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.
58 TF/01 Failure to get a lock on a hints database can have serious
59 consequences so log it to the panic log.
61 TF/02 Log LMTP confirmation messages in the same way as SMTP,
62 controlled using the smtp_confirmation log selector.
64 TF/03 Include the error message when we fail to unlink a spool file.
66 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
67 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
68 for maintaining out-of-tree patches for some time.
70 PP/01 Bugzilla 139: Documentation and portability issues.
71 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
72 Handle per-OS dynamic-module compilation flags.
74 PP/02 Let /dev/null have normal permissions.
75 The 4.73 fixes were a little too stringent and complained about the
76 permissions on /dev/null. Exempt it from some checks.
77 Reported by Andreas M. Kirchwitz.
79 PP/03 Report version information for many libraries, including
80 Exim version information for dynamically loaded libraries. Created
81 version.h, now support a version extension string for distributors
82 who patch heavily. Dynamic module ABI change.
84 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
85 privilege escalation vulnerability whereby the Exim run-time user
86 can cause root to append content of the attacker's choosing to
89 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
92 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
93 If dropping privileges for untrusted macros, we disabled normal logging
94 on the basis that it would fail; for the Exim run-time user, this is not
95 the case, and it resulted in successful deliveries going unlogged.
96 Fixed. Reported by Andreas Metzler.
102 PP/01 Date: & Message-Id: revert to normally being appended to a message,
103 only prepend for the Resent-* case. Fixes regression introduced in
104 Exim 4.70 by NM/22 for Bugzilla 607.
106 PP/02 Include check_rfc2047_length in configure.default because we're seeing
107 increasing numbers of administrators be bitten by this.
109 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
111 PP/03 Bugzilla 994: added openssl_options main configuration option.
113 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
115 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
117 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
119 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
120 without a peer certificate, leading to a segfault because of an
121 assumption that peers always have certificates. Be a little more
122 paranoid. Problem reported by Martin Tscholak.
124 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
125 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
126 NB: ClamAV planning to remove STREAM in "middle of 2010".
127 CL also introduces -bmalware, various -d+acl logging additions and
128 more caution in buffer sizes.
130 PP/09 Implemented reverse_ip expansion operator.
132 PP/10 Bugzilla 937: provide a "debug" ACL control.
134 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
136 PP/12 Bugzilla 973: Implement --version.
138 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
140 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
142 PP/15 Bugzilla 816: support multiple condition rules on Routers.
144 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
145 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
146 ignore trailing whitespace.
148 JJ/02 prevent non-panic DKIM error from being sent to paniclog
150 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
153 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
154 Notification from Dr Andrew Aitchison.
156 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
157 ExtendedDetectionInfo response format.
158 Notification from John Horne.
160 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
163 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
164 XSL and documented dependency on system catalogs, with examples of how
167 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
170 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
171 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
172 configuration file which is writeable by the Exim user or group.
174 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
175 of configuration files to cover files specified with the -C option if
176 they are going to be used with root privileges, not just the default
179 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
180 option (effectively making it always true).
182 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
183 files to be used while preserving root privileges.
185 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
186 that rogue child processes cannot use them.
188 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
189 run-time user, instead of root.
191 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
192 Exim run-time user without dropping privileges.
194 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
195 result string, instead of calling string_vformat() twice with the same
198 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
199 for other users. Others should always drop root privileges if they use
200 -C on the command line, even for a whitelisted configure file.
202 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
204 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
210 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
211 $data_path, and $header_path variables; fixed documentation bugs and
214 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
215 exipick to access non-standard spools, including the "frozen" queue
218 NM/01 Bugzilla 965: Support mysql stored procedures.
219 Patch from Alain Williams
221 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
223 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
224 Patch from Andreas Metzler
226 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
227 Patch from Kirill Miazine
229 NM/05 Bugzilla 671: Added umask to procmail example.
231 JJ/03 installed exipick 20100323.0, fixing doc bug
233 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
234 directory. Notification and patch from Dan Rosenberg.
236 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
238 TK/02 Improve log output when DKIM signing operation fails.
240 MH/01 Treat the transport option dkim_domain as a colon separated
241 list, not as a single string, and sign the message with each element,
242 omitting multiple occurences of the same signer.
244 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
245 Bugzilla 985, 986. Patch by Simon Arlott
247 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
248 Patch by Simon Arlott
250 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
251 MBX locking. Notification from Dan Rosenberg.
257 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
259 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
261 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
263 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
265 NM/04 Bugzilla 915: Flag broken perl installation during build.
271 TK/01 Added patch by Johannes Berg that expands the main option
272 "spamd_address" if it starts with a dollar sign.
274 TK/02 Write list of recipients to X-Envelope-Sender header when building
275 the mbox-format spool file for content scanning (suggested by Jakob
278 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
279 (http://www.dcc-servers.net/) support via dccifd. Activated by
280 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
282 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
283 by Mark Daniel Reidel <mr@df.eu>.
285 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
286 When building exim an external PCRE library is now needed -
287 PCRE is a system library on the majority of modern systems.
288 See entry on PCRE_LIBS in EDITME file.
290 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
291 conversation. Added nologin parameter to request.
292 Patch contributed by Kirill Miazine.
294 TF/01 Do not log submission mode rewrites if they do not change the address.
296 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
298 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
299 log files in place. Contributed by Roberto Lima.
301 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
303 TF/03 Bugzilla 615: When checking the local_parts router precondition
304 after a local_part_suffix or local_part_prefix option, Exim now
305 does not use the address's named list lookup cache, since this
306 contains cached lookups for the whole local part.
308 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
309 Robert Millan. Documentation is in experimental-spec.txt.
311 TF/04 Bugzilla 668: Fix parallel build (make -j).
313 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
315 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
316 Patch provided by Jan Srzednicki.
318 TF/05 Leading white space used to be stripped from $spam_report which
319 wrecked the formatting. Now it is preserved.
321 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
322 that they are available at delivery time.
324 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
326 TF/08 TLS error reporting now respects the incoming_interface and
327 incoming_port log selectors.
329 TF/09 Produce a more useful error message if an SMTP transport's hosts
330 setting expands to an empty string.
332 NM/06 Bugzilla 744: EXPN did not work under TLS.
333 Patch provided by Phil Pennock.
335 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
336 Patch provided by Richard Godbee.
338 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
339 acl_smtp_notquit, added index entry.
341 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
342 Patch provided by Eugene Bujak.
344 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
345 accept(). Patch provided by Maxim Dounin.
347 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
348 Patch provided by Phil Pennock.
350 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
352 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
353 Patch provided by Brad "anomie" Jorsch.
355 NM/14 Bugzilla 622: Added timeout setting to transport filter.
356 Patch provided by Dean Brooks.
358 TK/05 Add native DKIM support (does not depend on external libraries).
360 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
361 Patch provided by Graeme Fowler.
363 NM/16 Bugzilla 851: Documentation example syntax fix.
365 NM/17 Changed NOTICE file to remove references to embedded PCRE.
367 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
370 NM/19 Bugzilla 745: TLS version reporting.
371 Patch provided by Phil Pennock.
373 NM/20 Bugzilla 167: bool: condition support.
374 Patch provided by Phil Pennock.
376 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
377 clients. Patch provided by Phil Pennock.
379 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
380 Patch provided by Brad "anomie" Jorsch.
382 NM/23 Bugzilla 687: Fix misparses in eximstats.
383 Patch provided by Heiko Schlittermann.
385 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
386 Patch provided by Heiko Schlittermann.
388 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
389 plus update to original patch.
391 NM/26 Bugzilla 799: Documentation correction for ratelimit.
393 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
394 Patch provided by David Brownlee.
396 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
398 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
400 NM/30 Bugzilla 888: TLS documentation bugfixes.
402 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
404 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
405 Unlike the original bugzilla I have changed all shell scripts in src tree.
407 NM/33 Bugzilla 898: Transport filter timeout fix.
408 Patch by Todd Rinaldo.
410 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
411 Patch by Serge Demonchaux.
413 NM/35 Bugzilla 39: Base64 decode bug fixes.
414 Patch by Jakob Hirsch.
416 NM/36 Bugzilla 909: Correct connect() call in dcc code.
418 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
420 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
422 NM/39 Bugzilla 911: Fixed MakeLinks build script.
428 TK/01 Add preliminary DKIM support. Currently requires a forked version of
429 ALT-N's libdkim that I have put here:
430 http://duncanthrax.net/exim-experimental/
432 Note to Michael Haardt: I had to rename some vars in sieve.c. They
433 were called 'true' and it seems that C99 defines that as a reserved
434 keyword to be used with 'bool' variable types. That means you could
435 not include C99-style headers which use bools without triggering
436 build errors in sieve.c.
438 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
439 as mailq or other aliases. Changed the --help handling significantly
440 to do whats expected. exim_usage() emits usage/help information.
442 SC/01 Added the -bylocaldomain option to eximstats.
444 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
446 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
448 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
454 PH/01 Another patch from the Sieve maintainer.
456 PH/02 When an IPv6 address is converted to a string for single-key lookup
457 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
458 dots are used instead of colons so that keys in lsearch files need not
459 contain colons. This was done some time before quoting was made available
460 in lsearch files. However, iplsearch files do require colons in IPv6 keys
461 (notated using the quote facility) so as to distinguish them from IPv4
462 keys. This meant that lookups for IP addresses in host lists did not work
463 for iplsearch lookups.
465 This has been fixed by arranging for IPv6 addresses to be expressed with
466 colons if the lookup type is iplsearch. This is not incompatible, because
467 previously such lookups could never work.
469 The situation is now rather anomolous, since one *can* have colons in
470 ordinary lsearch keys. However, making the change in all cases is
471 incompatible and would probably break a number of configurations.
473 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
476 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
477 conversion specification without a maximum field width, thereby enabling
478 a rogue spamd server to cause a buffer overflow. While nobody in their
479 right mind would setup Exim to query an untrusted spamd server, an
480 attacker that gains access to a server running spamd could potentially
481 exploit this vulnerability to run arbitrary code as the Exim user.
483 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
484 $primary_hostname instead of what libspf2 thinks the hosts name is.
486 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
487 a directory entry by the name of the lookup key. Previously, if a
488 symlink pointed to a non-existing file or a file in a directory that
489 Exim lacked permissions to read, a lookup for a key matching that
490 symlink would fail. Now it is enough that a matching directory entry
491 exists, symlink or not. (Bugzilla 503.)
493 PH/03 The body_linecount and body_zerocount variables are now exported in the
496 PH/04 Added the $dnslist_matched variable.
498 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
499 This means they are set thereafter only if the connection becomes
502 PH/06 Added the client_condition to authenticators so that some can be skipped
503 by clients under certain conditions.
505 PH/07 The error message for a badly-placed control=no_multiline_responses left
506 "_responses" off the end of the name.
508 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
510 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
511 (without spaces) instead of just copying the configuration text.
513 PH/10 Added the /noupdate option to the ratelimit ACL condition.
515 PH/11 Added $max_received_linelength.
517 PH/12 Added +ignore_defer and +include_defer to host lists.
519 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
520 way in which PCRE > 7.0 is built.
522 PH/14 Implemented queue_only_load_latch.
524 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
525 MAIL command. The effect was to mangle the value on 64-bit systems.
527 PH/16 Another patch from the Sieve maintainer.
529 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
531 PH/18 If a system quota error occurred while trying to create the file for
532 a maildir delivery, the message "Mailbox is full" was not appended to the
533 bounce if the delivery eventually timed out. Change 4.67/27 below applied
534 only to a quota excession during the actual writing of the file.
536 PH/19 It seems that peer DN values may contain newlines (and other non-printing
537 characters?) which causes problems in log lines. The DN values are now
538 passed through string_printing() before being added to log lines.
540 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
541 and InterBase are left for another time.)
543 PH/21 Added message_body_newlines option.
545 PH/22 Guard against possible overflow in moan_check_errorcopy().
547 PH/23 POSIX allows open() to be a macro; guard against that.
549 PH/24 If the recipient of an error message contained an @ in the local part
550 (suitably quoted, of course), incorrect values were put in $domain and
551 $local_part during the evaluation of errors_copy.
557 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
558 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
561 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
562 issue a MAIL command.
564 PH/02 In an ACL statement such as
566 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
568 if a client was not listed at all, or was listed with a value other than
569 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
570 the condition was not true (as it should be), so access was not denied.
571 The bug was that the ! inversion was incorrectly passed on to the second
572 item. This has been fixed.
574 PH/03 Added additional dnslists conditions == and =& which are different from
575 = and & when the dns lookup returns more than one IP address.
577 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
578 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
580 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
581 FSYNC, which compiles an option called disable_fsync that allows for
582 bypassing fsync(). The documentation is heavily laced with warnings.
584 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
586 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
587 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
588 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
589 including adding "make clean"; (3) Added -fPIC when compiling the test
590 dynamically loaded module, to get rid of a warning.
592 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
593 message fails, move_frozen_messages = true and ignore_bounce_errors_after
594 = 0s. The bug is otherwise harmless.
596 PH/07 There was a bug in the dovecot authenticator such that the value of
597 $auth1 could be overwritten, and so not correctly preserved, after a
598 successful authentication. This usually meant that the value preserved by
599 the server_setid option was incorrect.
601 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
603 PH/09 Installed PCRE release 7.0.
605 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
606 run for batched SMTP input. It is now run at the start of every message
607 in the batch. While fixing this I discovered that the process information
608 (output by running exiwhat) was not always getting set for -bs and -bS
609 input. This is fixed, and it now also says "batched" for BSMTP.
611 PH/11 Added control=no_pipelining.
613 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
614 patch, slightly modified), and move the expansion of helo_data till after
615 the connection is made in the smtp transport (so it can use these
618 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
620 PH/14 Added log_selector = +pid.
622 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
624 PH/16 Add ${if forany and ${if forall.
626 PH/17 Added dsn_from option to vary the From: line in DSNs.
628 PH/18 Flush SMTP output before performing a callout, unless control =
629 no_callout_flush is set.
631 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
632 was true (the default) a successful delivery failed to delete the retry
633 item, thus causing premature timeout of the address. The bug is now
636 PH/20 Added hosts_avoid_pipelining to the smtp transport.
638 PH/21 Long custom messages for fakedefer and fakereject are now split up
639 into multiline reponses in the same way that messages for "deny" and
640 other ACL rejections are.
642 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
643 with slight modification.
645 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
646 draft, changing the syntax and factoring some duplicate code".
648 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
649 for deliveries of the second and subsequent messages over the same SMTP
652 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
653 ${reduce, with only minor "tidies".
655 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
657 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
658 expansion side effects.
660 PH/27 When a message times out after an over-quota error from an Exim-imposed
661 quota, the bounce message says "mailbox is full". This message was not
662 being given when it was a system quota that was exceeded. It now should
665 MH/03 Made $recipients available in local_scan(). local_scan() already has
666 better access to the recipient list through recipients_list[], but
667 $recipients can be useful in postmaster-provided expansion strings.
669 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
670 in the case of a MAIL command with additional options following the
671 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
672 were accidentally chopped off.
674 PH/29 SMTP synchronization checks are implemented when a command is read -
675 there is a check that no more input is waiting when there shouldn't be
676 any. However, for some commands, a delay in an ACL can mean that it is
677 some time before the response is written. In this time, more input might
678 arrive, invalidly. So now there are extra checks after an ACL has run for
679 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
680 pipelining has not been advertised.
682 PH/30 MH's patch to allow iscntrl() characters to be list separators.
684 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
685 returned in the SMTP response when smtp_return_error_details was false.
688 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
689 instead of the C I/O that was originally supplied, because problems were
692 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
693 Exim which did not show up earlier: it was assuming that a call to
694 SSL_CTX_set_info_callback() might give an error value. In fact, there is
695 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
696 was a macro that became an assignment, so it seemed to work. This has
697 changed to a proper function call with a void return, hence the compile
698 error. Exim's code has been fixed.
700 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
703 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
705 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
707 PH/37 If a message is not accepted after it has had an id assigned (e.g.
708 because it turns out to be too big or there is a timeout) there is no
709 "Completed" line in the log. When some messages of this type were
710 selected by exigrep, they were listed as "not completed". Others were
711 picked up by some special patterns. I have improved the selection
712 criteria to be more general.
714 PH/38 The host_find_failed option in the manualroute router can now be set
715 to "ignore", to completely ignore a host whose IP address cannot be
716 found. If all hosts are ignored, the behaviour is controlled by the new
717 host_all_ignored option.
719 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
720 homing or because of multiple MX records with /mx) generated more than
721 one IP address, and the following item turned out to be the local host,
722 all the secondary addresses of the first item were incorrectly removed
723 from the list, along with the local host and any following hosts (which
724 is what is supposed to happen).
726 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
727 whoever called Exim into the -H file. In the case of the daemon it was
728 behaving confusingly. When first started, it used values for whoever
729 started the daemon, but after a SIGHUP it used the Exim user (because it
730 calls itself on a restart). I have changed the code so that it now always
733 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
734 message are rejected with the same error (e.g. no authentication or bad
735 sender address), and a DATA command is nevertheless sent (as can happen
736 with PIPELINING or a stupid MUA), the error message that was given to the
737 RCPT commands is included in the rejection of the DATA command. This is
738 intended to be helpful for MUAs that show only the final error to their
741 PH/42 Another patch from the Sieve maintainer.
743 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
744 Eximstats - Fixed some broken HTML links and added missing column headers
746 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
747 columns for Rejects, Temp Rejects, Ham, and Spam rows.
749 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
751 PH/43 Yet another patch from the Sieve maintainer.
753 PH/44 I found a way to check for a TCP/IP connection going away before sending
754 the response to the final '.' that terminates a message, but only in the
755 case where the client has not sent further data following the '.'
756 (unfortunately, this is allowed). However, in many cases there won't be
757 any further data because there won't be any more messages to send. A call
758 to select() can be used: if it shows that the input is "ready", there is
759 either input waiting, or the socket has been closed. An attempt to read
760 the next input character can distinguish the two cases. Previously, Exim
761 would have sent an OK response which the client would never have see.
762 This could lead to message repetition. This fix should cure that, at
763 least in a lot of common cases.
765 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
766 advertised in response to EHLO.
772 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
773 fixed by 4.65/MH/01 (is this a record?) are fixed:
775 (i) An empty string was always treated as zero by the numeric comparison
776 operators. This behaviour has been restored.
778 (ii) It is documented that the numeric comparison operators always treat
779 their arguments as decimal numbers. This was broken in that numbers
780 starting with 0 were being interpreted as octal.
782 While fixing these problems I realized that there was another issue that
783 hadn't been noticed. Values of message_size_limit (both the global option
784 and the transport option) were treated as octal if they started with 0.
785 The documentation was vague. These values are now always treated as
786 decimal, and I will make that clear in the documentation.
792 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
793 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
796 MH/01 Don't check that the operands of numeric comparison operators are
797 integers when their expansion is in "skipping" mode (fixes bug
798 introduced by 4.64-PH/07).
800 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
801 child addresses, Exim now panics and dies. Previously, because the count
802 is held in a short int, deliveries were likely to be lost. As such a
803 large number of recipients for a single message is ridiculous
804 (performance will be very, very poor), I have chosen to impose a limit
805 rather than extend the field.
811 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
812 leftover -K file (the existence of which was triggered by #402).
813 While we were at it, introduced process PID as part of the -K
814 filename. This should rule out race conditions when creating
817 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
818 processing considerably. Previous code took too long for large mails,
819 triggering a timeout which in turn triggers #401.
821 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
822 in the DK code in transports.c. sendfile() is not really portable,
823 hence the _LINUX specificness.
825 TF/01 In the add_headers option to the mail command in an Exim filter,
826 there was a bug that Exim would claim a syntax error in any
827 header after the first one which had an odd number of characters
830 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
831 callout verification, Exim cached a "reject" for the entire domain. This
832 is correct for most verifications, but it is not correct for a recipient
833 verification with use_sender or use_postmaster set, because in that case
834 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
835 case of MAIL FROM:<> rejection from other early rejections (e.g.
836 rejection of HELO). When verifying a recipient using a non-null MAIL
837 address, the cache is ignored if it shows MAIL FROM:<> rejection.
838 Whatever the result of the callout, the value of the domain cache is
839 left unchanged (for any other kind of callout, getting as far as trying
840 RCPT means that the domain itself is ok).
842 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
845 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
846 manifest itself as EPIPE rather than ECONNECT. When tidying away a
847 session, the daemon ignores ECONNECT errors and logs others; it now
848 ignores EPIPE as well.
850 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
851 (quoted-printable decoding).
853 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
854 later the small subsequent patch to fix an introduced bug.
856 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
858 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
860 PH/08 An error is now given if message_size_limit is specified negative.
862 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
863 to be given (somewhat) arbitrary names.
865 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
868 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
869 miscellaneous code fixes
871 PH/10 Added the log_reject_target ACL modifier to specify where to log
874 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
875 hostname. This is wrong, because it relates to the incoming message (and
876 probably the interface on which it is arriving) and not to the outgoing
877 callout (which could be using a different interface). This has been
878 changed to use the value of the helo_data option from the smtp transport
879 instead - this is what is used when a message is actually being sent. If
880 there is no remote transport (possible with a router that sets up host
881 addresses), $smtp_active_hostname is used.
883 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
884 tweaks were necessary in order to get it to work (see also 21 below):
885 (a) The code assumed that strncpy() returns a negative number on buffer
886 overflow, which isn't the case. Replaced with Exim's string_format()
888 (b) There were several signed/unsigned issues. I just did the minimum
889 hacking in of casts. There is scope for a larger refactoring.
890 (c) The code used strcasecmp() which is not a standard C function.
891 Replaced with Exim's strcmpic() function.
892 (d) The code set only $1; it now sets $auth1 as well.
893 (e) A simple test gave the error "authentication client didn't specify
894 service in request". It would seem that Dovecot has changed its
895 interface. Fortunately there's a specification; I followed it and
896 changed what the client sends and it appears to be working now.
898 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
901 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
902 address A is aliased to B and C, where B exists and C does not. Without
903 -v the output is "A verified" because verification stops after a
904 successful redirection if more than one address is generated. However,
905 with -v the child addresses are also verified. Exim was outputting "A
906 failed to verify" and then showing the successful verification for C,
907 with its parentage. It now outputs "B failed to verify", showing B's
908 parentage before showing the successful verification of C.
910 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
911 look up a TXT record in a specific list after matching in a combined
914 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
915 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
916 they consult the DNS. I had assumed they would set it the way they
917 wanted; and indeed my experiments on Linux seem to show that in some
918 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
919 To be on the safe side, however, I have now made the interface to
920 host_find_byname() similar to host_find_bydns(), with an argument
921 containing the DNS resolver options. The host_find_byname() function now
922 sets these options at its start, just as host_find_bydns() does. The smtp
923 transport options dns_qualify_single and dns_search_parents are passed to
924 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
925 of host_find_byname() use the default settings of RES_DEFNAMES
926 (qualify_single) but not RES_DNSRCH (search_parents).
928 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
929 spool_read_header() do less string testing, by means of a preliminary
930 switch on the second character of optional "-foo" lines. (This is
931 overdue, caused by the large number of possibilities that now exist.
932 Originally there were few.) While I was there, I also converted the
933 str(n)cmp tests so they don't re-test the leading "-" and the first
934 character, in the hope this might squeeze out yet more improvement.
936 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
937 flag allowing group syntax was set by the header_syntax check but not
938 turned off, possible causing trouble later; (2) The flag was not being
939 set at all for the header_verify test, causing "group"-style headers to
940 be rejected. I have now set it in this case, and also caused header_
941 verify to ignore an empty address taken from a group. While doing this, I
942 came across some other cases where the code for allowing group syntax
943 while scanning a header line wasn't quite right (mostly, not resetting
944 the flag correctly in the right place). These bugs could have caused
945 trouble for malformed header lines. I hope it is now all correct.
947 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
948 with the "reply" argument non-NULL. The code, however (which originally
949 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
950 but it didn't always do it. This confused somebody who was copying the
951 code for some other use. I have removed all the tests.
953 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
954 feature that was used to support insecure browsers during the U.S. crypto
955 embargo. It requires special client support, and Exim is probably the
956 only MTA that supported it -- and would never use it because real RSA is
957 always available. This code has been removed, because it had the bad
958 effect of slowing Exim down by computing (never used) parameters for the
959 RSA_EXPORT functionality.
961 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
962 authenticator to fail if there's a tab character in the incoming data
963 (there should never be unless someone is messing about, as it's supposed
964 to be base64-encoded). Also added, on Timo's advice, the "secured" option
965 if the connection is using TLS or if the remote IP is the same as the
966 local IP, and the "valid-client-cert option" if a client certificate has
969 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
970 authenticators. This can be used for authorization after authentication
971 succeeds. (In the case of plaintext, it servers for both authentication
974 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
975 if any retry times were supplied.
977 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
978 connection, where there is no client IP address to check. In this
979 situation, the verify now always succeeds.
981 PH/25 Applied John Jetmore's -Mset patch.
983 PH/26 Added -bem to be like -Mset, but loading a message from a file.
985 PH/27 In a string expansion for a processed (not raw) header when multiple
986 headers of the same name were present, leading whitespace was being
987 removed from all of them, but trailing whitespace was being removed only
988 from the last one. Now trailing whitespace is removed from each header
989 before concatenation. Completely empty headers in a concatenation (as
992 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
993 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
995 PH/29 [Removed. This was a change that I later backed out, and forgot to
996 correct the ChangeLog entry (that I had efficiently created) before
997 committing the later change.]
999 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1000 address errors (4xx response to RCPT) over the same connection as other
1001 messages routed to the same hosts. Such deliveries are always "forced",
1002 so retry times are not inspected. This resulted in far too many retries
1003 for the affected addresses. The effect occurred only when there were more
1004 hosts than the hosts_max_try setting in the smtp transport when it had
1005 the 4xx errors. Those hosts that it had tried were not added to the list
1006 of hosts for which the message was waiting, so if all were tried, there
1007 was no problem. Two fixes have been applied:
1009 (i) If there are any address or message errors in an SMTP delivery, none
1010 of the hosts (tried or untried) are now added to the list of hosts
1011 for which the message is waiting, so the message should not be a
1012 candidate for sending over the same connection that was used for a
1013 successful delivery of some other message. This seems entirely
1014 reasonable: after all the message is NOT "waiting for some host".
1015 This is so "obvious" that I'm not sure why it wasn't done
1016 previously. Hope I haven't missed anything, but it can't do any
1017 harm, as the worst effect is to miss an optimization.
1019 (ii) If, despite (i), such a delivery is accidentally attempted, the
1020 routing retry time is respected, so at least it doesn't keep
1021 hammering the server.
1023 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1024 in ${readsocket because some servers need this prod.
1026 PH/32 Added some extra debug output when updating a wait-xxx database.
1028 PH/33 The hint "could be header name not terminated by colon", which has been
1029 given for certain expansion errors for a long time, was not being given
1030 for the ${if def:h_colon_omitted{... case.
1032 PH/34 The spec says: "With one important exception, whenever a domain list is
1033 being scanned, $domain contains the subject domain." There was at least
1034 one case where this was not true.
1036 PH/35 The error "getsockname() failed: connection reset by peer" was being
1037 written to the panic log as well as the main log, but it isn't really
1038 panic-worthy as it just means the connection died rather early on. I have
1039 removed the panic log writing for the ECONNRESET error when getsockname()
1042 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1043 runs only) independently of the message's sender address. This meant
1044 that, if the 4xx error was in fact related to the sender, a different
1045 message to the same recipient with a different sender could confuse
1046 things. In particualar, this can happen when sending to a greylisting
1047 server, but other circumstances could also provoke similar problems.
1048 I have changed the default so that the retry time for these errors is now
1049 based a combination of the sender and recipient addresses. This change
1050 can be overridden by setting address_retry_include_sender=false in the
1053 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1054 remote server are returned as part of bounce messages. This was not
1055 happening for LMTP over a pipe (the lmtp transport), but now it is the
1056 same for both kinds of LMTP.
1058 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1059 in header lines that were in fact CNAMEs. This is no longer the case.
1061 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1062 and queue runs started by the daemon processed all messages. This has
1063 been fixed so that -R and -S can now usefully be given with -q<time>.
1065 PH/40 Import PCRE release 6.7 (fixes some bugs).
1067 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1069 PH/42 Give an error if -q is specified more than once.
1071 PH/43 Renamed the variables $interface_address and $interface_port as
1072 $received_ip_address and $received_port, to make it clear that these
1073 values apply to message reception, and not to the outgoing interface when
1074 a message is delivered. (The old names remain recognized, of course.)
1076 PH/44 There was no timeout on the connect() call when using a Unix domain
1077 socket in the ${readsocket expansion. There now is.
1079 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1080 be meaningful with "accept".
1082 SC/01 Eximstats V1.43
1083 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1085 SC/02 Eximstats V1.44
1086 Use a glob alias rather than an array ref in the generated
1087 parser. This improves both readability and performance.
1089 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1090 Collect SpamAssassin and rejection statistics.
1091 Don't display local sender or destination tables unless
1092 there is data to show.
1093 Added average volumes into the top table text output.
1095 SC/04 Eximstats V1.46
1096 Collect data on the number of addresses (recipients)
1097 as well as the number of messages.
1099 SC/05 Eximstats V1.47
1100 Added 'Message too big' to the list of mail rejection
1101 reasons (thanks to Marco Gaiarin).
1103 SC/06 Eximstats V1.48
1104 Mainlog lines which have GMT offsets and are too short to
1105 have a flag are now skipped.
1107 SC/07 Eximstats V1.49 (Alain Williams)
1108 Added the -emptyok flag.
1110 SC/08 Eximstats V1.50
1111 Fixes for obtaining the IP address from reject messages.
1113 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1114 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1115 whitesspace changes from 4.64-PH/27
1117 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1120 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1121 are found, allow negative numbers in numeric criteria)
1123 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1125 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1126 to match changes made in 4.64-PH/43
1128 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1130 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1131 "domain" is actually an IP address, and force a failure. This locks out
1132 those revolvers/nameservers that support "A-for-A" lookups, in
1133 contravention of the specifications.
1135 PH/48 When a host name was looked up from an IP address, and the subsequent
1136 forward lookup of the name timed out, the host name was left in
1137 $sender_host_name, contrary to the specification.
1139 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1140 restricted to single-key lookups, Exim was not diagnosing an error if
1141 * or *@ was used with a query-style lookup.
1143 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1145 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1146 long ago, but noone interested enough thought of it. Let's just say that
1147 the "1.1" means that there are some new functions that weren't there at
1148 some point in the past.
1150 PH/51 Error processing for expansion failure of helo_data from an smtp
1151 transport during callout processing was broken.
1153 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1154 tested/used via the -bh/-bhc/-bs options.
1156 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1157 bug, fixed in subsequent PCRE releases).
1159 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1160 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1162 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1168 SC/01 Use a glob alias rather than an array ref in eximstats generated
1169 parser. This improves both readability and performance.
1171 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1172 Don't display local sender or destination tables in eximstats unless
1173 there is data to show.
1174 Added average volumes into the eximstats top table text output.
1176 SC/03 Collect data on the number of addresses (recipients) as well
1177 as the number of messages in eximstats.
1179 TF/01 Correct an error in the documentation for the redirect router. Exim
1180 does (usually) call initgroups() when daemonizing.
1182 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1183 with consistent privilege compared to when running as a daemon.
1185 TF/03 Note in the spec that $authenticated_id is not set for local
1186 submissions from trusted users.
1188 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1189 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1191 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1192 by adding some example configuration directives to the default
1193 configuration file. A little bit of work is required to uncomment the
1194 directives and define how usernames and passwords are checked, but
1195 there is now a framework to start from.
1197 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1198 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1199 without this. I don't know how relevant this is to other LDAP libraries.
1201 PH/02 Add the verb name to the "unknown ACL verb" error.
1203 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1205 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1207 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1208 directory not expanded when it should be if an expanded home directory
1209 was set for the address (which is overridden by the transport).
1211 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1214 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1215 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1216 because it is too late at that time, and has no effect.
1218 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1219 security issue with \' (bugzilla #107). I could not use the
1220 PQescapeStringConn() function, because it needs a PGconn value as one of
1223 PH/08 When testing addresses using -bt, indicate those final addresses that
1224 are duplicates that would not cause an additional delivery. At least one
1225 person was confused, thinking that -bt output corresponded to deliveries.
1226 (Suppressing duplicates isn't a good idea as you lose the information
1227 about possibly different redirections that led to the duplicates.)
1229 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1230 systems where poll() doesn't work, in particular OS X.
1232 PH/10 Added more information to debugging output for retry time not reached.
1234 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1235 operations in malware.c.
1237 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1240 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1241 syslog_duplication set false, log lines that would normally be written
1242 both the the main log and to the reject log were not written to syslog at
1245 PH/14 In the default configuration, change the use of "message" in ACL warn
1246 statements to "add_header".
1248 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1249 not followed by a command (e.g. "seen endif").
1251 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1252 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1255 PH/17 Added extra conditions to the default value of delay_warning_condition
1259 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1260 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1261 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1264 The Auto-Submitted: and various List- headers are standardised, whereas I
1265 don't think Precedence: ever was.
1267 PH/18 Refactored debugging code in route_finduser() to show more information,
1268 in particular, the error code if getpwnam() issues one.
1270 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1271 This is apparently needed in addition to the PH/07 change above to avoid
1272 any possible encoding problems.
1274 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1275 but not after initializing Perl.
1277 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1278 output them only if debugging. By default they are written stderr,
1279 apparently, which is not desirable.
1281 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1284 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1287 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1289 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1290 authenticated or an ident call has been made. Suppress the default
1291 values for $authenticated_id and $authenticated_sender (but permit -oMai
1292 and -oMas) when testing with -bh.
1294 PH/24 Re-jigged the order of the tests in the default configuration so that the
1295 tests for valid domains and recipients precede the DNS black list and CSA
1296 tests, on the grounds that those ones are more expensive.
1298 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1299 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1300 This bug exists in every version of Exim that I still have, right back to
1303 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1304 However, an attempt to turn on case-sensitivity in a regex key by
1305 including (?-i) didn't work because the subject string was already
1306 lowercased, and the effects were non-intuitive. It turns out that a
1307 one-line patch can be used to allow (?-i) to work as expected.
1313 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1314 other effects) broke the use of negated acl sub-conditions.
1316 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1319 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1320 "Deny" causes Exim to reject the incoming connection with a 554 error.
1321 Unfortunately, if there is a major crisis, such as a disk failure,
1322 tcp-wrappers gives "deny", whereas what one would like would be some
1323 kind of temporary error. A kludge has been added to help with this.
1324 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1325 554 error is used if errno is still zero or contains ENOENT (which occurs
1326 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1329 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1331 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1332 errors. Otherwise a message that provokes a temporary error (when other
1333 messages do not) can cause a whole host to time out.
1335 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1336 addresses were routed directly to files or pipes from a redirect router.
1337 File deliveries just didn't batch; pipe deliveries might have suffered
1340 PH/06 A failure to get a lock for a hints database would erroneously always say
1341 "Failed to get write lock", even when it was really a read lock.
1343 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1344 of 0600. This has been changed to use the value of the lockfile_mode
1345 option (which defaults to 0600).
1347 PH/08 Applied small patch from the Sieve maintainer.
1349 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1350 folder from quota calculations, a direct delivery into this folder messed
1351 up the contents of the maildirsize file. This was because the regex was
1352 used only to exclude .Trash (or whatever) when the size of the mailbox
1353 was calculated. There was no check that a delivery was happening into an
1354 excluded directory. This bug has been fixed by ignoring all quota
1355 processing for deliveries into excluded directories.
1357 PH/10 Added the maildirfolder_create_regex option to appendfile.
1363 PH/01 The code for finding all the local interface addresses on a FreeBSD
1364 system running IPv6 was broken. This may well have applied to all BSD
1365 systems, as well as to others that have similar system calls. The broken
1366 code found IPv4 interfaces correctly, but gave incorrect values for the
1367 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1368 that it would not match correctly against @[] and not recognize the IPv6
1371 PH/02 The ipliteral router was not recognizing addresses of the form user@
1372 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1374 PH/03 Added disable_ipv6.
1376 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1377 decoded form, because it is most often used to construct To: headers
1378 lines in autoreplies, and the decoded form may well be syntactically
1379 invalid. However, $reply_address has leading white space removed, and all
1380 newlines turned into spaces so that the autoreply transport does not
1383 PH/05 If group was specified without a user on a router, and no group or user
1384 was specified on a transport, the group from the router was ignored.
1386 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1387 for visible compile-time settings that can be used to change these
1388 numbers, for those that want even more. Backwards compatibility with old
1389 spool files has been maintained. However, going back to a previous Exim
1390 release will lost any variables that are in spool files.
1392 PH/07 Two small changes when running in the test harness: increase delay when
1393 passing a TCP/IP connection to a new process, in case the original
1394 process has to generate a bounce, and remove special handling of
1395 127.0.0.2 (sic), which is no longer necessary.
1397 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1398 be the same on different OS.
1400 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1403 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1404 whether --show-vars was specified or not
1406 JJ/02 exipick: Added support for new ACL variable spool format introduced
1409 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1410 syntactically invalid From: or Reply-to: line, and a filter used this to
1411 generate an autoreply, and therefore failed to obtain an address for the
1412 autoreply, Exim could try to deliver to a non-existent relative file
1413 name, causing unrelated and misleading errors. What now happens is that
1414 it logs this as a hard delivery error, but does not attempt to create a
1417 PH/11 The exinext utility has a -C option for testing purposes, but although
1418 the given file was scanned by exinext itself; it wasn't being passed on
1419 when Exim was called.
1421 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1422 an end-of-file indication when reading a command response.
1424 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1425 compiled. In many other places in Exim, IPv6 addresses are always
1426 recognized, so I have changed this. It also means that IPv4 domain
1427 literals of the form [IPV4:n.n.n.n] are now always recognized.
1429 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1430 used if the router is not running as root, for example, when verifying at
1431 ACL time, or when using -bh. The debugging output from this situation was
1432 non-existent - all you got was a failure to exec. I have made two
1435 (a) Failures to set uid/gid, the current directory, or a process leader
1436 in a subprocess such as that created by queryprogram now generate
1437 suitable debugging ouput when -d is set.
1439 (b) The queryprogram router detects when it is not running as root,
1440 outputs suitable debugging information if -d is set, and then runs
1441 the subprocess without attempting to change uid/gid.
1443 PH/15 Minor change to Makefile for building test_host (undocumented testing
1446 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1447 additional section of a DNS packet that returns MX or SRV records.
1448 Instead, it always explicitly searches for A/AAAA records. This avoids
1449 major problems that occur when a DNS server includes only records of one
1450 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1451 fixed another bug: if SRV records were looked up and the corresponding
1452 address records were *not* found in the additional section, the port
1453 values from the SRV records were lost.
1455 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1456 using the correct key (the original address) when searching the retry
1457 rules in order to find which one to use for generating the retry hint.
1459 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1460 adding the default one. Similarly, if it contains a Reply-To: header, the
1461 errors_reply_to option, if set, is not used.
1463 PH/19 When calculating a retry time, Exim used to measure the "time since
1464 failure" by looking at the "first failed" field in the retry record. Now
1465 it does not use this if it is later than than the arrival time of the
1466 message. Instead it uses the arrival time. This makes for better
1467 behaviour in cases where some deliveries succeed, thus re-setting the
1468 "first failed" field. An example is a quota failure for a huge message
1469 when small messages continue to be delivered. Without this change, the
1470 "time since failure" will always be short, possible causing more frequent
1471 delivery attempts for the huge message than are intended.
1472 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1474 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1475 $1, $2, $3) because the numerical variables can be reset during some
1476 expansion items (e.g. "match"), thereby losing the authentication data.
1478 PH/21 Make -bV show the size of off_t variables so that the test suite can
1479 decide whether to run tests for quotas > 2G.
1481 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1482 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1483 filecount value is greater than 2G or if a quota value is greater than 2G
1484 on a system where the size of off_t is not greater than 4, a panic error
1487 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1488 never match. The debug and -bh output now contains an explicit error
1489 message indicating a malformed IPv4 address or mask.
1491 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1492 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1493 PH/23 above applies.
1495 PH/25 Do not write to syslog when running in the test harness. The only
1496 occasion when this arises is a failure to open the main or panic logs
1497 (for which there is an explicit test).
1499 PH/26 Added the /no_tell option to "control=freeze".
1501 PH/27 If a host name lookup failed very early in a connection, for example, if
1502 the IP address matched host_lookup and the reverse lookup yielded a name
1503 that did not have a forward lookup, an error message of the form "no IP
1504 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1505 could be logged. Now it outputs the IP address instead of "NULL".
1507 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1508 allows the sender and the authenticated sender to be set when
1509 submitting a message from within Exim. Since child_open_exim() is
1510 documented for local_scan(), the new function should be too.
1512 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1513 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1514 results in an empty string is now treated as unset.
1516 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1518 PH/31 Added sender_verify_fail logging option.
1520 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1521 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1522 tidied the source and removed it altogether.
1524 PH/33 When a queue run was abandoned because the load average was too high, a
1525 log line was always written; now it is written only if the queue_run log
1526 selector is set. In addition, the log line for abandonment now contains
1527 information about the queue run such as the pid. This is always present
1528 in "start" and "stop" lines but was omitted from the "abandon" line.
1530 PH/34 Omit spaces between a header name and the colon in the error message that
1531 is given when verify = headers_syntax fails (if there are lots of them,
1532 the message gets confusing).
1534 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1535 names, as there are now some PTR records that contain slashes. This check
1536 is only to protect against broken name servers that fall over on strange
1537 characters, so the fact that it applies to all lookups doesn't matter.
1539 PH/36 Now that the new test suite is complete, we can remove some of the
1540 special code in Exim that was needed for the old test suite. For example,
1541 sorting DNS records because real resolvers return them in an arbitrary
1542 order. The new test suite's fake resolver always returns records in the
1545 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1546 bounces) except when queue_only is set, to avoid logging races between
1547 the different processes.
1549 PH/38 Panic-die if .include specifies a non-absolute path.
1551 PH/39 A tweak to the "H" retry rule from its user.
1553 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1554 a label. They prevented compilation on older perls.
1556 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1557 a warning to be raised on newish perls.
1559 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1560 on queue. Changes to match documented behaviour of showing count of
1561 messages matching specified criteria.
1563 PH/40 Changed the default ident timeout from 30s to 5s.
1565 PH/41 Added support for the use of login_cap features, on those BSD systems
1566 that have them, for controlling the resources used by pipe deliveries.
1568 PH/42 The content-scanning code uses fopen() to create files in which to put
1569 message data. Previously it was not paying any attention to the mode of
1570 the files. Exim runs with umask(0) because the rest of the code creates
1571 files with open(), and sets the required mode explicitly. Thus, these
1572 files were ending up world-writeable. This was not a big issue, because,
1573 being within the spool directory, they were not world-accessible. I have
1574 created a function called modefopen, which takes an additional mode
1575 argument. It sets umask(777), creates the file, chmods it to the required
1576 mode, then resets the umask. All the relevant calls to fopen() in the
1577 content scanning code have been changed to use this function.
1579 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1580 to 24 hours. This avoids potential overflow problems when processing G
1581 and H retry rules. I suspect nobody ever tinkers with this value.
1583 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1585 PH/45 When the plaintext authenticator is running as a client, the server's
1586 challenges are checked to ensure they are valid base64 strings. By
1587 default, the authentication attempt is cancelled if an invalid string is
1588 received. Setting client_ignore_invalid_base64 true ignores these errors.
1589 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1590 they are received. Thus, the responses can be made to depend on the
1591 challenges. If an invalid string is ignored, an empty string is placed in
1594 PH/46 Messages that are created by the autoreply transport now contains a
1595 References: header, in accordance with RFCs 2822 and 3834.
1597 PH/47 Added authenticated_sender_force to the smtp transport.
1599 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1601 PH/49 Installed latest patch from the Sieve maintainer.
1603 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1604 was also set, the appendfile transport was unnecessarily scanning a
1605 directory of message files (e.g. for maildir delivery) to find the count
1606 of files (along with the size), even though it did not need this
1607 information. It now does the scan only if it needs to find either the
1608 size of the count of files.
1610 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1612 PH/52 Two bugs concerned with error handling when the smtp transport is
1615 (i) Exim was not creating retry information for temporary errors given
1616 for individual recipients after the DATA command when the smtp transport
1617 was used in LMTP mode. This meant that they could be retried too
1618 frequently, and not timed out correctly.
1620 (ii) Exim was setting the flag that allows error details to be returned
1621 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1622 recipients that were returned after the DATA command.
1624 PH/53 This is related to PH/52, but is more general: for any failing address,
1625 when detailed error information was permitted to be returned to the
1626 sender, but the error was temporary, then after the final timeout, only
1627 "retry timeout exceeded" was returned. Now it returns the full error as
1628 well as "retry timeout exceeded".
1630 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1631 do this, and (what is worse) MTAs that accept it.
1633 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1634 will now be deprecated.
1636 PH/56 New os.c-cygwin from the Cygwin maintainer.
1638 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1639 formats (previously only available in exim formats via -bpr, -bpru,
1640 and -bpra. Now also available in native and exiqgrep formats)
1642 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1643 with very large, slow to parse queues
1645 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1647 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1649 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1650 responses to authentication challenges, though it was showing the
1651 challenges; (ii) I've removed the CR characters from the debug output for
1654 PH/58 Allow for the insertion of a newline as well as a space when a string
1655 is turned into more than one encoded-word during RFC 2047 encoding. The
1656 Sieve code now uses this.
1658 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1659 data_4xx, lost_connection, tls_required.
1661 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1662 message was being sent as an SMTP response.
1664 PH/61 Add -l and -k options to exicyclog.
1666 PH/62 When verifying, if an address was redirected to one new address, so that
1667 verification continued, and the new address failed or deferred after
1668 having set something in $address_data, the value of $address_data was not
1669 passed back to the ACL. This was different to the case when no
1670 redirection occurred. The value is now passed back in both cases.
1672 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1673 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1674 use login_cap.h, so on its own it isn't the distinguishing feature. The
1675 new name refers directly to the setclassresources() function.
1677 PH/65 Added configuration files for NetBSD3.
1679 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1681 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1682 is preferred over IPv4.
1684 PH/68 The bounce_return_message and bounce_return_body options were not being
1685 honoured for bounces generated during the reception of non-SMTP messages.
1686 In particular, this applied to messages rejected by the ACL. This bug has
1687 been fixed. However, if bounce_return_message is true and bounce_return_
1688 body is false, the headers that are returned for a non-SMTP message
1689 include only those that have been read before the error was detected.
1690 (In the case of an ACL rejection, they have all been read.)
1692 PH/69 The HTML version of the specification is now built in a directory called
1693 spec_html instead of spec.html, because the latter looks like a path with
1694 a MIME-type, and this confuses some software.
1696 PH/70 Catch two compiler warnings in sieve.c.
1698 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1699 function verify_get_ident() calls ip_connect() to connect a socket, but
1700 if the "connect()" function timed out, ip_connect() used to close the
1701 socket. However, verify_get_ident() also closes the socket later, and in
1702 between Exim writes to the log, which may get opened at this point. When
1703 the socket was closed in ip_connect(), the log could get the same file
1704 descriptor number as the socket. This naturally causes chaos. The fix is
1705 not to close the socket in ip_connect(); the socket should be closed by
1706 the function that creates it. There was only one place in the code where
1707 this was missing, in the iplookup router, which I don't think anybody now
1708 uses, but I've fixed it anyway.
1710 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1711 well as to direct DNS lookups. Otherwise the handling of names in host
1712 lists is inconsistent and therefore confusing.
1718 PH/01 Two changes to the default runtime configuration:
1720 (1) Move the checks for relay_from_hosts and authenticated clients from
1721 after to before the (commented out) DNS black list checks.
1723 (2) Add control=submission to the relay_from_hosts and authenticated
1724 clients checks, on the grounds that messages accepted by these
1725 statements are most likely to be submissions.
1727 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1729 (1) Generate an error if the third argument for the ${prvs expansion is
1732 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1735 (3) Reset the variables that are obtained from the first argument of
1736 ${prvscheck and used in the second argument before leaving the code,
1737 because their memory is reclaimed, so using them afterwards may do
1740 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1741 one (it's much easier than Tom thought :-).
1743 (5) Because of (4), we can now allow for the use of $prvscheck_result
1744 inside the third argument.
1746 PH/03 For some reason, the default setting of PATH when running a command from
1747 a pipe transport was just "/usr/bin". I have changed it to
1750 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1751 anything to be listed in the output from -bV.
1753 PH/05 When a filter generated an autoreply, the entire To: header line was
1754 quoted in the delivery log line, like this:
1756 => >A.N.Other <ano@some.domain> <original@ddress> ...
1758 This has been changed so that it extracts the operative address. There
1759 may be more than one such address. If so, they are comma-separated, like
1762 => >ano@some.domain,ona@other.domain <original@ddress> ...
1764 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1765 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1766 not being looked up in the rDNS to get a host name, Exim was showing the
1767 IP address twice in Received: lines, even though the IP addresses were
1768 identical. For example:
1770 Received: from [1.2.3.4] (helo=[1.2.3.4])
1772 However, if the real host name was known, it was omitting the HELO data
1773 if it matched the actual IP address. This has been tidied up so that it
1774 doesn't show the same IP address twice.
1776 PH/07 When both +timestamp and +memory debugging was on, the value given by
1777 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1778 called by the debug printing, thereby overwriting the timestamp buffer.
1779 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1781 PH/08 When the original message was included in an autoreply transport, it
1782 always said "this is a copy of the message, including all the headers",
1783 even if body_only or headers_only was set. It now gives an appropriate
1786 PH/09 Applied a patch from the Sieve maintainer which:
1788 o fixes some comments
1789 o adds the (disabled) notify extension core
1790 o adds some debug output for the result of if/elsif tests
1791 o points to the current vacation draft in the documentation
1792 and documents the missing references header update
1796 o fixes a bug in processing the envelope test (when testing
1797 multiple envelope elements, the last element determinted the
1800 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1801 Electronic Mail") by including:
1803 Auto-submitted: auto-generated
1805 in the messages that it generates (bounce messages and others, such as
1806 warnings). In the case of bounce messages for non-SMTP mesages, there was
1807 also a typo: it was using "Auto_submitted" (underscore instead of
1808 hyphen). Since every message generated by Exim is necessarily in response
1809 to another message, thes have all been changed to:
1811 Auto-Submitted: auto-replied
1813 in accordance with these statements in the RFC:
1815 The auto-replied keyword:
1817 - SHOULD be used on messages sent in direct response to another
1818 message by an automatic process,
1820 - MUST NOT be used on manually-generated messages,
1822 - MAY be used on Delivery Status Notifications (DSNs) and Message
1823 Disposition Notifications (MDNs),
1825 - MUST NOT be used on messages generated by automatic or periodic
1826 processes, except for messages which are automatic responses to
1829 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1830 to the default Received: header definition.
1832 PH/12 Added log selector acl_warn_skipped (default on).
1834 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1835 variables because (a) they are in the wrong storage pool and (b) even if
1836 they were copied, it wouldn't work properly because of the caching.
1838 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1839 checking when decoding. Apparently there are clients that generate
1840 overlong encoded strings. Why am I not surprised?
1842 PH/15 If the first argument of "${if match_address" was not empty, but did not
1843 contain an "@" character, Exim crashed. Now it writes a panic log message
1844 and treats the condition as false.
1846 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1848 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1849 extension "envelope-auth". The code is finished and in agreement with
1850 other implementations, but there is no documentation so far and in fact,
1851 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1852 not changing the active code.
1854 Print executed "if" and "elsif" statements when debugging is used. This
1855 helps a great deal to understand what a filter does.
1857 Document more things not specified clearly in RFC3028. I had all this
1858 sorted out, when out of a sudden new issues came to my mind. Oops."
1860 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1863 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1864 canonical form (as documented). However, after a host name lookup from
1865 the IP address, check_host() was doing a simple string comparison with
1866 addresses acquired from the DNS when checking that the found name did
1867 have the original IP as one of its addresses. Since any found IPv6
1868 addresses are likely to be in abbreviated form, the comparison could
1869 fail. Luckily, there already exists a function for doing the comparison
1870 by converting both addresses to binary, so now that is used instead of
1871 the text comparison.
1873 PH/20 There was another similar case to PH/19, when a complete host name was
1874 given in a host list; looking up its IP address could give an abbreviated
1875 form, whereas the current host's name might or might not be abbreviated.
1876 The same fix has been applied.
1882 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1883 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1886 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1887 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1889 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1891 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1892 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1893 Also, for "Auto-Submitted", treat anything other than "no" as
1894 non-personal, in accordance with RFC 3834. (Previously it treated
1895 anything starting "auto-" as non-personal.)
1897 TF/01 The control=submission/name=... option had a problem with syntax
1898 errors if the name included a slash character. The /name= option
1899 now slurps the rest of the string, so it can include any characters
1900 but it must come last in the list of options (after /sender_retain
1903 PH/05 Some modifications to the interface to the fake nameserver for the new
1911 TK/01 Added the "success_on_redirect" address verification option. See
1912 NewStuff for rationale and an example.
1914 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1916 PH/02 Patch to exigrep to allow it to work on syslog lines.
1918 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1919 fread() to skip over the body file's header line, because in Cygwin the
1920 header line is locked and is inaccessible.
1922 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1923 co-exist for some time) to make it clear that it is the Exim ID that is
1924 referenced, not the Message-ID: header line.
1926 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1927 string_format() function, because snprintf() does not exist on all
1930 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1931 ${stat: expansion item.
1933 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1934 protocol synchronization error", to keep the pedants happy.
1936 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1937 well as for IRIX systems, when gcc is being used. See the host.c source
1940 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1942 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1945 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1946 transport and to the smtp transport in LMTP mode.
1948 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1950 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1951 run a filter in a subprocess. This could lead to confusion in subsequent
1952 lookups in the parent process. There should also be a search_tidyup() at
1953 the end of the subprocess.
1955 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1956 only if the host matched helo_try_verify_hosts, which caused the
1957 verification to occur when the EHLO/HELO command was issued. The ACL just
1958 tested the remembered result. Now, if a previous verification attempt has
1959 not happened, "verify = helo" does it there and then.
1961 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1963 TK/03 Fix log output including CR from clamd.
1965 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1966 exist provoked a memory error which could cause a segfault.
1968 PH/15 Installed PCRE 6.2
1970 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1972 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1973 of the problem. Specifically, suggested +O2 rather than +O1 for the
1976 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1978 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1979 record, the port was not correctly logged when the outgoing_port log
1980 selector was set (it logged the transort's default port).
1982 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1983 fallback_hosts, and "hosts" in the smtp transport.
1985 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1986 host errors such as "Connection refused".
1988 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1989 authentication with radiusclient 0.4.9:
1991 - Error returned from rc_read_config was caught wrongly
1992 - Username/password not passed on to radius server due to wrong length.
1994 The presumption is that some radiusclient API changes for 4.51/PH/17
1995 were not taken care of correctly. The code is still untested by me (my
1996 Linux distribution still has 0.3.2 of radiusclient), but it was
1997 contributed by a Radius user.
1999 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2000 expanding the "port" option of the smtp transport.
2002 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2003 while reading a MIME header. Thanks to Tom Hughes for a patch.
2005 PH/24 Include config.h inside local_scan.h so that configuration settings are
2008 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2009 that in an ACL for RCPT (for example), you can examine exactly what was
2012 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2013 commands, but it was not correctly comparing the address with the actual
2014 client host address. Thus, it would show the EHLO address in Received:
2015 header lines when this was not necessary.
2017 PH/27 Added the % operator to ${eval:}.
2019 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2020 it should be ignoring failures (because with -C, for example, it has lost
2021 privilege). It wasn't ignoring creation failures other than "already
2024 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2027 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2028 because an input line was too long, either on its own, or by virtue of
2029 too many continuations, the temporary file was not being removed, and the
2030 return code was incorrect.
2032 PH/31 Missing "BOOL" in function definition in filtertest.c.
2034 PH/32 Applied Sieve patches from the maintainer.
2036 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2038 PH/33 Added "verify = not_blind".
2040 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2041 Local/Makefile (with some defaults set). These are used in built scripts
2042 such as exicyclog, but they have never been used in the exim_install
2043 script (though there are many overriding facilities there). I have
2044 arranged that the exim_install script now takes note of these two
2047 PH/35 Installed configuration files for Dragonfly.
2049 PH/36 When a locally submitted message by a trusted user did not contain a
2050 From: header, and the sender address was obtained from -f or from an SMTP
2051 MAIL command, and the trusted user did not use -F to supply a sender
2052 name, $originator_name was incorrectly used when constructing a From:
2053 header. Furthermore, $originator_name was used for submission mode
2054 messages from external hosts without From: headers in a similar way,
2055 which is clearly wrong.
2057 PH/37 Added control=suppress_local_fixups.
2059 PH/38 When log_selector = +received_sender was set, and the addition of the
2060 sender made the log line's construction buffer exactly full, or one byte
2061 less than full, an overflow happened when the terminating "\n" was
2064 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2065 when the result of a list match is failure because a DNS lookup failed.
2067 PH/40 RM_COMMAND is now used in the building process.
2069 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2070 the "build-* directories that it finds.
2072 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2073 address in a domain literal was a prefix of an interface address.
2075 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2076 when verifying a sender address, unless rewrite_headers is false.
2078 PH/44 Wrote a long comment about why errors_to addresses are verified as
2079 recipients, not senders.
2081 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2082 the ratelimit ACL was added.
2084 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2086 PH/46 Added extra information about PostgreSQL errors to the error string.
2088 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2089 avoiding the need to install special zones in a real server. This is
2090 backwards compatible; if it can't find the fake resolver, it drops back.
2091 Thus, both old and new test suites can be run.
2093 TF/02 Added util/ratelimit.pl
2095 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2096 clock is set back in time.
2098 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2099 Candler <B.Candler@pobox.com>.
2101 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2102 OK for addresses that are the result of redirections.
2104 PH/48 A number of further additions for the benefit of the new test suite,
2105 including a fake gethostbyname() that interfaces to the fake DNS resolver
2108 TF/06 The fix for widen_domains has also been applied to qualify_single and
2109 search_parents which are the other dnslookup options that can cause
2112 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2115 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2117 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2118 service" if it encountered an MX record with an empty target hostname.
2119 The message is now "an MX or SRV record indicated no SMTP service".
2121 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2122 if the DNS of the sending site is misconfigured. This is quite a
2123 common situation. This change restores the behaviour of treating a
2124 helo verification defer as a failure.
2126 PH/51 If self=fail was set on a router, the bounce message did not include the
2127 actual error message.
2133 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2135 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2136 command itself ended in error, the underlying message about the transport
2137 filter timeout was being overwritten with the pipe command error. Now the
2138 underlying error message should be appended to the second error message.
2140 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2142 PH/02 Reduce the amount of output that "make" produces by default. Full output
2143 can still be requested.
2145 PH/03 The warning log line about a condition test deferring for a "warn" verb
2146 was being output only once per connection, rather than after each
2147 occurrence (because it was using the same function as for successful
2148 "warn" verbs). This seems wrong, so I have changed it.
2150 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2151 it should not have, which might have caused a crash in the right
2152 circumstances, but probably never did.
2154 PH/04 Installed a modified version of Tony Finch's patch to make submission
2155 mode fix the return path as well as the Sender: header line, and to
2156 add a /name= option so that you can make the user's friendly name appear
2159 TF/03 Added the control = fakedefer ACL modifier.
2161 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2162 Mark Lowes for thorough testing.
2164 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2166 TK/03 Merged latest SRS patch from Miles Wilton.
2168 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2169 with the definition in sysexits.h (which is #included earlier).
2170 Fortunately, Exim does not actually use EX_OK. The code used to try to
2171 preserve the sysexits.h value, by assumimg that macro definitions were
2172 scanned for macro replacements. I have been disabused of this notion,
2173 so now the code just undefines EX_OK before #including unistd.h.
2175 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2176 in the smtp transport. When a block could not be written in a single
2177 write() function, the timeout was being re-applied to each part-write.
2178 This seems wrong - if the receiver was accepting one byte at a time it
2179 would take for ever. The timeout is now adjusted when this happens. It
2180 doesn't have to be particularly precise.
2182 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2183 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2185 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2186 without a domain if the check to <postmaster@domain> fails.
2188 SC/01 Eximstats: added -xls and the ability to specify output files
2189 (patch written by Frank Heydlauf).
2191 SC/02 Eximstats: use FileHandles for outputing results.
2193 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2195 SC/04 Eximstats: fixed display of large numbers with -nvr option
2197 SC/05 Eximstats: fixed merging of reports with empty tables.
2199 SC/06 Eximstats: added the -include_original_destination flag
2201 SC/07 Eximstats: removed tabs and trailing whitespace.
2203 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2205 TK/06 MBOX spool code: Add real "From " MBOX separator line
2206 so the .eml file is really in mbox format (even though
2207 most programs do not really care). Patch from Alex Miller.
2209 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2210 The latter is generated from $received_to and is only set if the
2211 message has one envelope recipient. SA can use these headers,
2212 obviously out-of-the-box. Patch from Alex Miller.
2214 PH/08 The ${def test on a variable was returning false if the variable's
2215 value was "0", contrary to what the specification has always said!
2216 The result should be true unless the variable is empty.
2218 PH/09 The syntax error of a character other than { following "${if
2219 def:variable_name" (after optional whitespace) was not being diagnosed.
2220 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2221 accidental colon was present, for example, could give incorrect results.
2223 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2224 result is used (not including appendfile, where other changes are about
2227 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2228 This involved changing a lot of size variables from int to off_t. It
2229 should work with maildirs and everything.
2231 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2232 spamd dying while we are connected to it.
2234 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2237 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2238 function for BDB 4.3.
2240 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2242 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2243 the sources for PCRE have been re-arranged and more files are now
2246 PH/15 The code I had for printing potentially long long variables in PH/11
2247 above was not the best (it lost precision). The length of off_t variables
2248 is now inspected at build time, and an appropriate printing format (%ld
2249 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2250 to be "long long int" or "long int". This is needed for the internal
2251 formatting function string_vformat().
2253 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2254 the configuration file to be ":syslog", then the script "guesses" where
2255 the logs files are, rather than using the compiled in default. In our
2256 case the guess is not the same as the compiled default, so the script
2257 suddenly stopped working when I started to use syslog. The patch checks
2258 to see if log_file_path is "". If so, it attempts to read it from exim
2259 with no configuration file to get the compiled in version, before it
2260 falls back to the previous guessing code."
2262 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2263 implementing BATV in an Exim configuration. See NewStuff for the gory
2266 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2267 Makefile that are specific to HP-UX.
2269 PH/18 If the "use_postmaster" option was set for a recipient callout together
2270 with the "random" option, the postmaster address was used as the MAIL
2271 FROM address for the random test, but not for the subsequent recipient
2272 test. It is now used for both.
2274 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2275 patch removes a few documentation additions to RFC 3028, because the
2276 latest draft now contains them. It adds the new en;ascii-case comparator
2277 and a new error check for 8bit text in MIME parts. Comparator and
2278 require names are now matched exactly. I enabled the subaddress
2279 extension, but it is not well tested yet (read: it works for me)."
2281 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2282 rework some of the code of TK/09 above to avoid the hardwired use of
2283 "%lld" and "long long". Replaced the call to snprintf() with a call to
2286 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2287 records point to non-existent hosts", "retry timeout exceeded", and
2288 "retry time not reached for any host after a long failure period".
2290 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2291 experimental DomainKeys support:
2293 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2294 (2) On an error such as an illegally used "control", the wrong name for
2295 the control was given.
2297 These problems did NOT occur unless DomainKeys support was compiled.
2299 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2301 PH/24 Added ${if match_ip condition.
2303 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2304 databases so that it will be absolutely obvious if a crash occurs in the
2305 DB library. This is a regular occurrence (often caused by mis-matched
2308 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2309 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2310 picked up on a user's system that detects such things. There doesn't seem
2311 to be a gcc warning option for this - only an attribute that has to be
2312 put on the function's prototype. It seems that in Fedora Core 4 they have
2313 set this on a number of new functions. No doubt there will be more in due
2316 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2317 specify a transport. However, if an address that was verified by such a
2318 router was the subject of a callout, Exim crashed because it tried to
2319 read the rcpt_include_affixes from the non-existent transport. Now it
2320 just assumes that the setting of that option is false. This bug was
2321 introduced by 4.51/PH/31.
2323 PH/28 Changed -d+all to exclude +memory, because that information is very
2324 rarely of interest, but it makes the output a lot bigger. People tend to
2325 do -d+all out of habit.
2327 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2328 code in os-type was giving problems when libc.so lives in lib64, like on
2331 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2332 aren't the modern standard, and it seems that some systems' include files
2333 don't always have them. Exim was already checking for some of the newer
2334 ones like T_AAAA, and defining it itself. I've added checks for all the
2335 record types that Exim uses.
2337 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2338 not automatically generating a new one, as it is supposed to. This
2339 prevented TLS from working. If the file did exist, but contained invalid
2340 data, a new version was generated, as expected. It was only the case of a
2341 non-existent file that was broken.
2343 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2344 with a change in libdomainkeys > 0.64.
2346 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2347 from DNS. If the selector record carries the flag, it now has
2348 precedence over the domain-wide flag.
2350 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2352 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2353 the use of a port name that isn't defined in /etc/services) occurred, the
2354 message was deferred as in a normal delivery, and thus remained on the
2355 spool, instead of being failed because of the mua_wrapper setting. This
2356 is now fixed, and I tidied up some of the mua_wrapper messages at the
2359 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2360 the messages in a hash of arrays rather than using individual hashes.
2361 This is a bit cleaner and results in dramatic memory savings, albeit
2362 at a slight CPU cost.
2364 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2365 as requested by Marc Sherman.
2367 SC/10 Eximstats: added histograms for user specified patterns as requested
2370 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2372 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2373 fopen() in the content-scanning modules that did not already have it.
2379 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2380 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2382 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2384 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2386 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2387 to test Sieve filters that use "vacation".
2389 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2390 that changes the way the GnuTLS parameters are stored in the cache file.
2391 The new format can be generated externally. For backward compatibility,
2392 if the data in the cache doesn't make sense, Exim assumes it has read an
2393 old-format file, and it generates new data and writes a new file. This
2394 means that you can't go back to an older release without removing the
2397 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2398 work if there are any delivery delays because "one_time" forces the
2399 parent to be marked "delivered", so its unseen clone is never tried
2400 again. For this reason, Exim now forbids the simultaneous setting of
2403 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2404 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2405 entry if you want to know the details. The fix, however, neglected to
2406 consider the case when local delivery batching is involved. The test for
2407 "previously delivered" was not happening when checking to see if an
2408 address could be batched with a previous (undelivered) one; under
2409 certain circumstances this could lead to multiple deliveries to the same
2412 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2413 in its include files, and this causes problems building Exim.
2415 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2416 header_syntax) but Exim was just ignoring anything given after a slash.
2417 In particular, this caused confusion with an attempt to use "verify =
2418 reverse_host_lookup/defer_ok". An error is now given when options are
2419 supplied for verify items that do not have them. (Maybe reverse_host_
2420 lookup should have a defer_ok option, but that's a different point.)
2422 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2423 defined by RFC 821) to 2048, because there were problems with some AUTH
2424 commands, and RFC 1869 says the size should be increased for extended
2425 SMTP commands that take arguments.
2427 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2430 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2431 "unknown" error; now it says that the functionality isn't in the binary.
2433 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2434 an address' error message when a string expansion fails (syntax or
2435 whatever). Otherwise the password may appear in the log. Following change
2436 PH/42 below, there is no longer a chance of it appearing in a bounce
2439 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2441 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2442 of its addresses was ever tried. (Bugzilla bug #2.)
2444 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2445 the result incorrectly in the debug output. (It correctly added a newline
2446 to what was transported.)
2448 TF/01 Added $received_time.
2450 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2451 commented out examples of how to interface to a virus scanner and to
2452 SpamAssassin. Also added commented examples of av_scanner and
2453 spamd_address settings.
2455 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2456 and controls are allowed in which ACLs. There were a couple of minor
2457 errors. Some of the entries in the conditions table (which is a table of
2458 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2459 negation of where the condition IS allowed.
2461 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2463 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2464 header file does not have a version number, so I've had to invent a new
2465 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2466 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2467 radiusclient), but it was contributed by a Radius user.
2469 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2470 files or directories, for OpenSSL.
2472 PH/19 When an Exim process that is running as root has to create an Exim log
2473 file, it does so in a subprocess that runs as exim:exim so as to get the
2474 ownership right at creation (otherwise, other Exim processes might see
2475 the file with the wrong ownership). There was no test for failure of this
2476 fork() call, which would lead to the process getting stuck as it waited
2477 for a non-existent subprocess. Forks do occasionally fail when resources
2478 run out. I reviewed all the other calls to fork(); they all seem to check
2481 PH/20 When checking for unexpected SMTP input at connect time (before writing
2482 the banner), Exim was not dealing correctly with a non-positive return
2483 from the read() function. If the client had disconnected by this time,
2484 the result was a log entry for a synchronization error with an empty
2485 string after "input=" when read() returned zero. If read() returned -1
2486 (an event I could not check), uninitialized data bytes were printed.
2487 There were reports of junk text (parts of files, etc) appearing after
2490 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2492 PH/22 Added support for macro redefinition, and (re)definition in between
2493 driver and ACL definitions.
2495 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2496 forgetting to use the resulting value; it was using the unexpanded value.
2498 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2499 hadn't been configured. The fix is from Juergen Kreileder, who
2500 understands it better than I do:
2502 "Here's what I see happening with three configured cyrus_sasl
2503 authenticators configured (plain, login, cram-md5):
2505 On startup auth_cyrus_sasl_init() gets called for each of these.
2506 This means three calls to sasl_listmech() without a specified mech_list.
2507 => SASL tests which mechs of all available mechs actually work
2508 => three warnings about OTP not working
2509 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2511 With the patch, sasl_listmech() also gets called three times. But now
2512 SASL's mech_list option is set to the server_mech specified in the the
2513 authenticator. Or in other words, the answer from sasl_listmech()
2514 gets limited to just the mech you're testing for (which is different
2516 => the return list contains just 'plain' or 'login', 'cram-md5' or
2517 nothing depending on the value of ob->server_mech.
2519 I've just tested the patch: Authentication still works fine,
2520 unavailable mechs specified in the exim configuration are still
2521 caught, and the auth.log warnings about OTP are gone."
2523 PH/25 When debugging is enabled, the contents of the command line are added
2524 to the debugging output, even when log_selector=+arguments is not
2527 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2528 answer is "GNU", and only if the return is "GNU/something" is the answer
2531 PH/27 $acl_verify_message is now set immediately after the failure of a
2532 verification in an ACL, and so is available in subsequent modifiers. In
2533 particular, the message can be preserved by coding like this:
2535 warn !verify = sender
2536 set acl_m0 = $acl_verify_message
2538 Previously, $acl_verify_message was set only while expanding "message"
2539 and "log_message" when a very denied access.
2541 PH/28 Modified OS/os.c-Linux with
2543 -#ifndef OS_LOAD_AVERAGE
2544 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2546 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2547 nomenclature these days.)
2549 PH/29 Installed patch from the Sieve maintainer that adds the options
2550 sieve_useraddress and sieve_subaddress to the redirect router.
2552 PH/30 In these circumstances:
2553 . Two addresses routed to the same list of hosts;
2554 . First host does not offer TLS;
2555 . First host accepts first address;
2556 . First host gives temporary error to second address;
2557 . Second host offers TLS and a TLS session is established;
2558 . Second host accepts second address.
2559 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2560 and peerdn, if requested) that were in fact used only for the second
2563 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2564 attention to any local part prefix or suffix that was matched by the
2565 router that accepted the address. It now behaves in the same way as it
2566 does for delivery: the affixes are removed from the local part unless
2567 rcpt_include_affixes is set on the transport.
2569 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2570 timeout during the DATA phase of an incoming message.
2572 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2573 applied a patch sanctioned by the Sieve maintainer.
2575 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2576 the uid or gid is negative. A case of a negative gid caused this to be
2577 noticed. The fix allows for either to be negative.
2579 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2580 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2583 PH/36 The change PH/12 above was broken. Fixed it.
2585 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2586 the grounds that routing the same address twice would always produce the
2587 same answer. This might have been true once, but it is certainly no
2588 longer true now. Routing a child address may depend on the previous
2589 routing that produced that child. Some complicated redirection strategies
2590 went wrong when messages had multiple recipients, and made Exim's
2591 behaviour dependent on the order in which the addresses were given.
2593 I have moved the duplicate checking until after the routing is complete.
2594 Exim scans the addresses that are assigned to local and remote
2595 transports, and removes any duplicates. This means that more work will be
2596 done, as duplicates will always all be routed, but duplicates are
2597 presumably rare, so I don't expect this is of any significance.
2599 For deliveries to pipes, files, and autoreplies, the duplicate checking
2600 still happens during the routing process, since they are not going to be
2603 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2604 It corrects a timeout issue with spamd. This is Ian's comment: "The
2605 background is that sometimes spamd either never reads data from a
2606 connection it has accepted, or it never writes response data. The exiscan
2607 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2608 blindly assumes that writes won't block so it may never time out."
2610 PH/39 Allow G after quota size as well as K and M.
2612 PH/40 The value set for $authenticated_id in an authenticator may not contain
2613 binary zeroes or newlines because the value is written to log lines and
2614 to spool files. There was no check on this. Now the value is run through
2615 the string_printing() function so that such characters are converted to
2616 printable escape sequences.
2618 PH/41 $message_linecount is a new variable that contains the total number of
2619 lines in the message. Compare $body_linecount, which is the count for the
2622 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2623 bounce and delay warning messages, except in certain special cases, which
2626 (a) An SMTP error message from a remote host;
2627 (b) A message specified in a :fail: redirection;
2628 (c) A message specified in a "fail" command in a system filter;
2629 (d) A message specified in a FAIL return from the queryprogram router;
2630 (e) A message specified by the cannot_route_message router option.
2632 In these cases only, Exim does include the error details in bounce and
2633 warning messages. There are also a few cases where bland messages such
2634 as "unrouteable address" or "local delivery error" are given.
2636 PH/43 $value is now also set for the "else" part of a ${run expansion.
2638 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2639 being worked on, but at least Exim now implements the latest version to
2642 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2643 process to complete was treated as a delivery failure, a timeout while
2644 writing the message to the pipe was logged, but erroneously treated as a
2645 successful delivery. Such timeouts include transport filter timeouts. For
2646 consistency with the overall process timeout, these timeouts are now
2647 treated as errors, giving rise to delivery failures by default. However,
2648 there is now a new Boolean option for the pipe transport called
2649 timeout_defer, which, if set TRUE, converts the failures into defers for
2650 both kinds of timeout. A transport filter timeout is now identified in
2653 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2654 systems where "make" and "gmake" are different, calling "gmake" at top
2655 level broke things. I've arranged for the value of $(MAKE) to be passed
2656 from the Makefile to this script so that it can call the same version of
2660 A note about Exim versions 4.44 and 4.50
2661 ----------------------------------------
2663 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2664 changes of various kinds. As a consequence, a big documentation update was
2665 needed. This delayed the release for rather longer than seemed good, especially
2666 in the light of a couple of (minor) security issues. Therefore, the changes
2667 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2668 release. So 4.44 and 4.50 are in effect two different branches that both start
2671 I have left the 4.50 change log unchanged; it contains all the changes since
2672 4.43. The change log for 4.44 is below; many of its items are identical to
2673 those for 4.50. This seems to be the most sensible way to preserve the
2674 historical information.
2680 1. Minor wording change to the doc/README.SIEVE file.
2682 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2683 computation of the current number of files was incorrect.
2685 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2686 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2689 4. Give more explanation in the error message when the command for a transport
2690 filter fails to execute.
2692 5. There are several places where Exim runs a non-Exim command in a
2693 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2694 was being done only for the command run by the queryprogram router. It is
2695 now done for all such subprocesses. The other cases are: ${run, transport
2696 filters, and the commands run by the lmtp and pipe transports.
2698 6. Added CONFIGURE_GROUP build-time option.
2700 7. Some older OS have a limit of 256 on the maximum number of file
2701 descriptors. Exim was using setrlimit() to set 1000 as a large value
2702 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2703 systems. I've change it so that if it can't get 1000, it tries for 256.
2705 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2706 was an oversight, and furthermore, ever since the addition of extra
2707 controls (e.g. 4.43/32), the checks on when to allow different forms of
2708 "control" were broken. There should now be diagnostics for all cases when a
2709 control that does not make sense is encountered.
2711 9. Added the /retain_sender option to "control=submission".
2713 10. $recipients is now available in the predata ACL (oversight).
2715 11. Tidy the search cache before the fork to do a delivery from a message
2716 received from the command line. Otherwise the child will trigger a lookup
2717 failure and thereby defer the delivery if it tries to use (for example) a
2718 cached ldap connection that the parent has called unbind on.
2720 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2721 of $address_data from the recipient verification was clobbered by the
2722 sender verification.
2724 13. The value of address_data from a sender verification is now available in
2725 $sender_address_data in subsequent conditions in the ACL statement.
2727 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2729 15. Added a new option "connect=<time>" to callout options, to set a different
2732 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2733 was its contents. (It was OK if the option was not defined at all.)
2735 17. A "Completed" log line is now written for messages that are removed from
2736 the spool by the -Mrm option.
2738 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2739 information about exactly what failed.
2741 19. Added -dd to debug only the daemon process.
2743 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2744 handles timeouts, both on the server side and network timeouts. Renamed the
2745 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2747 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2748 It is now set to "smtps".
2750 22. $host_address is now set to the target address during the checking of
2751 ignore_target_hosts.
2753 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2754 being passed; this would have caused $sender_host_name to have been used if
2755 matching the list had actually called for a host name (not very likely,
2756 since this list is usually IP addresses). A host name is now passed as
2759 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2760 code that specifies a non-restarting handler (typically sigaction() in
2761 modern systems) in an attempt to fix a rare and obscure crash bug.
2763 25. Narrowed the window for a race in the daemon that could cause it to ignore
2764 SIGCHLD signals. This is not a major problem, because they are used only to
2765 wake it up if nothing else does.
2767 26. A malformed maildirsize file could cause Exim to calculate negative values
2768 for the mailbox size or file count. Odd effects could occur as a result.
2769 The maildirsize information is now recalculated if the size or filecount
2772 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2773 support for a long time. Removed HAVE_SYS_VFS_H.
2775 28. Installed the latest version of exipick from John Jetmore.
2777 29. In an address list, if the pattern was not a regular expression, an empty
2778 subject address (from a bounce message) matched only if the pattern was an
2779 empty string. Non-empty patterns were not even tested. This was the wrong
2780 because it is perfectly reasonable to use an empty address as part of a
2781 database query. An empty address is now tested by patterns that are
2782 lookups. However, all the other forms of pattern expect the subject to
2783 contain a local part and a domain, and therefore, for them, an empty
2784 address still always fails if the pattern is not itself empty.
2786 30. Exim went into a mad DNS loop when attempting to do a callout where the
2787 host was specified on an smtp transport, and looking it up yielded more
2788 than one IP address.
2790 31. Re-factored the code for checking spool and log partition space into a
2791 function that finds that data and another that does the check. The former
2792 is then used to implement four new variables: $spool_space, $log_space,
2793 $spool_inodes, and $log_inodes.
2795 32. The RFC2047 encoding function was originally intended for short strings
2796 such as real names; it was not keeping to the 75-character limit for
2797 encoded words that the RFC imposes. It now respects the limit, and
2798 generates multiple encoded words if necessary. To be on the safe side, I
2799 have increased the buffer size for the ${rfc2047: expansion operator from
2802 33. It is now permitted to omit both strings after an "if" condition; if the
2803 condition is true, the result is "true". As before, when the second string
2804 is omitted, a false condition yields an empty string. This makes it less
2805 cumbersome to write custom ACL and router conditions.
2807 34. Failure to deliver a bounce message always caused it to be frozen, even if
2808 there was an errors_to setting on the router. The errors_to setting is now
2811 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2812 canonical form (fully expanded) before being placed in
2813 $sender_host_address.
2815 36. The table in the code that translates DNS record types into text (T_A to
2816 "A" for instance) was missing entries for NS and CNAME. It is just possible
2817 that this could have caused confusion if both these types were looked up
2818 for the same domain, because the text type is used as part of Exim's
2819 per-process caching. But the chance of anyone hitting this buglet seems
2822 37. The dnsdb lookup has been extended in a number of ways.
2824 (1) There is a new type, "zns", which walks up the domain tree until it
2825 finds some nameserver records. It should be used with care.
2827 (2) There is a new type, "mxh", which is like "mx" except that it returns
2828 just the host names, not the priorities.
2830 (3) It is now possible to give a list of domains (or IP addresses) to be
2831 looked up. The behaviour when one of the lookups defers can be
2832 controlled by a keyword.
2834 (4) It is now possible to specify the separator character for use when
2835 multiple records are returned.
2837 38. The dnslists ACL condition has been extended: it is now possible to supply
2838 a list of IP addresses and/or domains to be looked up in a particular DNS
2841 39. Added log_selector=+queue_time_overall.
2843 40. When running the queue in the test harness, wait just a tad after forking a
2844 delivery process, to get repeatability of debugging output.
2846 41. Include certificate and key file names in error message when GnuTLS fails
2847 to set them up, because the GnuTLS error message doesn't include the name
2848 of the failing file when there is a problem reading it.
2850 42. Allow both -bf and -bF in the same test run.
2852 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2854 44. The "Exiscan patch" is now merged into the mainline Exim source.
2856 45. Sometimes the final signoff response after QUIT could fail to get
2857 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2858 < 0 before doing a fflush(). This bug looks as though it goes back to the
2859 introduction of TLS in release 3.20, but "sometimes" must have been rare
2860 because the tests only now provoked it.
2862 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2863 (this can affect the format of dates).
2865 47. exim_tidydb, when checking for the continued existence of a message for
2866 which it has found a message-specific retry record, was not finding
2867 messages that were in split spool directories. Consequently, it was
2868 deleting retry records that should have stayed in existence.
2870 48. Steve fixed some bugs in eximstats.
2872 49. The SPA authentication driver was not abandoning authentication and moving
2873 on to the next authenticator when an expansion was forced to fail,
2874 contradicting the general specification for all authenticators. Instead it
2875 was generating a temporary error. It now behaves as specified.
2877 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2878 (the order specifies the preference for clients). The order is now AES256,
2879 AES128, 3DES, ARCFOUR128.
2881 51. Small patch to Sieve code - explicitly set From: when generating an
2884 52. Exim crashed if a remote delivery caused a very long error message to be
2885 recorded - for instance if somebody sent an entire SpamAssassin report back
2886 as a large number of 550 error lines. This bug was coincidentally fixed by
2887 increasing the size of one of Exim's internal buffers (big_buffer) that
2888 happened as part of the Exiscan merge. However, to be on the safe side, I
2889 have made the code more robust (and fixed the comments that describe what
2892 53. Now that there can be additional text after "Completed" in log lines (if
2893 the queue_time_overall log selector is set), a one-byte patch to exigrep
2894 was needed to allow it to recognize "Completed" as not the last thing in
2897 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2898 patch that reportedly fixes this has been added. I am not expert enough to
2899 create a test for it. This is what the patch creator wrote:
2901 "I found a little strange behaviour of ldap code when working with
2902 Windows 2003 AD Domain, where users was placed in more than one
2903 Organization Units. When I tried to give exim partial DN, the exit code
2904 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2905 But simultaneously result of request was absolutely normal ldap result,
2906 so I produce this patch..."
2908 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2909 REFERENCE, so I have modified the code to exclude the patch when that macro
2912 55. Some experimental protocols are using DNS PTR records for new purposes. The
2913 keys for these records are domain names, not reversed IP addresses. The
2914 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2915 leaves it alone. Component reversal etc. now happens only for IP addresses.
2918 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2920 57. Double the size of the debug message buffer (to 2048) so that more of very
2921 long debug lines gets shown.
2923 58. The exicyclog utility now does better if the number of log files to keep
2924 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2926 59. Two changes related to the smtp_active_hostname option:
2928 (1) $smtp_active_hostname is now available as a variable.
2929 (2) The default for smtp_banner uses $smtp_active_hostname instead
2930 of $primary_hostname.
2932 60. The host_aton() function is supposed to be passed a string that is known
2933 to be a valid IP address. However, in the case of IPv6 addresses, it was
2934 not checking this. This is a hostage to fortune. Exim now panics and dies
2935 if the condition is not met. A case was found where this could be provoked
2936 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2937 components; fortuitously, this particular loophole had already been fixed
2938 by change 4.50/55 above.
2940 If there are any other similar loopholes, the new check in host_aton()
2941 itself should stop them being exploited. The report I received stated that
2942 data on the command line could provoke the exploit when Exim was running as
2943 exim, but did not say which command line option was involved. All I could
2944 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2945 running as the user.
2948 61. There was a buffer overflow vulnerability in the SPA authentication code
2949 (which came originally from the Samba project). I have added a test to the
2950 spa_base64_to_bits() function which I hope fixes it.
2953 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2954 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2956 63. The daemon start-up calls getloadavg() while still root for those OS that
2957 need the first call to be done as root, but it missed one case: when
2958 deliver_queue_load_max is set with deliver_drop_privilege. This is
2959 necessary for the benefit of the queue runner, because there is no re-exec
2960 when deliver_drop_privilege is set.
2962 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2963 This has been fixed.
2965 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2966 was in use, was not putting the data itself into the right store pool;
2967 consequently, it could be overwritten for a subsequent message in the same
2968 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2971 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2973 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2974 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2975 the function were treating the return as a boolean value, which happened to
2976 work because 0=false and not-0=true, but is not correct code.
2978 68. The host_aton() function was not handling scoped IPv6 addresses (those
2979 with, for example, "%eth0" on the end) correctly.
2981 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2982 negated items (that is, ~something) in unsigned ints. Some compilers
2983 apparently mutter when there is no cast.
2985 70. If an address verification called from an ACL failed, and did not produce a
2986 user-specific message (i.e. there was only a "system" message), nothing was
2987 put in $acl_verify_message. In this situation, it now puts the system
2990 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2991 session; change 4.31/43 added the unwanted input to the log line - except
2992 that it did not do this in the start of session case. It now does.
2994 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2995 This is wrong and can cause the other end to generate a synchronization
2996 error if it is another Exim or anything else that does the synchronization
2997 check. A QUIT command is no longer sent after a timeout.
2999 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3000 during host lookups.
3002 74. The defer_ok option of callout verification was not working if it was used
3003 when verifying addresses in header lines, that is, for this case:
3005 verify = header_sender/callout=defer_ok
3007 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3008 those file descriptors could be used for SMTP connections. If anything
3009 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3010 could be sent to the SMTP client, causing chaos. The daemon now opens
3011 stdin, stdout, and stderr to /dev/null when it puts itself into the
3014 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3015 log by default. The user can override this with suitable Perl magic.
3017 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3018 the log message when discard triggers, was not working for the DATA ACL or
3019 for the non-SMTP ACL.
3021 78. Error message wording change in sieve.c.
3023 79. If smtp_accept_max_per_host was set, the number of connections could be
3024 restricted to fewer than expected, because the daemon was trying to set up
3025 a new connection before checking whether the processes handling previous
3026 connections had finished. The check for completed processes is now done
3027 earlier. On busy systems, this bug wouldn't be noticed because something
3028 else would have woken the daemon, and it would have reaped the completed
3031 80. If a message was submitted locally by a user whose login name contained one
3032 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3033 It caused a spool format error. I have fixed the spool reading code. A
3034 related problem was that the "from" clause in the Received: line became
3035 illegal because of the space(s). It is now covered by ${quote_local_part.
3037 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3040 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3042 83. Patch from Sieve maintainer to fix unterminated string problem in
3043 "vacation" handling.
3045 84. Some minor changes to the Linux configuration files to help with other
3046 OS variants using glibc.
3048 85. One more patch for Sieve to update vacation handling to latest spec.
3051 ----------------------------------------------------
3052 See the note above about the 4.44 and 4.50 releases.
3053 ----------------------------------------------------
3059 1. Change 4.43/35 introduced a bug that caused file counts to be
3060 incorrectly computed when quota_filecount was set in an appendfile
3063 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3064 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3067 3. Give more explanation in the error message when the command for a transport
3068 filter fails to execute.
3070 4. There are several places where Exim runs a non-Exim command in a
3071 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3072 was being done only for the command run by the queryprogram router. It is
3073 now done for all such subprocesses. The other cases are: ${run, transport
3074 filters, and the commands run by the lmtp and pipe transports.
3076 5. Some older OS have a limit of 256 on the maximum number of file
3077 descriptors. Exim was using setrlimit() to set 1000 as a large value
3078 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3079 systems. I've change it so that if it can't get 1000, it tries for 256.
3081 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3082 was an oversight, and furthermore, ever since the addition of extra
3083 controls (e.g. 4.43/32), the checks on when to allow different forms of
3084 "control" were broken. There should now be diagnostics for all cases when a
3085 control that does not make sense is encountered.
3087 7. $recipients is now available in the predata ACL (oversight).
3089 8. Tidy the search cache before the fork to do a delivery from a message
3090 received from the command line. Otherwise the child will trigger a lookup
3091 failure and thereby defer the delivery if it tries to use (for example) a
3092 cached ldap connection that the parent has called unbind on.
3094 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3095 of $address_data from the recipient verification was clobbered by the
3096 sender verification.
3098 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3099 was its contents. (It was OK if the option was not defined at all.)
3101 11. A "Completed" log line is now written for messages that are removed from
3102 the spool by the -Mrm option.
3104 12. $host_address is now set to the target address during the checking of
3105 ignore_target_hosts.
3107 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3108 being passed; this would have caused $sender_host_name to have been used if
3109 matching the list had actually called for a host name (not very likely,
3110 since this list is usually IP addresses). A host name is now passed as
3113 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3114 code that specifies a non-restarting handler (typically sigaction() in
3115 modern systems) in an attempt to fix a rare and obscure crash bug.
3117 15. Narrowed the window for a race in the daemon that could cause it to ignore
3118 SIGCHLD signals. This is not a major problem, because they are used only to
3119 wake it up if nothing else does.
3121 16. A malformed maildirsize file could cause Exim to calculate negative values
3122 for the mailbox size or file count. Odd effects could occur as a result.
3123 The maildirsize information is now recalculated if the size or filecount
3126 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3127 support for a long time. Removed HAVE_SYS_VFS_H.
3129 18. Updated exipick to current release from John Jetmore.
3131 19. Allow an empty sender to be matched against a lookup in an address list.
3132 Previously the only cases considered were a regular expression, or an
3135 20. Exim went into a mad DNS lookup loop when doing a callout where the
3136 host was specified on the transport, if the DNS lookup yielded more than
3139 21. The RFC2047 encoding function was originally intended for short strings
3140 such as real names; it was not keeping to the 75-character limit for
3141 encoded words that the RFC imposes. It now respects the limit, and
3142 generates multiple encoded words if necessary. To be on the safe side, I
3143 have increased the buffer size for the ${rfc2047: expansion operator from
3146 22. Failure to deliver a bounce message always caused it to be frozen, even if
3147 there was an errors_to setting on the router. The errors_to setting is now
3150 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3151 canonical form (fully expanded) before being placed in
3152 $sender_host_address.
3154 24. Updated eximstats to version 1.33
3156 25. Include certificate and key file names in error message when GnuTLS fails
3157 to set them up, because the GnuTLS error message doesn't include the name
3158 of the failing file when there is a problem reading it.
3160 26. Expand error message when OpenSSL has problems setting up cert/key files.
3163 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3164 (this can affect the format of dates).
3166 28. exim_tidydb, when checking for the continued existence of a message for
3167 which it has found a message-specific retry record, was not finding
3168 messages that were in split spool directories. Consequently, it was
3169 deleting retry records that should have stayed in existence.
3171 29. eximstats updated to version 1.35
3172 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3173 1.35 - bugfix such that pie charts by volume are generated correctly
3175 30. The SPA authentication driver was not abandoning authentication and moving
3176 on to the next authenticator when an expansion was forced to fail,
3177 contradicting the general specification for all authenticators. Instead it
3178 was generating a temporary error. It now behaves as specified.
3180 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3181 (the order specifies the preference for clients). The order is now AES256,
3182 AES128, 3DES, ARCFOUR128.
3184 31. Small patch to Sieve code - explicitly set From: when generating an
3187 32. Exim crashed if a remote delivery caused a very long error message to be
3188 recorded - for instance if somebody sent an entire SpamAssassin report back
3189 as a large number of 550 error lines. This bug was coincidentally fixed by
3190 increasing the size of one of Exim's internal buffers (big_buffer) that
3191 happened as part of the Exiscan merge. However, to be on the safe side, I
3192 have made the code more robust (and fixed the comments that describe what
3195 33. Some experimental protocols are using DNS PTR records for new purposes. The
3196 keys for these records are domain names, not reversed IP addresses. The
3197 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3198 leaves it alone. Component reversal etc. now happens only for IP addresses.
3201 34. The host_aton() function is supposed to be passed a string that is known
3202 to be a valid IP address. However, in the case of IPv6 addresses, it was
3203 not checking this. This is a hostage to fortune. Exim now panics and dies
3204 if the condition is not met. A case was found where this could be provoked
3205 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3206 components; fortuitously, this particular loophole had already been fixed
3207 by change 4.50/55 or 4.44/33 above.
3209 If there are any other similar loopholes, the new check in host_aton()
3210 itself should stop them being exploited. The report I received stated that
3211 data on the command line could provoke the exploit when Exim was running as
3212 exim, but did not say which command line option was involved. All I could
3213 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3214 running as the user.
3217 35. There was a buffer overflow vulnerability in the SPA authentication code
3218 (which came originally from the Samba project). I have added a test to the
3219 spa_base64_to_bits() function which I hope fixes it.
3222 36. The daemon start-up calls getloadavg() while still root for those OS that
3223 need the first call to be done as root, but it missed one case: when
3224 deliver_queue_load_max is set with deliver_drop_privilege. This is
3225 necessary for the benefit of the queue runner, because there is no re-exec
3226 when deliver_drop_privilege is set.
3228 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3229 was in use, was not putting the data itself into the right store pool;
3230 consequently, it could be overwritten for a subsequent message in the same
3231 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3234 38. Sometimes the final signoff response after QUIT could fail to get
3235 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3236 < 0 before doing a fflush(). This bug looks as though it goes back to the
3237 introduction of TLS in release 3.20, but "sometimes" must have been rare
3238 because the tests only now provoked it.
3244 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3245 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3246 Now it returns an integer. A number of calls were still expecting a T/F
3247 return. Fortuitously, in all cases, the tests worked in OK situations,
3248 which is the norm. However, things would have gone wrong on any write
3249 failures on the smtp file descriptor. This function is used when sending
3250 messages over SMTP and also when doing verify callouts.
3252 2. When Exim is called to do synchronous delivery of a locally submitted
3253 message (the -odf or -odi options), it no longer closes stderr before doing
3256 3. Implemented the mua_wrapper option.
3258 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3260 5. Implemented the functions header_remove(), header_testname(),
3261 header_add_at_position(), and receive_remove_recipient(), and exported them
3264 6. If an ACL "warn" statement specified the addition of headers, Exim already
3265 inserted X-ACL-Warn: at the start if there was no header name. However, it
3266 was not making this test for the second and subsequent header lines if
3267 there were newlines in the string. This meant that an invalid header could
3268 be inserted if Exim was badly configured.
3270 7. Allow an ACL "warn" statement to add header lines at the start or after all
3271 the Received: headers, as well as at the end.
3273 8. Added the rcpt_4xx retry error code.
3275 9. Added postmaster_mailfrom=xxx to callout verification option.
3277 10. Added mailfrom=xxxx to the callout verification option, for verify=
3280 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3281 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3283 12. Inserted some casts to stop certain compilers warning when using pointer
3284 differences as field lengths or precisions in printf-type calls (mostly
3285 affecting debugging statements).
3287 13. Added optional readline() support for -be (dynamically loaded).
3289 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3290 same clock tick as a message's arrival, so that its received time was the
3291 same as the "first fail" time on the retry record, and that message
3292 remained on the queue past the ultimate address timeout, every queue runner
3293 would try a delivery (because it was past the ultimate address timeout) but
3294 after another failure, the ultimate address timeout, which should have then
3295 bounced the address, did not kick in. This was a "< instead of <=" error;
3296 in most cases the first failure would have been in the next clock tick
3297 after the received time, and all would be well.
3299 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3300 being recognized when the domain list was tested by the match_domain
3301 condition in an expansion string.
3303 16. Added the ${str2b64: operator.
3305 17. Exim was always calling setrlimit() to set a large limit for the number of
3306 processes, without checking whether the existing limit was already
3307 adequate. (It did check for the limit on file descriptors.) Furthermore,
3308 errors from getrlimit() and setrlimit() were being ignored. Now they are
3309 logged to the main and panic logs, but Exim does carry on, to try to do its
3310 job under whatever limits there are.
3312 18. Imported PCRE 5.0.
3314 19. Trivial typo in log message " temporarily refused connection" (the leading
3317 20. If the log selector return_path_on_delivery was set and an address was
3318 redirected to /dev/null, the delivery process crashed because it assumed
3319 that a return path would always be set for a "successful" delivery. In this
3320 case, the whole delivery is bypassed as an optimization, and therefore no
3323 21. Internal re-arrangement: the function for sending a challenge and reading
3324 a response while authentication was assuming a zero-terminated challenge
3325 string. It's now changed to take a pointer and a length, to allow for
3326 binary data in such strings.
3328 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3330 23. Exim was not respecting finduser_retries when seeking the login of the
3331 uid under which it was called; it was always trying 10 times. (The default
3332 setting of finduser_retries is zero.) Also, it was sleeping after the final
3333 failure, which is pointless.
3335 24. Implemented tls_on_connect_ports.
3337 25. Implemented acl_smtp_predata.
3339 26. If the domain in control=submission is set empty, Exim assumes that the
3340 authenticated id is a complete email address when it generates From: or
3341 Sender: header lines.
3343 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3344 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3345 chown and chgrp in /bin and hostname in /usr/bin.
3347 28. Exim was keeping the "process log" file open after each use, just as it
3348 does for the main log. This opens the possibility of it remaining open for
3349 long periods when the USR1 signal hits a daemon. Occasional processlog
3350 errors were reported, that could have been caused by this. Anyway, it seems
3351 much more sensible not to leave this file open at all, so that is what now
3354 29. The long-running daemon process does not normally write to the log once it
3355 has entered its main loop, and it closes the log before doing so. This is
3356 so that log files can straightforwardly be renamed and moved. However,
3357 there are a couple of unusual error situations where the daemon does write
3358 log entries, and I had neglected to close the log afterwards.
3360 30. The text of an SMTP error response that was received during a remote
3361 delivery was being truncated at 512 bytes. This is too short for some of
3362 the long messages that one sometimes sees. I've increased the limit to
3365 31. It is now possible to make retry rules that apply only when a message has a
3366 specific sender, in particular, an empty sender.
3368 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3369 it possible to be selective about when SMTP synchronization is enforced.
3371 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3373 32. Implemented hosts_connection_nolog.
3375 33. Added an ACL for QUIT.
3377 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3380 35. Added mailbox_size and mailbox_filecount to appendfile.
3382 36. Added control = no_multiline_responses to ACLs.
3384 37. There was a bug in the logic of the code that waits for the clock to tick
3385 in the case where the clock went backwards by a substantial amount such
3386 that the microsecond fraction of "now" was more than the microsecond
3387 fraction of "then" (but the whole seconds number was less).
3389 38. Added support for the libradius Radius client library this is found on
3390 FreeBSD (previously only the radiusclient library was supported).
3396 1. When certain lookups returned multiple values in the form name=value, the
3397 quoting of the values was not always being done properly. Specifically:
3398 (a) If the value started with a double quote, but contained no whitespace,
3400 (b) If the value contained whitespace other than a space character (i.e.
3401 tabs or newlines or carriage returns) it was not quoted.
3402 This fix has been applied to the mysql and pgsql lookups by writing a
3403 separate quoting function and calling it from the lookup code. The fix
3404 should probably also be applied to nisplus, ibase and oracle lookups, but
3405 since I cannot test any of those, I have not disturbed their existing code.
3407 2. A hit in the callout cache for a specific address caused a log line with no
3408 reason for rejecting RCPT. Now it says "Previous (cached) callout
3409 verification failure".
3411 3. There was an off-by-one bug in the queryprogram router. An over-long
3412 return line was truncated at 256 instead of 255 characters, thereby
3413 overflowing its buffer with the terminating zero. As well as fixing this, I
3414 have increased the buffer size to 1024 (and made a note to document this).
3416 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3417 when Exim is waiting for an SMTP response from a remote server, Exim
3418 restarts its select() call on the socket, thereby resetting its timeout.
3419 This is not a problem when such interrupts are rare. Somebody set up a cron
3420 job to run exiwhat every 2 minutes, which is less than the normal select()
3421 timeout (5 or 10 minutes). This meant that the select() timeout never
3422 kicked in because it was always reset. I have fixed this by comparing the
3423 time when an interrupt arrives with the time at the start of the first call
3424 to select(). If more time than the timeout has elapsed, the interrupt is
3425 treated as a timeout.
3427 5. Some internal re-factoring in preparation for the addition of Sieve
3428 extensions (by MH). In particular, the "personal" test is moved to a
3429 separate function, and given an option for scanning Cc: and Bcc: (which is
3430 not set for Exim filters).
3432 6. When Exim created an email address using the login of the caller as the
3433 local part (e.g. when creating a From: or Sender: header line), it was not
3434 quoting the local part when it contained special characters such as @.
3436 7. Installed new OpenBSD configuration files.
3438 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3439 try to make them clearer.
3441 9. Callout options, other than the timeout value, were being ignored when
3442 verifying sender addresses in header lines. For example, when using
3444 verify = header_sender/callout=no_cache
3446 the cache was (incorrectly) being used.
3448 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3449 only the Cygwin environment.
3451 11. When return_path_on_delivery was set as a log selector, if different remote
3452 addresses in the same message used different return paths and parallel
3453 remote delivery occurred, the wrong values would sometimes be logged.
3454 (Whenever a remote delivery process finished, the return path value from
3455 the most recently started remote delivery process was logged.)
3457 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3458 lines when AUTH and/or TLS are in use. This is the "received protocol"
3459 field. Exim used to use "asmtp" for authenticated SMTP, without any
3460 indication (in the protocol name) for TLS use. Now it follows the RFC and
3461 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3462 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3463 names appear in log lines as well as in Received: header lines.
3465 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3466 extensions, and comparison tests, and to fix some bugs.
3468 14. Changes to the "personal" filter test:
3470 (1) The test was buggy in that it was just doing the equivalent of
3471 "contains" tests on header lines. For example, if a user's address was
3472 anne@some.where, the "personal" test would incorrectly be true for
3474 To: susanne@some.where
3476 This test is now done by extracting each address from the header in turn,
3477 and checking the entire address. Other tests that are part of "personal"
3478 are now done using regular expressions (for example, to check local parts
3479 of addresses in From: header lines).
3481 (2) The list of non-personal local parts in From: addresses has been
3482 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3483 taken from the Sieve specification recommendations.
3485 (3) If the message contains any header line starting with "List-" it is
3486 treated as non-personal.
3488 (4) The test for "circular" in the Subject: header line has been removed
3489 because it now seems ill-conceived.
3491 15. Minor typos in src/EDITME comments corrected.
3493 16. Installed latest exipick from John Jetmore.
3495 17. If headers_add on a router specified a text string that was too long for
3496 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3497 of string_sprintf() is now avoided.
3499 18. $message_body_size was not set (it was always zero) when running the DATA
3500 ACL and the local_scan() function.
3502 19. For the "mail" command in an Exim filter, no default was being set for
3503 the once_repeat time, causing a random time value to be used if "once" was
3504 specified. (If the value happened to be <= 0, no repeat happened.) The
3505 default is now 0s, meaning "never repeat". The "vacation" command was OK
3506 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3507 (I found it when inspecting the code).
3509 20. There is now an overall timeout for performing a callout verification. It
3510 defaults to 4 times the callout timeout, which applies to individual SMTP
3511 commands during the callout. The overall timeout applies when there is more
3512 than one host that can be tried. The timeout is checked before trying the
3513 next host. This prevents very long delays if there are a large number of
3514 hosts and all are timing out (e.g. when the network connections are timing
3515 out). The value of the overall timeout can be changed by specifying an
3516 additional sub-option for "callout", called "maxwait". For example:
3518 verify = sender/callout=5s,maxwait=20s
3520 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3521 to the end before writing, but this should make it even safer).
3523 22. Exim was forgetting that it had advertised PIPELINING for the second and
3524 subsequent messages on an SMTP connection. It was also not resetting its
3525 memory on STARTTLS and an internal HELO.
3527 23. When Exim logs an SMTP synchronization error within a session, it now
3528 records whether PIPELINING has been advertised or not.
3530 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3531 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3532 rather than long int.
3534 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3536 26. Added the never_mail option to autoreply.
3542 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3543 crash if the getsockname() call failed; this can happen if a connection is
3544 closed very soon after it is established. The problem was simply in the
3545 order in which certain operations were done, causing Exim to try to write
3546 to the SMTP stream before it had set up the file descriptor. The bug has
3547 been fixed by making things happen in the correct order.
3553 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3554 before the connection was closed, thus losing the rejection response.
3556 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3557 some early Solaris releases, but causes trouble in current releases where
3558 socklen_t is defined.
3560 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3563 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3566 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3567 that releasing the top of it at the end releases what was used for sub-
3568 expansions (unless the block got too big). However, discard this block if
3569 the first thing is a variable or header, so that we can use its block when
3570 it is dynamic (useful for very large $message_headers, for example).
3572 6. Lookups now cache *every* query, not just the most recent. A new, separate
3573 store pool is used for this. It can be recovered when all lookup caches are
3574 flushed. Lookups now release memory at the end of their result strings.
3575 This has involved some general refactoring of the lookup sources.
3577 7. Some code has been added to the store_xxx() functions to reduce the amount
3578 of flapping under certain conditions.
3580 8. log_incoming_interface used to affect only the <= reception log lines. Now
3581 it causes the local interface and port to be added to several more SMTP log
3582 lines, for example "SMTP connection from", and rejection lines.
3584 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3586 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3588 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3589 could be overwritten at the end of the current message (or the start of a
3590 new message if it was set in a HELO ACL). The value is now preserved for
3591 the duration of the SMTP connection.
3593 12. If a transport had a headers_rewrite setting, and a matching header line
3594 contained an unqualified address, that address was qualified, even if it
3595 did not match any rewriting rules. The underlying bug was that the values
3596 of the flags that permit the existence of unqualified sender and recipient
3597 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3598 non-local messages, and by -bnq for local messages) were not being
3599 preserved with the message after it was received.
3601 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3602 "next input=" as part of the text comprising the host identity instead of
3603 the correct text. The code was using the same buffer for two different
3604 strings. However, depending on which order the printing function evaluated
3605 its arguments, the bug did not always show up. Under Linux, for example, my
3606 test suite worked just fine.
3608 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3609 doesn't work with some older versions of Perl. It has been changed to "my",
3610 which in any case is probably the better facility to use.
3612 15. A really picky compiler found some instances of statements for creating
3613 error messages that either had too many or two few arguments for the format
3616 16. The size of the buffer for calls to the DNS resolver has been increased
3617 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3618 for addresses that have a lot of PTR records. This alleviates a problem; it
3619 does not fully solve it.
3621 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3622 buffer now truncates the list and logs the incident, which is the same
3623 action as happens when Exim is looking up a host name and its aliases.
3624 Previously in this situation something unpredictable would happen;
3625 sometimes it was "internal error: store_reset failed".
3627 18. If a server dropped the connection unexpectedly when an Exim client was
3628 using GnuTLS and trying to read a response, the client delivery process
3629 crashed while trying to generate an error log message.
3631 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3632 string, for example:
3634 warn message = H1: something\nH2: something
3636 the text was added as a single header line from Exim's point of view
3637 though it ended up OK in the delivered message. However, searching for the
3638 second and subsequent header lines using $h_h2: did not work. This has been
3639 fixed. Similarly, if a system filter added multiple headers in this way,
3640 the routers could not see them.
3642 20. Expanded the error message when iplsearch is called with an invalid key to
3643 suggest using net-iplsearch in a host list.
3645 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3646 ACLs are no longer actually imposed (and a message to that effect is
3649 22. If a "gecos" field in a passwd entry contained escaped characters, in
3650 particular, if it contained a \" sequence, Exim got it wrong when building
3651 a From: or a Sender: header from that name. A second bug also caused
3652 incorrect handling when an unquoted " was present following a character
3653 that needed quoting.
3655 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3656 was not being matched caselessly.
3658 24. Arranged for all hyphens in the exim.8 source to be escaped with
3661 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3662 properly. Recipient callout cache records were still being keyed to include
3663 the sender, even when use_sender was set false. This led to far more
3664 callouts that were necessary. The sender is no longer included in the key
3665 when use_sender is false.
3667 26. Added "control = submission" modifier to ACLs.
3669 27. Added the ${base62d: operator to decode base 62 numbers.
3671 28. dnsdb lookups can now access SRV records.
3673 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3674 the configuration file.
3676 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3677 (-v) mode. This makes the output for a verbose queue run more intelligible.
3679 31. Added a use_postmaster feature to recipient callouts.
3681 32. Added the $body_zerocount variable, containing the number of binary zero
3682 bytes in the message body.
3684 33. The time of last modification of the "new" subdirectory is now used as the
3685 "mailbox time last read" when there is a quota error for a maildir
3688 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3690 35. Added +ignore_unknown as a special item in host lists.
3692 36. Code for decoding IPv6 addresses in host lists is now included, even if
3693 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3694 address was recognized as an IP address, but was then not correctly decoded
3695 into binary, causing unexpected and incorrect effects when compared with
3702 1. Very minor rewording of debugging text in manualroute to say "list of
3703 hosts" instead of "hostlist".
3705 2. If verify=header_syntax was set, and a header line with an unqualified
3706 address (no domain) and a large number of spaces between the end of the
3707 name and the colon was received, the reception process suffered a buffer
3708 overflow, and (when I tested it) crashed. This was caused by some obsolete
3709 code that should have been removed. The fix is to remove it!
3711 3. When running in the test harness, delay a bit after writing a bounce
3712 message to get a bit more predictability in the log output.
3714 4. Added a call to search_tidyup() just before forking a reception process. In
3715 theory, someone could use a lookup in the expansion of smtp_accept_max_
3716 per_host which, without the tidyup, could leave open a database connection.
3718 5. Added the variables $recipient_data and $sender_data which get set from a
3719 lookup success in an ACL "recipients" or "senders" condition, or a router
3720 "senders" option, similar to $domain_data and $local_part_data.
3722 6. Moved the writing of debug_print from before to after the "senders" test
3725 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3726 problems for message scanning, either using a data ACL, or using
3727 local_scan() because the Received: header was not generated till after they
3728 were called (in order to set the time as the time of reception completion).
3729 I have revised the way this works. The header is now generated after the
3730 body is received, but before the ACL or local_scan() are called. After they
3731 are run, the timestamp in the header is updated.
3737 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3738 before starting a queue runner without re-exec. This happened only when
3739 deliver_drop_privilege was set or when the Exim user was set to root. The
3740 effect of the bug was that timeouts during subsequent deliveries caused
3741 crashes instead of being properly handled. The handler is now left at its
3742 default (and expected) setting.
3744 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3745 message, again when deliver_drop_privilege is set or Exim is run as root.
3746 The bug described in (1) was not present in this case, but the tidying up
3747 of the other signals was missing. I have made the two cases consistent.
3749 3. The ignore_target_hosts setting on a manualroute router was being ignored
3750 for hosts that were looked up using the /MX notation.
3752 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3755 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3756 operated on the sender address. After changing the $sender_address to <>
3757 for the sender address verify, Exim was re-instated it as the original
3758 (before rewriting) address, but remembering that it had rewritten it, so it
3759 wasn't rewriting it again. This bug also had the effect of breaking the
3760 sender address verification caching when the sender address was rewritten.
3762 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3763 This has been changed so that if the ip literal address matches
3764 ignore_target_hosts, the router declines.
3766 7. Added expansion conditions match_domain, match_address, and match_local_
3767 part (NOT match_host).
3769 8. The placeholder for the Received: header didn't have a length field set.
3771 9. Added code to Exim itself and to exim_lock to test for a specific race
3772 condition that could lead to file corruption when using MBX delivery. The
3773 issue is with the lockfile that is created in /tmp. If this file is removed
3774 after a process has opened it but before that process has acquired a lock,
3775 there is the potential for a second process to recreate the file and also
3776 acquire a lock. This could lead to two Exim processes writing to the file
3777 at the same time. The added code performs the same test as UW imapd; it
3778 checks after acquiring the lock that its file descriptor still refers to
3779 the same named file.
3781 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3782 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3785 11. Added the smtp_active_hostname option. If used, this will typically be made
3786 to depend on the incoming interface address. Because $interface_address is
3787 not set up until the daemon has forked a reception process, error responses
3788 that can happen earlier (such as "too many connections") no longer contain
3791 12. If an expansion in a condition on a "warn" statement fails because a lookup
3792 defers, the "warn" statement is abandoned, and the next ACL statement is
3793 processed. Previously this caused the whole ACL to be aborted.
3795 13. Added the iplsearch lookup type.
3797 14. Added ident_timeout as a log selector.
3799 15. Added tls_certificate_verified as a log selector.
3801 16. Added a global option tls_require_ciphers (compare the smtp transport
3802 option of the same name). This controls incoming TLS connections.
3804 17. I finally figured out how to make tls_require_ciphers do a similar thing
3805 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3806 before starting the TLS session.
3808 18. Tabs are now shown as \t in -bP output.
3810 19. If the log selector return_path_on_delivery was set, Exim crashed when
3811 bouncing a message because it had too many Received: header lines.
3813 20. If two routers both had headers_remove settings, and the first one included
3814 a superfluous trailing colon, the final name in the first list and the
3815 first name in the second list were incorrectly joined into one item (with a
3816 colon in the middle).
3822 1. Added -C and -D options to the exinext utility, mainly to make it easier
3823 to include in the automated testing, but these could be helpful when
3824 multiple configurations are in use.
3826 2. The exinext utility was not formatting the output nicely when there was
3827 an alternate port involved in the retry record key, nor when there was a
3828 message id as well (for retries that were specific to a specific message
3829 and a specific host). It was also confused by IPv6 addresses, because of
3830 the additional colons they contain. I have fixed the IPv4 problem, and
3831 patched it up to do a reasonable job for IPv6.
3833 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3834 delivery, the log line now contains "pipelined" if PIPELINING was used.
3836 4. An SMTP transport process used to panic and die if the bind() call to set
3837 an explicit outgoing interface failed. This has been changed; it is now
3838 treated in the same way as a connect() failure.
3840 5. A reference to $sender_host_name in the part of a conditional expansion
3841 that was being skipped was still causing a DNS lookup. This no longer
3844 6. The def: expansion condition was not recognizing references to header lines
3845 that used bh_ and bheader_.
3847 7. Added the _cache feature to named lists.
3849 8. The code for checking quota_filecount in the appendfile transport was
3850 allowing one more file than it should have been.
3852 9. For compatibility with Sendmail, the command line option
3860 and sets the incoming protocol and host name (for trusted callers). The
3861 host name and its colon can be omitted when only the protocol is to be set.
3862 Note the Exim already has two private options, -pd and -ps, that refer to
3863 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3864 "s", but I don't think that's a major issue.
3866 10. A number of refactoring changes to the code, none of which should affect
3869 (a) The number of logging options was getting close to filling up the
3870 32-bit word that was used as a bit map. I have split them into two classes:
3871 those that are passed in the argument to log_write(), and those that are
3872 only ever tested independently outside of that function. These are now in
3873 separate 32-bit words, so there is plenty of room for expansion again.
3874 There is no change in the user interface or the logging behaviour.
3876 (b) When building, for example, log lines, the code previously used a
3877 macro that called string_cat() twice, in order to add two strings. This is
3878 not really sufficiently general. Furthermore, there was one instance where
3879 it was actually wrong because one of the argument was used twice, and in
3880 one call a function was used. (As it happened, calling the function twice
3881 did not affect the overall behaviour.) The macro has been replaced by a
3882 function that can join an arbitrary number of extra strings onto a growing
3885 (c) The code for expansion conditions now uses a table and a binary chop
3886 instead of a serial search (which was left over from when there were very
3887 few conditions). Also, it now recognizes conditions like "pam" even when
3888 the relevant support is not compiled in: a suitably worded error message is
3889 given if an attempt is made to use such a condition.
3891 11. Added ${time_interval:xxxxx}.
3893 12. A bug was causing one of the ddress fields not to be passed back correctly
3894 from remote delivery subprocesses. The field in question was not being
3895 subsequently used, so this caused to problems in practice.
3897 13. Added new log selectors queue_time and deliver_time.
3899 14. Might have fixed a bug in maildirsizefile handling that threw up
3900 "unexpected character" debug warnings, and recalculated the data
3901 unnecessarily. In any case, I expanded the warning message to give more
3904 15. Added the message "Restricted characters in address" to the statements in
3905 the default ACL that block characters like @ and % in local parts.
3907 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3908 Three changes have been made:
3910 (a) There was a serious bug; a negative response to MAIL caused the whole
3911 recipient domain to be cached as invalid, thereby blocking all messages
3912 to all local parts at the same domain, from all senders. This bug has
3913 been fixed. The domain is no longer cached after a negative response to
3914 MAIL if the sender used is not empty.
3916 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3919 (c) A new callout option, "use_sender" has been added for people who want
3920 the modified behaviour.
3926 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3929 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3930 indeed breaks things for older releases.
3932 3. Added additional logging to the case where there is a problem reading data
3933 from a filter that is running in a subprocess using a pipe, in order to
3934 try to track down a specific problem.
3936 4. Testing facility fudge: when running in the test harness and attempting
3937 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3938 getting "No route to host". Convert this to a timeout.
3940 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3943 6. Some OS don't have socklen_t but use size_t instead. This affects the
3944 fifth argument of getsockopt() amongst other things. This is now
3945 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3946 can be set for individual OS. I have set it for SunOS5, OSF1, and
3947 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3948 some earlier ones do not.
3950 7. Change 4.30/15 was not doing the test caselessly.
3952 8. The standard form for an IPv6 address literal was being rejected by address
3953 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3954 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3955 this, as well as the form without the "IPv6" on the front (but only when
3956 address literals are enabled, of course).
3958 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3960 10. Exim crashed if a message with an empty sender address specified by -f
3961 encountered a router with an errors_to setting. This could be provoked only
3962 by a command such as
3966 where an empty string was supplied; "<>" did not hit this bug.
3968 11. Installed PCRE release 4.5.
3970 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3971 remained set. It is now erased.
3973 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3974 times from message ids (which are base 36 rather than the normal 62).
3976 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3977 were being counted as actual protocol errors, and logged if the log
3978 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3979 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3980 and DATA following a set of rejected RCPTs do not count as protocol errors.
3981 In other words, Exim assumes they were pipelined, though this may not
3982 actually be the case. Of course, in all cases the client gets an
3983 appropriate error code.
3985 15. If a lookup fails in an ACL condition, a message about the failure may
3986 be available; it is used if testing the ACL cannot continue, because most
3987 such messages specify what the cause of the deferral is. However, some
3988 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3989 that caused an old message to be retained and used if a later statement
3990 caused a defer, replacing the real cause of the deferral.
3992 16. If an IP address had so many PTR records that the DNS lookup buffer
3993 was not large enough to hold them, Exim could crash while trying to process
3994 the truncated data. It now detects and logs this case.
3996 17. Further to 4.21/58, another change has been made: if (and only if) the
3997 first line of a message (the first header line) ends with CRLF, a bare LF
3998 in a subsequent header line has a space inserted after it, so as not to
3999 terminate the header.
4001 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4002 unnecessarily, used atoi() instead of strtol(), and didn't check the
4003 termination when getting file sizes from file names by regex.
4005 19. Completely re-implemented the support for maildirsize files, in the light
4006 of a number of problems with the previous contributed implementation
4007 (4.30/29). In particular:
4009 . If the quota is zero, the maildirsize file is maintained, but no quota is
4012 . If the maildir directory does not exist, it is created before any attempt
4013 to write a maildirsize file.
4015 . The quota value in the file is just a cache; if the quota is changed in
4016 the transport, the new value overrides.
4018 . A regular expression is available for excluding directories from the
4021 20. The autoreply transport checks the characters in options that define the
4022 message's headers; it allows continued headers, but it was checking with
4023 isspace() after an embedded newline instead of explicitly looking for a
4026 21. If all the "regular" hosts to which an address was routed had passed their
4027 expiry times, and had not reached their retry times, the address was
4028 bounced, even if fallback hosts were defined. Now Exim should go on to try
4031 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4032 equivalent code in the SMTP transport. Some hosts send humungous responses
4033 to HELO/EHLO, more than 1024 it seems.
4035 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4036 gives compiler warnings in some environments. I've now done it "properly",
4039 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4040 (because of problems with the built-in one) was declared to return uschar *
4041 instead of char *, causing compiler failure.
4043 25. Fixed a file descriptor leak when processing alias/forward files.
4045 26. Fixed a minor format string issue in dbfn.c.
4047 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4049 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4050 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4053 29. When a sender address is verified, it is cached, to save repeating the test
4054 when there is more than one recipient in a message. However, when the
4055 verification involves a callout, it is possible for different callout
4056 options to be set for different recipients. It is too complicated to keep
4057 track of this in the cache, so now Exim always runs a verification when a
4058 callout is required, relying on the callout cache for the optimization.
4059 The overhead is duplication of the address routing, but this should not be
4062 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4063 to be verified with callout=postmaster, and the main callout worked but the
4064 postmaster check failed, the verification correctly failed. However, if a
4065 subsequent RCPT command asked for sender verification *without* the
4066 postmaster check, incorrect caching caused this verification also to fail,
4069 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4070 it was not caching the DNS options (qualify_single, search_parents) that
4071 were used when the lookup failed. A subsequent lookup with different
4072 options therefore always gave the same answer, though there were cases
4073 where it should not have. (Example: a "domains = !$mx_any" option on a
4074 dnslookup router: the "domains" option is always processed without any
4075 widening, but the router might have qualify_single set.) Now Exim uses the
4076 cached value only when the same options are set.
4078 32. Added John Jetmore's "exipick" utility to the distribution.
4080 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4081 than a timeout (e.g. a certificate is required, and is not provided), an
4082 Exim server now closes the connection immediately. Previously it waited for
4083 the client to close - but if the client is SSL, it seems that they each
4084 wait for each other, leading to a delay before one of them times out.
4086 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4087 maintained 0.8.x compatibility because I don't think many are using it, and
4088 it is clearly obsolete.
4090 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4093 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4094 client certificate was expired. A simple patch fixes this, though I don't
4095 understand the full logic of why the verify callback is called multiple
4098 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4099 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4100 which causes problems with some clients (such as the Certicom SSL Plus
4101 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4102 disables the coutermeasure allowing Eudora to connect."
4104 38. Exim was not checking that a write() to a log file succeeded. This could
4105 lead to Bad Things if a log got too big, in particular if it hit a file
4106 size limit. Exim now panics and dies if it cannot write to a log file, just
4107 as it does if it cannot open a log file.
4109 39. Modified OS/Makefile-Linux so that it now contains
4111 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4113 The two -D definitions ensure that Exim is compiled with large file
4114 support, which makes it possible to handle log files that are bigger than
4117 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4118 instance) a domain was checked against a named list that involved a lookup,
4119 causing $domain_data to be set, then another domain was checked against the
4120 same list, then the first domain was re-checked, the value of $domain_data
4121 after the final check could be wrong. In particular, if the second check
4122 failed, it could be set empty. This bug probably also applied to
4125 41. The strip_trailing_dot option was not being applied to the address given
4126 with the -f command-line option.
4128 42. The code for reading a message's header from the spool was incrementing
4129 $received_count, but never initializing it. This meant that the value was
4130 incorrect (doubled) while delivering a message in the same process in which
4131 it was received. In the most common configuration of Exim, this never
4132 happens - a fresh exec is done - but it can happen when
4133 deliver_drop_privilege is set.
4135 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4136 it now includes up to 150 characters of the unexpected data in the log
4139 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4140 and building data strings. The size of both of these buffers was 10 000
4141 bytes - far larger than anybody would *ever* want, thought I. Needless to
4142 say, somebody hit the limit. I have increased the maximum line length to
4143 20 000 and the maximum data length of concatenated lines to 100 000. I have
4144 also fixed two bugs, because there was no checking on these buffers. Tsk,
4145 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4146 buffer is too small.
4148 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4149 lsearch lookups. Now it does.
4151 46. When parsing a route_list item in a manualroute router, a fixed-length
4152 buffer was used for the list of hosts. I made this 1024 bytes long,
4153 thinking that nobody would ever have a list of hosts that long. Wrong.
4154 Somebody had a whole pile of complicated expansion conditions, and the
4155 string was silently truncated, leading to an expansion error. It turns out
4156 that it is easier to change to an unlimited length (owing to other changes
4157 that have happened since this code was originally written) than to build
4158 structure for giving a limitation error. The length of the item that
4159 expands into the list of hosts is now unlimited.
4161 47. The lsearch lookup could not handle data where the length of text line was
4162 more than 4095 characters. Such lines were truncated, leading to shortened
4163 data being returned. It should now handle lines of any length.
4165 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4166 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4169 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4170 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4171 no longer happens in comment lines. A list of replacements is now placed
4172 at the head of all of the source files, except those whose only change is
4173 to replace PERL_COMMAND in the very first #! line.
4175 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4176 messages on the queue, with a bottom-up merge sort, using code contributed
4177 by Michael Haardt. This should make operations like -bp somewhat faster on
4178 large queues. It won't affect queue runners, except when queue_run_in_order
4181 51. Installed eximstats 1.31 in the distribution.
4183 52. Added support for SRV lookups to the dnslookup router.
4185 53. If an ACL referred to $message_body or $message_body_end, the value was not
4186 reset for any messages that followed in the same SMTP session.
4188 54. The store-handling optimization for building very long strings was not
4189 differentiating between the different store pools. I don't think this
4190 actually made any difference in practice, but I've tidied it.
4192 55. While running the routers to verify a sender address, $sender_address
4193 was still set to the sender address. This is wrong, because when routing to
4194 send a bounce to the sender, it would be empty. Therefore, I have changed
4195 it so that, while verifying a sender address, $sender_address is set to <>.
4196 (There is no change to what happens when verifying a recipient address.)
4198 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4199 target A or AAAA records (if not already returned) without resetting the
4200 qualify_single or search_parents options of the DNS resolver. These are
4201 inappropriate in this case because the targets of MX and SRV records must
4202 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4203 target that, when qualified, matched something in the local domain. These
4204 two options are now turned off when doing these lookups.
4206 57. It seems that at least some releases of Reiserfs (which does not have the
4207 concept of a fixed number of inodes) returns zero and not -1 for the
4208 number of available inodes. This interacted badly with check_spool_inodes,
4209 which assumed that -1 was the "no such thing" setting. What I have done is
4210 to check that the total number of inodes is greater than zero before doing
4211 the test of how many are available.
4213 58. When a "warn" ACL statement has a log_message modifier, the message is
4214 remembered, and not repeated. This is to avoid a lot of repetition when a
4215 message has many recipients that cause the same warning to be written.
4216 Howewer, Exim was preserving the list of already written lines for an
4217 entire SMTP session, which doesn't seem right. The memory is now reset if a
4218 new message is started.
4220 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4221 debugging output unless log_rewrite was also set.
4223 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4224 of (int)(handle) when we know that handle contains (void *)(-1).
4226 61. The Exim daemon panic-logs an error return when it closes the incoming
4227 connection. However "connection reset by peer" seems to be common, and
4228 isn't really an error worthy of noting specially, so that particular error
4231 62. When Exim is trying to find all the local interfaces, it used to panic and
4232 die if the ioctl to get the interface flags failed. However, it seems that
4233 on at least one OS (Solaris 9) it is possible to have an interface that is
4234 included in the list of interfaces, but for which you get a failure error
4235 for this call. This happens when the interface is not "plumbed" into a
4236 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4237 failure of the "get flags" call assumes that the interface is down.
4239 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4240 makes life easier for people who are doing arithmetic on fields extracted
4241 from dates, where you often get leading zeros that should not be
4242 interpreted as octal.
4244 64. Added qualify_domain to the redirect router, to override the global
4247 65. If a pathologically long header line contained very many addresses (the
4248 report of this problem mentioned 10 000) and each of them was rewritten,
4249 Exim could use up a very large amount of memory. (It kept on making new
4250 copies of the header line as it rewrote, and never released the old ones.)
4251 At the expense of a bit more processing, the header rewriting function has
4252 been changed so that it no longer eats memory in this way.
4254 66. The generation of the Received: header has been moved from the time that a
4255 message starts to be received, to the time that it finishes. The timestamp
4256 in the Received: header should now be very close to that of the <= log
4257 line. There are two side-effects of this change:
4259 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4260 logged header lines no longer include the local Received: line, because
4261 it has not yet been created. The same applies to a copy of the message
4262 that is returned to a non-SMTP sender when a message is rejected.
4264 (b) When a filter file is tested using -bf, no additional Received: header
4265 is added to the test message. After some thought, I decided that this
4268 This change does not affect the value of $received_for. It is still set
4269 after address rewriting, but before local_scan() is called.
4271 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4273 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4274 gave an unhelpful panic error message, and a defer error. I have managed to
4275 change this behaviour so that it now rejects any supplied certificate,
4276 which seems right, as the list of acceptable certificates is empty.
4278 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4279 gave an unhelpful defer error. I have not managed to make this reject any
4280 supplied certificates, but the error message it gives is "no certificate
4281 supplied", which is not helpful.
4283 70. exigrep's output now also includes lines that are not associated with any
4284 message, but which match the given pattern. Implemented by a patch from
4285 Martin Sluka, which also tidied up the Perl a bit.
4287 71. Recipient callout verification, like sender verification, was using <> in
4288 the MAIL FROM command. This isn't really the right thing, since the actual
4289 sender may affect whether the remote host accepts the recipient or not. I
4290 have changed it to use the actual sender in the callout; this means that
4291 the cache record is now keyed on a recipient/sender pair, not just the
4292 recipient address. There doesn't seem to be a real danger of callout loops,
4293 since a callout by the remote host to check the sender would use <>.
4294 [SEE ABOVE: changed after hitting problems.]
4296 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4297 temporary errors. However, in the case of such a code being given after
4298 the end of a data transmission (i.e. after ".") Exim was failing to write
4299 a retry record for the message. (Yes, there was some broken host that was
4300 actually sending 8xx at this point.)
4302 73. An unknown lookup type in a host list could cause Exim to panic-die when
4303 the list was checked. (An example that provoked this was putting <; in the
4304 middle of a list instead of at the start.) If this happened during a DATA
4305 ACL check, a -D file could be left lying around. This kind of configuration
4306 error no longer causes Exim to die; instead it causes a defer errror. The
4307 incident is still logged to the main and panic logs.
4309 74. Buglet left over from Exim 3 conversion. The message "too many messages
4310 in one connection" was written to the rejectlog but not the mainlog, except
4311 when address rewriting (yes!) was being logged.
4313 75. Added write_rejectlog option.
4315 76. When a system filter was run not as root (that is, when system_filter_user
4316 was set), the values of the $n variables were not being returned to the
4317 main process; thus, they were not subsequently available in the $sn
4320 77. Added +return_path_on_delivery log selector.
4322 78. A connection timeout was being treated differently from recipients deferred
4323 when testing hosts_max_try with a message that was older than the host's
4324 retry timeout. (The host should not be counted, thus allowing all hosts to
4325 be tried at least once before bouncing.) This may have been the cause of an
4326 occasionally reported bug whereby a message would remain on the queue
4327 longer than the retry timeout, but would be bounced if a delivery was
4328 forced. I say "may" because I never totally pinned down the problem;
4329 setting up timeout/retry tests is difficult. See also the next item.
4331 79. The ultimate address timeout was not being applied to errors that involved
4332 a combination of host plus message (for example, a timeout on a MAIL
4333 command). When an address resolved to a number of possible hosts, and they
4334 were not all tried for each delivery (e.g. because of hosts_max_try), a
4335 message could remain on the queue longer than the retry timeout.
4337 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4340 81. Fixed an obscure SMTP outgoing bug which required at least the following
4341 conditions: (a) there was another message waiting for the same server;
4342 (b) the server returned 5xx to all RCPT commands in the first message so
4343 that the message was not completed; (c) the server dropped the connection
4344 or gave a negative response to the RSET that Exim sends to abort the
4345 transaction. The observed case was a dropped connection after DATA that had
4346 been sent in pipelining mode. That is, the server had advertised PIPELINING
4347 but was not implementing it correctly. The effect of the bug was incorrect
4348 behaviour, such as trying another host, and this could lead to a crash.
4354 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4355 and daemon.c were passed as pointers to ints; they should have been
4356 pointers to socklen_t variables (which are typically unsigned ints).
4358 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4361 3. Fixed a really odd bug that affected only the testing scheme; patching a
4362 certain fixed string in the binary changed the value of another string that
4363 happened to be identical to the end of the original first string.
4365 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4366 name", it returns that address as the IP address. On some operating
4367 systems (e.g. Solaris), it also passes back the IP address string as the
4368 "host name". However, on others (e.g. Linux), it passes back an empty
4369 string. Exim wasn't checking for this, and was changing the host name to an
4370 empty string, assuming it had been canonicized.
4372 5. Although rare, it is permitted to have more than one PTR record for a given
4373 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4374 all the names associated with an address, because they do in Solaris.
4375 However, it seems that they do not in Linux for data that comes from the
4376 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4377 I found this out when I moved to a new Linux workstation and tried to run
4378 the Exim test suite.
4380 To get round this problem I have changed the code so that it now does its
4381 own call to the DNS to look up PTR records when searching for a host name.
4382 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4383 addresses that are only in /etc/hosts are still found.
4385 This behaviour is, however, controlled by an option called host_lookup_
4386 order, which defaults to "bydns:byaddr". If people want to use the other
4387 order, or indeed, just use one or the other means of lookup, they can
4388 specify it in this variable.
4390 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4391 some operating systems, this comes back from gethostbyaddr() as an empty
4392 string, and this is what Exim used to test for. However, it seems that in
4393 other systems, "." is yielded. Exim now tests for this case too.
4395 7. The values of check_spool_space and check_log_space are now held internally
4396 as a number of kilobytes instead of an absolute number of bytes. If a
4397 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4398 kilobyte. This means that much larger values can be stored.
4400 8. Exim monitor: an attempt to get the action menu when not actually pointing
4401 at a message produces an empty menu entitled "No message selected". This
4402 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4403 no entries in it ("Shell widget menu has zero width and/or height"). So I
4404 have added a single, blank menu entry in this case.
4406 9. Added ${quote_local_part.
4408 10. MIME decoding is now applied to the contents of Subject: header lines when
4411 11. Now that a reference to $sender_host_address automatically causes a reverse
4412 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4413 host lookup before query-style lookups in lists that might use this
4414 variable. This has therefore been abolished, and the "net-" prefix is no
4415 longer necessary for query-style lookups.
4417 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4418 have been a typo for LFLAGS, so it has been changed.
4420 13. The install script calls Exim with "-C /dev/null" in order to find the
4421 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4422 to be output. Howeve, since Exim outputs its version number before the
4423 error, it didn't break the script. It just looked ugly. I fixed this by
4424 always allowing "-C /dev/null" if the caller is root.
4426 14. Ignore overlarge ACL variable number when reading spool file - insurance
4427 against a later release with more variables having written the file.
4429 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4430 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4431 this, as well as the form without the "IPv6" on the front.
4433 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4434 OS/Makefile-Darwin file.
4436 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4437 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4439 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4440 variable. However, some OS use ints for inodes. I've added cast to long int
4441 to get rid of the compiler warning.
4443 19. I had forgotten to lock out "/../" in configuration file names when
4444 ALT_CONFIG_PREFIX was set.
4446 20. Routers used for verification do not need to specify transports. However,
4447 if such a router generated a host list, and callout was configured, Exim
4448 crashed, because it could not find a port number from the (non-existent)
4449 transport. It now assumes port 25 in this circumstance.
4451 21. Added the -t option to exigrep.
4453 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4454 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4455 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4456 them is set, LOOKUP_LSEARCH is forced.
4458 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4460 24. Added sender and host information to the "rejected by local_scan()" log
4461 line; previously there was no indication of these.
4463 25. Added .include_if_exists.
4465 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4466 Linux. It turns out that not all file systems support this. Apparently some
4467 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4468 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4469 not supported on the file descriptor, is now ignored when Exim is trying to
4470 sync a directory. This applies only to Linux.
4472 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4474 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4475 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4478 29. Added support for maildirsize files from supplied patch (modified a bit).
4480 30. The use of :fail: followed by an empty string could lead Exim to respond to
4481 sender verification failures with (e.g.):
4483 550 Verification failed for <xxx>
4484 550 Sender verify failed
4486 where the first response line was missing the '-' that indicates it is not
4487 the final line of the response.
4489 31. The loop for finding the name of the user that called Exim had a hardwired
4490 limit of 10; it now uses the value of finduser_retries, which is used for
4491 all other user lookups.
4493 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4496 33. Exim was neglecting to zero errno before one call of strtol() when
4497 expanding a string and expecting an integer value. On some systems this
4498 resulted in spurious "integer overflow" errors. Also, it was casting the
4499 result into an int without checking.
4501 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4502 did not work. The code looks as if it has *never* worked, though it appears
4503 to have been documented since at least releast 1.62. I have made it work.
4505 35. The "timeout_DNS" error in retry rules, also documented since at least
4506 1.62, also never worked. As it isn't clear exactly what this means, and
4507 clearly it isn't a major issue, I have abolished the feature by treating it
4508 as "timeout", and writing a warning to the main and panic logs.
4510 36. The display of retry rules for -brt wasn't always showing the error code
4513 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4514 timeout_connect_A, timeout_connect_MX.
4516 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4517 to the empty sender.
4519 39. The daemon was not analysing the content of -oX till after it had closed
4520 stderr and disconnected from the controlling terminal. This meant that any
4521 syntax errors were only noted on the panic log, and the return code from
4522 the command was 0. By re-arranging the code a little, I've made the
4523 decoding happen first, so such errors now appear on stderr, and the return
4524 code is 1. However, the actual setting up of the sockets still happens in
4525 the disconnected process, so errors there are still only recorded on the
4528 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4529 connections (as happens on some IP stacks) was logged at start up time as
4530 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4531 it from "IPv6 and IPv4", which means that two separate sockets are being
4534 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4535 says whether AF_INET or AF_INET6 was passed as an argument.
4537 42. Exiwhat output was messed up when time zones were included in log
4540 43. Exiwhat now gives more information about the daemon's listening ports,
4541 and whether -tls-on-connect was used.
4543 44. The "port" option of the smtp transport is now expanded.
4545 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4546 silently ignored. Now an error message is written to the main and panic
4549 46. There's a new ACL modifier called "logwrite" which writes to a log file
4550 as soon as it is encountered.
4552 47. Added $local_user_uid and $local_user_gid at routing time.
4554 48. Exim crashed when trying to verify a sender address that was being
4557 49. Exim was recognizing only a space character after ".include". It now also
4558 recognizes a tab character.
4560 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4561 extracting the relevant information from the specification. The man page no
4562 longer contains scrambled data for the -d option, and I've added a section
4563 at the front about calling Exim under different names.
4565 51. Added "extra_headers" argument to the "mail" command in filter files.
4567 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4570 53. Installed eximstats 1.29.
4572 54. Added transport_filter_timeout as a generic transport option.
4574 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4575 Cc: header lines. This was required by RFC 822, but it not required by RFC
4578 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4579 incoming messages that did not have them. Now it does so only if the
4580 message originates locally, that is, if there is no associated remote host
4581 address. When Resent- header lines are present, this applies to the Resent-
4582 lines rather than the non-Resent- lines.
4584 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4585 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4587 58. Messages for configuration errors now include the name of the main
4588 configuration file - useful now that there may be more than one file in a
4589 list (.included file names were always shown).
4591 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4592 for those rare installations that do not start the daemon as root or run it
4593 setuid root. I've cut out the call to initgroups() if the daemon is not
4596 60. The Exim user and group can now be bound into the binary as text strings
4597 that are looked up at the start of Exim's processing.
4599 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4601 62. Added $mailstore_basename variable.
4603 63. Installed patch to sieve.c from Michael Haardt.
4605 64. When Exim failed to open the panic log after failing to open the main log,
4606 the original message it was trying to log was written to stderr and debug
4607 output, but if they were not available (the usual case in production), it
4608 was lost. Now it is written to syslog before the two lines that record the
4609 failures to open the logs.
4611 65. Users' Exim filters run in subprocesses under the user's uid. It is
4612 possible for a "deliver" command or an alias in a "personal" command to
4613 provoke an address rewrite. If logging of address rewriting is configured,
4614 this fails because the process is not running as root or exim. There may be
4615 a better way of dealing with this, but for the moment (because 4.30 needs
4616 to be released), I have disabled address rewrite logging when running a
4617 filter in a non-root, non-exim process.
4623 1. The buildconfig auxiliary program wasn't quoting the value set for
4624 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4625 not defined. This bug was masked in 4.22 by the effect that was fixed in
4628 2. Some messages that were rejected after a message id was allocated were
4629 shown as "incomplete" by exigrep. It no longer does this for messages that
4630 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4632 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4633 have allow_domain_literals set, the ID did not get logged in the <= line.
4634 Domain literals are now always recognized in Message-ID: header lines.
4636 4. The first argument for a ${extract expansion item is the key name or field
4637 number. Leading and trailing spaces in this item were not being ignored,
4638 causing some misleading effects.
4640 5. When deliver_drop_privilege was set, single queue runner processes started
4641 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4642 same command in the process it spins off) were not dropping privilege.
4644 6. When the daemon running as "exim" started a queue runner, it always
4645 re-executed Exim in the spun-off process. This is a waste of effort when
4646 deliver_drop_privilege is set. The new process now just calls the
4647 queue-runner function directly.
4653 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4656 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4657 ignored. Though the use of -oP was forcing the writing of a pid file, it
4658 was always written to the default place.
4660 3. If the message "no IP address found for host xxxx" is generated during
4661 incoming verification, it is now followed by identification of the incoming
4662 connection (so you can more easily find what provoked it).
4664 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4666 5. Added some features to "harden" Exim a bit more against certain attacks:
4668 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4669 be put in Local/Makefile. This is like the never_users runtime option,
4670 but it cannot be overridden. The default setting is "root".
4672 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4673 prefix string with which any file named in a -C command line option
4676 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4677 is retained for -C and -D only if the caller of Exim is root. Without
4678 it, the exim user may also use -C and -D and retain privilege.
4680 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4681 command line option is disabled.
4683 6. Macro names set by the -D option must start with an upper case letter, just
4684 like macro names defined in the configuration file.
4686 7. Added "dereference=" facility to LDAP.
4688 8. Two instances of the typo "uknown" in the source files are fixed.
4690 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4691 the Configure-Makefile script screwed up while processing it.
4693 10. Incorporated PCRE 4.4.
4695 11. The SMTP synchronization check was not operating right at the start of an
4696 SMTP session. For example, it could not catch a HELO sent before the client
4697 waited for the greeting. There is now a check for outstanding input at the
4698 point when the greeting is written. Because of the duplex, asynchronous
4699 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4700 way, but not yet received, when the check is performed.
4702 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4703 on TCP/IP sockets, because this apparently causes some broken clients to
4706 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4707 unchanged) from the Cygwin maintainer.
4709 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4710 was supported instead of testing for mbx. Effectively a typo.
4712 15. The spa authenticator server code was not checking that the input it
4713 received was valid base64.
4715 16. The debug output line for the "set" modifier in ACLs was not showing the
4716 name of the variable that was being set.
4718 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4720 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4721 lookup on its own. Something else was needed to trigger the lookup. For
4722 example, a match in host_lookup or the need for a host name in a host list.
4723 Now, if $sender_host_name is referenced and the host name has not yet been
4724 looked up, a lookup is performed. If the lookup fails, the variable remains
4725 empty, and $host_lookup_failed is set to "1".
4727 19. Added "eqi" as a case-independent comparison operator.
4729 20. The saslauthd authentication condition could segfault if neither service
4730 nor realm was specified.
4732 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4733 error message that was logged was misleading, and incoming SMTP
4734 connections were dropped. The message is now more accurate, and temporary
4735 errors are given to SMTP connections.
4737 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4738 (except RSET) with a 421 error, until QUIT is received. However, it was
4739 failing to send a response to QUIT.
4741 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4742 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4743 ACL is now run after the helo_try_verify_hosts code.
4745 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4746 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4747 case-independent, so other case variants are also recognized). Apparently
4748 some systems use these upper case variants.
4750 25. If more than two messages were waiting for the same host, and a transport
4751 filter was specified for the transport, Exim sent two messages over the
4752 same TCP/IP connection, and then failed with "socket operation on non-
4753 socket" when it tried to send the third.
4755 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4757 27. The extern definition of crypt16() in expand.c was not being excluded when
4758 the OS had its own crypt16() function.
4760 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4761 as a preferred synonym for return_size_limit, both as an option and as an
4764 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4766 30. Changed the default configuration ACL to relax the local part checking rule
4767 for addresses that are not in any local domains. For these addresses,
4768 slashes and pipe symbols are allowed within local parts, but the sequence
4769 /../ is explicitly forbidden.
4771 31. SPA server authentication was not clearing the challenge buffer before
4774 32. log_message in a "warn" ACL statement was writing to the reject log as
4775 well as to the main log, which contradicts the documentation and doesn't
4776 seem right (because no rejection is happening). So I have stopped it.
4778 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4779 However, I am unable to do any testing of this.
4781 34. Fixed an infelicity in the appendfile transport. When checking directories
4782 for a mailbox, to see if any needed to be created, it was accidentally
4783 using path names with one or more superfluous leading slashes; tracing
4784 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4786 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4787 contents are added to the log line that is written for every discarded
4788 recipient. (Previously a log_message setting was ignored.)
4790 36. The ${quote: operator now quotes the string if it is empty.
4792 37. The install script runs exim in order to find its version number. If for
4793 some reason other than non-existence or emptiness, which it checks, it
4794 could not run './exim', it was installing it with an empty version number,
4795 i.e. as "exim-". This error state is now caught, and the installation is
4798 38. An argument was missing from the function that creates an error message
4799 when Exim fails to connect to the socket for saslauthd authentication.
4800 This could cause Exim to crash, or give a corrupted message.
4802 39. Added isip, isip4, and isip6 to ${if conditions.
4804 40. The ACL variables $acl_xx are now saved with the message, and can be
4805 accessed later in routers, transports, and filters.
4807 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4808 strings in the file are not string-expanded.
4810 42. If a MAIL command specified a SIZE value that was too large to fit into an
4811 int variable, the check against message_size_limit failed. Such values are
4812 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4813 day this will have to be increased, but I don't think I want to be around
4814 when emails are that large.
4821 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4822 iconv() is not standard in FreeBSD.
4824 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4825 IPv6 enabled. The observed symptom was a segmentation fault on return from
4826 the function os_common_find_running_interfaces() in src/os.c.
4828 3. In the check_special_case() function in daemon.c I had used "errno" as an
4829 argument name, which causes warnings on some systems. This was basically a
4830 typo, since it was named "eno" in the comments!
4832 4. The code that waits for the clock to tick (at a resolution of some fraction
4833 of a second) so as to ensure message-id uniqueness was always waiting for
4834 at least one whole tick, when it could have waited for less. [This is
4835 almost certainly not relevant at current processor speeds, where it is
4836 unlikely to ever wait at all. But we try to future-proof.]
4838 5. The function that sleeps for a time interval that includes fractions of a
4839 second contained a race. It did not block SIGALRM between setting the
4840 timer, and suspending (a couple of lines later). If the interval was short
4841 and the sigsuspend() was delayed until after it had expired, the suspension
4842 never ended. On busy systems this could lead to processes getting stuck for
4845 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4846 process, before it forks any delivery processes. The open lookup caching
4847 mechanism meant that the open file or database connection was passed into
4848 the delivery process. The problem was that delivery processes always tidy
4849 up cached lookup data. This could cause a problem for the next delivery
4850 process started by the queue runner, because the external queue runner
4851 process does not know about the closure. So the next delivery process
4852 still has data in the lookup cache. In the case of a file lookup, there was
4853 no problem because closing a file descriptor in a subprocess doesn't affect
4854 the parent. However, if the lookup was caching a connection to a database,
4855 the connection was closed, and the second delivery process was likely to
4856 see errors such as "PGSQL: query failed: server closed the connection
4857 unexpectedly". The problem has been fixed by closing all cached lookups
4858 in a queue runner before running a delivery process.
4860 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4861 seem to have the "const" qualifier which it has on other OS. I've
4864 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4865 *of the same type* (client or server) with the same public name that an
4866 error should be diagnosed.
4868 9. When Exim looked up a host name for an IP address, but failed to find the
4869 original IP address when looking up the host name (a safety check), it
4870 output the message "<ip address> does not match any IP for NULL", which was
4871 confusing, to say the least. The bug was that the host name should have
4872 appeared instead of "NULL".
4874 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4875 user that is built into the binary, and the -C or -D options is used, root
4876 privilege is dropped before the configuration file is read. In addition,
4877 logging is switched to stderr instead of the normal log files. If the
4878 configuration then re-defines the Exim user, the unprivileged environment
4879 is probably not what is expected, so Exim logs a panic warning message (but
4882 However, if deliver_drop_privilege is set, the unprivileged state may well
4883 be exactly what is intended, so the warning has been cut out in that case,
4884 and Exim is allowed to try to write to its normal log files.
4890 1. smtp_return_error_details was not giving details for temporary sender
4891 or receiver verification errors.
4893 2. Diagnose a configuration error if two authenticators have the same public
4896 3. Exim used not to create the message log file for a message until the first
4897 delivery attempt. This could be confusing when incoming messages were held
4898 for policy or load reasons. The message log file is now created at the time
4899 the message is received, and an initial "Received" line is written to it.
4901 4. The automatically generated man page for command line options had a minor
4902 bug that caused no ill effects; however, a more serious problem was that
4903 the procedure for building the man page automatically didn't always
4904 operate. Consequently, release 4.20 contains an out-of-date version. This
4905 shouldn't happen again.
4907 5. When building Exim with embedded Perl support, the script that builds the
4908 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4909 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4911 6. The freeze_tell option was not being used for messages that were frozen on
4912 arrival, either by an ACL or by local_scan().
4914 7. Added the smtp_incomplete_transaction log selector.
4916 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4917 was accepting AUTH without a new EHLO.
4919 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4920 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4923 10. Logging of TCP/IP connections (when configured) now happens in the main
4924 daemon process instead of the child process, so that the TCP/IP connection
4925 count is more accurate (but it can never be perfect).
4927 11. The use of "drop" in a nested ACL was not being handled correctly in the
4928 outer ACL. Now, if condition failure induced by the nested "drop" causes
4929 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4930 or "require"), the connection is dropped.
4932 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4933 that yield "discard" can now be used with an "accept" or a "discard" verb,
4934 but an error is generated for any others (because I can't see a useful way
4935 to define what should happen).
4937 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4938 are now processed in the same way as lines in the Exim configuration file.
4939 In particular, continuation lines are supported.
4941 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4943 15. Added -ti meaning -t -i.
4945 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4946 domains, and warn by logging if others are found.
4948 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4949 returned from GIFCONF when Exim is trying to find the list of interfaces on
4950 a host. The code in os.c has been modified to copy each ifreq to an aligned
4951 structure in all cases.
4953 Also, in some cases, the returned ifreq's were being copied to a 'struct
4954 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4955 means the last couple of bytes of an IPv6 address could be chopped if the
4956 ifreq contained only a normal sockaddr (14 bytes storage).
4958 18. Named domain lists were not supported in the hosts_treat_as_local option.
4959 An entry such as +xxxx was not recognized, and was treated as a literal
4962 19. Ensure that header lines added by a DATA ACL are included in the reject log
4963 if the ACL subsequently rejects the message.
4965 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4966 MD5 (which is deprecated).
4968 21. When testing a filter file using -bf, Exim was writing a message when it
4969 took the sender from a "From " line in the message, but it was not doing so
4970 when it took $return_path from a Return-Path: header line. It now does.
4972 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4973 with a valid header line field name (a series of printing characters
4974 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4976 23. Changed "disc" in the source to "disk" to conform to the documentation and
4977 the book and for uniformity.
4979 24. Ignore Sendmail's -Ooption=value command line item.
4981 25. When execve() failed while trying to run a command in a pipe transport,
4982 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4983 could be confused with a return value of 69 from the command itself. This
4984 has been changed to 127, the value the shell returns if it is asked to run
4985 a non-existent command. The wording for the related log line suggests a
4986 non-existent command as the problem.
4988 26. If received_header_text expands to an empty string, do not add a Received:
4989 header line to the message. (Well, it adds a token one on the spool, but
4990 marks it "old" so that it doesn't get used or transmitted.)
4992 27. Installed eximstats 1.28 (addition of -nt option).
4994 28. There was no check for failure on the call to getsockname() in the daemon
4995 code. This can fail if there is a shortage of resources on the system, with
4996 ENOMEM, for example. A temporary error is now given on failure.
4998 29. Contrary to the C standard, it seems that in some environments, the
4999 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5000 program. Exim now does this explicitly; it affects the formatting of
5001 timestamps using strftime().
5003 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5004 complaints. I've now initialized all the variables, to avoid this.
5006 32. Header lines added by a system filter were not being "seen" during
5007 transport-time rewrites.
5009 33. The info_callback() function passed to OpenSSL is set up with type void
5010 (*)(SSL *, int, int), as described somewhere. However, when calling the
5011 function (actually a macro) that sets it up, the type void(*)() is
5012 expected. I've put in a cast to prevent warnings from picky compilers.
5014 34. If a DNS black list lookup found a CNAME record, but there were no A
5015 records associated with the domain it pointed at, Exim crashed.
5017 35. If a DNS black list lookup returned more than one A record, Exim ignored
5018 all but the first. It now scans all returned addresses if a particular IP
5019 value is being sought. In this situation, the contents of the
5020 $dnslist_value variable are a list of all the addresses, separated by a
5023 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5024 to accept a host name and all its aliases if the forward lookup for any of
5025 them yielded the IP address of the incoming connection. Now it accepts only
5026 those names whose forward lookup yields the correct IP address. Any other
5027 names are discarded. This closes a loophole whereby a rogue DNS
5028 administrator could create reverse DNS records to break through a
5029 wildcarded host restriction in an ACL.
5031 37. If a user filter or a system filter that ran in a subprocess used any of
5032 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5033 the wrong values were passed to the pipe command ($thisaddress had the
5034 value of $0, $0 had the value of $1, etc). This bug was introduced by
5035 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5037 38. Improved the line breaking for long SMTP error messages from ACLs.
5038 Previously, if there was no break point between 40 and 75 characters, Exim
5039 left the rest of the message alone. Two changes have been made: (a) I've
5040 reduced the minimum length to 35 characters; (b) if it can't find a break
5041 point between 35 and 75 characters, it looks ahead and uses the first one
5042 that it finds. This may give the occasional overlong line, but at least the
5043 remaining text gets split now.
5045 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5046 file descriptors might be low, and that setting 1000 would always raise it.
5047 It turns out that in some environments, the limit is already over 1000 and
5048 that lowering it causes trouble. So now Exim takes care not to decrease it.
5050 40. When delivering a message, the value of $return_path is set to $sender_
5051 address at the start of routing (routers may change the value). By an
5052 oversight, this default was not being set up when an address was tested by
5053 -bt or -bv, which affected the outcome if any router or filter referred to
5056 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5057 log only when "message" or "log_message" are set. However, if one of the
5058 conditions was an address verification, or a call to a nested ACL, the
5059 messages generated by the underlying test were being passed through. This
5060 no longer happens. The underlying message is available in $acl_verify_
5061 message for both "message" and "log_message" expansions, so it can be
5062 passed through if needed.
5064 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5065 new expansion $bh_ to give the encoded byte string without charset
5066 translation. Translation happens only if iconv() is available; HAVE_ICONV
5067 indicates this at build time. HEADERS_CHARSET gives the charset to
5068 translate to; headers_charset can change it in the configuration, and
5069 "headers charset" can change it in an individual filter file.
5071 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5072 that creates RFC 2047 encoded "words" labels them as that charset instead
5073 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5074 expansion operator; (ii) when Exim creates a From: line for a local
5075 message; (iii) when a header line is rewritten to include a "phrase" part.
5077 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5078 buggy, causing it to skip the first lines of messages whose message ID
5079 ended in 'D'. This would not have bitten before Exim release 4.14, because
5080 message IDs were unlikely to end in 'D' before then. The effect was to have
5081 incorrect size information for certain domains.
5083 45. #include "config.h" was missing at the start of the crypt16.c module. This
5084 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5087 46. If there was a timeout during a "random" callout check, Exim treated it as
5088 a failure of the random address, and carried on sending RSET and the real
5089 address. If the delay was just some slowness somewhere, the response to the
5090 original RCPT would be taken as a response to RSET and so on, causing
5091 mayhem of various kinds.
5093 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5094 when I implemented it. It didn't allow for the fact that some option values
5095 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5096 the right test for positive values.
5098 48. Domain names in DNS records are case-independent. Exim always looks them up
5099 in lower case. Some resolvers return domain names in exactly the case they
5100 appear in the zone file, that is, they may contain uppercase letters. Not
5101 all resolvers do this - some return always lower case. Exim was treating a
5102 change of case by a resolver as a change of domain, similar to a widening
5103 of a domain abbreviation. This triggered its re-routing code and so it was
5104 trying to route what was effectively the same domain again. This normally
5105 caused routing to fail (because the router wouldn't handle the domain
5106 twice). Now Exim checks for this case specially, and just changes the
5107 casing of the domain that it ultimately uses when it transmits the message
5110 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5113 50. If a filter generated a file delivery with a non-absolute name (possible if
5114 no home directory exists for the router), the forbid_file option was not
5117 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5118 the existing equality matching.
5120 52. Exim was using ints instead of ino_t variables in some places where it was
5121 dealing with inode numbers.
5123 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5124 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5125 called TMPDIR, and if it finds it is different, it changes its value.
5127 54. The smtp_printf() function is now made available to local_scan() so
5128 additional output lines can be written before returning. There is also an
5129 smtp_fflush() function to enable the detection of a dropped connection.
5130 The variables smtp_input and smtp_batched_input are exported to
5133 55. Changed the default runtime configuration: the message "Unknown user"
5134 has been removed from the ACL, and instead placed on the localuser router,
5135 using the cannot_route_message feature. This means that any verification
5136 failures that generate their own messages won't get overridden. Similarly,
5137 the "Unrouteable address" message that was in the ACL for unverifiable
5138 relay addresses has also been removed.
5140 56. Added hosts_avoid_esmtp to the smtp transport.
5142 57. The exicyclog script was not checking for the esoteric option
5143 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5144 will work only if exicyclog is run under the appropriate euid.
5146 58. Following a discussion on the list, the rules by which Exim recognises line
5147 endings on incoming messages have been changed. The -dropcr and drop_cr
5148 options are now no-ops, retained only for backwards compatibility. The
5149 following line terminators are recognized: LF CRLF CR. However, special
5150 processing applies to CR:
5152 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5153 nor a local message in the state where . is a terminator.
5155 (ii) If a bare CR is encountered in a header line, an extra space is added
5156 after the line terminator so as not to end the header. The reasoning
5157 behind this is that bare CRs in header lines are most likely either
5158 to be mistakes, or people trying to play silly games.
5160 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5161 was being incorrectly given as 18 bytes larger than it should have been.
5162 This is a VOB (very old bug).
5164 60. This may never have affected anything current, but just in case it has:
5165 When the local host is found other than at the start of a list of hosts,
5166 the local host, those with the same MX, and any that follow, are discarded.
5167 When the list in question was part of a longer list of hosts, the following
5168 hosts (not currently being processed) were also being discarded. This no
5169 longer happens. I'm not sure if this situation could ever has previously
5172 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5175 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5176 header. This is something that is recommended in a new Internet Draft, and
5177 is something that is documented as being done by Sendmail. There are two
5178 possible values. For messages generated by the autoreply transport, Exim
5181 Auto-Submitted: auto-replied
5183 whereas for all other generated messages (e.g. bounces) it adds
5185 Auto-Submitted: auto-generated
5187 63. The "personal" condition in filters now includes a test for the
5188 Auto-Submitted: header. If it contains the string "auto-" the message it
5189 not considered personal.
5191 64. Added rcpt_include_affixes as a generic transport option.
5193 65. Added queue_only_override (default true).
5195 66. Added the syslog_duplication option.
5197 67. If what should have been the first header line of a message consisted of
5198 a space followed by a colon, Exim was mis-interpreting it as a header line.
5199 It isn't of course - it is syntactically invalid and should therefore be
5200 treated as the start of the message body. The misbehaviour could have
5201 caused a number of strange effects, including loss of data in subsequent
5202 header lines, and spool format errors.
5204 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5205 client host had authenticated. This control can now be exercised by an ACL
5206 for more flexibility.
5208 69. By default, callouts do not happen when testing with -bh. There is now a
5209 variant, -bhc, which does actually run the callout code, including
5210 consulting and updating the callout cache.
5212 70. Added support for saslauthd authentication, courtesy of Alexander
5215 71. If statvfs() failed on the spool or log directories while checking their
5216 size for availability, Exim confusingly gave the error "space shortage".
5217 Furthermore, in debugging mode it crashed with a floating point exception.
5218 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5219 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5220 serious problem, Exim now writes to the main and panic logs when this
5221 happens, with details of the failure. It then refuses to accept the
5222 incoming message, giving the message "spool directory problem" or "log
5223 directory problem" with a 421 code for SMTP messages.
5225 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5226 0, 1, and 2 exist, because some OS complain for execs without them (see
5227 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5228 descriptors. However, the code omitted to check that the open succeeded,
5229 causing mysterious errors if for some reason the permissions on /dev/null
5230 got screwed. Now Exim writes a message to the main and panic logs, and
5231 bombs out if it can't open /dev/null.
5233 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5234 interact so that it is all more flexible. It is supposed to remain
5235 backwards compatible. Also added extra_local_interfaces.
5237 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5238 to bomb out with an assertion failure - to the client this appears as a
5239 connection drop. This problem occurs in the part of the code that was taken
5240 from the Samba project. Fortunately, the assertion is in a very simple
5241 function, so I have fixed this by reproducing the function inline in the
5242 one place where it is called, and arranging for authentication to fail
5243 instead of killing the process with assert().
5245 75. The SPA client code was not working when the server requested OEM rather
5246 than Unicode encoding.
5248 76. Added code to make require_files with a specific uid setting more usable in
5249 the case where statting the file as root fails - usually a non-root-mounted
5250 NFS file system. When this happens and the failure is EACCES, Exim now
5251 forks a subprocess and does the per-uid checking as the relevant uid.
5253 77. Added process_log_path.
5255 78. If log_file_path was not explicitly set, a setting of check_log_space or
5256 check_log_inodes was ignored.
5258 79. If a space check for the spool or log partitions fails, the incident is now
5259 logged. Of course, in the latter case the data may get lost...
5261 80. Added the %p formatting code to string_format() so that it can be used to
5262 print addresses in debug_print(). Adjusted all the address printing in the
5263 debugging in store.c to use %p rather than %d.
5265 81. There was a concern that a line of code in smtp_in.c could overflow a
5266 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5267 initially expressed, the concern was not well-founded, because trailing
5268 spaces are removed early. However, if the trailing spaces were followed by
5269 a NULL, they did not get removed, so the overflow was possible. Two fixes
5272 (a) I re-wrote the offending code in a cleaner fashion.
5273 (b) If an incoming SMTP command contains a NULL character, it is rejected
5276 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5277 runs initgroups(), so that if the Exim user is in any additional groups,
5278 they will be used during message reception.
5284 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob