1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependancy in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
44 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
45 Patch from Jakob Hirsch.
47 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
48 SQL string expansion failure details.
49 Patch from Andrey Oktyabrski.
51 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
52 Patch from Simon Arlott.
54 PP/09 New lookups functionality failed to compile on old gcc which rejects
55 extern declarations in function scope.
56 Patch from Oliver Fleischmann
58 PP/10 Use sig_atomic_t for flags set from signal handlers.
59 Check getgroups() return and improve debugging.
60 Fixed developed for diagnosis in bug 927 (which turned out to be
67 TF/01 Failure to get a lock on a hints database can have serious
68 consequences so log it to the panic log.
70 TF/02 Log LMTP confirmation messages in the same way as SMTP,
71 controlled using the smtp_confirmation log selector.
73 TF/03 Include the error message when we fail to unlink a spool file.
75 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
76 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
77 for maintaining out-of-tree patches for some time.
79 PP/01 Bugzilla 139: Documentation and portability issues.
80 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
81 Handle per-OS dynamic-module compilation flags.
83 PP/02 Let /dev/null have normal permissions.
84 The 4.73 fixes were a little too stringent and complained about the
85 permissions on /dev/null. Exempt it from some checks.
86 Reported by Andreas M. Kirchwitz.
88 PP/03 Report version information for many libraries, including
89 Exim version information for dynamically loaded libraries. Created
90 version.h, now support a version extension string for distributors
91 who patch heavily. Dynamic module ABI change.
93 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
94 privilege escalation vulnerability whereby the Exim run-time user
95 can cause root to append content of the attacker's choosing to
98 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
101 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
102 If dropping privileges for untrusted macros, we disabled normal logging
103 on the basis that it would fail; for the Exim run-time user, this is not
104 the case, and it resulted in successful deliveries going unlogged.
105 Fixed. Reported by Andreas Metzler.
111 PP/01 Date: & Message-Id: revert to normally being appended to a message,
112 only prepend for the Resent-* case. Fixes regression introduced in
113 Exim 4.70 by NM/22 for Bugzilla 607.
115 PP/02 Include check_rfc2047_length in configure.default because we're seeing
116 increasing numbers of administrators be bitten by this.
118 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
120 PP/03 Bugzilla 994: added openssl_options main configuration option.
122 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
124 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
126 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
128 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
129 without a peer certificate, leading to a segfault because of an
130 assumption that peers always have certificates. Be a little more
131 paranoid. Problem reported by Martin Tscholak.
133 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
134 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
135 NB: ClamAV planning to remove STREAM in "middle of 2010".
136 CL also introduces -bmalware, various -d+acl logging additions and
137 more caution in buffer sizes.
139 PP/09 Implemented reverse_ip expansion operator.
141 PP/10 Bugzilla 937: provide a "debug" ACL control.
143 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
145 PP/12 Bugzilla 973: Implement --version.
147 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
149 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
151 PP/15 Bugzilla 816: support multiple condition rules on Routers.
153 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
154 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
155 ignore trailing whitespace.
157 JJ/02 prevent non-panic DKIM error from being sent to paniclog
159 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
162 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
163 Notification from Dr Andrew Aitchison.
165 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
166 ExtendedDetectionInfo response format.
167 Notification from John Horne.
169 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
172 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
173 XSL and documented dependency on system catalogs, with examples of how
176 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
179 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
180 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
181 configuration file which is writeable by the Exim user or group.
183 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
184 of configuration files to cover files specified with the -C option if
185 they are going to be used with root privileges, not just the default
188 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
189 option (effectively making it always true).
191 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
192 files to be used while preserving root privileges.
194 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
195 that rogue child processes cannot use them.
197 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
198 run-time user, instead of root.
200 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
201 Exim run-time user without dropping privileges.
203 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
204 result string, instead of calling string_vformat() twice with the same
207 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
208 for other users. Others should always drop root privileges if they use
209 -C on the command line, even for a whitelisted configure file.
211 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
213 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
219 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
220 $data_path, and $header_path variables; fixed documentation bugs and
223 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
224 exipick to access non-standard spools, including the "frozen" queue
227 NM/01 Bugzilla 965: Support mysql stored procedures.
228 Patch from Alain Williams
230 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
232 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
233 Patch from Andreas Metzler
235 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
236 Patch from Kirill Miazine
238 NM/05 Bugzilla 671: Added umask to procmail example.
240 JJ/03 installed exipick 20100323.0, fixing doc bug
242 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
243 directory. Notification and patch from Dan Rosenberg.
245 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
247 TK/02 Improve log output when DKIM signing operation fails.
249 MH/01 Treat the transport option dkim_domain as a colon separated
250 list, not as a single string, and sign the message with each element,
251 omitting multiple occurences of the same signer.
253 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
254 Bugzilla 985, 986. Patch by Simon Arlott
256 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
257 Patch by Simon Arlott
259 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
260 MBX locking. Notification from Dan Rosenberg.
266 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
268 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
270 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
272 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
274 NM/04 Bugzilla 915: Flag broken perl installation during build.
280 TK/01 Added patch by Johannes Berg that expands the main option
281 "spamd_address" if it starts with a dollar sign.
283 TK/02 Write list of recipients to X-Envelope-Sender header when building
284 the mbox-format spool file for content scanning (suggested by Jakob
287 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
288 (http://www.dcc-servers.net/) support via dccifd. Activated by
289 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
291 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
292 by Mark Daniel Reidel <mr@df.eu>.
294 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
295 When building exim an external PCRE library is now needed -
296 PCRE is a system library on the majority of modern systems.
297 See entry on PCRE_LIBS in EDITME file.
299 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
300 conversation. Added nologin parameter to request.
301 Patch contributed by Kirill Miazine.
303 TF/01 Do not log submission mode rewrites if they do not change the address.
305 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
307 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
308 log files in place. Contributed by Roberto Lima.
310 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
312 TF/03 Bugzilla 615: When checking the local_parts router precondition
313 after a local_part_suffix or local_part_prefix option, Exim now
314 does not use the address's named list lookup cache, since this
315 contains cached lookups for the whole local part.
317 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
318 Robert Millan. Documentation is in experimental-spec.txt.
320 TF/04 Bugzilla 668: Fix parallel build (make -j).
322 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
324 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
325 Patch provided by Jan Srzednicki.
327 TF/05 Leading white space used to be stripped from $spam_report which
328 wrecked the formatting. Now it is preserved.
330 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
331 that they are available at delivery time.
333 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
335 TF/08 TLS error reporting now respects the incoming_interface and
336 incoming_port log selectors.
338 TF/09 Produce a more useful error message if an SMTP transport's hosts
339 setting expands to an empty string.
341 NM/06 Bugzilla 744: EXPN did not work under TLS.
342 Patch provided by Phil Pennock.
344 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
345 Patch provided by Richard Godbee.
347 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
348 acl_smtp_notquit, added index entry.
350 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
351 Patch provided by Eugene Bujak.
353 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
354 accept(). Patch provided by Maxim Dounin.
356 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
357 Patch provided by Phil Pennock.
359 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
361 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
362 Patch provided by Brad "anomie" Jorsch.
364 NM/14 Bugzilla 622: Added timeout setting to transport filter.
365 Patch provided by Dean Brooks.
367 TK/05 Add native DKIM support (does not depend on external libraries).
369 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
370 Patch provided by Graeme Fowler.
372 NM/16 Bugzilla 851: Documentation example syntax fix.
374 NM/17 Changed NOTICE file to remove references to embedded PCRE.
376 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
379 NM/19 Bugzilla 745: TLS version reporting.
380 Patch provided by Phil Pennock.
382 NM/20 Bugzilla 167: bool: condition support.
383 Patch provided by Phil Pennock.
385 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
386 clients. Patch provided by Phil Pennock.
388 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
389 Patch provided by Brad "anomie" Jorsch.
391 NM/23 Bugzilla 687: Fix misparses in eximstats.
392 Patch provided by Heiko Schlittermann.
394 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
395 Patch provided by Heiko Schlittermann.
397 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
398 plus update to original patch.
400 NM/26 Bugzilla 799: Documentation correction for ratelimit.
402 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
403 Patch provided by David Brownlee.
405 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
407 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
409 NM/30 Bugzilla 888: TLS documentation bugfixes.
411 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
413 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
414 Unlike the original bugzilla I have changed all shell scripts in src tree.
416 NM/33 Bugzilla 898: Transport filter timeout fix.
417 Patch by Todd Rinaldo.
419 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
420 Patch by Serge Demonchaux.
422 NM/35 Bugzilla 39: Base64 decode bug fixes.
423 Patch by Jakob Hirsch.
425 NM/36 Bugzilla 909: Correct connect() call in dcc code.
427 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
429 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
431 NM/39 Bugzilla 911: Fixed MakeLinks build script.
437 TK/01 Add preliminary DKIM support. Currently requires a forked version of
438 ALT-N's libdkim that I have put here:
439 http://duncanthrax.net/exim-experimental/
441 Note to Michael Haardt: I had to rename some vars in sieve.c. They
442 were called 'true' and it seems that C99 defines that as a reserved
443 keyword to be used with 'bool' variable types. That means you could
444 not include C99-style headers which use bools without triggering
445 build errors in sieve.c.
447 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
448 as mailq or other aliases. Changed the --help handling significantly
449 to do whats expected. exim_usage() emits usage/help information.
451 SC/01 Added the -bylocaldomain option to eximstats.
453 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
455 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
457 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
463 PH/01 Another patch from the Sieve maintainer.
465 PH/02 When an IPv6 address is converted to a string for single-key lookup
466 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
467 dots are used instead of colons so that keys in lsearch files need not
468 contain colons. This was done some time before quoting was made available
469 in lsearch files. However, iplsearch files do require colons in IPv6 keys
470 (notated using the quote facility) so as to distinguish them from IPv4
471 keys. This meant that lookups for IP addresses in host lists did not work
472 for iplsearch lookups.
474 This has been fixed by arranging for IPv6 addresses to be expressed with
475 colons if the lookup type is iplsearch. This is not incompatible, because
476 previously such lookups could never work.
478 The situation is now rather anomolous, since one *can* have colons in
479 ordinary lsearch keys. However, making the change in all cases is
480 incompatible and would probably break a number of configurations.
482 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
485 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
486 conversion specification without a maximum field width, thereby enabling
487 a rogue spamd server to cause a buffer overflow. While nobody in their
488 right mind would setup Exim to query an untrusted spamd server, an
489 attacker that gains access to a server running spamd could potentially
490 exploit this vulnerability to run arbitrary code as the Exim user.
492 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
493 $primary_hostname instead of what libspf2 thinks the hosts name is.
495 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
496 a directory entry by the name of the lookup key. Previously, if a
497 symlink pointed to a non-existing file or a file in a directory that
498 Exim lacked permissions to read, a lookup for a key matching that
499 symlink would fail. Now it is enough that a matching directory entry
500 exists, symlink or not. (Bugzilla 503.)
502 PH/03 The body_linecount and body_zerocount variables are now exported in the
505 PH/04 Added the $dnslist_matched variable.
507 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
508 This means they are set thereafter only if the connection becomes
511 PH/06 Added the client_condition to authenticators so that some can be skipped
512 by clients under certain conditions.
514 PH/07 The error message for a badly-placed control=no_multiline_responses left
515 "_responses" off the end of the name.
517 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
519 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
520 (without spaces) instead of just copying the configuration text.
522 PH/10 Added the /noupdate option to the ratelimit ACL condition.
524 PH/11 Added $max_received_linelength.
526 PH/12 Added +ignore_defer and +include_defer to host lists.
528 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
529 way in which PCRE > 7.0 is built.
531 PH/14 Implemented queue_only_load_latch.
533 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
534 MAIL command. The effect was to mangle the value on 64-bit systems.
536 PH/16 Another patch from the Sieve maintainer.
538 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
540 PH/18 If a system quota error occurred while trying to create the file for
541 a maildir delivery, the message "Mailbox is full" was not appended to the
542 bounce if the delivery eventually timed out. Change 4.67/27 below applied
543 only to a quota excession during the actual writing of the file.
545 PH/19 It seems that peer DN values may contain newlines (and other non-printing
546 characters?) which causes problems in log lines. The DN values are now
547 passed through string_printing() before being added to log lines.
549 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
550 and InterBase are left for another time.)
552 PH/21 Added message_body_newlines option.
554 PH/22 Guard against possible overflow in moan_check_errorcopy().
556 PH/23 POSIX allows open() to be a macro; guard against that.
558 PH/24 If the recipient of an error message contained an @ in the local part
559 (suitably quoted, of course), incorrect values were put in $domain and
560 $local_part during the evaluation of errors_copy.
566 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
567 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
570 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
571 issue a MAIL command.
573 PH/02 In an ACL statement such as
575 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
577 if a client was not listed at all, or was listed with a value other than
578 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
579 the condition was not true (as it should be), so access was not denied.
580 The bug was that the ! inversion was incorrectly passed on to the second
581 item. This has been fixed.
583 PH/03 Added additional dnslists conditions == and =& which are different from
584 = and & when the dns lookup returns more than one IP address.
586 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
587 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
589 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
590 FSYNC, which compiles an option called disable_fsync that allows for
591 bypassing fsync(). The documentation is heavily laced with warnings.
593 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
595 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
596 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
597 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
598 including adding "make clean"; (3) Added -fPIC when compiling the test
599 dynamically loaded module, to get rid of a warning.
601 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
602 message fails, move_frozen_messages = true and ignore_bounce_errors_after
603 = 0s. The bug is otherwise harmless.
605 PH/07 There was a bug in the dovecot authenticator such that the value of
606 $auth1 could be overwritten, and so not correctly preserved, after a
607 successful authentication. This usually meant that the value preserved by
608 the server_setid option was incorrect.
610 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
612 PH/09 Installed PCRE release 7.0.
614 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
615 run for batched SMTP input. It is now run at the start of every message
616 in the batch. While fixing this I discovered that the process information
617 (output by running exiwhat) was not always getting set for -bs and -bS
618 input. This is fixed, and it now also says "batched" for BSMTP.
620 PH/11 Added control=no_pipelining.
622 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
623 patch, slightly modified), and move the expansion of helo_data till after
624 the connection is made in the smtp transport (so it can use these
627 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
629 PH/14 Added log_selector = +pid.
631 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
633 PH/16 Add ${if forany and ${if forall.
635 PH/17 Added dsn_from option to vary the From: line in DSNs.
637 PH/18 Flush SMTP output before performing a callout, unless control =
638 no_callout_flush is set.
640 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
641 was true (the default) a successful delivery failed to delete the retry
642 item, thus causing premature timeout of the address. The bug is now
645 PH/20 Added hosts_avoid_pipelining to the smtp transport.
647 PH/21 Long custom messages for fakedefer and fakereject are now split up
648 into multiline reponses in the same way that messages for "deny" and
649 other ACL rejections are.
651 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
652 with slight modification.
654 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
655 draft, changing the syntax and factoring some duplicate code".
657 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
658 for deliveries of the second and subsequent messages over the same SMTP
661 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
662 ${reduce, with only minor "tidies".
664 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
666 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
667 expansion side effects.
669 PH/27 When a message times out after an over-quota error from an Exim-imposed
670 quota, the bounce message says "mailbox is full". This message was not
671 being given when it was a system quota that was exceeded. It now should
674 MH/03 Made $recipients available in local_scan(). local_scan() already has
675 better access to the recipient list through recipients_list[], but
676 $recipients can be useful in postmaster-provided expansion strings.
678 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
679 in the case of a MAIL command with additional options following the
680 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
681 were accidentally chopped off.
683 PH/29 SMTP synchronization checks are implemented when a command is read -
684 there is a check that no more input is waiting when there shouldn't be
685 any. However, for some commands, a delay in an ACL can mean that it is
686 some time before the response is written. In this time, more input might
687 arrive, invalidly. So now there are extra checks after an ACL has run for
688 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
689 pipelining has not been advertised.
691 PH/30 MH's patch to allow iscntrl() characters to be list separators.
693 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
694 returned in the SMTP response when smtp_return_error_details was false.
697 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
698 instead of the C I/O that was originally supplied, because problems were
701 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
702 Exim which did not show up earlier: it was assuming that a call to
703 SSL_CTX_set_info_callback() might give an error value. In fact, there is
704 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
705 was a macro that became an assignment, so it seemed to work. This has
706 changed to a proper function call with a void return, hence the compile
707 error. Exim's code has been fixed.
709 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
712 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
714 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
716 PH/37 If a message is not accepted after it has had an id assigned (e.g.
717 because it turns out to be too big or there is a timeout) there is no
718 "Completed" line in the log. When some messages of this type were
719 selected by exigrep, they were listed as "not completed". Others were
720 picked up by some special patterns. I have improved the selection
721 criteria to be more general.
723 PH/38 The host_find_failed option in the manualroute router can now be set
724 to "ignore", to completely ignore a host whose IP address cannot be
725 found. If all hosts are ignored, the behaviour is controlled by the new
726 host_all_ignored option.
728 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
729 homing or because of multiple MX records with /mx) generated more than
730 one IP address, and the following item turned out to be the local host,
731 all the secondary addresses of the first item were incorrectly removed
732 from the list, along with the local host and any following hosts (which
733 is what is supposed to happen).
735 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
736 whoever called Exim into the -H file. In the case of the daemon it was
737 behaving confusingly. When first started, it used values for whoever
738 started the daemon, but after a SIGHUP it used the Exim user (because it
739 calls itself on a restart). I have changed the code so that it now always
742 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
743 message are rejected with the same error (e.g. no authentication or bad
744 sender address), and a DATA command is nevertheless sent (as can happen
745 with PIPELINING or a stupid MUA), the error message that was given to the
746 RCPT commands is included in the rejection of the DATA command. This is
747 intended to be helpful for MUAs that show only the final error to their
750 PH/42 Another patch from the Sieve maintainer.
752 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
753 Eximstats - Fixed some broken HTML links and added missing column headers
755 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
756 columns for Rejects, Temp Rejects, Ham, and Spam rows.
758 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
760 PH/43 Yet another patch from the Sieve maintainer.
762 PH/44 I found a way to check for a TCP/IP connection going away before sending
763 the response to the final '.' that terminates a message, but only in the
764 case where the client has not sent further data following the '.'
765 (unfortunately, this is allowed). However, in many cases there won't be
766 any further data because there won't be any more messages to send. A call
767 to select() can be used: if it shows that the input is "ready", there is
768 either input waiting, or the socket has been closed. An attempt to read
769 the next input character can distinguish the two cases. Previously, Exim
770 would have sent an OK response which the client would never have see.
771 This could lead to message repetition. This fix should cure that, at
772 least in a lot of common cases.
774 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
775 advertised in response to EHLO.
781 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
782 fixed by 4.65/MH/01 (is this a record?) are fixed:
784 (i) An empty string was always treated as zero by the numeric comparison
785 operators. This behaviour has been restored.
787 (ii) It is documented that the numeric comparison operators always treat
788 their arguments as decimal numbers. This was broken in that numbers
789 starting with 0 were being interpreted as octal.
791 While fixing these problems I realized that there was another issue that
792 hadn't been noticed. Values of message_size_limit (both the global option
793 and the transport option) were treated as octal if they started with 0.
794 The documentation was vague. These values are now always treated as
795 decimal, and I will make that clear in the documentation.
801 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
802 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
805 MH/01 Don't check that the operands of numeric comparison operators are
806 integers when their expansion is in "skipping" mode (fixes bug
807 introduced by 4.64-PH/07).
809 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
810 child addresses, Exim now panics and dies. Previously, because the count
811 is held in a short int, deliveries were likely to be lost. As such a
812 large number of recipients for a single message is ridiculous
813 (performance will be very, very poor), I have chosen to impose a limit
814 rather than extend the field.
820 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
821 leftover -K file (the existence of which was triggered by #402).
822 While we were at it, introduced process PID as part of the -K
823 filename. This should rule out race conditions when creating
826 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
827 processing considerably. Previous code took too long for large mails,
828 triggering a timeout which in turn triggers #401.
830 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
831 in the DK code in transports.c. sendfile() is not really portable,
832 hence the _LINUX specificness.
834 TF/01 In the add_headers option to the mail command in an Exim filter,
835 there was a bug that Exim would claim a syntax error in any
836 header after the first one which had an odd number of characters
839 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
840 callout verification, Exim cached a "reject" for the entire domain. This
841 is correct for most verifications, but it is not correct for a recipient
842 verification with use_sender or use_postmaster set, because in that case
843 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
844 case of MAIL FROM:<> rejection from other early rejections (e.g.
845 rejection of HELO). When verifying a recipient using a non-null MAIL
846 address, the cache is ignored if it shows MAIL FROM:<> rejection.
847 Whatever the result of the callout, the value of the domain cache is
848 left unchanged (for any other kind of callout, getting as far as trying
849 RCPT means that the domain itself is ok).
851 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
854 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
855 manifest itself as EPIPE rather than ECONNECT. When tidying away a
856 session, the daemon ignores ECONNECT errors and logs others; it now
857 ignores EPIPE as well.
859 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
860 (quoted-printable decoding).
862 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
863 later the small subsequent patch to fix an introduced bug.
865 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
867 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
869 PH/08 An error is now given if message_size_limit is specified negative.
871 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
872 to be given (somewhat) arbitrary names.
874 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
877 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
878 miscellaneous code fixes
880 PH/10 Added the log_reject_target ACL modifier to specify where to log
883 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
884 hostname. This is wrong, because it relates to the incoming message (and
885 probably the interface on which it is arriving) and not to the outgoing
886 callout (which could be using a different interface). This has been
887 changed to use the value of the helo_data option from the smtp transport
888 instead - this is what is used when a message is actually being sent. If
889 there is no remote transport (possible with a router that sets up host
890 addresses), $smtp_active_hostname is used.
892 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
893 tweaks were necessary in order to get it to work (see also 21 below):
894 (a) The code assumed that strncpy() returns a negative number on buffer
895 overflow, which isn't the case. Replaced with Exim's string_format()
897 (b) There were several signed/unsigned issues. I just did the minimum
898 hacking in of casts. There is scope for a larger refactoring.
899 (c) The code used strcasecmp() which is not a standard C function.
900 Replaced with Exim's strcmpic() function.
901 (d) The code set only $1; it now sets $auth1 as well.
902 (e) A simple test gave the error "authentication client didn't specify
903 service in request". It would seem that Dovecot has changed its
904 interface. Fortunately there's a specification; I followed it and
905 changed what the client sends and it appears to be working now.
907 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
910 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
911 address A is aliased to B and C, where B exists and C does not. Without
912 -v the output is "A verified" because verification stops after a
913 successful redirection if more than one address is generated. However,
914 with -v the child addresses are also verified. Exim was outputting "A
915 failed to verify" and then showing the successful verification for C,
916 with its parentage. It now outputs "B failed to verify", showing B's
917 parentage before showing the successful verification of C.
919 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
920 look up a TXT record in a specific list after matching in a combined
923 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
924 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
925 they consult the DNS. I had assumed they would set it the way they
926 wanted; and indeed my experiments on Linux seem to show that in some
927 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
928 To be on the safe side, however, I have now made the interface to
929 host_find_byname() similar to host_find_bydns(), with an argument
930 containing the DNS resolver options. The host_find_byname() function now
931 sets these options at its start, just as host_find_bydns() does. The smtp
932 transport options dns_qualify_single and dns_search_parents are passed to
933 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
934 of host_find_byname() use the default settings of RES_DEFNAMES
935 (qualify_single) but not RES_DNSRCH (search_parents).
937 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
938 spool_read_header() do less string testing, by means of a preliminary
939 switch on the second character of optional "-foo" lines. (This is
940 overdue, caused by the large number of possibilities that now exist.
941 Originally there were few.) While I was there, I also converted the
942 str(n)cmp tests so they don't re-test the leading "-" and the first
943 character, in the hope this might squeeze out yet more improvement.
945 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
946 flag allowing group syntax was set by the header_syntax check but not
947 turned off, possible causing trouble later; (2) The flag was not being
948 set at all for the header_verify test, causing "group"-style headers to
949 be rejected. I have now set it in this case, and also caused header_
950 verify to ignore an empty address taken from a group. While doing this, I
951 came across some other cases where the code for allowing group syntax
952 while scanning a header line wasn't quite right (mostly, not resetting
953 the flag correctly in the right place). These bugs could have caused
954 trouble for malformed header lines. I hope it is now all correct.
956 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
957 with the "reply" argument non-NULL. The code, however (which originally
958 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
959 but it didn't always do it. This confused somebody who was copying the
960 code for some other use. I have removed all the tests.
962 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
963 feature that was used to support insecure browsers during the U.S. crypto
964 embargo. It requires special client support, and Exim is probably the
965 only MTA that supported it -- and would never use it because real RSA is
966 always available. This code has been removed, because it had the bad
967 effect of slowing Exim down by computing (never used) parameters for the
968 RSA_EXPORT functionality.
970 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
971 authenticator to fail if there's a tab character in the incoming data
972 (there should never be unless someone is messing about, as it's supposed
973 to be base64-encoded). Also added, on Timo's advice, the "secured" option
974 if the connection is using TLS or if the remote IP is the same as the
975 local IP, and the "valid-client-cert option" if a client certificate has
978 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
979 authenticators. This can be used for authorization after authentication
980 succeeds. (In the case of plaintext, it servers for both authentication
983 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
984 if any retry times were supplied.
986 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
987 connection, where there is no client IP address to check. In this
988 situation, the verify now always succeeds.
990 PH/25 Applied John Jetmore's -Mset patch.
992 PH/26 Added -bem to be like -Mset, but loading a message from a file.
994 PH/27 In a string expansion for a processed (not raw) header when multiple
995 headers of the same name were present, leading whitespace was being
996 removed from all of them, but trailing whitespace was being removed only
997 from the last one. Now trailing whitespace is removed from each header
998 before concatenation. Completely empty headers in a concatenation (as
1001 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1002 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1004 PH/29 [Removed. This was a change that I later backed out, and forgot to
1005 correct the ChangeLog entry (that I had efficiently created) before
1006 committing the later change.]
1008 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1009 address errors (4xx response to RCPT) over the same connection as other
1010 messages routed to the same hosts. Such deliveries are always "forced",
1011 so retry times are not inspected. This resulted in far too many retries
1012 for the affected addresses. The effect occurred only when there were more
1013 hosts than the hosts_max_try setting in the smtp transport when it had
1014 the 4xx errors. Those hosts that it had tried were not added to the list
1015 of hosts for which the message was waiting, so if all were tried, there
1016 was no problem. Two fixes have been applied:
1018 (i) If there are any address or message errors in an SMTP delivery, none
1019 of the hosts (tried or untried) are now added to the list of hosts
1020 for which the message is waiting, so the message should not be a
1021 candidate for sending over the same connection that was used for a
1022 successful delivery of some other message. This seems entirely
1023 reasonable: after all the message is NOT "waiting for some host".
1024 This is so "obvious" that I'm not sure why it wasn't done
1025 previously. Hope I haven't missed anything, but it can't do any
1026 harm, as the worst effect is to miss an optimization.
1028 (ii) If, despite (i), such a delivery is accidentally attempted, the
1029 routing retry time is respected, so at least it doesn't keep
1030 hammering the server.
1032 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1033 in ${readsocket because some servers need this prod.
1035 PH/32 Added some extra debug output when updating a wait-xxx database.
1037 PH/33 The hint "could be header name not terminated by colon", which has been
1038 given for certain expansion errors for a long time, was not being given
1039 for the ${if def:h_colon_omitted{... case.
1041 PH/34 The spec says: "With one important exception, whenever a domain list is
1042 being scanned, $domain contains the subject domain." There was at least
1043 one case where this was not true.
1045 PH/35 The error "getsockname() failed: connection reset by peer" was being
1046 written to the panic log as well as the main log, but it isn't really
1047 panic-worthy as it just means the connection died rather early on. I have
1048 removed the panic log writing for the ECONNRESET error when getsockname()
1051 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1052 runs only) independently of the message's sender address. This meant
1053 that, if the 4xx error was in fact related to the sender, a different
1054 message to the same recipient with a different sender could confuse
1055 things. In particualar, this can happen when sending to a greylisting
1056 server, but other circumstances could also provoke similar problems.
1057 I have changed the default so that the retry time for these errors is now
1058 based a combination of the sender and recipient addresses. This change
1059 can be overridden by setting address_retry_include_sender=false in the
1062 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1063 remote server are returned as part of bounce messages. This was not
1064 happening for LMTP over a pipe (the lmtp transport), but now it is the
1065 same for both kinds of LMTP.
1067 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1068 in header lines that were in fact CNAMEs. This is no longer the case.
1070 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1071 and queue runs started by the daemon processed all messages. This has
1072 been fixed so that -R and -S can now usefully be given with -q<time>.
1074 PH/40 Import PCRE release 6.7 (fixes some bugs).
1076 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1078 PH/42 Give an error if -q is specified more than once.
1080 PH/43 Renamed the variables $interface_address and $interface_port as
1081 $received_ip_address and $received_port, to make it clear that these
1082 values apply to message reception, and not to the outgoing interface when
1083 a message is delivered. (The old names remain recognized, of course.)
1085 PH/44 There was no timeout on the connect() call when using a Unix domain
1086 socket in the ${readsocket expansion. There now is.
1088 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1089 be meaningful with "accept".
1091 SC/01 Eximstats V1.43
1092 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1094 SC/02 Eximstats V1.44
1095 Use a glob alias rather than an array ref in the generated
1096 parser. This improves both readability and performance.
1098 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1099 Collect SpamAssassin and rejection statistics.
1100 Don't display local sender or destination tables unless
1101 there is data to show.
1102 Added average volumes into the top table text output.
1104 SC/04 Eximstats V1.46
1105 Collect data on the number of addresses (recipients)
1106 as well as the number of messages.
1108 SC/05 Eximstats V1.47
1109 Added 'Message too big' to the list of mail rejection
1110 reasons (thanks to Marco Gaiarin).
1112 SC/06 Eximstats V1.48
1113 Mainlog lines which have GMT offsets and are too short to
1114 have a flag are now skipped.
1116 SC/07 Eximstats V1.49 (Alain Williams)
1117 Added the -emptyok flag.
1119 SC/08 Eximstats V1.50
1120 Fixes for obtaining the IP address from reject messages.
1122 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1123 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1124 whitesspace changes from 4.64-PH/27
1126 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1129 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1130 are found, allow negative numbers in numeric criteria)
1132 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1134 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1135 to match changes made in 4.64-PH/43
1137 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1139 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1140 "domain" is actually an IP address, and force a failure. This locks out
1141 those revolvers/nameservers that support "A-for-A" lookups, in
1142 contravention of the specifications.
1144 PH/48 When a host name was looked up from an IP address, and the subsequent
1145 forward lookup of the name timed out, the host name was left in
1146 $sender_host_name, contrary to the specification.
1148 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1149 restricted to single-key lookups, Exim was not diagnosing an error if
1150 * or *@ was used with a query-style lookup.
1152 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1154 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1155 long ago, but noone interested enough thought of it. Let's just say that
1156 the "1.1" means that there are some new functions that weren't there at
1157 some point in the past.
1159 PH/51 Error processing for expansion failure of helo_data from an smtp
1160 transport during callout processing was broken.
1162 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1163 tested/used via the -bh/-bhc/-bs options.
1165 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1166 bug, fixed in subsequent PCRE releases).
1168 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1169 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1171 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1177 SC/01 Use a glob alias rather than an array ref in eximstats generated
1178 parser. This improves both readability and performance.
1180 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1181 Don't display local sender or destination tables in eximstats unless
1182 there is data to show.
1183 Added average volumes into the eximstats top table text output.
1185 SC/03 Collect data on the number of addresses (recipients) as well
1186 as the number of messages in eximstats.
1188 TF/01 Correct an error in the documentation for the redirect router. Exim
1189 does (usually) call initgroups() when daemonizing.
1191 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1192 with consistent privilege compared to when running as a daemon.
1194 TF/03 Note in the spec that $authenticated_id is not set for local
1195 submissions from trusted users.
1197 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1198 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1200 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1201 by adding some example configuration directives to the default
1202 configuration file. A little bit of work is required to uncomment the
1203 directives and define how usernames and passwords are checked, but
1204 there is now a framework to start from.
1206 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1207 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1208 without this. I don't know how relevant this is to other LDAP libraries.
1210 PH/02 Add the verb name to the "unknown ACL verb" error.
1212 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1214 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1216 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1217 directory not expanded when it should be if an expanded home directory
1218 was set for the address (which is overridden by the transport).
1220 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1223 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1224 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1225 because it is too late at that time, and has no effect.
1227 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1228 security issue with \' (bugzilla #107). I could not use the
1229 PQescapeStringConn() function, because it needs a PGconn value as one of
1232 PH/08 When testing addresses using -bt, indicate those final addresses that
1233 are duplicates that would not cause an additional delivery. At least one
1234 person was confused, thinking that -bt output corresponded to deliveries.
1235 (Suppressing duplicates isn't a good idea as you lose the information
1236 about possibly different redirections that led to the duplicates.)
1238 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1239 systems where poll() doesn't work, in particular OS X.
1241 PH/10 Added more information to debugging output for retry time not reached.
1243 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1244 operations in malware.c.
1246 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1249 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1250 syslog_duplication set false, log lines that would normally be written
1251 both the the main log and to the reject log were not written to syslog at
1254 PH/14 In the default configuration, change the use of "message" in ACL warn
1255 statements to "add_header".
1257 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1258 not followed by a command (e.g. "seen endif").
1260 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1261 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1264 PH/17 Added extra conditions to the default value of delay_warning_condition
1268 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1269 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1270 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1273 The Auto-Submitted: and various List- headers are standardised, whereas I
1274 don't think Precedence: ever was.
1276 PH/18 Refactored debugging code in route_finduser() to show more information,
1277 in particular, the error code if getpwnam() issues one.
1279 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1280 This is apparently needed in addition to the PH/07 change above to avoid
1281 any possible encoding problems.
1283 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1284 but not after initializing Perl.
1286 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1287 output them only if debugging. By default they are written stderr,
1288 apparently, which is not desirable.
1290 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1293 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1296 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1298 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1299 authenticated or an ident call has been made. Suppress the default
1300 values for $authenticated_id and $authenticated_sender (but permit -oMai
1301 and -oMas) when testing with -bh.
1303 PH/24 Re-jigged the order of the tests in the default configuration so that the
1304 tests for valid domains and recipients precede the DNS black list and CSA
1305 tests, on the grounds that those ones are more expensive.
1307 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1308 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1309 This bug exists in every version of Exim that I still have, right back to
1312 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1313 However, an attempt to turn on case-sensitivity in a regex key by
1314 including (?-i) didn't work because the subject string was already
1315 lowercased, and the effects were non-intuitive. It turns out that a
1316 one-line patch can be used to allow (?-i) to work as expected.
1322 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1323 other effects) broke the use of negated acl sub-conditions.
1325 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1328 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1329 "Deny" causes Exim to reject the incoming connection with a 554 error.
1330 Unfortunately, if there is a major crisis, such as a disk failure,
1331 tcp-wrappers gives "deny", whereas what one would like would be some
1332 kind of temporary error. A kludge has been added to help with this.
1333 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1334 554 error is used if errno is still zero or contains ENOENT (which occurs
1335 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1338 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1340 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1341 errors. Otherwise a message that provokes a temporary error (when other
1342 messages do not) can cause a whole host to time out.
1344 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1345 addresses were routed directly to files or pipes from a redirect router.
1346 File deliveries just didn't batch; pipe deliveries might have suffered
1349 PH/06 A failure to get a lock for a hints database would erroneously always say
1350 "Failed to get write lock", even when it was really a read lock.
1352 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1353 of 0600. This has been changed to use the value of the lockfile_mode
1354 option (which defaults to 0600).
1356 PH/08 Applied small patch from the Sieve maintainer.
1358 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1359 folder from quota calculations, a direct delivery into this folder messed
1360 up the contents of the maildirsize file. This was because the regex was
1361 used only to exclude .Trash (or whatever) when the size of the mailbox
1362 was calculated. There was no check that a delivery was happening into an
1363 excluded directory. This bug has been fixed by ignoring all quota
1364 processing for deliveries into excluded directories.
1366 PH/10 Added the maildirfolder_create_regex option to appendfile.
1372 PH/01 The code for finding all the local interface addresses on a FreeBSD
1373 system running IPv6 was broken. This may well have applied to all BSD
1374 systems, as well as to others that have similar system calls. The broken
1375 code found IPv4 interfaces correctly, but gave incorrect values for the
1376 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1377 that it would not match correctly against @[] and not recognize the IPv6
1380 PH/02 The ipliteral router was not recognizing addresses of the form user@
1381 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1383 PH/03 Added disable_ipv6.
1385 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1386 decoded form, because it is most often used to construct To: headers
1387 lines in autoreplies, and the decoded form may well be syntactically
1388 invalid. However, $reply_address has leading white space removed, and all
1389 newlines turned into spaces so that the autoreply transport does not
1392 PH/05 If group was specified without a user on a router, and no group or user
1393 was specified on a transport, the group from the router was ignored.
1395 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1396 for visible compile-time settings that can be used to change these
1397 numbers, for those that want even more. Backwards compatibility with old
1398 spool files has been maintained. However, going back to a previous Exim
1399 release will lost any variables that are in spool files.
1401 PH/07 Two small changes when running in the test harness: increase delay when
1402 passing a TCP/IP connection to a new process, in case the original
1403 process has to generate a bounce, and remove special handling of
1404 127.0.0.2 (sic), which is no longer necessary.
1406 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1407 be the same on different OS.
1409 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1412 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1413 whether --show-vars was specified or not
1415 JJ/02 exipick: Added support for new ACL variable spool format introduced
1418 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1419 syntactically invalid From: or Reply-to: line, and a filter used this to
1420 generate an autoreply, and therefore failed to obtain an address for the
1421 autoreply, Exim could try to deliver to a non-existent relative file
1422 name, causing unrelated and misleading errors. What now happens is that
1423 it logs this as a hard delivery error, but does not attempt to create a
1426 PH/11 The exinext utility has a -C option for testing purposes, but although
1427 the given file was scanned by exinext itself; it wasn't being passed on
1428 when Exim was called.
1430 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1431 an end-of-file indication when reading a command response.
1433 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1434 compiled. In many other places in Exim, IPv6 addresses are always
1435 recognized, so I have changed this. It also means that IPv4 domain
1436 literals of the form [IPV4:n.n.n.n] are now always recognized.
1438 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1439 used if the router is not running as root, for example, when verifying at
1440 ACL time, or when using -bh. The debugging output from this situation was
1441 non-existent - all you got was a failure to exec. I have made two
1444 (a) Failures to set uid/gid, the current directory, or a process leader
1445 in a subprocess such as that created by queryprogram now generate
1446 suitable debugging ouput when -d is set.
1448 (b) The queryprogram router detects when it is not running as root,
1449 outputs suitable debugging information if -d is set, and then runs
1450 the subprocess without attempting to change uid/gid.
1452 PH/15 Minor change to Makefile for building test_host (undocumented testing
1455 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1456 additional section of a DNS packet that returns MX or SRV records.
1457 Instead, it always explicitly searches for A/AAAA records. This avoids
1458 major problems that occur when a DNS server includes only records of one
1459 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1460 fixed another bug: if SRV records were looked up and the corresponding
1461 address records were *not* found in the additional section, the port
1462 values from the SRV records were lost.
1464 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1465 using the correct key (the original address) when searching the retry
1466 rules in order to find which one to use for generating the retry hint.
1468 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1469 adding the default one. Similarly, if it contains a Reply-To: header, the
1470 errors_reply_to option, if set, is not used.
1472 PH/19 When calculating a retry time, Exim used to measure the "time since
1473 failure" by looking at the "first failed" field in the retry record. Now
1474 it does not use this if it is later than than the arrival time of the
1475 message. Instead it uses the arrival time. This makes for better
1476 behaviour in cases where some deliveries succeed, thus re-setting the
1477 "first failed" field. An example is a quota failure for a huge message
1478 when small messages continue to be delivered. Without this change, the
1479 "time since failure" will always be short, possible causing more frequent
1480 delivery attempts for the huge message than are intended.
1481 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1483 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1484 $1, $2, $3) because the numerical variables can be reset during some
1485 expansion items (e.g. "match"), thereby losing the authentication data.
1487 PH/21 Make -bV show the size of off_t variables so that the test suite can
1488 decide whether to run tests for quotas > 2G.
1490 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1491 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1492 filecount value is greater than 2G or if a quota value is greater than 2G
1493 on a system where the size of off_t is not greater than 4, a panic error
1496 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1497 never match. The debug and -bh output now contains an explicit error
1498 message indicating a malformed IPv4 address or mask.
1500 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1501 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1502 PH/23 above applies.
1504 PH/25 Do not write to syslog when running in the test harness. The only
1505 occasion when this arises is a failure to open the main or panic logs
1506 (for which there is an explicit test).
1508 PH/26 Added the /no_tell option to "control=freeze".
1510 PH/27 If a host name lookup failed very early in a connection, for example, if
1511 the IP address matched host_lookup and the reverse lookup yielded a name
1512 that did not have a forward lookup, an error message of the form "no IP
1513 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1514 could be logged. Now it outputs the IP address instead of "NULL".
1516 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1517 allows the sender and the authenticated sender to be set when
1518 submitting a message from within Exim. Since child_open_exim() is
1519 documented for local_scan(), the new function should be too.
1521 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1522 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1523 results in an empty string is now treated as unset.
1525 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1527 PH/31 Added sender_verify_fail logging option.
1529 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1530 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1531 tidied the source and removed it altogether.
1533 PH/33 When a queue run was abandoned because the load average was too high, a
1534 log line was always written; now it is written only if the queue_run log
1535 selector is set. In addition, the log line for abandonment now contains
1536 information about the queue run such as the pid. This is always present
1537 in "start" and "stop" lines but was omitted from the "abandon" line.
1539 PH/34 Omit spaces between a header name and the colon in the error message that
1540 is given when verify = headers_syntax fails (if there are lots of them,
1541 the message gets confusing).
1543 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1544 names, as there are now some PTR records that contain slashes. This check
1545 is only to protect against broken name servers that fall over on strange
1546 characters, so the fact that it applies to all lookups doesn't matter.
1548 PH/36 Now that the new test suite is complete, we can remove some of the
1549 special code in Exim that was needed for the old test suite. For example,
1550 sorting DNS records because real resolvers return them in an arbitrary
1551 order. The new test suite's fake resolver always returns records in the
1554 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1555 bounces) except when queue_only is set, to avoid logging races between
1556 the different processes.
1558 PH/38 Panic-die if .include specifies a non-absolute path.
1560 PH/39 A tweak to the "H" retry rule from its user.
1562 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1563 a label. They prevented compilation on older perls.
1565 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1566 a warning to be raised on newish perls.
1568 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1569 on queue. Changes to match documented behaviour of showing count of
1570 messages matching specified criteria.
1572 PH/40 Changed the default ident timeout from 30s to 5s.
1574 PH/41 Added support for the use of login_cap features, on those BSD systems
1575 that have them, for controlling the resources used by pipe deliveries.
1577 PH/42 The content-scanning code uses fopen() to create files in which to put
1578 message data. Previously it was not paying any attention to the mode of
1579 the files. Exim runs with umask(0) because the rest of the code creates
1580 files with open(), and sets the required mode explicitly. Thus, these
1581 files were ending up world-writeable. This was not a big issue, because,
1582 being within the spool directory, they were not world-accessible. I have
1583 created a function called modefopen, which takes an additional mode
1584 argument. It sets umask(777), creates the file, chmods it to the required
1585 mode, then resets the umask. All the relevant calls to fopen() in the
1586 content scanning code have been changed to use this function.
1588 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1589 to 24 hours. This avoids potential overflow problems when processing G
1590 and H retry rules. I suspect nobody ever tinkers with this value.
1592 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1594 PH/45 When the plaintext authenticator is running as a client, the server's
1595 challenges are checked to ensure they are valid base64 strings. By
1596 default, the authentication attempt is cancelled if an invalid string is
1597 received. Setting client_ignore_invalid_base64 true ignores these errors.
1598 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1599 they are received. Thus, the responses can be made to depend on the
1600 challenges. If an invalid string is ignored, an empty string is placed in
1603 PH/46 Messages that are created by the autoreply transport now contains a
1604 References: header, in accordance with RFCs 2822 and 3834.
1606 PH/47 Added authenticated_sender_force to the smtp transport.
1608 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1610 PH/49 Installed latest patch from the Sieve maintainer.
1612 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1613 was also set, the appendfile transport was unnecessarily scanning a
1614 directory of message files (e.g. for maildir delivery) to find the count
1615 of files (along with the size), even though it did not need this
1616 information. It now does the scan only if it needs to find either the
1617 size of the count of files.
1619 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1621 PH/52 Two bugs concerned with error handling when the smtp transport is
1624 (i) Exim was not creating retry information for temporary errors given
1625 for individual recipients after the DATA command when the smtp transport
1626 was used in LMTP mode. This meant that they could be retried too
1627 frequently, and not timed out correctly.
1629 (ii) Exim was setting the flag that allows error details to be returned
1630 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1631 recipients that were returned after the DATA command.
1633 PH/53 This is related to PH/52, but is more general: for any failing address,
1634 when detailed error information was permitted to be returned to the
1635 sender, but the error was temporary, then after the final timeout, only
1636 "retry timeout exceeded" was returned. Now it returns the full error as
1637 well as "retry timeout exceeded".
1639 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1640 do this, and (what is worse) MTAs that accept it.
1642 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1643 will now be deprecated.
1645 PH/56 New os.c-cygwin from the Cygwin maintainer.
1647 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1648 formats (previously only available in exim formats via -bpr, -bpru,
1649 and -bpra. Now also available in native and exiqgrep formats)
1651 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1652 with very large, slow to parse queues
1654 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1656 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1658 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1659 responses to authentication challenges, though it was showing the
1660 challenges; (ii) I've removed the CR characters from the debug output for
1663 PH/58 Allow for the insertion of a newline as well as a space when a string
1664 is turned into more than one encoded-word during RFC 2047 encoding. The
1665 Sieve code now uses this.
1667 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1668 data_4xx, lost_connection, tls_required.
1670 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1671 message was being sent as an SMTP response.
1673 PH/61 Add -l and -k options to exicyclog.
1675 PH/62 When verifying, if an address was redirected to one new address, so that
1676 verification continued, and the new address failed or deferred after
1677 having set something in $address_data, the value of $address_data was not
1678 passed back to the ACL. This was different to the case when no
1679 redirection occurred. The value is now passed back in both cases.
1681 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1682 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1683 use login_cap.h, so on its own it isn't the distinguishing feature. The
1684 new name refers directly to the setclassresources() function.
1686 PH/65 Added configuration files for NetBSD3.
1688 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1690 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1691 is preferred over IPv4.
1693 PH/68 The bounce_return_message and bounce_return_body options were not being
1694 honoured for bounces generated during the reception of non-SMTP messages.
1695 In particular, this applied to messages rejected by the ACL. This bug has
1696 been fixed. However, if bounce_return_message is true and bounce_return_
1697 body is false, the headers that are returned for a non-SMTP message
1698 include only those that have been read before the error was detected.
1699 (In the case of an ACL rejection, they have all been read.)
1701 PH/69 The HTML version of the specification is now built in a directory called
1702 spec_html instead of spec.html, because the latter looks like a path with
1703 a MIME-type, and this confuses some software.
1705 PH/70 Catch two compiler warnings in sieve.c.
1707 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1708 function verify_get_ident() calls ip_connect() to connect a socket, but
1709 if the "connect()" function timed out, ip_connect() used to close the
1710 socket. However, verify_get_ident() also closes the socket later, and in
1711 between Exim writes to the log, which may get opened at this point. When
1712 the socket was closed in ip_connect(), the log could get the same file
1713 descriptor number as the socket. This naturally causes chaos. The fix is
1714 not to close the socket in ip_connect(); the socket should be closed by
1715 the function that creates it. There was only one place in the code where
1716 this was missing, in the iplookup router, which I don't think anybody now
1717 uses, but I've fixed it anyway.
1719 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1720 well as to direct DNS lookups. Otherwise the handling of names in host
1721 lists is inconsistent and therefore confusing.
1727 PH/01 Two changes to the default runtime configuration:
1729 (1) Move the checks for relay_from_hosts and authenticated clients from
1730 after to before the (commented out) DNS black list checks.
1732 (2) Add control=submission to the relay_from_hosts and authenticated
1733 clients checks, on the grounds that messages accepted by these
1734 statements are most likely to be submissions.
1736 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1738 (1) Generate an error if the third argument for the ${prvs expansion is
1741 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1744 (3) Reset the variables that are obtained from the first argument of
1745 ${prvscheck and used in the second argument before leaving the code,
1746 because their memory is reclaimed, so using them afterwards may do
1749 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1750 one (it's much easier than Tom thought :-).
1752 (5) Because of (4), we can now allow for the use of $prvscheck_result
1753 inside the third argument.
1755 PH/03 For some reason, the default setting of PATH when running a command from
1756 a pipe transport was just "/usr/bin". I have changed it to
1759 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1760 anything to be listed in the output from -bV.
1762 PH/05 When a filter generated an autoreply, the entire To: header line was
1763 quoted in the delivery log line, like this:
1765 => >A.N.Other <ano@some.domain> <original@ddress> ...
1767 This has been changed so that it extracts the operative address. There
1768 may be more than one such address. If so, they are comma-separated, like
1771 => >ano@some.domain,ona@other.domain <original@ddress> ...
1773 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1774 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1775 not being looked up in the rDNS to get a host name, Exim was showing the
1776 IP address twice in Received: lines, even though the IP addresses were
1777 identical. For example:
1779 Received: from [1.2.3.4] (helo=[1.2.3.4])
1781 However, if the real host name was known, it was omitting the HELO data
1782 if it matched the actual IP address. This has been tidied up so that it
1783 doesn't show the same IP address twice.
1785 PH/07 When both +timestamp and +memory debugging was on, the value given by
1786 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1787 called by the debug printing, thereby overwriting the timestamp buffer.
1788 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1790 PH/08 When the original message was included in an autoreply transport, it
1791 always said "this is a copy of the message, including all the headers",
1792 even if body_only or headers_only was set. It now gives an appropriate
1795 PH/09 Applied a patch from the Sieve maintainer which:
1797 o fixes some comments
1798 o adds the (disabled) notify extension core
1799 o adds some debug output for the result of if/elsif tests
1800 o points to the current vacation draft in the documentation
1801 and documents the missing references header update
1805 o fixes a bug in processing the envelope test (when testing
1806 multiple envelope elements, the last element determinted the
1809 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1810 Electronic Mail") by including:
1812 Auto-submitted: auto-generated
1814 in the messages that it generates (bounce messages and others, such as
1815 warnings). In the case of bounce messages for non-SMTP mesages, there was
1816 also a typo: it was using "Auto_submitted" (underscore instead of
1817 hyphen). Since every message generated by Exim is necessarily in response
1818 to another message, thes have all been changed to:
1820 Auto-Submitted: auto-replied
1822 in accordance with these statements in the RFC:
1824 The auto-replied keyword:
1826 - SHOULD be used on messages sent in direct response to another
1827 message by an automatic process,
1829 - MUST NOT be used on manually-generated messages,
1831 - MAY be used on Delivery Status Notifications (DSNs) and Message
1832 Disposition Notifications (MDNs),
1834 - MUST NOT be used on messages generated by automatic or periodic
1835 processes, except for messages which are automatic responses to
1838 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1839 to the default Received: header definition.
1841 PH/12 Added log selector acl_warn_skipped (default on).
1843 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1844 variables because (a) they are in the wrong storage pool and (b) even if
1845 they were copied, it wouldn't work properly because of the caching.
1847 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1848 checking when decoding. Apparently there are clients that generate
1849 overlong encoded strings. Why am I not surprised?
1851 PH/15 If the first argument of "${if match_address" was not empty, but did not
1852 contain an "@" character, Exim crashed. Now it writes a panic log message
1853 and treats the condition as false.
1855 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1857 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1858 extension "envelope-auth". The code is finished and in agreement with
1859 other implementations, but there is no documentation so far and in fact,
1860 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1861 not changing the active code.
1863 Print executed "if" and "elsif" statements when debugging is used. This
1864 helps a great deal to understand what a filter does.
1866 Document more things not specified clearly in RFC3028. I had all this
1867 sorted out, when out of a sudden new issues came to my mind. Oops."
1869 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1872 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1873 canonical form (as documented). However, after a host name lookup from
1874 the IP address, check_host() was doing a simple string comparison with
1875 addresses acquired from the DNS when checking that the found name did
1876 have the original IP as one of its addresses. Since any found IPv6
1877 addresses are likely to be in abbreviated form, the comparison could
1878 fail. Luckily, there already exists a function for doing the comparison
1879 by converting both addresses to binary, so now that is used instead of
1880 the text comparison.
1882 PH/20 There was another similar case to PH/19, when a complete host name was
1883 given in a host list; looking up its IP address could give an abbreviated
1884 form, whereas the current host's name might or might not be abbreviated.
1885 The same fix has been applied.
1891 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1892 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1895 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1896 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1898 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1900 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1901 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1902 Also, for "Auto-Submitted", treat anything other than "no" as
1903 non-personal, in accordance with RFC 3834. (Previously it treated
1904 anything starting "auto-" as non-personal.)
1906 TF/01 The control=submission/name=... option had a problem with syntax
1907 errors if the name included a slash character. The /name= option
1908 now slurps the rest of the string, so it can include any characters
1909 but it must come last in the list of options (after /sender_retain
1912 PH/05 Some modifications to the interface to the fake nameserver for the new
1920 TK/01 Added the "success_on_redirect" address verification option. See
1921 NewStuff for rationale and an example.
1923 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1925 PH/02 Patch to exigrep to allow it to work on syslog lines.
1927 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1928 fread() to skip over the body file's header line, because in Cygwin the
1929 header line is locked and is inaccessible.
1931 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1932 co-exist for some time) to make it clear that it is the Exim ID that is
1933 referenced, not the Message-ID: header line.
1935 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1936 string_format() function, because snprintf() does not exist on all
1939 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1940 ${stat: expansion item.
1942 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1943 protocol synchronization error", to keep the pedants happy.
1945 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1946 well as for IRIX systems, when gcc is being used. See the host.c source
1949 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1951 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1954 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1955 transport and to the smtp transport in LMTP mode.
1957 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1959 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1960 run a filter in a subprocess. This could lead to confusion in subsequent
1961 lookups in the parent process. There should also be a search_tidyup() at
1962 the end of the subprocess.
1964 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1965 only if the host matched helo_try_verify_hosts, which caused the
1966 verification to occur when the EHLO/HELO command was issued. The ACL just
1967 tested the remembered result. Now, if a previous verification attempt has
1968 not happened, "verify = helo" does it there and then.
1970 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1972 TK/03 Fix log output including CR from clamd.
1974 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1975 exist provoked a memory error which could cause a segfault.
1977 PH/15 Installed PCRE 6.2
1979 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1981 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1982 of the problem. Specifically, suggested +O2 rather than +O1 for the
1985 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1987 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1988 record, the port was not correctly logged when the outgoing_port log
1989 selector was set (it logged the transort's default port).
1991 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1992 fallback_hosts, and "hosts" in the smtp transport.
1994 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1995 host errors such as "Connection refused".
1997 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1998 authentication with radiusclient 0.4.9:
2000 - Error returned from rc_read_config was caught wrongly
2001 - Username/password not passed on to radius server due to wrong length.
2003 The presumption is that some radiusclient API changes for 4.51/PH/17
2004 were not taken care of correctly. The code is still untested by me (my
2005 Linux distribution still has 0.3.2 of radiusclient), but it was
2006 contributed by a Radius user.
2008 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2009 expanding the "port" option of the smtp transport.
2011 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2012 while reading a MIME header. Thanks to Tom Hughes for a patch.
2014 PH/24 Include config.h inside local_scan.h so that configuration settings are
2017 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2018 that in an ACL for RCPT (for example), you can examine exactly what was
2021 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2022 commands, but it was not correctly comparing the address with the actual
2023 client host address. Thus, it would show the EHLO address in Received:
2024 header lines when this was not necessary.
2026 PH/27 Added the % operator to ${eval:}.
2028 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2029 it should be ignoring failures (because with -C, for example, it has lost
2030 privilege). It wasn't ignoring creation failures other than "already
2033 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2036 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2037 because an input line was too long, either on its own, or by virtue of
2038 too many continuations, the temporary file was not being removed, and the
2039 return code was incorrect.
2041 PH/31 Missing "BOOL" in function definition in filtertest.c.
2043 PH/32 Applied Sieve patches from the maintainer.
2045 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2047 PH/33 Added "verify = not_blind".
2049 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2050 Local/Makefile (with some defaults set). These are used in built scripts
2051 such as exicyclog, but they have never been used in the exim_install
2052 script (though there are many overriding facilities there). I have
2053 arranged that the exim_install script now takes note of these two
2056 PH/35 Installed configuration files for Dragonfly.
2058 PH/36 When a locally submitted message by a trusted user did not contain a
2059 From: header, and the sender address was obtained from -f or from an SMTP
2060 MAIL command, and the trusted user did not use -F to supply a sender
2061 name, $originator_name was incorrectly used when constructing a From:
2062 header. Furthermore, $originator_name was used for submission mode
2063 messages from external hosts without From: headers in a similar way,
2064 which is clearly wrong.
2066 PH/37 Added control=suppress_local_fixups.
2068 PH/38 When log_selector = +received_sender was set, and the addition of the
2069 sender made the log line's construction buffer exactly full, or one byte
2070 less than full, an overflow happened when the terminating "\n" was
2073 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2074 when the result of a list match is failure because a DNS lookup failed.
2076 PH/40 RM_COMMAND is now used in the building process.
2078 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2079 the "build-* directories that it finds.
2081 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2082 address in a domain literal was a prefix of an interface address.
2084 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2085 when verifying a sender address, unless rewrite_headers is false.
2087 PH/44 Wrote a long comment about why errors_to addresses are verified as
2088 recipients, not senders.
2090 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2091 the ratelimit ACL was added.
2093 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2095 PH/46 Added extra information about PostgreSQL errors to the error string.
2097 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2098 avoiding the need to install special zones in a real server. This is
2099 backwards compatible; if it can't find the fake resolver, it drops back.
2100 Thus, both old and new test suites can be run.
2102 TF/02 Added util/ratelimit.pl
2104 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2105 clock is set back in time.
2107 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2108 Candler <B.Candler@pobox.com>.
2110 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2111 OK for addresses that are the result of redirections.
2113 PH/48 A number of further additions for the benefit of the new test suite,
2114 including a fake gethostbyname() that interfaces to the fake DNS resolver
2117 TF/06 The fix for widen_domains has also been applied to qualify_single and
2118 search_parents which are the other dnslookup options that can cause
2121 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2124 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2126 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2127 service" if it encountered an MX record with an empty target hostname.
2128 The message is now "an MX or SRV record indicated no SMTP service".
2130 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2131 if the DNS of the sending site is misconfigured. This is quite a
2132 common situation. This change restores the behaviour of treating a
2133 helo verification defer as a failure.
2135 PH/51 If self=fail was set on a router, the bounce message did not include the
2136 actual error message.
2142 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2144 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2145 command itself ended in error, the underlying message about the transport
2146 filter timeout was being overwritten with the pipe command error. Now the
2147 underlying error message should be appended to the second error message.
2149 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2151 PH/02 Reduce the amount of output that "make" produces by default. Full output
2152 can still be requested.
2154 PH/03 The warning log line about a condition test deferring for a "warn" verb
2155 was being output only once per connection, rather than after each
2156 occurrence (because it was using the same function as for successful
2157 "warn" verbs). This seems wrong, so I have changed it.
2159 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2160 it should not have, which might have caused a crash in the right
2161 circumstances, but probably never did.
2163 PH/04 Installed a modified version of Tony Finch's patch to make submission
2164 mode fix the return path as well as the Sender: header line, and to
2165 add a /name= option so that you can make the user's friendly name appear
2168 TF/03 Added the control = fakedefer ACL modifier.
2170 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2171 Mark Lowes for thorough testing.
2173 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2175 TK/03 Merged latest SRS patch from Miles Wilton.
2177 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2178 with the definition in sysexits.h (which is #included earlier).
2179 Fortunately, Exim does not actually use EX_OK. The code used to try to
2180 preserve the sysexits.h value, by assumimg that macro definitions were
2181 scanned for macro replacements. I have been disabused of this notion,
2182 so now the code just undefines EX_OK before #including unistd.h.
2184 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2185 in the smtp transport. When a block could not be written in a single
2186 write() function, the timeout was being re-applied to each part-write.
2187 This seems wrong - if the receiver was accepting one byte at a time it
2188 would take for ever. The timeout is now adjusted when this happens. It
2189 doesn't have to be particularly precise.
2191 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2192 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2194 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2195 without a domain if the check to <postmaster@domain> fails.
2197 SC/01 Eximstats: added -xls and the ability to specify output files
2198 (patch written by Frank Heydlauf).
2200 SC/02 Eximstats: use FileHandles for outputing results.
2202 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2204 SC/04 Eximstats: fixed display of large numbers with -nvr option
2206 SC/05 Eximstats: fixed merging of reports with empty tables.
2208 SC/06 Eximstats: added the -include_original_destination flag
2210 SC/07 Eximstats: removed tabs and trailing whitespace.
2212 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2214 TK/06 MBOX spool code: Add real "From " MBOX separator line
2215 so the .eml file is really in mbox format (even though
2216 most programs do not really care). Patch from Alex Miller.
2218 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2219 The latter is generated from $received_to and is only set if the
2220 message has one envelope recipient. SA can use these headers,
2221 obviously out-of-the-box. Patch from Alex Miller.
2223 PH/08 The ${def test on a variable was returning false if the variable's
2224 value was "0", contrary to what the specification has always said!
2225 The result should be true unless the variable is empty.
2227 PH/09 The syntax error of a character other than { following "${if
2228 def:variable_name" (after optional whitespace) was not being diagnosed.
2229 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2230 accidental colon was present, for example, could give incorrect results.
2232 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2233 result is used (not including appendfile, where other changes are about
2236 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2237 This involved changing a lot of size variables from int to off_t. It
2238 should work with maildirs and everything.
2240 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2241 spamd dying while we are connected to it.
2243 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2246 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2247 function for BDB 4.3.
2249 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2251 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2252 the sources for PCRE have been re-arranged and more files are now
2255 PH/15 The code I had for printing potentially long long variables in PH/11
2256 above was not the best (it lost precision). The length of off_t variables
2257 is now inspected at build time, and an appropriate printing format (%ld
2258 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2259 to be "long long int" or "long int". This is needed for the internal
2260 formatting function string_vformat().
2262 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2263 the configuration file to be ":syslog", then the script "guesses" where
2264 the logs files are, rather than using the compiled in default. In our
2265 case the guess is not the same as the compiled default, so the script
2266 suddenly stopped working when I started to use syslog. The patch checks
2267 to see if log_file_path is "". If so, it attempts to read it from exim
2268 with no configuration file to get the compiled in version, before it
2269 falls back to the previous guessing code."
2271 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2272 implementing BATV in an Exim configuration. See NewStuff for the gory
2275 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2276 Makefile that are specific to HP-UX.
2278 PH/18 If the "use_postmaster" option was set for a recipient callout together
2279 with the "random" option, the postmaster address was used as the MAIL
2280 FROM address for the random test, but not for the subsequent recipient
2281 test. It is now used for both.
2283 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2284 patch removes a few documentation additions to RFC 3028, because the
2285 latest draft now contains them. It adds the new en;ascii-case comparator
2286 and a new error check for 8bit text in MIME parts. Comparator and
2287 require names are now matched exactly. I enabled the subaddress
2288 extension, but it is not well tested yet (read: it works for me)."
2290 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2291 rework some of the code of TK/09 above to avoid the hardwired use of
2292 "%lld" and "long long". Replaced the call to snprintf() with a call to
2295 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2296 records point to non-existent hosts", "retry timeout exceeded", and
2297 "retry time not reached for any host after a long failure period".
2299 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2300 experimental DomainKeys support:
2302 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2303 (2) On an error such as an illegally used "control", the wrong name for
2304 the control was given.
2306 These problems did NOT occur unless DomainKeys support was compiled.
2308 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2310 PH/24 Added ${if match_ip condition.
2312 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2313 databases so that it will be absolutely obvious if a crash occurs in the
2314 DB library. This is a regular occurrence (often caused by mis-matched
2317 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2318 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2319 picked up on a user's system that detects such things. There doesn't seem
2320 to be a gcc warning option for this - only an attribute that has to be
2321 put on the function's prototype. It seems that in Fedora Core 4 they have
2322 set this on a number of new functions. No doubt there will be more in due
2325 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2326 specify a transport. However, if an address that was verified by such a
2327 router was the subject of a callout, Exim crashed because it tried to
2328 read the rcpt_include_affixes from the non-existent transport. Now it
2329 just assumes that the setting of that option is false. This bug was
2330 introduced by 4.51/PH/31.
2332 PH/28 Changed -d+all to exclude +memory, because that information is very
2333 rarely of interest, but it makes the output a lot bigger. People tend to
2334 do -d+all out of habit.
2336 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2337 code in os-type was giving problems when libc.so lives in lib64, like on
2340 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2341 aren't the modern standard, and it seems that some systems' include files
2342 don't always have them. Exim was already checking for some of the newer
2343 ones like T_AAAA, and defining it itself. I've added checks for all the
2344 record types that Exim uses.
2346 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2347 not automatically generating a new one, as it is supposed to. This
2348 prevented TLS from working. If the file did exist, but contained invalid
2349 data, a new version was generated, as expected. It was only the case of a
2350 non-existent file that was broken.
2352 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2353 with a change in libdomainkeys > 0.64.
2355 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2356 from DNS. If the selector record carries the flag, it now has
2357 precedence over the domain-wide flag.
2359 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2361 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2362 the use of a port name that isn't defined in /etc/services) occurred, the
2363 message was deferred as in a normal delivery, and thus remained on the
2364 spool, instead of being failed because of the mua_wrapper setting. This
2365 is now fixed, and I tidied up some of the mua_wrapper messages at the
2368 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2369 the messages in a hash of arrays rather than using individual hashes.
2370 This is a bit cleaner and results in dramatic memory savings, albeit
2371 at a slight CPU cost.
2373 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2374 as requested by Marc Sherman.
2376 SC/10 Eximstats: added histograms for user specified patterns as requested
2379 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2381 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2382 fopen() in the content-scanning modules that did not already have it.
2388 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2389 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2391 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2393 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2395 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2396 to test Sieve filters that use "vacation".
2398 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2399 that changes the way the GnuTLS parameters are stored in the cache file.
2400 The new format can be generated externally. For backward compatibility,
2401 if the data in the cache doesn't make sense, Exim assumes it has read an
2402 old-format file, and it generates new data and writes a new file. This
2403 means that you can't go back to an older release without removing the
2406 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2407 work if there are any delivery delays because "one_time" forces the
2408 parent to be marked "delivered", so its unseen clone is never tried
2409 again. For this reason, Exim now forbids the simultaneous setting of
2412 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2413 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2414 entry if you want to know the details. The fix, however, neglected to
2415 consider the case when local delivery batching is involved. The test for
2416 "previously delivered" was not happening when checking to see if an
2417 address could be batched with a previous (undelivered) one; under
2418 certain circumstances this could lead to multiple deliveries to the same
2421 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2422 in its include files, and this causes problems building Exim.
2424 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2425 header_syntax) but Exim was just ignoring anything given after a slash.
2426 In particular, this caused confusion with an attempt to use "verify =
2427 reverse_host_lookup/defer_ok". An error is now given when options are
2428 supplied for verify items that do not have them. (Maybe reverse_host_
2429 lookup should have a defer_ok option, but that's a different point.)
2431 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2432 defined by RFC 821) to 2048, because there were problems with some AUTH
2433 commands, and RFC 1869 says the size should be increased for extended
2434 SMTP commands that take arguments.
2436 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2439 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2440 "unknown" error; now it says that the functionality isn't in the binary.
2442 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2443 an address' error message when a string expansion fails (syntax or
2444 whatever). Otherwise the password may appear in the log. Following change
2445 PH/42 below, there is no longer a chance of it appearing in a bounce
2448 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2450 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2451 of its addresses was ever tried. (Bugzilla bug #2.)
2453 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2454 the result incorrectly in the debug output. (It correctly added a newline
2455 to what was transported.)
2457 TF/01 Added $received_time.
2459 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2460 commented out examples of how to interface to a virus scanner and to
2461 SpamAssassin. Also added commented examples of av_scanner and
2462 spamd_address settings.
2464 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2465 and controls are allowed in which ACLs. There were a couple of minor
2466 errors. Some of the entries in the conditions table (which is a table of
2467 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2468 negation of where the condition IS allowed.
2470 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2472 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2473 header file does not have a version number, so I've had to invent a new
2474 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2475 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2476 radiusclient), but it was contributed by a Radius user.
2478 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2479 files or directories, for OpenSSL.
2481 PH/19 When an Exim process that is running as root has to create an Exim log
2482 file, it does so in a subprocess that runs as exim:exim so as to get the
2483 ownership right at creation (otherwise, other Exim processes might see
2484 the file with the wrong ownership). There was no test for failure of this
2485 fork() call, which would lead to the process getting stuck as it waited
2486 for a non-existent subprocess. Forks do occasionally fail when resources
2487 run out. I reviewed all the other calls to fork(); they all seem to check
2490 PH/20 When checking for unexpected SMTP input at connect time (before writing
2491 the banner), Exim was not dealing correctly with a non-positive return
2492 from the read() function. If the client had disconnected by this time,
2493 the result was a log entry for a synchronization error with an empty
2494 string after "input=" when read() returned zero. If read() returned -1
2495 (an event I could not check), uninitialized data bytes were printed.
2496 There were reports of junk text (parts of files, etc) appearing after
2499 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2501 PH/22 Added support for macro redefinition, and (re)definition in between
2502 driver and ACL definitions.
2504 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2505 forgetting to use the resulting value; it was using the unexpanded value.
2507 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2508 hadn't been configured. The fix is from Juergen Kreileder, who
2509 understands it better than I do:
2511 "Here's what I see happening with three configured cyrus_sasl
2512 authenticators configured (plain, login, cram-md5):
2514 On startup auth_cyrus_sasl_init() gets called for each of these.
2515 This means three calls to sasl_listmech() without a specified mech_list.
2516 => SASL tests which mechs of all available mechs actually work
2517 => three warnings about OTP not working
2518 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2520 With the patch, sasl_listmech() also gets called three times. But now
2521 SASL's mech_list option is set to the server_mech specified in the the
2522 authenticator. Or in other words, the answer from sasl_listmech()
2523 gets limited to just the mech you're testing for (which is different
2525 => the return list contains just 'plain' or 'login', 'cram-md5' or
2526 nothing depending on the value of ob->server_mech.
2528 I've just tested the patch: Authentication still works fine,
2529 unavailable mechs specified in the exim configuration are still
2530 caught, and the auth.log warnings about OTP are gone."
2532 PH/25 When debugging is enabled, the contents of the command line are added
2533 to the debugging output, even when log_selector=+arguments is not
2536 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2537 answer is "GNU", and only if the return is "GNU/something" is the answer
2540 PH/27 $acl_verify_message is now set immediately after the failure of a
2541 verification in an ACL, and so is available in subsequent modifiers. In
2542 particular, the message can be preserved by coding like this:
2544 warn !verify = sender
2545 set acl_m0 = $acl_verify_message
2547 Previously, $acl_verify_message was set only while expanding "message"
2548 and "log_message" when a very denied access.
2550 PH/28 Modified OS/os.c-Linux with
2552 -#ifndef OS_LOAD_AVERAGE
2553 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2555 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2556 nomenclature these days.)
2558 PH/29 Installed patch from the Sieve maintainer that adds the options
2559 sieve_useraddress and sieve_subaddress to the redirect router.
2561 PH/30 In these circumstances:
2562 . Two addresses routed to the same list of hosts;
2563 . First host does not offer TLS;
2564 . First host accepts first address;
2565 . First host gives temporary error to second address;
2566 . Second host offers TLS and a TLS session is established;
2567 . Second host accepts second address.
2568 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2569 and peerdn, if requested) that were in fact used only for the second
2572 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2573 attention to any local part prefix or suffix that was matched by the
2574 router that accepted the address. It now behaves in the same way as it
2575 does for delivery: the affixes are removed from the local part unless
2576 rcpt_include_affixes is set on the transport.
2578 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2579 timeout during the DATA phase of an incoming message.
2581 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2582 applied a patch sanctioned by the Sieve maintainer.
2584 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2585 the uid or gid is negative. A case of a negative gid caused this to be
2586 noticed. The fix allows for either to be negative.
2588 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2589 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2592 PH/36 The change PH/12 above was broken. Fixed it.
2594 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2595 the grounds that routing the same address twice would always produce the
2596 same answer. This might have been true once, but it is certainly no
2597 longer true now. Routing a child address may depend on the previous
2598 routing that produced that child. Some complicated redirection strategies
2599 went wrong when messages had multiple recipients, and made Exim's
2600 behaviour dependent on the order in which the addresses were given.
2602 I have moved the duplicate checking until after the routing is complete.
2603 Exim scans the addresses that are assigned to local and remote
2604 transports, and removes any duplicates. This means that more work will be
2605 done, as duplicates will always all be routed, but duplicates are
2606 presumably rare, so I don't expect this is of any significance.
2608 For deliveries to pipes, files, and autoreplies, the duplicate checking
2609 still happens during the routing process, since they are not going to be
2612 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2613 It corrects a timeout issue with spamd. This is Ian's comment: "The
2614 background is that sometimes spamd either never reads data from a
2615 connection it has accepted, or it never writes response data. The exiscan
2616 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2617 blindly assumes that writes won't block so it may never time out."
2619 PH/39 Allow G after quota size as well as K and M.
2621 PH/40 The value set for $authenticated_id in an authenticator may not contain
2622 binary zeroes or newlines because the value is written to log lines and
2623 to spool files. There was no check on this. Now the value is run through
2624 the string_printing() function so that such characters are converted to
2625 printable escape sequences.
2627 PH/41 $message_linecount is a new variable that contains the total number of
2628 lines in the message. Compare $body_linecount, which is the count for the
2631 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2632 bounce and delay warning messages, except in certain special cases, which
2635 (a) An SMTP error message from a remote host;
2636 (b) A message specified in a :fail: redirection;
2637 (c) A message specified in a "fail" command in a system filter;
2638 (d) A message specified in a FAIL return from the queryprogram router;
2639 (e) A message specified by the cannot_route_message router option.
2641 In these cases only, Exim does include the error details in bounce and
2642 warning messages. There are also a few cases where bland messages such
2643 as "unrouteable address" or "local delivery error" are given.
2645 PH/43 $value is now also set for the "else" part of a ${run expansion.
2647 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2648 being worked on, but at least Exim now implements the latest version to
2651 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2652 process to complete was treated as a delivery failure, a timeout while
2653 writing the message to the pipe was logged, but erroneously treated as a
2654 successful delivery. Such timeouts include transport filter timeouts. For
2655 consistency with the overall process timeout, these timeouts are now
2656 treated as errors, giving rise to delivery failures by default. However,
2657 there is now a new Boolean option for the pipe transport called
2658 timeout_defer, which, if set TRUE, converts the failures into defers for
2659 both kinds of timeout. A transport filter timeout is now identified in
2662 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2663 systems where "make" and "gmake" are different, calling "gmake" at top
2664 level broke things. I've arranged for the value of $(MAKE) to be passed
2665 from the Makefile to this script so that it can call the same version of
2669 A note about Exim versions 4.44 and 4.50
2670 ----------------------------------------
2672 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2673 changes of various kinds. As a consequence, a big documentation update was
2674 needed. This delayed the release for rather longer than seemed good, especially
2675 in the light of a couple of (minor) security issues. Therefore, the changes
2676 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2677 release. So 4.44 and 4.50 are in effect two different branches that both start
2680 I have left the 4.50 change log unchanged; it contains all the changes since
2681 4.43. The change log for 4.44 is below; many of its items are identical to
2682 those for 4.50. This seems to be the most sensible way to preserve the
2683 historical information.
2689 1. Minor wording change to the doc/README.SIEVE file.
2691 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2692 computation of the current number of files was incorrect.
2694 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2695 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2698 4. Give more explanation in the error message when the command for a transport
2699 filter fails to execute.
2701 5. There are several places where Exim runs a non-Exim command in a
2702 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2703 was being done only for the command run by the queryprogram router. It is
2704 now done for all such subprocesses. The other cases are: ${run, transport
2705 filters, and the commands run by the lmtp and pipe transports.
2707 6. Added CONFIGURE_GROUP build-time option.
2709 7. Some older OS have a limit of 256 on the maximum number of file
2710 descriptors. Exim was using setrlimit() to set 1000 as a large value
2711 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2712 systems. I've change it so that if it can't get 1000, it tries for 256.
2714 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2715 was an oversight, and furthermore, ever since the addition of extra
2716 controls (e.g. 4.43/32), the checks on when to allow different forms of
2717 "control" were broken. There should now be diagnostics for all cases when a
2718 control that does not make sense is encountered.
2720 9. Added the /retain_sender option to "control=submission".
2722 10. $recipients is now available in the predata ACL (oversight).
2724 11. Tidy the search cache before the fork to do a delivery from a message
2725 received from the command line. Otherwise the child will trigger a lookup
2726 failure and thereby defer the delivery if it tries to use (for example) a
2727 cached ldap connection that the parent has called unbind on.
2729 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2730 of $address_data from the recipient verification was clobbered by the
2731 sender verification.
2733 13. The value of address_data from a sender verification is now available in
2734 $sender_address_data in subsequent conditions in the ACL statement.
2736 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2738 15. Added a new option "connect=<time>" to callout options, to set a different
2741 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2742 was its contents. (It was OK if the option was not defined at all.)
2744 17. A "Completed" log line is now written for messages that are removed from
2745 the spool by the -Mrm option.
2747 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2748 information about exactly what failed.
2750 19. Added -dd to debug only the daemon process.
2752 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2753 handles timeouts, both on the server side and network timeouts. Renamed the
2754 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2756 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2757 It is now set to "smtps".
2759 22. $host_address is now set to the target address during the checking of
2760 ignore_target_hosts.
2762 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2763 being passed; this would have caused $sender_host_name to have been used if
2764 matching the list had actually called for a host name (not very likely,
2765 since this list is usually IP addresses). A host name is now passed as
2768 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2769 code that specifies a non-restarting handler (typically sigaction() in
2770 modern systems) in an attempt to fix a rare and obscure crash bug.
2772 25. Narrowed the window for a race in the daemon that could cause it to ignore
2773 SIGCHLD signals. This is not a major problem, because they are used only to
2774 wake it up if nothing else does.
2776 26. A malformed maildirsize file could cause Exim to calculate negative values
2777 for the mailbox size or file count. Odd effects could occur as a result.
2778 The maildirsize information is now recalculated if the size or filecount
2781 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2782 support for a long time. Removed HAVE_SYS_VFS_H.
2784 28. Installed the latest version of exipick from John Jetmore.
2786 29. In an address list, if the pattern was not a regular expression, an empty
2787 subject address (from a bounce message) matched only if the pattern was an
2788 empty string. Non-empty patterns were not even tested. This was the wrong
2789 because it is perfectly reasonable to use an empty address as part of a
2790 database query. An empty address is now tested by patterns that are
2791 lookups. However, all the other forms of pattern expect the subject to
2792 contain a local part and a domain, and therefore, for them, an empty
2793 address still always fails if the pattern is not itself empty.
2795 30. Exim went into a mad DNS loop when attempting to do a callout where the
2796 host was specified on an smtp transport, and looking it up yielded more
2797 than one IP address.
2799 31. Re-factored the code for checking spool and log partition space into a
2800 function that finds that data and another that does the check. The former
2801 is then used to implement four new variables: $spool_space, $log_space,
2802 $spool_inodes, and $log_inodes.
2804 32. The RFC2047 encoding function was originally intended for short strings
2805 such as real names; it was not keeping to the 75-character limit for
2806 encoded words that the RFC imposes. It now respects the limit, and
2807 generates multiple encoded words if necessary. To be on the safe side, I
2808 have increased the buffer size for the ${rfc2047: expansion operator from
2811 33. It is now permitted to omit both strings after an "if" condition; if the
2812 condition is true, the result is "true". As before, when the second string
2813 is omitted, a false condition yields an empty string. This makes it less
2814 cumbersome to write custom ACL and router conditions.
2816 34. Failure to deliver a bounce message always caused it to be frozen, even if
2817 there was an errors_to setting on the router. The errors_to setting is now
2820 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2821 canonical form (fully expanded) before being placed in
2822 $sender_host_address.
2824 36. The table in the code that translates DNS record types into text (T_A to
2825 "A" for instance) was missing entries for NS and CNAME. It is just possible
2826 that this could have caused confusion if both these types were looked up
2827 for the same domain, because the text type is used as part of Exim's
2828 per-process caching. But the chance of anyone hitting this buglet seems
2831 37. The dnsdb lookup has been extended in a number of ways.
2833 (1) There is a new type, "zns", which walks up the domain tree until it
2834 finds some nameserver records. It should be used with care.
2836 (2) There is a new type, "mxh", which is like "mx" except that it returns
2837 just the host names, not the priorities.
2839 (3) It is now possible to give a list of domains (or IP addresses) to be
2840 looked up. The behaviour when one of the lookups defers can be
2841 controlled by a keyword.
2843 (4) It is now possible to specify the separator character for use when
2844 multiple records are returned.
2846 38. The dnslists ACL condition has been extended: it is now possible to supply
2847 a list of IP addresses and/or domains to be looked up in a particular DNS
2850 39. Added log_selector=+queue_time_overall.
2852 40. When running the queue in the test harness, wait just a tad after forking a
2853 delivery process, to get repeatability of debugging output.
2855 41. Include certificate and key file names in error message when GnuTLS fails
2856 to set them up, because the GnuTLS error message doesn't include the name
2857 of the failing file when there is a problem reading it.
2859 42. Allow both -bf and -bF in the same test run.
2861 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2863 44. The "Exiscan patch" is now merged into the mainline Exim source.
2865 45. Sometimes the final signoff response after QUIT could fail to get
2866 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2867 < 0 before doing a fflush(). This bug looks as though it goes back to the
2868 introduction of TLS in release 3.20, but "sometimes" must have been rare
2869 because the tests only now provoked it.
2871 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2872 (this can affect the format of dates).
2874 47. exim_tidydb, when checking for the continued existence of a message for
2875 which it has found a message-specific retry record, was not finding
2876 messages that were in split spool directories. Consequently, it was
2877 deleting retry records that should have stayed in existence.
2879 48. Steve fixed some bugs in eximstats.
2881 49. The SPA authentication driver was not abandoning authentication and moving
2882 on to the next authenticator when an expansion was forced to fail,
2883 contradicting the general specification for all authenticators. Instead it
2884 was generating a temporary error. It now behaves as specified.
2886 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2887 (the order specifies the preference for clients). The order is now AES256,
2888 AES128, 3DES, ARCFOUR128.
2890 51. Small patch to Sieve code - explicitly set From: when generating an
2893 52. Exim crashed if a remote delivery caused a very long error message to be
2894 recorded - for instance if somebody sent an entire SpamAssassin report back
2895 as a large number of 550 error lines. This bug was coincidentally fixed by
2896 increasing the size of one of Exim's internal buffers (big_buffer) that
2897 happened as part of the Exiscan merge. However, to be on the safe side, I
2898 have made the code more robust (and fixed the comments that describe what
2901 53. Now that there can be additional text after "Completed" in log lines (if
2902 the queue_time_overall log selector is set), a one-byte patch to exigrep
2903 was needed to allow it to recognize "Completed" as not the last thing in
2906 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2907 patch that reportedly fixes this has been added. I am not expert enough to
2908 create a test for it. This is what the patch creator wrote:
2910 "I found a little strange behaviour of ldap code when working with
2911 Windows 2003 AD Domain, where users was placed in more than one
2912 Organization Units. When I tried to give exim partial DN, the exit code
2913 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2914 But simultaneously result of request was absolutely normal ldap result,
2915 so I produce this patch..."
2917 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2918 REFERENCE, so I have modified the code to exclude the patch when that macro
2921 55. Some experimental protocols are using DNS PTR records for new purposes. The
2922 keys for these records are domain names, not reversed IP addresses. The
2923 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2924 leaves it alone. Component reversal etc. now happens only for IP addresses.
2927 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2929 57. Double the size of the debug message buffer (to 2048) so that more of very
2930 long debug lines gets shown.
2932 58. The exicyclog utility now does better if the number of log files to keep
2933 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2935 59. Two changes related to the smtp_active_hostname option:
2937 (1) $smtp_active_hostname is now available as a variable.
2938 (2) The default for smtp_banner uses $smtp_active_hostname instead
2939 of $primary_hostname.
2941 60. The host_aton() function is supposed to be passed a string that is known
2942 to be a valid IP address. However, in the case of IPv6 addresses, it was
2943 not checking this. This is a hostage to fortune. Exim now panics and dies
2944 if the condition is not met. A case was found where this could be provoked
2945 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2946 components; fortuitously, this particular loophole had already been fixed
2947 by change 4.50/55 above.
2949 If there are any other similar loopholes, the new check in host_aton()
2950 itself should stop them being exploited. The report I received stated that
2951 data on the command line could provoke the exploit when Exim was running as
2952 exim, but did not say which command line option was involved. All I could
2953 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2954 running as the user.
2957 61. There was a buffer overflow vulnerability in the SPA authentication code
2958 (which came originally from the Samba project). I have added a test to the
2959 spa_base64_to_bits() function which I hope fixes it.
2962 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2963 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2965 63. The daemon start-up calls getloadavg() while still root for those OS that
2966 need the first call to be done as root, but it missed one case: when
2967 deliver_queue_load_max is set with deliver_drop_privilege. This is
2968 necessary for the benefit of the queue runner, because there is no re-exec
2969 when deliver_drop_privilege is set.
2971 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2972 This has been fixed.
2974 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2975 was in use, was not putting the data itself into the right store pool;
2976 consequently, it could be overwritten for a subsequent message in the same
2977 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2980 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2982 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2983 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2984 the function were treating the return as a boolean value, which happened to
2985 work because 0=false and not-0=true, but is not correct code.
2987 68. The host_aton() function was not handling scoped IPv6 addresses (those
2988 with, for example, "%eth0" on the end) correctly.
2990 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2991 negated items (that is, ~something) in unsigned ints. Some compilers
2992 apparently mutter when there is no cast.
2994 70. If an address verification called from an ACL failed, and did not produce a
2995 user-specific message (i.e. there was only a "system" message), nothing was
2996 put in $acl_verify_message. In this situation, it now puts the system
2999 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3000 session; change 4.31/43 added the unwanted input to the log line - except
3001 that it did not do this in the start of session case. It now does.
3003 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3004 This is wrong and can cause the other end to generate a synchronization
3005 error if it is another Exim or anything else that does the synchronization
3006 check. A QUIT command is no longer sent after a timeout.
3008 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3009 during host lookups.
3011 74. The defer_ok option of callout verification was not working if it was used
3012 when verifying addresses in header lines, that is, for this case:
3014 verify = header_sender/callout=defer_ok
3016 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3017 those file descriptors could be used for SMTP connections. If anything
3018 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3019 could be sent to the SMTP client, causing chaos. The daemon now opens
3020 stdin, stdout, and stderr to /dev/null when it puts itself into the
3023 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3024 log by default. The user can override this with suitable Perl magic.
3026 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3027 the log message when discard triggers, was not working for the DATA ACL or
3028 for the non-SMTP ACL.
3030 78. Error message wording change in sieve.c.
3032 79. If smtp_accept_max_per_host was set, the number of connections could be
3033 restricted to fewer than expected, because the daemon was trying to set up
3034 a new connection before checking whether the processes handling previous
3035 connections had finished. The check for completed processes is now done
3036 earlier. On busy systems, this bug wouldn't be noticed because something
3037 else would have woken the daemon, and it would have reaped the completed
3040 80. If a message was submitted locally by a user whose login name contained one
3041 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3042 It caused a spool format error. I have fixed the spool reading code. A
3043 related problem was that the "from" clause in the Received: line became
3044 illegal because of the space(s). It is now covered by ${quote_local_part.
3046 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3049 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3051 83. Patch from Sieve maintainer to fix unterminated string problem in
3052 "vacation" handling.
3054 84. Some minor changes to the Linux configuration files to help with other
3055 OS variants using glibc.
3057 85. One more patch for Sieve to update vacation handling to latest spec.
3060 ----------------------------------------------------
3061 See the note above about the 4.44 and 4.50 releases.
3062 ----------------------------------------------------
3068 1. Change 4.43/35 introduced a bug that caused file counts to be
3069 incorrectly computed when quota_filecount was set in an appendfile
3072 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3073 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3076 3. Give more explanation in the error message when the command for a transport
3077 filter fails to execute.
3079 4. There are several places where Exim runs a non-Exim command in a
3080 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3081 was being done only for the command run by the queryprogram router. It is
3082 now done for all such subprocesses. The other cases are: ${run, transport
3083 filters, and the commands run by the lmtp and pipe transports.
3085 5. Some older OS have a limit of 256 on the maximum number of file
3086 descriptors. Exim was using setrlimit() to set 1000 as a large value
3087 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3088 systems. I've change it so that if it can't get 1000, it tries for 256.
3090 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3091 was an oversight, and furthermore, ever since the addition of extra
3092 controls (e.g. 4.43/32), the checks on when to allow different forms of
3093 "control" were broken. There should now be diagnostics for all cases when a
3094 control that does not make sense is encountered.
3096 7. $recipients is now available in the predata ACL (oversight).
3098 8. Tidy the search cache before the fork to do a delivery from a message
3099 received from the command line. Otherwise the child will trigger a lookup
3100 failure and thereby defer the delivery if it tries to use (for example) a
3101 cached ldap connection that the parent has called unbind on.
3103 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3104 of $address_data from the recipient verification was clobbered by the
3105 sender verification.
3107 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3108 was its contents. (It was OK if the option was not defined at all.)
3110 11. A "Completed" log line is now written for messages that are removed from
3111 the spool by the -Mrm option.
3113 12. $host_address is now set to the target address during the checking of
3114 ignore_target_hosts.
3116 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3117 being passed; this would have caused $sender_host_name to have been used if
3118 matching the list had actually called for a host name (not very likely,
3119 since this list is usually IP addresses). A host name is now passed as
3122 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3123 code that specifies a non-restarting handler (typically sigaction() in
3124 modern systems) in an attempt to fix a rare and obscure crash bug.
3126 15. Narrowed the window for a race in the daemon that could cause it to ignore
3127 SIGCHLD signals. This is not a major problem, because they are used only to
3128 wake it up if nothing else does.
3130 16. A malformed maildirsize file could cause Exim to calculate negative values
3131 for the mailbox size or file count. Odd effects could occur as a result.
3132 The maildirsize information is now recalculated if the size or filecount
3135 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3136 support for a long time. Removed HAVE_SYS_VFS_H.
3138 18. Updated exipick to current release from John Jetmore.
3140 19. Allow an empty sender to be matched against a lookup in an address list.
3141 Previously the only cases considered were a regular expression, or an
3144 20. Exim went into a mad DNS lookup loop when doing a callout where the
3145 host was specified on the transport, if the DNS lookup yielded more than
3148 21. The RFC2047 encoding function was originally intended for short strings
3149 such as real names; it was not keeping to the 75-character limit for
3150 encoded words that the RFC imposes. It now respects the limit, and
3151 generates multiple encoded words if necessary. To be on the safe side, I
3152 have increased the buffer size for the ${rfc2047: expansion operator from
3155 22. Failure to deliver a bounce message always caused it to be frozen, even if
3156 there was an errors_to setting on the router. The errors_to setting is now
3159 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3160 canonical form (fully expanded) before being placed in
3161 $sender_host_address.
3163 24. Updated eximstats to version 1.33
3165 25. Include certificate and key file names in error message when GnuTLS fails
3166 to set them up, because the GnuTLS error message doesn't include the name
3167 of the failing file when there is a problem reading it.
3169 26. Expand error message when OpenSSL has problems setting up cert/key files.
3172 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3173 (this can affect the format of dates).
3175 28. exim_tidydb, when checking for the continued existence of a message for
3176 which it has found a message-specific retry record, was not finding
3177 messages that were in split spool directories. Consequently, it was
3178 deleting retry records that should have stayed in existence.
3180 29. eximstats updated to version 1.35
3181 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3182 1.35 - bugfix such that pie charts by volume are generated correctly
3184 30. The SPA authentication driver was not abandoning authentication and moving
3185 on to the next authenticator when an expansion was forced to fail,
3186 contradicting the general specification for all authenticators. Instead it
3187 was generating a temporary error. It now behaves as specified.
3189 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3190 (the order specifies the preference for clients). The order is now AES256,
3191 AES128, 3DES, ARCFOUR128.
3193 31. Small patch to Sieve code - explicitly set From: when generating an
3196 32. Exim crashed if a remote delivery caused a very long error message to be
3197 recorded - for instance if somebody sent an entire SpamAssassin report back
3198 as a large number of 550 error lines. This bug was coincidentally fixed by
3199 increasing the size of one of Exim's internal buffers (big_buffer) that
3200 happened as part of the Exiscan merge. However, to be on the safe side, I
3201 have made the code more robust (and fixed the comments that describe what
3204 33. Some experimental protocols are using DNS PTR records for new purposes. The
3205 keys for these records are domain names, not reversed IP addresses. The
3206 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3207 leaves it alone. Component reversal etc. now happens only for IP addresses.
3210 34. The host_aton() function is supposed to be passed a string that is known
3211 to be a valid IP address. However, in the case of IPv6 addresses, it was
3212 not checking this. This is a hostage to fortune. Exim now panics and dies
3213 if the condition is not met. A case was found where this could be provoked
3214 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3215 components; fortuitously, this particular loophole had already been fixed
3216 by change 4.50/55 or 4.44/33 above.
3218 If there are any other similar loopholes, the new check in host_aton()
3219 itself should stop them being exploited. The report I received stated that
3220 data on the command line could provoke the exploit when Exim was running as
3221 exim, but did not say which command line option was involved. All I could
3222 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3223 running as the user.
3226 35. There was a buffer overflow vulnerability in the SPA authentication code
3227 (which came originally from the Samba project). I have added a test to the
3228 spa_base64_to_bits() function which I hope fixes it.
3231 36. The daemon start-up calls getloadavg() while still root for those OS that
3232 need the first call to be done as root, but it missed one case: when
3233 deliver_queue_load_max is set with deliver_drop_privilege. This is
3234 necessary for the benefit of the queue runner, because there is no re-exec
3235 when deliver_drop_privilege is set.
3237 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3238 was in use, was not putting the data itself into the right store pool;
3239 consequently, it could be overwritten for a subsequent message in the same
3240 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3243 38. Sometimes the final signoff response after QUIT could fail to get
3244 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3245 < 0 before doing a fflush(). This bug looks as though it goes back to the
3246 introduction of TLS in release 3.20, but "sometimes" must have been rare
3247 because the tests only now provoked it.
3253 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3254 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3255 Now it returns an integer. A number of calls were still expecting a T/F
3256 return. Fortuitously, in all cases, the tests worked in OK situations,
3257 which is the norm. However, things would have gone wrong on any write
3258 failures on the smtp file descriptor. This function is used when sending
3259 messages over SMTP and also when doing verify callouts.
3261 2. When Exim is called to do synchronous delivery of a locally submitted
3262 message (the -odf or -odi options), it no longer closes stderr before doing
3265 3. Implemented the mua_wrapper option.
3267 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3269 5. Implemented the functions header_remove(), header_testname(),
3270 header_add_at_position(), and receive_remove_recipient(), and exported them
3273 6. If an ACL "warn" statement specified the addition of headers, Exim already
3274 inserted X-ACL-Warn: at the start if there was no header name. However, it
3275 was not making this test for the second and subsequent header lines if
3276 there were newlines in the string. This meant that an invalid header could
3277 be inserted if Exim was badly configured.
3279 7. Allow an ACL "warn" statement to add header lines at the start or after all
3280 the Received: headers, as well as at the end.
3282 8. Added the rcpt_4xx retry error code.
3284 9. Added postmaster_mailfrom=xxx to callout verification option.
3286 10. Added mailfrom=xxxx to the callout verification option, for verify=
3289 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3290 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3292 12. Inserted some casts to stop certain compilers warning when using pointer
3293 differences as field lengths or precisions in printf-type calls (mostly
3294 affecting debugging statements).
3296 13. Added optional readline() support for -be (dynamically loaded).
3298 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3299 same clock tick as a message's arrival, so that its received time was the
3300 same as the "first fail" time on the retry record, and that message
3301 remained on the queue past the ultimate address timeout, every queue runner
3302 would try a delivery (because it was past the ultimate address timeout) but
3303 after another failure, the ultimate address timeout, which should have then
3304 bounced the address, did not kick in. This was a "< instead of <=" error;
3305 in most cases the first failure would have been in the next clock tick
3306 after the received time, and all would be well.
3308 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3309 being recognized when the domain list was tested by the match_domain
3310 condition in an expansion string.
3312 16. Added the ${str2b64: operator.
3314 17. Exim was always calling setrlimit() to set a large limit for the number of
3315 processes, without checking whether the existing limit was already
3316 adequate. (It did check for the limit on file descriptors.) Furthermore,
3317 errors from getrlimit() and setrlimit() were being ignored. Now they are
3318 logged to the main and panic logs, but Exim does carry on, to try to do its
3319 job under whatever limits there are.
3321 18. Imported PCRE 5.0.
3323 19. Trivial typo in log message " temporarily refused connection" (the leading
3326 20. If the log selector return_path_on_delivery was set and an address was
3327 redirected to /dev/null, the delivery process crashed because it assumed
3328 that a return path would always be set for a "successful" delivery. In this
3329 case, the whole delivery is bypassed as an optimization, and therefore no
3332 21. Internal re-arrangement: the function for sending a challenge and reading
3333 a response while authentication was assuming a zero-terminated challenge
3334 string. It's now changed to take a pointer and a length, to allow for
3335 binary data in such strings.
3337 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3339 23. Exim was not respecting finduser_retries when seeking the login of the
3340 uid under which it was called; it was always trying 10 times. (The default
3341 setting of finduser_retries is zero.) Also, it was sleeping after the final
3342 failure, which is pointless.
3344 24. Implemented tls_on_connect_ports.
3346 25. Implemented acl_smtp_predata.
3348 26. If the domain in control=submission is set empty, Exim assumes that the
3349 authenticated id is a complete email address when it generates From: or
3350 Sender: header lines.
3352 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3353 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3354 chown and chgrp in /bin and hostname in /usr/bin.
3356 28. Exim was keeping the "process log" file open after each use, just as it
3357 does for the main log. This opens the possibility of it remaining open for
3358 long periods when the USR1 signal hits a daemon. Occasional processlog
3359 errors were reported, that could have been caused by this. Anyway, it seems
3360 much more sensible not to leave this file open at all, so that is what now
3363 29. The long-running daemon process does not normally write to the log once it
3364 has entered its main loop, and it closes the log before doing so. This is
3365 so that log files can straightforwardly be renamed and moved. However,
3366 there are a couple of unusual error situations where the daemon does write
3367 log entries, and I had neglected to close the log afterwards.
3369 30. The text of an SMTP error response that was received during a remote
3370 delivery was being truncated at 512 bytes. This is too short for some of
3371 the long messages that one sometimes sees. I've increased the limit to
3374 31. It is now possible to make retry rules that apply only when a message has a
3375 specific sender, in particular, an empty sender.
3377 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3378 it possible to be selective about when SMTP synchronization is enforced.
3380 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3382 32. Implemented hosts_connection_nolog.
3384 33. Added an ACL for QUIT.
3386 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3389 35. Added mailbox_size and mailbox_filecount to appendfile.
3391 36. Added control = no_multiline_responses to ACLs.
3393 37. There was a bug in the logic of the code that waits for the clock to tick
3394 in the case where the clock went backwards by a substantial amount such
3395 that the microsecond fraction of "now" was more than the microsecond
3396 fraction of "then" (but the whole seconds number was less).
3398 38. Added support for the libradius Radius client library this is found on
3399 FreeBSD (previously only the radiusclient library was supported).
3405 1. When certain lookups returned multiple values in the form name=value, the
3406 quoting of the values was not always being done properly. Specifically:
3407 (a) If the value started with a double quote, but contained no whitespace,
3409 (b) If the value contained whitespace other than a space character (i.e.
3410 tabs or newlines or carriage returns) it was not quoted.
3411 This fix has been applied to the mysql and pgsql lookups by writing a
3412 separate quoting function and calling it from the lookup code. The fix
3413 should probably also be applied to nisplus, ibase and oracle lookups, but
3414 since I cannot test any of those, I have not disturbed their existing code.
3416 2. A hit in the callout cache for a specific address caused a log line with no
3417 reason for rejecting RCPT. Now it says "Previous (cached) callout
3418 verification failure".
3420 3. There was an off-by-one bug in the queryprogram router. An over-long
3421 return line was truncated at 256 instead of 255 characters, thereby
3422 overflowing its buffer with the terminating zero. As well as fixing this, I
3423 have increased the buffer size to 1024 (and made a note to document this).
3425 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3426 when Exim is waiting for an SMTP response from a remote server, Exim
3427 restarts its select() call on the socket, thereby resetting its timeout.
3428 This is not a problem when such interrupts are rare. Somebody set up a cron
3429 job to run exiwhat every 2 minutes, which is less than the normal select()
3430 timeout (5 or 10 minutes). This meant that the select() timeout never
3431 kicked in because it was always reset. I have fixed this by comparing the
3432 time when an interrupt arrives with the time at the start of the first call
3433 to select(). If more time than the timeout has elapsed, the interrupt is
3434 treated as a timeout.
3436 5. Some internal re-factoring in preparation for the addition of Sieve
3437 extensions (by MH). In particular, the "personal" test is moved to a
3438 separate function, and given an option for scanning Cc: and Bcc: (which is
3439 not set for Exim filters).
3441 6. When Exim created an email address using the login of the caller as the
3442 local part (e.g. when creating a From: or Sender: header line), it was not
3443 quoting the local part when it contained special characters such as @.
3445 7. Installed new OpenBSD configuration files.
3447 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3448 try to make them clearer.
3450 9. Callout options, other than the timeout value, were being ignored when
3451 verifying sender addresses in header lines. For example, when using
3453 verify = header_sender/callout=no_cache
3455 the cache was (incorrectly) being used.
3457 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3458 only the Cygwin environment.
3460 11. When return_path_on_delivery was set as a log selector, if different remote
3461 addresses in the same message used different return paths and parallel
3462 remote delivery occurred, the wrong values would sometimes be logged.
3463 (Whenever a remote delivery process finished, the return path value from
3464 the most recently started remote delivery process was logged.)
3466 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3467 lines when AUTH and/or TLS are in use. This is the "received protocol"
3468 field. Exim used to use "asmtp" for authenticated SMTP, without any
3469 indication (in the protocol name) for TLS use. Now it follows the RFC and
3470 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3471 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3472 names appear in log lines as well as in Received: header lines.
3474 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3475 extensions, and comparison tests, and to fix some bugs.
3477 14. Changes to the "personal" filter test:
3479 (1) The test was buggy in that it was just doing the equivalent of
3480 "contains" tests on header lines. For example, if a user's address was
3481 anne@some.where, the "personal" test would incorrectly be true for
3483 To: susanne@some.where
3485 This test is now done by extracting each address from the header in turn,
3486 and checking the entire address. Other tests that are part of "personal"
3487 are now done using regular expressions (for example, to check local parts
3488 of addresses in From: header lines).
3490 (2) The list of non-personal local parts in From: addresses has been
3491 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3492 taken from the Sieve specification recommendations.
3494 (3) If the message contains any header line starting with "List-" it is
3495 treated as non-personal.
3497 (4) The test for "circular" in the Subject: header line has been removed
3498 because it now seems ill-conceived.
3500 15. Minor typos in src/EDITME comments corrected.
3502 16. Installed latest exipick from John Jetmore.
3504 17. If headers_add on a router specified a text string that was too long for
3505 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3506 of string_sprintf() is now avoided.
3508 18. $message_body_size was not set (it was always zero) when running the DATA
3509 ACL and the local_scan() function.
3511 19. For the "mail" command in an Exim filter, no default was being set for
3512 the once_repeat time, causing a random time value to be used if "once" was
3513 specified. (If the value happened to be <= 0, no repeat happened.) The
3514 default is now 0s, meaning "never repeat". The "vacation" command was OK
3515 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3516 (I found it when inspecting the code).
3518 20. There is now an overall timeout for performing a callout verification. It
3519 defaults to 4 times the callout timeout, which applies to individual SMTP
3520 commands during the callout. The overall timeout applies when there is more
3521 than one host that can be tried. The timeout is checked before trying the
3522 next host. This prevents very long delays if there are a large number of
3523 hosts and all are timing out (e.g. when the network connections are timing
3524 out). The value of the overall timeout can be changed by specifying an
3525 additional sub-option for "callout", called "maxwait". For example:
3527 verify = sender/callout=5s,maxwait=20s
3529 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3530 to the end before writing, but this should make it even safer).
3532 22. Exim was forgetting that it had advertised PIPELINING for the second and
3533 subsequent messages on an SMTP connection. It was also not resetting its
3534 memory on STARTTLS and an internal HELO.
3536 23. When Exim logs an SMTP synchronization error within a session, it now
3537 records whether PIPELINING has been advertised or not.
3539 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3540 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3541 rather than long int.
3543 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3545 26. Added the never_mail option to autoreply.
3551 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3552 crash if the getsockname() call failed; this can happen if a connection is
3553 closed very soon after it is established. The problem was simply in the
3554 order in which certain operations were done, causing Exim to try to write
3555 to the SMTP stream before it had set up the file descriptor. The bug has
3556 been fixed by making things happen in the correct order.
3562 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3563 before the connection was closed, thus losing the rejection response.
3565 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3566 some early Solaris releases, but causes trouble in current releases where
3567 socklen_t is defined.
3569 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3572 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3575 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3576 that releasing the top of it at the end releases what was used for sub-
3577 expansions (unless the block got too big). However, discard this block if
3578 the first thing is a variable or header, so that we can use its block when
3579 it is dynamic (useful for very large $message_headers, for example).
3581 6. Lookups now cache *every* query, not just the most recent. A new, separate
3582 store pool is used for this. It can be recovered when all lookup caches are
3583 flushed. Lookups now release memory at the end of their result strings.
3584 This has involved some general refactoring of the lookup sources.
3586 7. Some code has been added to the store_xxx() functions to reduce the amount
3587 of flapping under certain conditions.
3589 8. log_incoming_interface used to affect only the <= reception log lines. Now
3590 it causes the local interface and port to be added to several more SMTP log
3591 lines, for example "SMTP connection from", and rejection lines.
3593 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3595 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3597 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3598 could be overwritten at the end of the current message (or the start of a
3599 new message if it was set in a HELO ACL). The value is now preserved for
3600 the duration of the SMTP connection.
3602 12. If a transport had a headers_rewrite setting, and a matching header line
3603 contained an unqualified address, that address was qualified, even if it
3604 did not match any rewriting rules. The underlying bug was that the values
3605 of the flags that permit the existence of unqualified sender and recipient
3606 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3607 non-local messages, and by -bnq for local messages) were not being
3608 preserved with the message after it was received.
3610 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3611 "next input=" as part of the text comprising the host identity instead of
3612 the correct text. The code was using the same buffer for two different
3613 strings. However, depending on which order the printing function evaluated
3614 its arguments, the bug did not always show up. Under Linux, for example, my
3615 test suite worked just fine.
3617 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3618 doesn't work with some older versions of Perl. It has been changed to "my",
3619 which in any case is probably the better facility to use.
3621 15. A really picky compiler found some instances of statements for creating
3622 error messages that either had too many or two few arguments for the format
3625 16. The size of the buffer for calls to the DNS resolver has been increased
3626 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3627 for addresses that have a lot of PTR records. This alleviates a problem; it
3628 does not fully solve it.
3630 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3631 buffer now truncates the list and logs the incident, which is the same
3632 action as happens when Exim is looking up a host name and its aliases.
3633 Previously in this situation something unpredictable would happen;
3634 sometimes it was "internal error: store_reset failed".
3636 18. If a server dropped the connection unexpectedly when an Exim client was
3637 using GnuTLS and trying to read a response, the client delivery process
3638 crashed while trying to generate an error log message.
3640 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3641 string, for example:
3643 warn message = H1: something\nH2: something
3645 the text was added as a single header line from Exim's point of view
3646 though it ended up OK in the delivered message. However, searching for the
3647 second and subsequent header lines using $h_h2: did not work. This has been
3648 fixed. Similarly, if a system filter added multiple headers in this way,
3649 the routers could not see them.
3651 20. Expanded the error message when iplsearch is called with an invalid key to
3652 suggest using net-iplsearch in a host list.
3654 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3655 ACLs are no longer actually imposed (and a message to that effect is
3658 22. If a "gecos" field in a passwd entry contained escaped characters, in
3659 particular, if it contained a \" sequence, Exim got it wrong when building
3660 a From: or a Sender: header from that name. A second bug also caused
3661 incorrect handling when an unquoted " was present following a character
3662 that needed quoting.
3664 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3665 was not being matched caselessly.
3667 24. Arranged for all hyphens in the exim.8 source to be escaped with
3670 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3671 properly. Recipient callout cache records were still being keyed to include
3672 the sender, even when use_sender was set false. This led to far more
3673 callouts that were necessary. The sender is no longer included in the key
3674 when use_sender is false.
3676 26. Added "control = submission" modifier to ACLs.
3678 27. Added the ${base62d: operator to decode base 62 numbers.
3680 28. dnsdb lookups can now access SRV records.
3682 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3683 the configuration file.
3685 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3686 (-v) mode. This makes the output for a verbose queue run more intelligible.
3688 31. Added a use_postmaster feature to recipient callouts.
3690 32. Added the $body_zerocount variable, containing the number of binary zero
3691 bytes in the message body.
3693 33. The time of last modification of the "new" subdirectory is now used as the
3694 "mailbox time last read" when there is a quota error for a maildir
3697 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3699 35. Added +ignore_unknown as a special item in host lists.
3701 36. Code for decoding IPv6 addresses in host lists is now included, even if
3702 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3703 address was recognized as an IP address, but was then not correctly decoded
3704 into binary, causing unexpected and incorrect effects when compared with
3711 1. Very minor rewording of debugging text in manualroute to say "list of
3712 hosts" instead of "hostlist".
3714 2. If verify=header_syntax was set, and a header line with an unqualified
3715 address (no domain) and a large number of spaces between the end of the
3716 name and the colon was received, the reception process suffered a buffer
3717 overflow, and (when I tested it) crashed. This was caused by some obsolete
3718 code that should have been removed. The fix is to remove it!
3720 3. When running in the test harness, delay a bit after writing a bounce
3721 message to get a bit more predictability in the log output.
3723 4. Added a call to search_tidyup() just before forking a reception process. In
3724 theory, someone could use a lookup in the expansion of smtp_accept_max_
3725 per_host which, without the tidyup, could leave open a database connection.
3727 5. Added the variables $recipient_data and $sender_data which get set from a
3728 lookup success in an ACL "recipients" or "senders" condition, or a router
3729 "senders" option, similar to $domain_data and $local_part_data.
3731 6. Moved the writing of debug_print from before to after the "senders" test
3734 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3735 problems for message scanning, either using a data ACL, or using
3736 local_scan() because the Received: header was not generated till after they
3737 were called (in order to set the time as the time of reception completion).
3738 I have revised the way this works. The header is now generated after the
3739 body is received, but before the ACL or local_scan() are called. After they
3740 are run, the timestamp in the header is updated.
3746 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3747 before starting a queue runner without re-exec. This happened only when
3748 deliver_drop_privilege was set or when the Exim user was set to root. The
3749 effect of the bug was that timeouts during subsequent deliveries caused
3750 crashes instead of being properly handled. The handler is now left at its
3751 default (and expected) setting.
3753 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3754 message, again when deliver_drop_privilege is set or Exim is run as root.
3755 The bug described in (1) was not present in this case, but the tidying up
3756 of the other signals was missing. I have made the two cases consistent.
3758 3. The ignore_target_hosts setting on a manualroute router was being ignored
3759 for hosts that were looked up using the /MX notation.
3761 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3764 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3765 operated on the sender address. After changing the $sender_address to <>
3766 for the sender address verify, Exim was re-instated it as the original
3767 (before rewriting) address, but remembering that it had rewritten it, so it
3768 wasn't rewriting it again. This bug also had the effect of breaking the
3769 sender address verification caching when the sender address was rewritten.
3771 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3772 This has been changed so that if the ip literal address matches
3773 ignore_target_hosts, the router declines.
3775 7. Added expansion conditions match_domain, match_address, and match_local_
3776 part (NOT match_host).
3778 8. The placeholder for the Received: header didn't have a length field set.
3780 9. Added code to Exim itself and to exim_lock to test for a specific race
3781 condition that could lead to file corruption when using MBX delivery. The
3782 issue is with the lockfile that is created in /tmp. If this file is removed
3783 after a process has opened it but before that process has acquired a lock,
3784 there is the potential for a second process to recreate the file and also
3785 acquire a lock. This could lead to two Exim processes writing to the file
3786 at the same time. The added code performs the same test as UW imapd; it
3787 checks after acquiring the lock that its file descriptor still refers to
3788 the same named file.
3790 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3791 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3794 11. Added the smtp_active_hostname option. If used, this will typically be made
3795 to depend on the incoming interface address. Because $interface_address is
3796 not set up until the daemon has forked a reception process, error responses
3797 that can happen earlier (such as "too many connections") no longer contain
3800 12. If an expansion in a condition on a "warn" statement fails because a lookup
3801 defers, the "warn" statement is abandoned, and the next ACL statement is
3802 processed. Previously this caused the whole ACL to be aborted.
3804 13. Added the iplsearch lookup type.
3806 14. Added ident_timeout as a log selector.
3808 15. Added tls_certificate_verified as a log selector.
3810 16. Added a global option tls_require_ciphers (compare the smtp transport
3811 option of the same name). This controls incoming TLS connections.
3813 17. I finally figured out how to make tls_require_ciphers do a similar thing
3814 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3815 before starting the TLS session.
3817 18. Tabs are now shown as \t in -bP output.
3819 19. If the log selector return_path_on_delivery was set, Exim crashed when
3820 bouncing a message because it had too many Received: header lines.
3822 20. If two routers both had headers_remove settings, and the first one included
3823 a superfluous trailing colon, the final name in the first list and the
3824 first name in the second list were incorrectly joined into one item (with a
3825 colon in the middle).
3831 1. Added -C and -D options to the exinext utility, mainly to make it easier
3832 to include in the automated testing, but these could be helpful when
3833 multiple configurations are in use.
3835 2. The exinext utility was not formatting the output nicely when there was
3836 an alternate port involved in the retry record key, nor when there was a
3837 message id as well (for retries that were specific to a specific message
3838 and a specific host). It was also confused by IPv6 addresses, because of
3839 the additional colons they contain. I have fixed the IPv4 problem, and
3840 patched it up to do a reasonable job for IPv6.
3842 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3843 delivery, the log line now contains "pipelined" if PIPELINING was used.
3845 4. An SMTP transport process used to panic and die if the bind() call to set
3846 an explicit outgoing interface failed. This has been changed; it is now
3847 treated in the same way as a connect() failure.
3849 5. A reference to $sender_host_name in the part of a conditional expansion
3850 that was being skipped was still causing a DNS lookup. This no longer
3853 6. The def: expansion condition was not recognizing references to header lines
3854 that used bh_ and bheader_.
3856 7. Added the _cache feature to named lists.
3858 8. The code for checking quota_filecount in the appendfile transport was
3859 allowing one more file than it should have been.
3861 9. For compatibility with Sendmail, the command line option
3869 and sets the incoming protocol and host name (for trusted callers). The
3870 host name and its colon can be omitted when only the protocol is to be set.
3871 Note the Exim already has two private options, -pd and -ps, that refer to
3872 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3873 "s", but I don't think that's a major issue.
3875 10. A number of refactoring changes to the code, none of which should affect
3878 (a) The number of logging options was getting close to filling up the
3879 32-bit word that was used as a bit map. I have split them into two classes:
3880 those that are passed in the argument to log_write(), and those that are
3881 only ever tested independently outside of that function. These are now in
3882 separate 32-bit words, so there is plenty of room for expansion again.
3883 There is no change in the user interface or the logging behaviour.
3885 (b) When building, for example, log lines, the code previously used a
3886 macro that called string_cat() twice, in order to add two strings. This is
3887 not really sufficiently general. Furthermore, there was one instance where
3888 it was actually wrong because one of the argument was used twice, and in
3889 one call a function was used. (As it happened, calling the function twice
3890 did not affect the overall behaviour.) The macro has been replaced by a
3891 function that can join an arbitrary number of extra strings onto a growing
3894 (c) The code for expansion conditions now uses a table and a binary chop
3895 instead of a serial search (which was left over from when there were very
3896 few conditions). Also, it now recognizes conditions like "pam" even when
3897 the relevant support is not compiled in: a suitably worded error message is
3898 given if an attempt is made to use such a condition.
3900 11. Added ${time_interval:xxxxx}.
3902 12. A bug was causing one of the ddress fields not to be passed back correctly
3903 from remote delivery subprocesses. The field in question was not being
3904 subsequently used, so this caused to problems in practice.
3906 13. Added new log selectors queue_time and deliver_time.
3908 14. Might have fixed a bug in maildirsizefile handling that threw up
3909 "unexpected character" debug warnings, and recalculated the data
3910 unnecessarily. In any case, I expanded the warning message to give more
3913 15. Added the message "Restricted characters in address" to the statements in
3914 the default ACL that block characters like @ and % in local parts.
3916 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3917 Three changes have been made:
3919 (a) There was a serious bug; a negative response to MAIL caused the whole
3920 recipient domain to be cached as invalid, thereby blocking all messages
3921 to all local parts at the same domain, from all senders. This bug has
3922 been fixed. The domain is no longer cached after a negative response to
3923 MAIL if the sender used is not empty.
3925 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3928 (c) A new callout option, "use_sender" has been added for people who want
3929 the modified behaviour.
3935 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3938 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3939 indeed breaks things for older releases.
3941 3. Added additional logging to the case where there is a problem reading data
3942 from a filter that is running in a subprocess using a pipe, in order to
3943 try to track down a specific problem.
3945 4. Testing facility fudge: when running in the test harness and attempting
3946 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3947 getting "No route to host". Convert this to a timeout.
3949 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3952 6. Some OS don't have socklen_t but use size_t instead. This affects the
3953 fifth argument of getsockopt() amongst other things. This is now
3954 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3955 can be set for individual OS. I have set it for SunOS5, OSF1, and
3956 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3957 some earlier ones do not.
3959 7. Change 4.30/15 was not doing the test caselessly.
3961 8. The standard form for an IPv6 address literal was being rejected by address
3962 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3963 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3964 this, as well as the form without the "IPv6" on the front (but only when
3965 address literals are enabled, of course).
3967 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3969 10. Exim crashed if a message with an empty sender address specified by -f
3970 encountered a router with an errors_to setting. This could be provoked only
3971 by a command such as
3975 where an empty string was supplied; "<>" did not hit this bug.
3977 11. Installed PCRE release 4.5.
3979 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3980 remained set. It is now erased.
3982 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3983 times from message ids (which are base 36 rather than the normal 62).
3985 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3986 were being counted as actual protocol errors, and logged if the log
3987 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3988 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3989 and DATA following a set of rejected RCPTs do not count as protocol errors.
3990 In other words, Exim assumes they were pipelined, though this may not
3991 actually be the case. Of course, in all cases the client gets an
3992 appropriate error code.
3994 15. If a lookup fails in an ACL condition, a message about the failure may
3995 be available; it is used if testing the ACL cannot continue, because most
3996 such messages specify what the cause of the deferral is. However, some
3997 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3998 that caused an old message to be retained and used if a later statement
3999 caused a defer, replacing the real cause of the deferral.
4001 16. If an IP address had so many PTR records that the DNS lookup buffer
4002 was not large enough to hold them, Exim could crash while trying to process
4003 the truncated data. It now detects and logs this case.
4005 17. Further to 4.21/58, another change has been made: if (and only if) the
4006 first line of a message (the first header line) ends with CRLF, a bare LF
4007 in a subsequent header line has a space inserted after it, so as not to
4008 terminate the header.
4010 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4011 unnecessarily, used atoi() instead of strtol(), and didn't check the
4012 termination when getting file sizes from file names by regex.
4014 19. Completely re-implemented the support for maildirsize files, in the light
4015 of a number of problems with the previous contributed implementation
4016 (4.30/29). In particular:
4018 . If the quota is zero, the maildirsize file is maintained, but no quota is
4021 . If the maildir directory does not exist, it is created before any attempt
4022 to write a maildirsize file.
4024 . The quota value in the file is just a cache; if the quota is changed in
4025 the transport, the new value overrides.
4027 . A regular expression is available for excluding directories from the
4030 20. The autoreply transport checks the characters in options that define the
4031 message's headers; it allows continued headers, but it was checking with
4032 isspace() after an embedded newline instead of explicitly looking for a
4035 21. If all the "regular" hosts to which an address was routed had passed their
4036 expiry times, and had not reached their retry times, the address was
4037 bounced, even if fallback hosts were defined. Now Exim should go on to try
4040 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4041 equivalent code in the SMTP transport. Some hosts send humungous responses
4042 to HELO/EHLO, more than 1024 it seems.
4044 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4045 gives compiler warnings in some environments. I've now done it "properly",
4048 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4049 (because of problems with the built-in one) was declared to return uschar *
4050 instead of char *, causing compiler failure.
4052 25. Fixed a file descriptor leak when processing alias/forward files.
4054 26. Fixed a minor format string issue in dbfn.c.
4056 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4058 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4059 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4062 29. When a sender address is verified, it is cached, to save repeating the test
4063 when there is more than one recipient in a message. However, when the
4064 verification involves a callout, it is possible for different callout
4065 options to be set for different recipients. It is too complicated to keep
4066 track of this in the cache, so now Exim always runs a verification when a
4067 callout is required, relying on the callout cache for the optimization.
4068 The overhead is duplication of the address routing, but this should not be
4071 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4072 to be verified with callout=postmaster, and the main callout worked but the
4073 postmaster check failed, the verification correctly failed. However, if a
4074 subsequent RCPT command asked for sender verification *without* the
4075 postmaster check, incorrect caching caused this verification also to fail,
4078 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4079 it was not caching the DNS options (qualify_single, search_parents) that
4080 were used when the lookup failed. A subsequent lookup with different
4081 options therefore always gave the same answer, though there were cases
4082 where it should not have. (Example: a "domains = !$mx_any" option on a
4083 dnslookup router: the "domains" option is always processed without any
4084 widening, but the router might have qualify_single set.) Now Exim uses the
4085 cached value only when the same options are set.
4087 32. Added John Jetmore's "exipick" utility to the distribution.
4089 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4090 than a timeout (e.g. a certificate is required, and is not provided), an
4091 Exim server now closes the connection immediately. Previously it waited for
4092 the client to close - but if the client is SSL, it seems that they each
4093 wait for each other, leading to a delay before one of them times out.
4095 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4096 maintained 0.8.x compatibility because I don't think many are using it, and
4097 it is clearly obsolete.
4099 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4102 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4103 client certificate was expired. A simple patch fixes this, though I don't
4104 understand the full logic of why the verify callback is called multiple
4107 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4108 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4109 which causes problems with some clients (such as the Certicom SSL Plus
4110 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4111 disables the coutermeasure allowing Eudora to connect."
4113 38. Exim was not checking that a write() to a log file succeeded. This could
4114 lead to Bad Things if a log got too big, in particular if it hit a file
4115 size limit. Exim now panics and dies if it cannot write to a log file, just
4116 as it does if it cannot open a log file.
4118 39. Modified OS/Makefile-Linux so that it now contains
4120 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4122 The two -D definitions ensure that Exim is compiled with large file
4123 support, which makes it possible to handle log files that are bigger than
4126 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4127 instance) a domain was checked against a named list that involved a lookup,
4128 causing $domain_data to be set, then another domain was checked against the
4129 same list, then the first domain was re-checked, the value of $domain_data
4130 after the final check could be wrong. In particular, if the second check
4131 failed, it could be set empty. This bug probably also applied to
4134 41. The strip_trailing_dot option was not being applied to the address given
4135 with the -f command-line option.
4137 42. The code for reading a message's header from the spool was incrementing
4138 $received_count, but never initializing it. This meant that the value was
4139 incorrect (doubled) while delivering a message in the same process in which
4140 it was received. In the most common configuration of Exim, this never
4141 happens - a fresh exec is done - but it can happen when
4142 deliver_drop_privilege is set.
4144 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4145 it now includes up to 150 characters of the unexpected data in the log
4148 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4149 and building data strings. The size of both of these buffers was 10 000
4150 bytes - far larger than anybody would *ever* want, thought I. Needless to
4151 say, somebody hit the limit. I have increased the maximum line length to
4152 20 000 and the maximum data length of concatenated lines to 100 000. I have
4153 also fixed two bugs, because there was no checking on these buffers. Tsk,
4154 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4155 buffer is too small.
4157 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4158 lsearch lookups. Now it does.
4160 46. When parsing a route_list item in a manualroute router, a fixed-length
4161 buffer was used for the list of hosts. I made this 1024 bytes long,
4162 thinking that nobody would ever have a list of hosts that long. Wrong.
4163 Somebody had a whole pile of complicated expansion conditions, and the
4164 string was silently truncated, leading to an expansion error. It turns out
4165 that it is easier to change to an unlimited length (owing to other changes
4166 that have happened since this code was originally written) than to build
4167 structure for giving a limitation error. The length of the item that
4168 expands into the list of hosts is now unlimited.
4170 47. The lsearch lookup could not handle data where the length of text line was
4171 more than 4095 characters. Such lines were truncated, leading to shortened
4172 data being returned. It should now handle lines of any length.
4174 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4175 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4178 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4179 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4180 no longer happens in comment lines. A list of replacements is now placed
4181 at the head of all of the source files, except those whose only change is
4182 to replace PERL_COMMAND in the very first #! line.
4184 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4185 messages on the queue, with a bottom-up merge sort, using code contributed
4186 by Michael Haardt. This should make operations like -bp somewhat faster on
4187 large queues. It won't affect queue runners, except when queue_run_in_order
4190 51. Installed eximstats 1.31 in the distribution.
4192 52. Added support for SRV lookups to the dnslookup router.
4194 53. If an ACL referred to $message_body or $message_body_end, the value was not
4195 reset for any messages that followed in the same SMTP session.
4197 54. The store-handling optimization for building very long strings was not
4198 differentiating between the different store pools. I don't think this
4199 actually made any difference in practice, but I've tidied it.
4201 55. While running the routers to verify a sender address, $sender_address
4202 was still set to the sender address. This is wrong, because when routing to
4203 send a bounce to the sender, it would be empty. Therefore, I have changed
4204 it so that, while verifying a sender address, $sender_address is set to <>.
4205 (There is no change to what happens when verifying a recipient address.)
4207 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4208 target A or AAAA records (if not already returned) without resetting the
4209 qualify_single or search_parents options of the DNS resolver. These are
4210 inappropriate in this case because the targets of MX and SRV records must
4211 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4212 target that, when qualified, matched something in the local domain. These
4213 two options are now turned off when doing these lookups.
4215 57. It seems that at least some releases of Reiserfs (which does not have the
4216 concept of a fixed number of inodes) returns zero and not -1 for the
4217 number of available inodes. This interacted badly with check_spool_inodes,
4218 which assumed that -1 was the "no such thing" setting. What I have done is
4219 to check that the total number of inodes is greater than zero before doing
4220 the test of how many are available.
4222 58. When a "warn" ACL statement has a log_message modifier, the message is
4223 remembered, and not repeated. This is to avoid a lot of repetition when a
4224 message has many recipients that cause the same warning to be written.
4225 Howewer, Exim was preserving the list of already written lines for an
4226 entire SMTP session, which doesn't seem right. The memory is now reset if a
4227 new message is started.
4229 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4230 debugging output unless log_rewrite was also set.
4232 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4233 of (int)(handle) when we know that handle contains (void *)(-1).
4235 61. The Exim daemon panic-logs an error return when it closes the incoming
4236 connection. However "connection reset by peer" seems to be common, and
4237 isn't really an error worthy of noting specially, so that particular error
4240 62. When Exim is trying to find all the local interfaces, it used to panic and
4241 die if the ioctl to get the interface flags failed. However, it seems that
4242 on at least one OS (Solaris 9) it is possible to have an interface that is
4243 included in the list of interfaces, but for which you get a failure error
4244 for this call. This happens when the interface is not "plumbed" into a
4245 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4246 failure of the "get flags" call assumes that the interface is down.
4248 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4249 makes life easier for people who are doing arithmetic on fields extracted
4250 from dates, where you often get leading zeros that should not be
4251 interpreted as octal.
4253 64. Added qualify_domain to the redirect router, to override the global
4256 65. If a pathologically long header line contained very many addresses (the
4257 report of this problem mentioned 10 000) and each of them was rewritten,
4258 Exim could use up a very large amount of memory. (It kept on making new
4259 copies of the header line as it rewrote, and never released the old ones.)
4260 At the expense of a bit more processing, the header rewriting function has
4261 been changed so that it no longer eats memory in this way.
4263 66. The generation of the Received: header has been moved from the time that a
4264 message starts to be received, to the time that it finishes. The timestamp
4265 in the Received: header should now be very close to that of the <= log
4266 line. There are two side-effects of this change:
4268 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4269 logged header lines no longer include the local Received: line, because
4270 it has not yet been created. The same applies to a copy of the message
4271 that is returned to a non-SMTP sender when a message is rejected.
4273 (b) When a filter file is tested using -bf, no additional Received: header
4274 is added to the test message. After some thought, I decided that this
4277 This change does not affect the value of $received_for. It is still set
4278 after address rewriting, but before local_scan() is called.
4280 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4282 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4283 gave an unhelpful panic error message, and a defer error. I have managed to
4284 change this behaviour so that it now rejects any supplied certificate,
4285 which seems right, as the list of acceptable certificates is empty.
4287 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4288 gave an unhelpful defer error. I have not managed to make this reject any
4289 supplied certificates, but the error message it gives is "no certificate
4290 supplied", which is not helpful.
4292 70. exigrep's output now also includes lines that are not associated with any
4293 message, but which match the given pattern. Implemented by a patch from
4294 Martin Sluka, which also tidied up the Perl a bit.
4296 71. Recipient callout verification, like sender verification, was using <> in
4297 the MAIL FROM command. This isn't really the right thing, since the actual
4298 sender may affect whether the remote host accepts the recipient or not. I
4299 have changed it to use the actual sender in the callout; this means that
4300 the cache record is now keyed on a recipient/sender pair, not just the
4301 recipient address. There doesn't seem to be a real danger of callout loops,
4302 since a callout by the remote host to check the sender would use <>.
4303 [SEE ABOVE: changed after hitting problems.]
4305 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4306 temporary errors. However, in the case of such a code being given after
4307 the end of a data transmission (i.e. after ".") Exim was failing to write
4308 a retry record for the message. (Yes, there was some broken host that was
4309 actually sending 8xx at this point.)
4311 73. An unknown lookup type in a host list could cause Exim to panic-die when
4312 the list was checked. (An example that provoked this was putting <; in the
4313 middle of a list instead of at the start.) If this happened during a DATA
4314 ACL check, a -D file could be left lying around. This kind of configuration
4315 error no longer causes Exim to die; instead it causes a defer errror. The
4316 incident is still logged to the main and panic logs.
4318 74. Buglet left over from Exim 3 conversion. The message "too many messages
4319 in one connection" was written to the rejectlog but not the mainlog, except
4320 when address rewriting (yes!) was being logged.
4322 75. Added write_rejectlog option.
4324 76. When a system filter was run not as root (that is, when system_filter_user
4325 was set), the values of the $n variables were not being returned to the
4326 main process; thus, they were not subsequently available in the $sn
4329 77. Added +return_path_on_delivery log selector.
4331 78. A connection timeout was being treated differently from recipients deferred
4332 when testing hosts_max_try with a message that was older than the host's
4333 retry timeout. (The host should not be counted, thus allowing all hosts to
4334 be tried at least once before bouncing.) This may have been the cause of an
4335 occasionally reported bug whereby a message would remain on the queue
4336 longer than the retry timeout, but would be bounced if a delivery was
4337 forced. I say "may" because I never totally pinned down the problem;
4338 setting up timeout/retry tests is difficult. See also the next item.
4340 79. The ultimate address timeout was not being applied to errors that involved
4341 a combination of host plus message (for example, a timeout on a MAIL
4342 command). When an address resolved to a number of possible hosts, and they
4343 were not all tried for each delivery (e.g. because of hosts_max_try), a
4344 message could remain on the queue longer than the retry timeout.
4346 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4349 81. Fixed an obscure SMTP outgoing bug which required at least the following
4350 conditions: (a) there was another message waiting for the same server;
4351 (b) the server returned 5xx to all RCPT commands in the first message so
4352 that the message was not completed; (c) the server dropped the connection
4353 or gave a negative response to the RSET that Exim sends to abort the
4354 transaction. The observed case was a dropped connection after DATA that had
4355 been sent in pipelining mode. That is, the server had advertised PIPELINING
4356 but was not implementing it correctly. The effect of the bug was incorrect
4357 behaviour, such as trying another host, and this could lead to a crash.
4363 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4364 and daemon.c were passed as pointers to ints; they should have been
4365 pointers to socklen_t variables (which are typically unsigned ints).
4367 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4370 3. Fixed a really odd bug that affected only the testing scheme; patching a
4371 certain fixed string in the binary changed the value of another string that
4372 happened to be identical to the end of the original first string.
4374 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4375 name", it returns that address as the IP address. On some operating
4376 systems (e.g. Solaris), it also passes back the IP address string as the
4377 "host name". However, on others (e.g. Linux), it passes back an empty
4378 string. Exim wasn't checking for this, and was changing the host name to an
4379 empty string, assuming it had been canonicized.
4381 5. Although rare, it is permitted to have more than one PTR record for a given
4382 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4383 all the names associated with an address, because they do in Solaris.
4384 However, it seems that they do not in Linux for data that comes from the
4385 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4386 I found this out when I moved to a new Linux workstation and tried to run
4387 the Exim test suite.
4389 To get round this problem I have changed the code so that it now does its
4390 own call to the DNS to look up PTR records when searching for a host name.
4391 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4392 addresses that are only in /etc/hosts are still found.
4394 This behaviour is, however, controlled by an option called host_lookup_
4395 order, which defaults to "bydns:byaddr". If people want to use the other
4396 order, or indeed, just use one or the other means of lookup, they can
4397 specify it in this variable.
4399 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4400 some operating systems, this comes back from gethostbyaddr() as an empty
4401 string, and this is what Exim used to test for. However, it seems that in
4402 other systems, "." is yielded. Exim now tests for this case too.
4404 7. The values of check_spool_space and check_log_space are now held internally
4405 as a number of kilobytes instead of an absolute number of bytes. If a
4406 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4407 kilobyte. This means that much larger values can be stored.
4409 8. Exim monitor: an attempt to get the action menu when not actually pointing
4410 at a message produces an empty menu entitled "No message selected". This
4411 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4412 no entries in it ("Shell widget menu has zero width and/or height"). So I
4413 have added a single, blank menu entry in this case.
4415 9. Added ${quote_local_part.
4417 10. MIME decoding is now applied to the contents of Subject: header lines when
4420 11. Now that a reference to $sender_host_address automatically causes a reverse
4421 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4422 host lookup before query-style lookups in lists that might use this
4423 variable. This has therefore been abolished, and the "net-" prefix is no
4424 longer necessary for query-style lookups.
4426 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4427 have been a typo for LFLAGS, so it has been changed.
4429 13. The install script calls Exim with "-C /dev/null" in order to find the
4430 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4431 to be output. Howeve, since Exim outputs its version number before the
4432 error, it didn't break the script. It just looked ugly. I fixed this by
4433 always allowing "-C /dev/null" if the caller is root.
4435 14. Ignore overlarge ACL variable number when reading spool file - insurance
4436 against a later release with more variables having written the file.
4438 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4439 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4440 this, as well as the form without the "IPv6" on the front.
4442 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4443 OS/Makefile-Darwin file.
4445 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4446 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4448 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4449 variable. However, some OS use ints for inodes. I've added cast to long int
4450 to get rid of the compiler warning.
4452 19. I had forgotten to lock out "/../" in configuration file names when
4453 ALT_CONFIG_PREFIX was set.
4455 20. Routers used for verification do not need to specify transports. However,
4456 if such a router generated a host list, and callout was configured, Exim
4457 crashed, because it could not find a port number from the (non-existent)
4458 transport. It now assumes port 25 in this circumstance.
4460 21. Added the -t option to exigrep.
4462 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4463 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4464 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4465 them is set, LOOKUP_LSEARCH is forced.
4467 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4469 24. Added sender and host information to the "rejected by local_scan()" log
4470 line; previously there was no indication of these.
4472 25. Added .include_if_exists.
4474 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4475 Linux. It turns out that not all file systems support this. Apparently some
4476 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4477 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4478 not supported on the file descriptor, is now ignored when Exim is trying to
4479 sync a directory. This applies only to Linux.
4481 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4483 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4484 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4487 29. Added support for maildirsize files from supplied patch (modified a bit).
4489 30. The use of :fail: followed by an empty string could lead Exim to respond to
4490 sender verification failures with (e.g.):
4492 550 Verification failed for <xxx>
4493 550 Sender verify failed
4495 where the first response line was missing the '-' that indicates it is not
4496 the final line of the response.
4498 31. The loop for finding the name of the user that called Exim had a hardwired
4499 limit of 10; it now uses the value of finduser_retries, which is used for
4500 all other user lookups.
4502 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4505 33. Exim was neglecting to zero errno before one call of strtol() when
4506 expanding a string and expecting an integer value. On some systems this
4507 resulted in spurious "integer overflow" errors. Also, it was casting the
4508 result into an int without checking.
4510 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4511 did not work. The code looks as if it has *never* worked, though it appears
4512 to have been documented since at least releast 1.62. I have made it work.
4514 35. The "timeout_DNS" error in retry rules, also documented since at least
4515 1.62, also never worked. As it isn't clear exactly what this means, and
4516 clearly it isn't a major issue, I have abolished the feature by treating it
4517 as "timeout", and writing a warning to the main and panic logs.
4519 36. The display of retry rules for -brt wasn't always showing the error code
4522 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4523 timeout_connect_A, timeout_connect_MX.
4525 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4526 to the empty sender.
4528 39. The daemon was not analysing the content of -oX till after it had closed
4529 stderr and disconnected from the controlling terminal. This meant that any
4530 syntax errors were only noted on the panic log, and the return code from
4531 the command was 0. By re-arranging the code a little, I've made the
4532 decoding happen first, so such errors now appear on stderr, and the return
4533 code is 1. However, the actual setting up of the sockets still happens in
4534 the disconnected process, so errors there are still only recorded on the
4537 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4538 connections (as happens on some IP stacks) was logged at start up time as
4539 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4540 it from "IPv6 and IPv4", which means that two separate sockets are being
4543 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4544 says whether AF_INET or AF_INET6 was passed as an argument.
4546 42. Exiwhat output was messed up when time zones were included in log
4549 43. Exiwhat now gives more information about the daemon's listening ports,
4550 and whether -tls-on-connect was used.
4552 44. The "port" option of the smtp transport is now expanded.
4554 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4555 silently ignored. Now an error message is written to the main and panic
4558 46. There's a new ACL modifier called "logwrite" which writes to a log file
4559 as soon as it is encountered.
4561 47. Added $local_user_uid and $local_user_gid at routing time.
4563 48. Exim crashed when trying to verify a sender address that was being
4566 49. Exim was recognizing only a space character after ".include". It now also
4567 recognizes a tab character.
4569 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4570 extracting the relevant information from the specification. The man page no
4571 longer contains scrambled data for the -d option, and I've added a section
4572 at the front about calling Exim under different names.
4574 51. Added "extra_headers" argument to the "mail" command in filter files.
4576 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4579 53. Installed eximstats 1.29.
4581 54. Added transport_filter_timeout as a generic transport option.
4583 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4584 Cc: header lines. This was required by RFC 822, but it not required by RFC
4587 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4588 incoming messages that did not have them. Now it does so only if the
4589 message originates locally, that is, if there is no associated remote host
4590 address. When Resent- header lines are present, this applies to the Resent-
4591 lines rather than the non-Resent- lines.
4593 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4594 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4596 58. Messages for configuration errors now include the name of the main
4597 configuration file - useful now that there may be more than one file in a
4598 list (.included file names were always shown).
4600 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4601 for those rare installations that do not start the daemon as root or run it
4602 setuid root. I've cut out the call to initgroups() if the daemon is not
4605 60. The Exim user and group can now be bound into the binary as text strings
4606 that are looked up at the start of Exim's processing.
4608 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4610 62. Added $mailstore_basename variable.
4612 63. Installed patch to sieve.c from Michael Haardt.
4614 64. When Exim failed to open the panic log after failing to open the main log,
4615 the original message it was trying to log was written to stderr and debug
4616 output, but if they were not available (the usual case in production), it
4617 was lost. Now it is written to syslog before the two lines that record the
4618 failures to open the logs.
4620 65. Users' Exim filters run in subprocesses under the user's uid. It is
4621 possible for a "deliver" command or an alias in a "personal" command to
4622 provoke an address rewrite. If logging of address rewriting is configured,
4623 this fails because the process is not running as root or exim. There may be
4624 a better way of dealing with this, but for the moment (because 4.30 needs
4625 to be released), I have disabled address rewrite logging when running a
4626 filter in a non-root, non-exim process.
4632 1. The buildconfig auxiliary program wasn't quoting the value set for
4633 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4634 not defined. This bug was masked in 4.22 by the effect that was fixed in
4637 2. Some messages that were rejected after a message id was allocated were
4638 shown as "incomplete" by exigrep. It no longer does this for messages that
4639 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4641 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4642 have allow_domain_literals set, the ID did not get logged in the <= line.
4643 Domain literals are now always recognized in Message-ID: header lines.
4645 4. The first argument for a ${extract expansion item is the key name or field
4646 number. Leading and trailing spaces in this item were not being ignored,
4647 causing some misleading effects.
4649 5. When deliver_drop_privilege was set, single queue runner processes started
4650 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4651 same command in the process it spins off) were not dropping privilege.
4653 6. When the daemon running as "exim" started a queue runner, it always
4654 re-executed Exim in the spun-off process. This is a waste of effort when
4655 deliver_drop_privilege is set. The new process now just calls the
4656 queue-runner function directly.
4662 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4665 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4666 ignored. Though the use of -oP was forcing the writing of a pid file, it
4667 was always written to the default place.
4669 3. If the message "no IP address found for host xxxx" is generated during
4670 incoming verification, it is now followed by identification of the incoming
4671 connection (so you can more easily find what provoked it).
4673 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4675 5. Added some features to "harden" Exim a bit more against certain attacks:
4677 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4678 be put in Local/Makefile. This is like the never_users runtime option,
4679 but it cannot be overridden. The default setting is "root".
4681 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4682 prefix string with which any file named in a -C command line option
4685 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4686 is retained for -C and -D only if the caller of Exim is root. Without
4687 it, the exim user may also use -C and -D and retain privilege.
4689 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4690 command line option is disabled.
4692 6. Macro names set by the -D option must start with an upper case letter, just
4693 like macro names defined in the configuration file.
4695 7. Added "dereference=" facility to LDAP.
4697 8. Two instances of the typo "uknown" in the source files are fixed.
4699 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4700 the Configure-Makefile script screwed up while processing it.
4702 10. Incorporated PCRE 4.4.
4704 11. The SMTP synchronization check was not operating right at the start of an
4705 SMTP session. For example, it could not catch a HELO sent before the client
4706 waited for the greeting. There is now a check for outstanding input at the
4707 point when the greeting is written. Because of the duplex, asynchronous
4708 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4709 way, but not yet received, when the check is performed.
4711 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4712 on TCP/IP sockets, because this apparently causes some broken clients to
4715 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4716 unchanged) from the Cygwin maintainer.
4718 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4719 was supported instead of testing for mbx. Effectively a typo.
4721 15. The spa authenticator server code was not checking that the input it
4722 received was valid base64.
4724 16. The debug output line for the "set" modifier in ACLs was not showing the
4725 name of the variable that was being set.
4727 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4729 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4730 lookup on its own. Something else was needed to trigger the lookup. For
4731 example, a match in host_lookup or the need for a host name in a host list.
4732 Now, if $sender_host_name is referenced and the host name has not yet been
4733 looked up, a lookup is performed. If the lookup fails, the variable remains
4734 empty, and $host_lookup_failed is set to "1".
4736 19. Added "eqi" as a case-independent comparison operator.
4738 20. The saslauthd authentication condition could segfault if neither service
4739 nor realm was specified.
4741 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4742 error message that was logged was misleading, and incoming SMTP
4743 connections were dropped. The message is now more accurate, and temporary
4744 errors are given to SMTP connections.
4746 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4747 (except RSET) with a 421 error, until QUIT is received. However, it was
4748 failing to send a response to QUIT.
4750 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4751 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4752 ACL is now run after the helo_try_verify_hosts code.
4754 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4755 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4756 case-independent, so other case variants are also recognized). Apparently
4757 some systems use these upper case variants.
4759 25. If more than two messages were waiting for the same host, and a transport
4760 filter was specified for the transport, Exim sent two messages over the
4761 same TCP/IP connection, and then failed with "socket operation on non-
4762 socket" when it tried to send the third.
4764 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4766 27. The extern definition of crypt16() in expand.c was not being excluded when
4767 the OS had its own crypt16() function.
4769 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4770 as a preferred synonym for return_size_limit, both as an option and as an
4773 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4775 30. Changed the default configuration ACL to relax the local part checking rule
4776 for addresses that are not in any local domains. For these addresses,
4777 slashes and pipe symbols are allowed within local parts, but the sequence
4778 /../ is explicitly forbidden.
4780 31. SPA server authentication was not clearing the challenge buffer before
4783 32. log_message in a "warn" ACL statement was writing to the reject log as
4784 well as to the main log, which contradicts the documentation and doesn't
4785 seem right (because no rejection is happening). So I have stopped it.
4787 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4788 However, I am unable to do any testing of this.
4790 34. Fixed an infelicity in the appendfile transport. When checking directories
4791 for a mailbox, to see if any needed to be created, it was accidentally
4792 using path names with one or more superfluous leading slashes; tracing
4793 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4795 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4796 contents are added to the log line that is written for every discarded
4797 recipient. (Previously a log_message setting was ignored.)
4799 36. The ${quote: operator now quotes the string if it is empty.
4801 37. The install script runs exim in order to find its version number. If for
4802 some reason other than non-existence or emptiness, which it checks, it
4803 could not run './exim', it was installing it with an empty version number,
4804 i.e. as "exim-". This error state is now caught, and the installation is
4807 38. An argument was missing from the function that creates an error message
4808 when Exim fails to connect to the socket for saslauthd authentication.
4809 This could cause Exim to crash, or give a corrupted message.
4811 39. Added isip, isip4, and isip6 to ${if conditions.
4813 40. The ACL variables $acl_xx are now saved with the message, and can be
4814 accessed later in routers, transports, and filters.
4816 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4817 strings in the file are not string-expanded.
4819 42. If a MAIL command specified a SIZE value that was too large to fit into an
4820 int variable, the check against message_size_limit failed. Such values are
4821 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4822 day this will have to be increased, but I don't think I want to be around
4823 when emails are that large.
4830 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4831 iconv() is not standard in FreeBSD.
4833 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4834 IPv6 enabled. The observed symptom was a segmentation fault on return from
4835 the function os_common_find_running_interfaces() in src/os.c.
4837 3. In the check_special_case() function in daemon.c I had used "errno" as an
4838 argument name, which causes warnings on some systems. This was basically a
4839 typo, since it was named "eno" in the comments!
4841 4. The code that waits for the clock to tick (at a resolution of some fraction
4842 of a second) so as to ensure message-id uniqueness was always waiting for
4843 at least one whole tick, when it could have waited for less. [This is
4844 almost certainly not relevant at current processor speeds, where it is
4845 unlikely to ever wait at all. But we try to future-proof.]
4847 5. The function that sleeps for a time interval that includes fractions of a
4848 second contained a race. It did not block SIGALRM between setting the
4849 timer, and suspending (a couple of lines later). If the interval was short
4850 and the sigsuspend() was delayed until after it had expired, the suspension
4851 never ended. On busy systems this could lead to processes getting stuck for
4854 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4855 process, before it forks any delivery processes. The open lookup caching
4856 mechanism meant that the open file or database connection was passed into
4857 the delivery process. The problem was that delivery processes always tidy
4858 up cached lookup data. This could cause a problem for the next delivery
4859 process started by the queue runner, because the external queue runner
4860 process does not know about the closure. So the next delivery process
4861 still has data in the lookup cache. In the case of a file lookup, there was
4862 no problem because closing a file descriptor in a subprocess doesn't affect
4863 the parent. However, if the lookup was caching a connection to a database,
4864 the connection was closed, and the second delivery process was likely to
4865 see errors such as "PGSQL: query failed: server closed the connection
4866 unexpectedly". The problem has been fixed by closing all cached lookups
4867 in a queue runner before running a delivery process.
4869 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4870 seem to have the "const" qualifier which it has on other OS. I've
4873 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4874 *of the same type* (client or server) with the same public name that an
4875 error should be diagnosed.
4877 9. When Exim looked up a host name for an IP address, but failed to find the
4878 original IP address when looking up the host name (a safety check), it
4879 output the message "<ip address> does not match any IP for NULL", which was
4880 confusing, to say the least. The bug was that the host name should have
4881 appeared instead of "NULL".
4883 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4884 user that is built into the binary, and the -C or -D options is used, root
4885 privilege is dropped before the configuration file is read. In addition,
4886 logging is switched to stderr instead of the normal log files. If the
4887 configuration then re-defines the Exim user, the unprivileged environment
4888 is probably not what is expected, so Exim logs a panic warning message (but
4891 However, if deliver_drop_privilege is set, the unprivileged state may well
4892 be exactly what is intended, so the warning has been cut out in that case,
4893 and Exim is allowed to try to write to its normal log files.
4899 1. smtp_return_error_details was not giving details for temporary sender
4900 or receiver verification errors.
4902 2. Diagnose a configuration error if two authenticators have the same public
4905 3. Exim used not to create the message log file for a message until the first
4906 delivery attempt. This could be confusing when incoming messages were held
4907 for policy or load reasons. The message log file is now created at the time
4908 the message is received, and an initial "Received" line is written to it.
4910 4. The automatically generated man page for command line options had a minor
4911 bug that caused no ill effects; however, a more serious problem was that
4912 the procedure for building the man page automatically didn't always
4913 operate. Consequently, release 4.20 contains an out-of-date version. This
4914 shouldn't happen again.
4916 5. When building Exim with embedded Perl support, the script that builds the
4917 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4918 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4920 6. The freeze_tell option was not being used for messages that were frozen on
4921 arrival, either by an ACL or by local_scan().
4923 7. Added the smtp_incomplete_transaction log selector.
4925 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4926 was accepting AUTH without a new EHLO.
4928 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4929 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4932 10. Logging of TCP/IP connections (when configured) now happens in the main
4933 daemon process instead of the child process, so that the TCP/IP connection
4934 count is more accurate (but it can never be perfect).
4936 11. The use of "drop" in a nested ACL was not being handled correctly in the
4937 outer ACL. Now, if condition failure induced by the nested "drop" causes
4938 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4939 or "require"), the connection is dropped.
4941 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4942 that yield "discard" can now be used with an "accept" or a "discard" verb,
4943 but an error is generated for any others (because I can't see a useful way
4944 to define what should happen).
4946 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4947 are now processed in the same way as lines in the Exim configuration file.
4948 In particular, continuation lines are supported.
4950 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4952 15. Added -ti meaning -t -i.
4954 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4955 domains, and warn by logging if others are found.
4957 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4958 returned from GIFCONF when Exim is trying to find the list of interfaces on
4959 a host. The code in os.c has been modified to copy each ifreq to an aligned
4960 structure in all cases.
4962 Also, in some cases, the returned ifreq's were being copied to a 'struct
4963 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4964 means the last couple of bytes of an IPv6 address could be chopped if the
4965 ifreq contained only a normal sockaddr (14 bytes storage).
4967 18. Named domain lists were not supported in the hosts_treat_as_local option.
4968 An entry such as +xxxx was not recognized, and was treated as a literal
4971 19. Ensure that header lines added by a DATA ACL are included in the reject log
4972 if the ACL subsequently rejects the message.
4974 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4975 MD5 (which is deprecated).
4977 21. When testing a filter file using -bf, Exim was writing a message when it
4978 took the sender from a "From " line in the message, but it was not doing so
4979 when it took $return_path from a Return-Path: header line. It now does.
4981 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4982 with a valid header line field name (a series of printing characters
4983 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4985 23. Changed "disc" in the source to "disk" to conform to the documentation and
4986 the book and for uniformity.
4988 24. Ignore Sendmail's -Ooption=value command line item.
4990 25. When execve() failed while trying to run a command in a pipe transport,
4991 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4992 could be confused with a return value of 69 from the command itself. This
4993 has been changed to 127, the value the shell returns if it is asked to run
4994 a non-existent command. The wording for the related log line suggests a
4995 non-existent command as the problem.
4997 26. If received_header_text expands to an empty string, do not add a Received:
4998 header line to the message. (Well, it adds a token one on the spool, but
4999 marks it "old" so that it doesn't get used or transmitted.)
5001 27. Installed eximstats 1.28 (addition of -nt option).
5003 28. There was no check for failure on the call to getsockname() in the daemon
5004 code. This can fail if there is a shortage of resources on the system, with
5005 ENOMEM, for example. A temporary error is now given on failure.
5007 29. Contrary to the C standard, it seems that in some environments, the
5008 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5009 program. Exim now does this explicitly; it affects the formatting of
5010 timestamps using strftime().
5012 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5013 complaints. I've now initialized all the variables, to avoid this.
5015 32. Header lines added by a system filter were not being "seen" during
5016 transport-time rewrites.
5018 33. The info_callback() function passed to OpenSSL is set up with type void
5019 (*)(SSL *, int, int), as described somewhere. However, when calling the
5020 function (actually a macro) that sets it up, the type void(*)() is
5021 expected. I've put in a cast to prevent warnings from picky compilers.
5023 34. If a DNS black list lookup found a CNAME record, but there were no A
5024 records associated with the domain it pointed at, Exim crashed.
5026 35. If a DNS black list lookup returned more than one A record, Exim ignored
5027 all but the first. It now scans all returned addresses if a particular IP
5028 value is being sought. In this situation, the contents of the
5029 $dnslist_value variable are a list of all the addresses, separated by a
5032 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5033 to accept a host name and all its aliases if the forward lookup for any of
5034 them yielded the IP address of the incoming connection. Now it accepts only
5035 those names whose forward lookup yields the correct IP address. Any other
5036 names are discarded. This closes a loophole whereby a rogue DNS
5037 administrator could create reverse DNS records to break through a
5038 wildcarded host restriction in an ACL.
5040 37. If a user filter or a system filter that ran in a subprocess used any of
5041 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5042 the wrong values were passed to the pipe command ($thisaddress had the
5043 value of $0, $0 had the value of $1, etc). This bug was introduced by
5044 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5046 38. Improved the line breaking for long SMTP error messages from ACLs.
5047 Previously, if there was no break point between 40 and 75 characters, Exim
5048 left the rest of the message alone. Two changes have been made: (a) I've
5049 reduced the minimum length to 35 characters; (b) if it can't find a break
5050 point between 35 and 75 characters, it looks ahead and uses the first one
5051 that it finds. This may give the occasional overlong line, but at least the
5052 remaining text gets split now.
5054 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5055 file descriptors might be low, and that setting 1000 would always raise it.
5056 It turns out that in some environments, the limit is already over 1000 and
5057 that lowering it causes trouble. So now Exim takes care not to decrease it.
5059 40. When delivering a message, the value of $return_path is set to $sender_
5060 address at the start of routing (routers may change the value). By an
5061 oversight, this default was not being set up when an address was tested by
5062 -bt or -bv, which affected the outcome if any router or filter referred to
5065 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5066 log only when "message" or "log_message" are set. However, if one of the
5067 conditions was an address verification, or a call to a nested ACL, the
5068 messages generated by the underlying test were being passed through. This
5069 no longer happens. The underlying message is available in $acl_verify_
5070 message for both "message" and "log_message" expansions, so it can be
5071 passed through if needed.
5073 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5074 new expansion $bh_ to give the encoded byte string without charset
5075 translation. Translation happens only if iconv() is available; HAVE_ICONV
5076 indicates this at build time. HEADERS_CHARSET gives the charset to
5077 translate to; headers_charset can change it in the configuration, and
5078 "headers charset" can change it in an individual filter file.
5080 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5081 that creates RFC 2047 encoded "words" labels them as that charset instead
5082 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5083 expansion operator; (ii) when Exim creates a From: line for a local
5084 message; (iii) when a header line is rewritten to include a "phrase" part.
5086 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5087 buggy, causing it to skip the first lines of messages whose message ID
5088 ended in 'D'. This would not have bitten before Exim release 4.14, because
5089 message IDs were unlikely to end in 'D' before then. The effect was to have
5090 incorrect size information for certain domains.
5092 45. #include "config.h" was missing at the start of the crypt16.c module. This
5093 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5096 46. If there was a timeout during a "random" callout check, Exim treated it as
5097 a failure of the random address, and carried on sending RSET and the real
5098 address. If the delay was just some slowness somewhere, the response to the
5099 original RCPT would be taken as a response to RSET and so on, causing
5100 mayhem of various kinds.
5102 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5103 when I implemented it. It didn't allow for the fact that some option values
5104 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5105 the right test for positive values.
5107 48. Domain names in DNS records are case-independent. Exim always looks them up
5108 in lower case. Some resolvers return domain names in exactly the case they
5109 appear in the zone file, that is, they may contain uppercase letters. Not
5110 all resolvers do this - some return always lower case. Exim was treating a
5111 change of case by a resolver as a change of domain, similar to a widening
5112 of a domain abbreviation. This triggered its re-routing code and so it was
5113 trying to route what was effectively the same domain again. This normally
5114 caused routing to fail (because the router wouldn't handle the domain
5115 twice). Now Exim checks for this case specially, and just changes the
5116 casing of the domain that it ultimately uses when it transmits the message
5119 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5122 50. If a filter generated a file delivery with a non-absolute name (possible if
5123 no home directory exists for the router), the forbid_file option was not
5126 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5127 the existing equality matching.
5129 52. Exim was using ints instead of ino_t variables in some places where it was
5130 dealing with inode numbers.
5132 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5133 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5134 called TMPDIR, and if it finds it is different, it changes its value.
5136 54. The smtp_printf() function is now made available to local_scan() so
5137 additional output lines can be written before returning. There is also an
5138 smtp_fflush() function to enable the detection of a dropped connection.
5139 The variables smtp_input and smtp_batched_input are exported to
5142 55. Changed the default runtime configuration: the message "Unknown user"
5143 has been removed from the ACL, and instead placed on the localuser router,
5144 using the cannot_route_message feature. This means that any verification
5145 failures that generate their own messages won't get overridden. Similarly,
5146 the "Unrouteable address" message that was in the ACL for unverifiable
5147 relay addresses has also been removed.
5149 56. Added hosts_avoid_esmtp to the smtp transport.
5151 57. The exicyclog script was not checking for the esoteric option
5152 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5153 will work only if exicyclog is run under the appropriate euid.
5155 58. Following a discussion on the list, the rules by which Exim recognises line
5156 endings on incoming messages have been changed. The -dropcr and drop_cr
5157 options are now no-ops, retained only for backwards compatibility. The
5158 following line terminators are recognized: LF CRLF CR. However, special
5159 processing applies to CR:
5161 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5162 nor a local message in the state where . is a terminator.
5164 (ii) If a bare CR is encountered in a header line, an extra space is added
5165 after the line terminator so as not to end the header. The reasoning
5166 behind this is that bare CRs in header lines are most likely either
5167 to be mistakes, or people trying to play silly games.
5169 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5170 was being incorrectly given as 18 bytes larger than it should have been.
5171 This is a VOB (very old bug).
5173 60. This may never have affected anything current, but just in case it has:
5174 When the local host is found other than at the start of a list of hosts,
5175 the local host, those with the same MX, and any that follow, are discarded.
5176 When the list in question was part of a longer list of hosts, the following
5177 hosts (not currently being processed) were also being discarded. This no
5178 longer happens. I'm not sure if this situation could ever has previously
5181 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5184 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5185 header. This is something that is recommended in a new Internet Draft, and
5186 is something that is documented as being done by Sendmail. There are two
5187 possible values. For messages generated by the autoreply transport, Exim
5190 Auto-Submitted: auto-replied
5192 whereas for all other generated messages (e.g. bounces) it adds
5194 Auto-Submitted: auto-generated
5196 63. The "personal" condition in filters now includes a test for the
5197 Auto-Submitted: header. If it contains the string "auto-" the message it
5198 not considered personal.
5200 64. Added rcpt_include_affixes as a generic transport option.
5202 65. Added queue_only_override (default true).
5204 66. Added the syslog_duplication option.
5206 67. If what should have been the first header line of a message consisted of
5207 a space followed by a colon, Exim was mis-interpreting it as a header line.
5208 It isn't of course - it is syntactically invalid and should therefore be
5209 treated as the start of the message body. The misbehaviour could have
5210 caused a number of strange effects, including loss of data in subsequent
5211 header lines, and spool format errors.
5213 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5214 client host had authenticated. This control can now be exercised by an ACL
5215 for more flexibility.
5217 69. By default, callouts do not happen when testing with -bh. There is now a
5218 variant, -bhc, which does actually run the callout code, including
5219 consulting and updating the callout cache.
5221 70. Added support for saslauthd authentication, courtesy of Alexander
5224 71. If statvfs() failed on the spool or log directories while checking their
5225 size for availability, Exim confusingly gave the error "space shortage".
5226 Furthermore, in debugging mode it crashed with a floating point exception.
5227 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5228 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5229 serious problem, Exim now writes to the main and panic logs when this
5230 happens, with details of the failure. It then refuses to accept the
5231 incoming message, giving the message "spool directory problem" or "log
5232 directory problem" with a 421 code for SMTP messages.
5234 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5235 0, 1, and 2 exist, because some OS complain for execs without them (see
5236 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5237 descriptors. However, the code omitted to check that the open succeeded,
5238 causing mysterious errors if for some reason the permissions on /dev/null
5239 got screwed. Now Exim writes a message to the main and panic logs, and
5240 bombs out if it can't open /dev/null.
5242 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5243 interact so that it is all more flexible. It is supposed to remain
5244 backwards compatible. Also added extra_local_interfaces.
5246 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5247 to bomb out with an assertion failure - to the client this appears as a
5248 connection drop. This problem occurs in the part of the code that was taken
5249 from the Samba project. Fortunately, the assertion is in a very simple
5250 function, so I have fixed this by reproducing the function inline in the
5251 one place where it is called, and arranging for authentication to fail
5252 instead of killing the process with assert().
5254 75. The SPA client code was not working when the server requested OEM rather
5255 than Unicode encoding.
5257 76. Added code to make require_files with a specific uid setting more usable in
5258 the case where statting the file as root fails - usually a non-root-mounted
5259 NFS file system. When this happens and the failure is EACCES, Exim now
5260 forks a subprocess and does the per-uid checking as the relevant uid.
5262 77. Added process_log_path.
5264 78. If log_file_path was not explicitly set, a setting of check_log_space or
5265 check_log_inodes was ignored.
5267 79. If a space check for the spool or log partitions fails, the incident is now
5268 logged. Of course, in the latter case the data may get lost...
5270 80. Added the %p formatting code to string_format() so that it can be used to
5271 print addresses in debug_print(). Adjusted all the address printing in the
5272 debugging in store.c to use %p rather than %d.
5274 81. There was a concern that a line of code in smtp_in.c could overflow a
5275 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5276 initially expressed, the concern was not well-founded, because trailing
5277 spaces are removed early. However, if the trailing spaces were followed by
5278 a NULL, they did not get removed, so the overflow was possible. Two fixes
5281 (a) I re-wrote the offending code in a cleaner fashion.
5282 (b) If an incoming SMTP command contains a NULL character, it is rejected
5285 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5286 runs initgroups(), so that if the Exim user is in any additional groups,
5287 they will be used during message reception.
5293 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob