRM between setting the
4830 timer, and suspending (a couple of lines later). If the interval was short
4831 and the sigsuspend() was delayed until after it had expired, the suspension
4832 never ended. On busy systems this could lead to processes getting stuck for
4835 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4836 process, before it forks any delivery processes. The open lookup caching
4837 mechanism meant that the open file or database connection was passed into
4838 the delivery process. The problem was that delivery processes always tidy
4839 up cached lookup data. This could cause a problem for the next delivery
4840 process started by the queue runner, because the external queue runner
4841 process does not know about the closure. So the next delivery process
4842 still has data in the lookup cache. In the case of a file lookup, there was
4843 no problem because closing a file descriptor in a subprocess doesn't affect
4844 the parent. However, if the lookup was caching a connection to a database,
4845 the connection was closed, and the second delivery process was likely to
4846 see errors such as "PGSQL: query failed: server closed the connection
4847 unexpectedly". The problem has been fixed by closing all cached lookups
4848 in a queue runner before running a delivery process.
4850 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4851 seem to have the "const" qualifier which it has on other OS. I've
4854 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4855 *of the same type* (client or server) with the same public name that an
4856 error should be diagnosed.
4858 9. When Exim looked up a host name for an IP address, but failed to find the
4859 original IP address when looking up the host name (a safety check), it
4860 output the message "<ip address> does not match any IP for NULL", which was
4861 confusing, to say the least. The bug was that the host name should have
4862 appeared instead of "NULL".
4864 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4865 user that is built into the binary, and the -C or -D options is used, root
4866 privilege is dropped before the configuration file is read. In addition,
4867 logging is switched to stderr instead of the normal log files. If the
4868 configuration then re-defines the Exim user, the unprivileged environment
4869 is probably not what is expected, so Exim logs a panic warning message (but
4872 However, if deliver_drop_privilege is set, the unprivileged state may well
4873 be exactly what is intended, so the warning has been cut out in that case,
4874 and Exim is allowed to try to write to its normal log files.
4880 1. smtp_return_error_details was not giving details for temporary sender
4881 or receiver verification errors.
4883 2. Diagnose a configuration error if two authenticators have the same public
4886 3. Exim used not to create the message log file for a message until the first
4887 delivery attempt. This could be confusing when incoming messages were held
4888 for policy or load reasons. The message log file is now created at the time
4889 the message is received, and an initial "Received" line is written to it.
4891 4. The automatically generated man page for command line options had a minor
4892 bug that caused no ill effects; however, a more serious problem was that
4893 the procedure for building the man page automatically didn't always
4894 operate. Consequently, release 4.20 contains an out-of-date version. This
4895 shouldn't happen again.
4897 5. When building Exim with embedded Perl support, the script that builds the
4898 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4899 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4901 6. The freeze_tell option was not being used for messages that were frozen on
4902 arrival, either by an ACL or by local_scan().
4904 7. Added the smtp_incomplete_transaction log selector.
4906 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4907 was accepting AUTH without a new EHLO.
4909 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4910 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4913 10. Logging of TCP/IP connections (when configured) now happens in the main
4914 daemon process instead of the child process, so that the TCP/IP connection
4915 count is more accurate (but it can never be perfect).
4917 11. The use of "drop" in a nested ACL was not being handled correctly in the
4918 outer ACL. Now, if condition failure induced by the nested "drop" causes
4919 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4920 or "require"), the connection is dropped.
4922 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4923 that yield "discard" can now be used with an "accept" or a "discard" verb,
4924 but an error is generated for any others (because I can't see a useful way
4925 to define what should happen).
4927 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4928 are now processed in the same way as lines in the Exim configuration file.
4929 In particular, continuation lines are supported.
4931 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4933 15. Added -ti meaning -t -i.
4935 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4936 domains, and warn by logging if others are found.
4938 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4939 returned from GIFCONF when Exim is trying to find the list of interfaces on
4940 a host. The code in os.c has been modified to copy each ifreq to an aligned
4941 structure in all cases.
4943 Also, in some cases, the returned ifreq's were being copied to a 'struct
4944 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4945 means the last couple of bytes of an IPv6 address could be chopped if the
4946 ifreq contained only a normal sockaddr (14 bytes storage).
4948 18. Named domain lists were not supported in the hosts_treat_as_local option.
4949 An entry such as +xxxx was not recognized, and was treated as a literal
4952 19. Ensure that header lines added by a DATA ACL are included in the reject log
4953 if the ACL subsequently rejects the message.
4955 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4956 MD5 (which is deprecated).
4958 21. When testing a filter file using -bf, Exim was writing a message when it
4959 took the sender from a "From " line in the message, but it was not doing so
4960 when it took $return_path from a Return-Path: header line. It now does.
4962 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4963 with a valid header line field name (a series of printing characters
4964 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4966 23. Changed "disc" in the source to "disk" to conform to the documentation and
4967 the book and for uniformity.
4969 24. Ignore Sendmail's -Ooption=value command line item.
4971 25. When execve() failed while trying to run a command in a pipe transport,
4972 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4973 could be confused with a return value of 69 from the command itself. This
4974 has been changed to 127, the value the shell returns if it is asked to run
4975 a non-existent command. The wording for the related log line suggests a
4976 non-existent command as the problem.
4978 26. If received_header_text expands to an empty string, do not add a Received:
4979 header line to the message. (Well, it adds a token one on the spool, but
4980 marks it "old" so that it doesn't get used or transmitted.)
4982 27. Installed eximstats 1.28 (addition of -nt option).
4984 28. There was no check for failure on the call to getsockname() in the daemon
4985 code. This can fail if there is a shortage of resources on the system, with
4986 ENOMEM, for example. A temporary error is now given on failure.
4988 29. Contrary to the C standard, it seems that in some environments, the
4989 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4990 program. Exim now does this explicitly; it affects the formatting of
4991 timestamps using strftime().
4993 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4994 complaints. I've now initialized all the variables, to avoid this.
4996 32. Header lines added by a system filter were not being "seen" during
4997 transport-time rewrites.
4999 33. The info_callback() function passed to OpenSSL is set up with type void
5000 (*)(SSL *, int, int), as described somewhere. However, when calling the
5001 function (actually a macro) that sets it up, the type void(*)() is
5002 expected. I've put in a cast to prevent warnings from picky compilers.
5004 34. If a DNS black list lookup found a CNAME record, but there were no A
5005 records associated with the domain it pointed at, Exim crashed.
5007 35. If a DNS black list lookup returned more than one A record, Exim ignored
5008 all but the first. It now scans all returned addresses if a particular IP
5009 value is being sought. In this situation, the contents of the
5010 $dnslist_value variable are a list of all the addresses, separated by a
5013 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5014 to accept a host name and all its aliases if the forward lookup for any of
5015 them yielded the IP address of the incoming connection. Now it accepts only
5016 those names whose forward lookup yields the correct IP address. Any other
5017 names are discarded. This closes a loophole whereby a rogue DNS
5018 administrator could create reverse DNS records to break through a
5019 wildcarded host restriction in an ACL.
5021 37. If a user filter or a system filter that ran in a subprocess used any of
5022 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5023 the wrong values were passed to the pipe command ($thisaddress had the
5024 value of $0, $0 had the value of $1, etc). This bug was introduced by
5025 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5027 38. Improved the line breaking for long SMTP error messages from ACLs.
5028 Previously, if there was no break point between 40 and 75 characters, Exim
5029 left the rest of the message alone. Two changes have been made: (a) I've
5030 reduced the minimum length to 35 characters; (b) if it can't find a break
5031 point between 35 and 75 characters, it looks ahead and uses the first one
5032 that it finds. This may give the occasional overlong line, but at least the
5033 remaining text gets split now.
5035 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5036 file descriptors might be low, and that setting 1000 would always raise it.
5037 It turns out that in some environments, the limit is already over 1000 and
5038 that lowering it causes trouble. So now Exim takes care not to decrease it.
5040 40. When delivering a message, the value of $return_path is set to $sender_
5041 address at the start of routing (routers may change the value). By an
5042 oversight, this default was not being set up when an address was tested by
5043 -bt or -bv, which affected the outcome if any router or filter referred to
5046 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5047 log only when "message" or "log_message" are set. However, if one of the
5048 conditions was an address verification, or a call to a nested ACL, the
5049 messages generated by the underlying test were being passed through. This
5050 no longer happens. The underlying message is available in $acl_verify_
5051 message for both "message" and "log_message" expansions, so it can be
5052 passed through if needed.
5054 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5055 new expansion $bh_ to give the encoded byte string without charset
5056 translation. Translation happens only if iconv() is available; HAVE_ICONV
5057 indicates this at build time. HEADERS_CHARSET gives the charset to
5058 translate to; headers_charset can change it in the configuration, and
5059 "headers charset" can change it in an individual filter file.
5061 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5062 that creates RFC 2047 encoded "words" labels them as that charset instead
5063 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5064 expansion operator; (ii) when Exim creates a From: line for a local
5065 message; (iii) when a header line is rewritten to include a "phrase" part.
5067 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5068 buggy, causing it to skip the first lines of messages whose message ID
5069 ended in 'D'. This would not have bitten before Exim release 4.14, because
5070 message IDs were unlikely to end in 'D' before then. The effect was to have
5071 incorrect size information for certain domains.
5073 45. #include "config.h" was missing at the start of the crypt16.c module. This
5074 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5077 46. If there was a timeout during a "random" callout check, Exim treated it as
5078 a failure of the random address, and carried on sending RSET and the real
5079 address. If the delay was just some slowness somewhere, the response to the
5080 original RCPT would be taken as a response to RSET and so on, causing
5081 mayhem of various kinds.
5083 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5084 when I implemented it. It didn't allow for the fact that some option values
5085 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5086 the right test for positive values.
5088 48. Domain names in DNS records are case-independent. Exim always looks them up
5089 in lower case. Some resolvers return domain names in exactly the case they
5090 appear in the zone file, that is, they may contain uppercase letters. Not
5091 all resolvers do this - some return always lower case. Exim was treating a
5092 change of case by a resolver as a change of domain, similar to a widening
5093 of a domain abbreviation. This triggered its re-routing code and so it was
5094 trying to route what was effectively the same domain again. This normally
5095 caused routing to fail (because the router wouldn't handle the domain
5096 twice). Now Exim checks for this case specially, and just changes the
5097 casing of the domain that it ultimately uses when it transmits the message
5100 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5103 50. If a filter generated a file delivery with a non-absolute name (possible if
5104 no home directory exists for the router), the forbid_file option was not
5107 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5108 the existing equality matching.
5110 52. Exim was using ints instead of ino_t variables in some places where it was
5111 dealing with inode numbers.
5113 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5114 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5115 called TMPDIR, and if it finds it is different, it changes its value.
5117 54. The smtp_printf() function is now made available to local_scan() so
5118 additional output lines can be written before returning. There is also an
5119 smtp_fflush() function to enable the detection of a dropped connection.
5120 The variables smtp_input and smtp_batched_input are exported to
5123 55. Changed the default runtime configuration: the message "Unknown user"
5124 has been removed from the ACL, and instead placed on the localuser router,
5125 using the cannot_route_message feature. This means that any verification
5126 failures that generate their own messages won't get overridden. Similarly,
5127 the "Unrouteable address" message that was in the ACL for unverifiable
5128 relay addresses has also been removed.
5130 56. Added hosts_avoid_esmtp to the smtp transport.
5132 57. The exicyclog script was not checking for the esoteric option
5133 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5134 will work only if exicyclog is run under the appropriate euid.
5136 58. Following a discussion on the list, the rules by which Exim recognises line
5137 endings on incoming messages have been changed. The -dropcr and drop_cr
5138 options are now no-ops, retained only for backwards compatibility. The
5139 following line terminators are recognized: LF CRLF CR. However, special
5140 processing applies to CR:
5142 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5143 nor a local message in the state where . is a terminator.
5145 (ii) If a bare CR is encountered in a header line, an extra space is added
5146 after the line terminator so as not to end the header. The reasoning
5147 behind this is that bare CRs in header lines are most likely either
5148 to be mistakes, or people trying to play silly games.
5150 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5151 was being incorrectly given as 18 bytes larger than it should have been.
5152 This is a VOB (very old bug).
5154 60. This may never have affected anything current, but just in case it has:
5155 When the local host is found other than at the start of a list of hosts,
5156 the local host, those with the same MX, and any that follow, are discarded.
5157 When the list in question was part of a longer list of hosts, the following
5158 hosts (not currently being processed) were also being discarded. This no
5159 longer happens. I'm not sure if this situation could ever has previously
5162 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5165 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5166 header. This is something that is recommended in a new Internet Draft, and
5167 is something that is documented as being done by Sendmail. There are two
5168 possible values. For messages generated by the autoreply transport, Exim
5171 Auto-Submitted: auto-replied
5173 whereas for all other generated messages (e.g. bounces) it adds
5175 Auto-Submitted: auto-generated
5177 63. The "personal" condition in filters now includes a test for the
5178 Auto-Submitted: header. If it contains the string "auto-" the message it
5179 not considered personal.
5181 64. Added rcpt_include_affixes as a generic transport option.
5183 65. Added queue_only_override (default true).
5185 66. Added the syslog_duplication option.
5187 67. If what should have been the first header line of a message consisted of
5188 a space followed by a colon, Exim was mis-interpreting it as a header line.
5189 It isn't of course - it is syntactically invalid and should therefore be
5190 treated as the start of the message body. The misbehaviour could have
5191 caused a number of strange effects, including loss of data in subsequent
5192 header lines, and spool format errors.
5194 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5195 client host had authenticated. This control can now be exercised by an ACL
5196 for more flexibility.
5198 69. By default, callouts do not happen when testing with -bh. There is now a
5199 variant, -bhc, which does actually run the callout code, including
5200 consulting and updating the callout cache.
5202 70. Added support for saslauthd authentication, courtesy of Alexander
5205 71. If statvfs() failed on the spool or log directories while checking their
5206 size for availability, Exim confusingly gave the error "space shortage".
5207 Furthermore, in debugging mode it crashed with a floating point exception.
5208 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5209 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5210 serious problem, Exim now writes to the main and panic logs when this
5211 happens, with details of the failure. It then refuses to accept the
5212 incoming message, giving the message "spool directory problem" or "log
5213 directory problem" with a 421 code for SMTP messages.
5215 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5216 0, 1, and 2 exist, because some OS complain for execs without them (see
5217 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5218 descriptors. However, the code omitted to check that the open succeeded,
5219 causing mysterious errors if for some reason the permissions on /dev/null
5220 got screwed. Now Exim writes a message to the main and panic logs, and
5221 bombs out if it can't open /dev/null.
5223 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5224 interact so that it is all more flexible. It is supposed to remain
5225 backwards compatible. Also added extra_local_interfaces.
5227 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5228 to bomb out with an assertion failure - to the client this appears as a
5229 connection drop. This problem occurs in the part of the code that was taken
5230 from the Samba project. Fortunately, the assertion is in a very simple
5231 function, so I have fixed this by reproducing the function inline in the
5232 one place where it is called, and arranging for authentication to fail
5233 instead of killing the process with assert().
5235 75. The SPA client code was not working when the server requested OEM rather
5236 than Unicode encoding.
5238 76. Added code to make require_files with a specific uid setting more usable in
5239 the case where statting the file as root fails - usually a non-root-mounted
5240 NFS file system. When this happens and the failure is EACCES, Exim now
5241 forks a subprocess and does the per-uid checking as the relevant uid.
5243 77. Added process_log_path.
5245 78. If log_file_path was not explicitly set, a setting of check_log_space or
5246 check_log_inodes was ignored.
5248 79. If a space check for the spool or log partitions fails, the incident is now
5249 logged. Of course, in the latter case the data may get lost...
5251 80. Added the %p formatting code to string_format() so that it can be used to
5252 print addresses in debug_print(). Adjusted all the address printing in the
5253 debugging in store.c to use %p rather than %d.
5255 81. There was a concern that a line of code in smtp_in.c could overflow a
5256 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5257 initially expressed, the concern was not well-founded, because trailing
5258 spaces are removed early. However, if the trailing spaces were followed by
5259 a NULL, they did not get removed, so the overflow was possible. Two fixes
5262 (a) I re-wrote the offending code in a cleaner fashion.
5263 (b) If an incoming SMTP command contains a NULL character, it is rejected
5266 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5267 runs initgroups(), so that if the Exim user is in any additional groups,
5268 they will be used during message reception.
5274 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob