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 PP/01 Date: & Message-Id: revert to normally being appended to a message,
10 only prepend for the Resent-* case. Fixes regression introduced in
11 Exim 4.70 by NM/22 for Bugzilla 607.
13 PP/02 Include check_rfc2047_length in configure.default because we're seeing
14 increasing numbers of administrators be bitten by this.
16 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
18 PP/03 Bugzilla 994: added openssl_options main configuration option.
20 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
22 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
24 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
26 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
27 without a peer certificate, leading to a segfault because of an
28 assumption that peers always have certificates. Be a little more
29 paranoid. Problem reported by Martin Tscholak.
31 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
32 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
33 NB: ClamAV planning to remove STREAM in "middle of 2010".
34 CL also introduces -bmalware, various -d+acl logging additions and
35 more caution in buffer sizes.
37 PP/09 Implemented reverse_ip expansion operator.
39 PP/10 Bugzilla 937: provide a "debug" ACL control.
41 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
43 PP/12 Bugzilla 973: Implement --version.
45 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
47 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
49 PP/15 Bugzilla 816: support multiple condition rules on Routers.
51 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
52 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
53 ignore trailing whitespace.
55 JJ/02 prevent non-panic DKIM error from being sent to paniclog
57 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
60 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
61 Notification from Dr Andrew Aitchison.
63 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
64 ExtendedDetectionInfo response format.
65 Notification from John Horne.
67 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
70 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
71 XSL and documented dependency on system catalogs, with examples of how
74 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
77 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
78 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
79 configuration file which is writeable by the Exim user or group.
81 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
82 of configuration files to cover files specified with the -C option if
83 they are going to be used with root privileges, not just the default
86 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
87 option (effectively making it always true).
89 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
90 files to be used while preserving root privileges.
92 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
93 that rogue child processes cannot use them.
95 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
96 run-time user, instead of root.
98 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
99 Exim run-time user without dropping privileges.
101 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
102 result string, instead of calling string_vformat() twice with the same
105 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
106 for other users. Others should always drop root privileges if they use
107 -C on the command line, even for a whitelisted configure file.
109 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
115 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
116 $data_path, and $header_path variables; fixed documentation bugs and
119 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
120 exipick to access non-standard spools, including the "frozen" queue
123 NM/01 Bugzilla 965: Support mysql stored procedures.
124 Patch from Alain Williams
126 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
128 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
129 Patch from Andreas Metzler
131 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
132 Patch from Kirill Miazine
134 NM/05 Bugzilla 671: Added umask to procmail example.
136 JJ/03 installed exipick 20100323.0, fixing doc bug
138 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
139 directory. Notification and patch from Dan Rosenberg.
141 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
143 TK/02 Improve log output when DKIM signing operation fails.
145 MH/01 Treat the transport option dkim_domain as a colon separated
146 list, not as a single string, and sign the message with each element,
147 omitting multiple occurences of the same signer.
149 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
150 Bugzilla 985, 986. Patch by Simon Arlott
152 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
153 Patch by Simon Arlott
155 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
156 MBX locking. Notification from Dan Rosenberg.
162 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
164 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
166 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
168 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
170 NM/04 Bugzilla 915: Flag broken perl installation during build.
176 TK/01 Added patch by Johannes Berg that expands the main option
177 "spamd_address" if it starts with a dollar sign.
179 TK/02 Write list of recipients to X-Envelope-Sender header when building
180 the mbox-format spool file for content scanning (suggested by Jakob
183 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
184 (http://www.dcc-servers.net/) support via dccifd. Activated by
185 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
187 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
188 by Mark Daniel Reidel <mr@df.eu>.
190 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
191 When building exim an external PCRE library is now needed -
192 PCRE is a system library on the majority of modern systems.
193 See entry on PCRE_LIBS in EDITME file.
195 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
196 conversation. Added nologin parameter to request.
197 Patch contributed by Kirill Miazine.
199 TF/01 Do not log submission mode rewrites if they do not change the address.
201 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
203 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
204 log files in place. Contributed by Roberto Lima.
206 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
208 TF/03 Bugzilla 615: When checking the local_parts router precondition
209 after a local_part_suffix or local_part_prefix option, Exim now
210 does not use the address's named list lookup cache, since this
211 contains cached lookups for the whole local part.
213 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
214 Robert Millan. Documentation is in experimental-spec.txt.
216 TF/04 Bugzilla 668: Fix parallel build (make -j).
218 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
220 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
221 Patch provided by Jan Srzednicki.
223 TF/05 Leading white space used to be stripped from $spam_report which
224 wrecked the formatting. Now it is preserved.
226 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
227 that they are available at delivery time.
229 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
231 TF/08 TLS error reporting now respects the incoming_interface and
232 incoming_port log selectors.
234 TF/09 Produce a more useful error message if an SMTP transport's hosts
235 setting expands to an empty string.
237 NM/06 Bugzilla 744: EXPN did not work under TLS.
238 Patch provided by Phil Pennock.
240 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
241 Patch provided by Richard Godbee.
243 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
244 acl_smtp_notquit, added index entry.
246 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
247 Patch provided by Eugene Bujak.
249 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
250 accept(). Patch provided by Maxim Dounin.
252 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
253 Patch provided by Phil Pennock.
255 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
257 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
258 Patch provided by Brad "anomie" Jorsch.
260 NM/14 Bugzilla 622: Added timeout setting to transport filter.
261 Patch provided by Dean Brooks.
263 TK/05 Add native DKIM support (does not depend on external libraries).
265 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
266 Patch provided by Graeme Fowler.
268 NM/16 Bugzilla 851: Documentation example syntax fix.
270 NM/17 Changed NOTICE file to remove references to embedded PCRE.
272 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
275 NM/19 Bugzilla 745: TLS version reporting.
276 Patch provided by Phil Pennock.
278 NM/20 Bugzilla 167: bool: condition support.
279 Patch provided by Phil Pennock.
281 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
282 clients. Patch provided by Phil Pennock.
284 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
285 Patch provided by Brad "anomie" Jorsch.
287 NM/23 Bugzilla 687: Fix misparses in eximstats.
288 Patch provided by Heiko Schlittermann.
290 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
291 Patch provided by Heiko Schlittermann.
293 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
294 plus update to original patch.
296 NM/26 Bugzilla 799: Documentation correction for ratelimit.
298 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
299 Patch provided by David Brownlee.
301 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
303 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
305 NM/30 Bugzilla 888: TLS documentation bugfixes.
307 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
309 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
310 Unlike the original bugzilla I have changed all shell scripts in src tree.
312 NM/33 Bugzilla 898: Transport filter timeout fix.
313 Patch by Todd Rinaldo.
315 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
316 Patch by Serge Demonchaux.
318 NM/35 Bugzilla 39: Base64 decode bug fixes.
319 Patch by Jakob Hirsch.
321 NM/36 Bugzilla 909: Correct connect() call in dcc code.
323 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
325 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
327 NM/39 Bugzilla 911: Fixed MakeLinks build script.
333 TK/01 Add preliminary DKIM support. Currently requires a forked version of
334 ALT-N's libdkim that I have put here:
335 http://duncanthrax.net/exim-experimental/
337 Note to Michael Haardt: I had to rename some vars in sieve.c. They
338 were called 'true' and it seems that C99 defines that as a reserved
339 keyword to be used with 'bool' variable types. That means you could
340 not include C99-style headers which use bools without triggering
341 build errors in sieve.c.
343 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
344 as mailq or other aliases. Changed the --help handling significantly
345 to do whats expected. exim_usage() emits usage/help information.
347 SC/01 Added the -bylocaldomain option to eximstats.
349 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
351 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
353 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
359 PH/01 Another patch from the Sieve maintainer.
361 PH/02 When an IPv6 address is converted to a string for single-key lookup
362 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
363 dots are used instead of colons so that keys in lsearch files need not
364 contain colons. This was done some time before quoting was made available
365 in lsearch files. However, iplsearch files do require colons in IPv6 keys
366 (notated using the quote facility) so as to distinguish them from IPv4
367 keys. This meant that lookups for IP addresses in host lists did not work
368 for iplsearch lookups.
370 This has been fixed by arranging for IPv6 addresses to be expressed with
371 colons if the lookup type is iplsearch. This is not incompatible, because
372 previously such lookups could never work.
374 The situation is now rather anomolous, since one *can* have colons in
375 ordinary lsearch keys. However, making the change in all cases is
376 incompatible and would probably break a number of configurations.
378 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
381 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
382 conversion specification without a maximum field width, thereby enabling
383 a rogue spamd server to cause a buffer overflow. While nobody in their
384 right mind would setup Exim to query an untrusted spamd server, an
385 attacker that gains access to a server running spamd could potentially
386 exploit this vulnerability to run arbitrary code as the Exim user.
388 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
389 $primary_hostname instead of what libspf2 thinks the hosts name is.
391 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
392 a directory entry by the name of the lookup key. Previously, if a
393 symlink pointed to a non-existing file or a file in a directory that
394 Exim lacked permissions to read, a lookup for a key matching that
395 symlink would fail. Now it is enough that a matching directory entry
396 exists, symlink or not. (Bugzilla 503.)
398 PH/03 The body_linecount and body_zerocount variables are now exported in the
401 PH/04 Added the $dnslist_matched variable.
403 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
404 This means they are set thereafter only if the connection becomes
407 PH/06 Added the client_condition to authenticators so that some can be skipped
408 by clients under certain conditions.
410 PH/07 The error message for a badly-placed control=no_multiline_responses left
411 "_responses" off the end of the name.
413 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
415 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
416 (without spaces) instead of just copying the configuration text.
418 PH/10 Added the /noupdate option to the ratelimit ACL condition.
420 PH/11 Added $max_received_linelength.
422 PH/12 Added +ignore_defer and +include_defer to host lists.
424 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
425 way in which PCRE > 7.0 is built.
427 PH/14 Implemented queue_only_load_latch.
429 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
430 MAIL command. The effect was to mangle the value on 64-bit systems.
432 PH/16 Another patch from the Sieve maintainer.
434 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
436 PH/18 If a system quota error occurred while trying to create the file for
437 a maildir delivery, the message "Mailbox is full" was not appended to the
438 bounce if the delivery eventually timed out. Change 4.67/27 below applied
439 only to a quota excession during the actual writing of the file.
441 PH/19 It seems that peer DN values may contain newlines (and other non-printing
442 characters?) which causes problems in log lines. The DN values are now
443 passed through string_printing() before being added to log lines.
445 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
446 and InterBase are left for another time.)
448 PH/21 Added message_body_newlines option.
450 PH/22 Guard against possible overflow in moan_check_errorcopy().
452 PH/23 POSIX allows open() to be a macro; guard against that.
454 PH/24 If the recipient of an error message contained an @ in the local part
455 (suitably quoted, of course), incorrect values were put in $domain and
456 $local_part during the evaluation of errors_copy.
462 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
463 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
466 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
467 issue a MAIL command.
469 PH/02 In an ACL statement such as
471 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
473 if a client was not listed at all, or was listed with a value other than
474 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
475 the condition was not true (as it should be), so access was not denied.
476 The bug was that the ! inversion was incorrectly passed on to the second
477 item. This has been fixed.
479 PH/03 Added additional dnslists conditions == and =& which are different from
480 = and & when the dns lookup returns more than one IP address.
482 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
483 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
485 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
486 FSYNC, which compiles an option called disable_fsync that allows for
487 bypassing fsync(). The documentation is heavily laced with warnings.
489 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
491 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
492 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
493 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
494 including adding "make clean"; (3) Added -fPIC when compiling the test
495 dynamically loaded module, to get rid of a warning.
497 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
498 message fails, move_frozen_messages = true and ignore_bounce_errors_after
499 = 0s. The bug is otherwise harmless.
501 PH/07 There was a bug in the dovecot authenticator such that the value of
502 $auth1 could be overwritten, and so not correctly preserved, after a
503 successful authentication. This usually meant that the value preserved by
504 the server_setid option was incorrect.
506 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
508 PH/09 Installed PCRE release 7.0.
510 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
511 run for batched SMTP input. It is now run at the start of every message
512 in the batch. While fixing this I discovered that the process information
513 (output by running exiwhat) was not always getting set for -bs and -bS
514 input. This is fixed, and it now also says "batched" for BSMTP.
516 PH/11 Added control=no_pipelining.
518 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
519 patch, slightly modified), and move the expansion of helo_data till after
520 the connection is made in the smtp transport (so it can use these
523 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
525 PH/14 Added log_selector = +pid.
527 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
529 PH/16 Add ${if forany and ${if forall.
531 PH/17 Added dsn_from option to vary the From: line in DSNs.
533 PH/18 Flush SMTP output before performing a callout, unless control =
534 no_callout_flush is set.
536 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
537 was true (the default) a successful delivery failed to delete the retry
538 item, thus causing premature timeout of the address. The bug is now
541 PH/20 Added hosts_avoid_pipelining to the smtp transport.
543 PH/21 Long custom messages for fakedefer and fakereject are now split up
544 into multiline reponses in the same way that messages for "deny" and
545 other ACL rejections are.
547 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
548 with slight modification.
550 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
551 draft, changing the syntax and factoring some duplicate code".
553 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
554 for deliveries of the second and subsequent messages over the same SMTP
557 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
558 ${reduce, with only minor "tidies".
560 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
562 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
563 expansion side effects.
565 PH/27 When a message times out after an over-quota error from an Exim-imposed
566 quota, the bounce message says "mailbox is full". This message was not
567 being given when it was a system quota that was exceeded. It now should
570 MH/03 Made $recipients available in local_scan(). local_scan() already has
571 better access to the recipient list through recipients_list[], but
572 $recipients can be useful in postmaster-provided expansion strings.
574 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
575 in the case of a MAIL command with additional options following the
576 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
577 were accidentally chopped off.
579 PH/29 SMTP synchronization checks are implemented when a command is read -
580 there is a check that no more input is waiting when there shouldn't be
581 any. However, for some commands, a delay in an ACL can mean that it is
582 some time before the response is written. In this time, more input might
583 arrive, invalidly. So now there are extra checks after an ACL has run for
584 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
585 pipelining has not been advertised.
587 PH/30 MH's patch to allow iscntrl() characters to be list separators.
589 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
590 returned in the SMTP response when smtp_return_error_details was false.
593 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
594 instead of the C I/O that was originally supplied, because problems were
597 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
598 Exim which did not show up earlier: it was assuming that a call to
599 SSL_CTX_set_info_callback() might give an error value. In fact, there is
600 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
601 was a macro that became an assignment, so it seemed to work. This has
602 changed to a proper function call with a void return, hence the compile
603 error. Exim's code has been fixed.
605 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
608 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
610 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
612 PH/37 If a message is not accepted after it has had an id assigned (e.g.
613 because it turns out to be too big or there is a timeout) there is no
614 "Completed" line in the log. When some messages of this type were
615 selected by exigrep, they were listed as "not completed". Others were
616 picked up by some special patterns. I have improved the selection
617 criteria to be more general.
619 PH/38 The host_find_failed option in the manualroute router can now be set
620 to "ignore", to completely ignore a host whose IP address cannot be
621 found. If all hosts are ignored, the behaviour is controlled by the new
622 host_all_ignored option.
624 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
625 homing or because of multiple MX records with /mx) generated more than
626 one IP address, and the following item turned out to be the local host,
627 all the secondary addresses of the first item were incorrectly removed
628 from the list, along with the local host and any following hosts (which
629 is what is supposed to happen).
631 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
632 whoever called Exim into the -H file. In the case of the daemon it was
633 behaving confusingly. When first started, it used values for whoever
634 started the daemon, but after a SIGHUP it used the Exim user (because it
635 calls itself on a restart). I have changed the code so that it now always
638 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
639 message are rejected with the same error (e.g. no authentication or bad
640 sender address), and a DATA command is nevertheless sent (as can happen
641 with PIPELINING or a stupid MUA), the error message that was given to the
642 RCPT commands is included in the rejection of the DATA command. This is
643 intended to be helpful for MUAs that show only the final error to their
646 PH/42 Another patch from the Sieve maintainer.
648 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
649 Eximstats - Fixed some broken HTML links and added missing column headers
651 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
652 columns for Rejects, Temp Rejects, Ham, and Spam rows.
654 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
656 PH/43 Yet another patch from the Sieve maintainer.
658 PH/44 I found a way to check for a TCP/IP connection going away before sending
659 the response to the final '.' that terminates a message, but only in the
660 case where the client has not sent further data following the '.'
661 (unfortunately, this is allowed). However, in many cases there won't be
662 any further data because there won't be any more messages to send. A call
663 to select() can be used: if it shows that the input is "ready", there is
664 either input waiting, or the socket has been closed. An attempt to read
665 the next input character can distinguish the two cases. Previously, Exim
666 would have sent an OK response which the client would never have see.
667 This could lead to message repetition. This fix should cure that, at
668 least in a lot of common cases.
670 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
671 advertised in response to EHLO.
677 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
678 fixed by 4.65/MH/01 (is this a record?) are fixed:
680 (i) An empty string was always treated as zero by the numeric comparison
681 operators. This behaviour has been restored.
683 (ii) It is documented that the numeric comparison operators always treat
684 their arguments as decimal numbers. This was broken in that numbers
685 starting with 0 were being interpreted as octal.
687 While fixing these problems I realized that there was another issue that
688 hadn't been noticed. Values of message_size_limit (both the global option
689 and the transport option) were treated as octal if they started with 0.
690 The documentation was vague. These values are now always treated as
691 decimal, and I will make that clear in the documentation.
697 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
698 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
701 MH/01 Don't check that the operands of numeric comparison operators are
702 integers when their expansion is in "skipping" mode (fixes bug
703 introduced by 4.64-PH/07).
705 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
706 child addresses, Exim now panics and dies. Previously, because the count
707 is held in a short int, deliveries were likely to be lost. As such a
708 large number of recipients for a single message is ridiculous
709 (performance will be very, very poor), I have chosen to impose a limit
710 rather than extend the field.
716 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
717 leftover -K file (the existence of which was triggered by #402).
718 While we were at it, introduced process PID as part of the -K
719 filename. This should rule out race conditions when creating
722 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
723 processing considerably. Previous code took too long for large mails,
724 triggering a timeout which in turn triggers #401.
726 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
727 in the DK code in transports.c. sendfile() is not really portable,
728 hence the _LINUX specificness.
730 TF/01 In the add_headers option to the mail command in an Exim filter,
731 there was a bug that Exim would claim a syntax error in any
732 header after the first one which had an odd number of characters
735 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
736 callout verification, Exim cached a "reject" for the entire domain. This
737 is correct for most verifications, but it is not correct for a recipient
738 verification with use_sender or use_postmaster set, because in that case
739 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
740 case of MAIL FROM:<> rejection from other early rejections (e.g.
741 rejection of HELO). When verifying a recipient using a non-null MAIL
742 address, the cache is ignored if it shows MAIL FROM:<> rejection.
743 Whatever the result of the callout, the value of the domain cache is
744 left unchanged (for any other kind of callout, getting as far as trying
745 RCPT means that the domain itself is ok).
747 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
750 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
751 manifest itself as EPIPE rather than ECONNECT. When tidying away a
752 session, the daemon ignores ECONNECT errors and logs others; it now
753 ignores EPIPE as well.
755 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
756 (quoted-printable decoding).
758 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
759 later the small subsequent patch to fix an introduced bug.
761 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
763 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
765 PH/08 An error is now given if message_size_limit is specified negative.
767 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
768 to be given (somewhat) arbitrary names.
770 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
773 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
774 miscellaneous code fixes
776 PH/10 Added the log_reject_target ACL modifier to specify where to log
779 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
780 hostname. This is wrong, because it relates to the incoming message (and
781 probably the interface on which it is arriving) and not to the outgoing
782 callout (which could be using a different interface). This has been
783 changed to use the value of the helo_data option from the smtp transport
784 instead - this is what is used when a message is actually being sent. If
785 there is no remote transport (possible with a router that sets up host
786 addresses), $smtp_active_hostname is used.
788 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
789 tweaks were necessary in order to get it to work (see also 21 below):
790 (a) The code assumed that strncpy() returns a negative number on buffer
791 overflow, which isn't the case. Replaced with Exim's string_format()
793 (b) There were several signed/unsigned issues. I just did the minimum
794 hacking in of casts. There is scope for a larger refactoring.
795 (c) The code used strcasecmp() which is not a standard C function.
796 Replaced with Exim's strcmpic() function.
797 (d) The code set only $1; it now sets $auth1 as well.
798 (e) A simple test gave the error "authentication client didn't specify
799 service in request". It would seem that Dovecot has changed its
800 interface. Fortunately there's a specification; I followed it and
801 changed what the client sends and it appears to be working now.
803 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
806 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
807 address A is aliased to B and C, where B exists and C does not. Without
808 -v the output is "A verified" because verification stops after a
809 successful redirection if more than one address is generated. However,
810 with -v the child addresses are also verified. Exim was outputting "A
811 failed to verify" and then showing the successful verification for C,
812 with its parentage. It now outputs "B failed to verify", showing B's
813 parentage before showing the successful verification of C.
815 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
816 look up a TXT record in a specific list after matching in a combined
819 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
820 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
821 they consult the DNS. I had assumed they would set it the way they
822 wanted; and indeed my experiments on Linux seem to show that in some
823 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
824 To be on the safe side, however, I have now made the interface to
825 host_find_byname() similar to host_find_bydns(), with an argument
826 containing the DNS resolver options. The host_find_byname() function now
827 sets these options at its start, just as host_find_bydns() does. The smtp
828 transport options dns_qualify_single and dns_search_parents are passed to
829 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
830 of host_find_byname() use the default settings of RES_DEFNAMES
831 (qualify_single) but not RES_DNSRCH (search_parents).
833 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
834 spool_read_header() do less string testing, by means of a preliminary
835 switch on the second character of optional "-foo" lines. (This is
836 overdue, caused by the large number of possibilities that now exist.
837 Originally there were few.) While I was there, I also converted the
838 str(n)cmp tests so they don't re-test the leading "-" and the first
839 character, in the hope this might squeeze out yet more improvement.
841 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
842 flag allowing group syntax was set by the header_syntax check but not
843 turned off, possible causing trouble later; (2) The flag was not being
844 set at all for the header_verify test, causing "group"-style headers to
845 be rejected. I have now set it in this case, and also caused header_
846 verify to ignore an empty address taken from a group. While doing this, I
847 came across some other cases where the code for allowing group syntax
848 while scanning a header line wasn't quite right (mostly, not resetting
849 the flag correctly in the right place). These bugs could have caused
850 trouble for malformed header lines. I hope it is now all correct.
852 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
853 with the "reply" argument non-NULL. The code, however (which originally
854 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
855 but it didn't always do it. This confused somebody who was copying the
856 code for some other use. I have removed all the tests.
858 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
859 feature that was used to support insecure browsers during the U.S. crypto
860 embargo. It requires special client support, and Exim is probably the
861 only MTA that supported it -- and would never use it because real RSA is
862 always available. This code has been removed, because it had the bad
863 effect of slowing Exim down by computing (never used) parameters for the
864 RSA_EXPORT functionality.
866 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
867 authenticator to fail if there's a tab character in the incoming data
868 (there should never be unless someone is messing about, as it's supposed
869 to be base64-encoded). Also added, on Timo's advice, the "secured" option
870 if the connection is using TLS or if the remote IP is the same as the
871 local IP, and the "valid-client-cert option" if a client certificate has
874 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
875 authenticators. This can be used for authorization after authentication
876 succeeds. (In the case of plaintext, it servers for both authentication
879 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
880 if any retry times were supplied.
882 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
883 connection, where there is no client IP address to check. In this
884 situation, the verify now always succeeds.
886 PH/25 Applied John Jetmore's -Mset patch.
888 PH/26 Added -bem to be like -Mset, but loading a message from a file.
890 PH/27 In a string expansion for a processed (not raw) header when multiple
891 headers of the same name were present, leading whitespace was being
892 removed from all of them, but trailing whitespace was being removed only
893 from the last one. Now trailing whitespace is removed from each header
894 before concatenation. Completely empty headers in a concatenation (as
897 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
898 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
900 PH/29 [Removed. This was a change that I later backed out, and forgot to
901 correct the ChangeLog entry (that I had efficiently created) before
902 committing the later change.]
904 PH/30 Exim was sometimes attempting to deliver messages that had suffered
905 address errors (4xx response to RCPT) over the same connection as other
906 messages routed to the same hosts. Such deliveries are always "forced",
907 so retry times are not inspected. This resulted in far too many retries
908 for the affected addresses. The effect occurred only when there were more
909 hosts than the hosts_max_try setting in the smtp transport when it had
910 the 4xx errors. Those hosts that it had tried were not added to the list
911 of hosts for which the message was waiting, so if all were tried, there
912 was no problem. Two fixes have been applied:
914 (i) If there are any address or message errors in an SMTP delivery, none
915 of the hosts (tried or untried) are now added to the list of hosts
916 for which the message is waiting, so the message should not be a
917 candidate for sending over the same connection that was used for a
918 successful delivery of some other message. This seems entirely
919 reasonable: after all the message is NOT "waiting for some host".
920 This is so "obvious" that I'm not sure why it wasn't done
921 previously. Hope I haven't missed anything, but it can't do any
922 harm, as the worst effect is to miss an optimization.
924 (ii) If, despite (i), such a delivery is accidentally attempted, the
925 routing retry time is respected, so at least it doesn't keep
926 hammering the server.
928 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
929 in ${readsocket because some servers need this prod.
931 PH/32 Added some extra debug output when updating a wait-xxx database.
933 PH/33 The hint "could be header name not terminated by colon", which has been
934 given for certain expansion errors for a long time, was not being given
935 for the ${if def:h_colon_omitted{... case.
937 PH/34 The spec says: "With one important exception, whenever a domain list is
938 being scanned, $domain contains the subject domain." There was at least
939 one case where this was not true.
941 PH/35 The error "getsockname() failed: connection reset by peer" was being
942 written to the panic log as well as the main log, but it isn't really
943 panic-worthy as it just means the connection died rather early on. I have
944 removed the panic log writing for the ECONNRESET error when getsockname()
947 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
948 runs only) independently of the message's sender address. This meant
949 that, if the 4xx error was in fact related to the sender, a different
950 message to the same recipient with a different sender could confuse
951 things. In particualar, this can happen when sending to a greylisting
952 server, but other circumstances could also provoke similar problems.
953 I have changed the default so that the retry time for these errors is now
954 based a combination of the sender and recipient addresses. This change
955 can be overridden by setting address_retry_include_sender=false in the
958 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
959 remote server are returned as part of bounce messages. This was not
960 happening for LMTP over a pipe (the lmtp transport), but now it is the
961 same for both kinds of LMTP.
963 PH/38 Despite being documented as not happening, Exim was rewriting addresses
964 in header lines that were in fact CNAMEs. This is no longer the case.
966 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
967 and queue runs started by the daemon processed all messages. This has
968 been fixed so that -R and -S can now usefully be given with -q<time>.
970 PH/40 Import PCRE release 6.7 (fixes some bugs).
972 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
974 PH/42 Give an error if -q is specified more than once.
976 PH/43 Renamed the variables $interface_address and $interface_port as
977 $received_ip_address and $received_port, to make it clear that these
978 values apply to message reception, and not to the outgoing interface when
979 a message is delivered. (The old names remain recognized, of course.)
981 PH/44 There was no timeout on the connect() call when using a Unix domain
982 socket in the ${readsocket expansion. There now is.
984 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
985 be meaningful with "accept".
987 SC/01 Eximstats V1.43
988 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
990 SC/02 Eximstats V1.44
991 Use a glob alias rather than an array ref in the generated
992 parser. This improves both readability and performance.
994 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
995 Collect SpamAssassin and rejection statistics.
996 Don't display local sender or destination tables unless
997 there is data to show.
998 Added average volumes into the top table text output.
1000 SC/04 Eximstats V1.46
1001 Collect data on the number of addresses (recipients)
1002 as well as the number of messages.
1004 SC/05 Eximstats V1.47
1005 Added 'Message too big' to the list of mail rejection
1006 reasons (thanks to Marco Gaiarin).
1008 SC/06 Eximstats V1.48
1009 Mainlog lines which have GMT offsets and are too short to
1010 have a flag are now skipped.
1012 SC/07 Eximstats V1.49 (Alain Williams)
1013 Added the -emptyok flag.
1015 SC/08 Eximstats V1.50
1016 Fixes for obtaining the IP address from reject messages.
1018 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1019 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1020 whitesspace changes from 4.64-PH/27
1022 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1025 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1026 are found, allow negative numbers in numeric criteria)
1028 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1030 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1031 to match changes made in 4.64-PH/43
1033 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1035 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1036 "domain" is actually an IP address, and force a failure. This locks out
1037 those revolvers/nameservers that support "A-for-A" lookups, in
1038 contravention of the specifications.
1040 PH/48 When a host name was looked up from an IP address, and the subsequent
1041 forward lookup of the name timed out, the host name was left in
1042 $sender_host_name, contrary to the specification.
1044 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1045 restricted to single-key lookups, Exim was not diagnosing an error if
1046 * or *@ was used with a query-style lookup.
1048 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1050 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1051 long ago, but noone interested enough thought of it. Let's just say that
1052 the "1.1" means that there are some new functions that weren't there at
1053 some point in the past.
1055 PH/51 Error processing for expansion failure of helo_data from an smtp
1056 transport during callout processing was broken.
1058 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1059 tested/used via the -bh/-bhc/-bs options.
1061 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1062 bug, fixed in subsequent PCRE releases).
1064 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1065 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1067 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1073 SC/01 Use a glob alias rather than an array ref in eximstats generated
1074 parser. This improves both readability and performance.
1076 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1077 Don't display local sender or destination tables in eximstats unless
1078 there is data to show.
1079 Added average volumes into the eximstats top table text output.
1081 SC/03 Collect data on the number of addresses (recipients) as well
1082 as the number of messages in eximstats.
1084 TF/01 Correct an error in the documentation for the redirect router. Exim
1085 does (usually) call initgroups() when daemonizing.
1087 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1088 with consistent privilege compared to when running as a daemon.
1090 TF/03 Note in the spec that $authenticated_id is not set for local
1091 submissions from trusted users.
1093 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1094 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1096 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1097 by adding some example configuration directives to the default
1098 configuration file. A little bit of work is required to uncomment the
1099 directives and define how usernames and passwords are checked, but
1100 there is now a framework to start from.
1102 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1103 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1104 without this. I don't know how relevant this is to other LDAP libraries.
1106 PH/02 Add the verb name to the "unknown ACL verb" error.
1108 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1110 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1112 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1113 directory not expanded when it should be if an expanded home directory
1114 was set for the address (which is overridden by the transport).
1116 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1119 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1120 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1121 because it is too late at that time, and has no effect.
1123 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1124 security issue with \' (bugzilla #107). I could not use the
1125 PQescapeStringConn() function, because it needs a PGconn value as one of
1128 PH/08 When testing addresses using -bt, indicate those final addresses that
1129 are duplicates that would not cause an additional delivery. At least one
1130 person was confused, thinking that -bt output corresponded to deliveries.
1131 (Suppressing duplicates isn't a good idea as you lose the information
1132 about possibly different redirections that led to the duplicates.)
1134 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1135 systems where poll() doesn't work, in particular OS X.
1137 PH/10 Added more information to debugging output for retry time not reached.
1139 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1140 operations in malware.c.
1142 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1145 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1146 syslog_duplication set false, log lines that would normally be written
1147 both the the main log and to the reject log were not written to syslog at
1150 PH/14 In the default configuration, change the use of "message" in ACL warn
1151 statements to "add_header".
1153 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1154 not followed by a command (e.g. "seen endif").
1156 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1157 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1160 PH/17 Added extra conditions to the default value of delay_warning_condition
1164 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1165 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1166 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1169 The Auto-Submitted: and various List- headers are standardised, whereas I
1170 don't think Precedence: ever was.
1172 PH/18 Refactored debugging code in route_finduser() to show more information,
1173 in particular, the error code if getpwnam() issues one.
1175 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1176 This is apparently needed in addition to the PH/07 change above to avoid
1177 any possible encoding problems.
1179 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1180 but not after initializing Perl.
1182 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1183 output them only if debugging. By default they are written stderr,
1184 apparently, which is not desirable.
1186 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1189 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1192 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1194 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1195 authenticated or an ident call has been made. Suppress the default
1196 values for $authenticated_id and $authenticated_sender (but permit -oMai
1197 and -oMas) when testing with -bh.
1199 PH/24 Re-jigged the order of the tests in the default configuration so that the
1200 tests for valid domains and recipients precede the DNS black list and CSA
1201 tests, on the grounds that those ones are more expensive.
1203 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1204 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1205 This bug exists in every version of Exim that I still have, right back to
1208 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1209 However, an attempt to turn on case-sensitivity in a regex key by
1210 including (?-i) didn't work because the subject string was already
1211 lowercased, and the effects were non-intuitive. It turns out that a
1212 one-line patch can be used to allow (?-i) to work as expected.
1218 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1219 other effects) broke the use of negated acl sub-conditions.
1221 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1224 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1225 "Deny" causes Exim to reject the incoming connection with a 554 error.
1226 Unfortunately, if there is a major crisis, such as a disk failure,
1227 tcp-wrappers gives "deny", whereas what one would like would be some
1228 kind of temporary error. A kludge has been added to help with this.
1229 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1230 554 error is used if errno is still zero or contains ENOENT (which occurs
1231 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1234 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1236 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1237 errors. Otherwise a message that provokes a temporary error (when other
1238 messages do not) can cause a whole host to time out.
1240 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1241 addresses were routed directly to files or pipes from a redirect router.
1242 File deliveries just didn't batch; pipe deliveries might have suffered
1245 PH/06 A failure to get a lock for a hints database would erroneously always say
1246 "Failed to get write lock", even when it was really a read lock.
1248 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1249 of 0600. This has been changed to use the value of the lockfile_mode
1250 option (which defaults to 0600).
1252 PH/08 Applied small patch from the Sieve maintainer.
1254 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1255 folder from quota calculations, a direct delivery into this folder messed
1256 up the contents of the maildirsize file. This was because the regex was
1257 used only to exclude .Trash (or whatever) when the size of the mailbox
1258 was calculated. There was no check that a delivery was happening into an
1259 excluded directory. This bug has been fixed by ignoring all quota
1260 processing for deliveries into excluded directories.
1262 PH/10 Added the maildirfolder_create_regex option to appendfile.
1268 PH/01 The code for finding all the local interface addresses on a FreeBSD
1269 system running IPv6 was broken. This may well have applied to all BSD
1270 systems, as well as to others that have similar system calls. The broken
1271 code found IPv4 interfaces correctly, but gave incorrect values for the
1272 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1273 that it would not match correctly against @[] and not recognize the IPv6
1276 PH/02 The ipliteral router was not recognizing addresses of the form user@
1277 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1279 PH/03 Added disable_ipv6.
1281 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1282 decoded form, because it is most often used to construct To: headers
1283 lines in autoreplies, and the decoded form may well be syntactically
1284 invalid. However, $reply_address has leading white space removed, and all
1285 newlines turned into spaces so that the autoreply transport does not
1288 PH/05 If group was specified without a user on a router, and no group or user
1289 was specified on a transport, the group from the router was ignored.
1291 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1292 for visible compile-time settings that can be used to change these
1293 numbers, for those that want even more. Backwards compatibility with old
1294 spool files has been maintained. However, going back to a previous Exim
1295 release will lost any variables that are in spool files.
1297 PH/07 Two small changes when running in the test harness: increase delay when
1298 passing a TCP/IP connection to a new process, in case the original
1299 process has to generate a bounce, and remove special handling of
1300 127.0.0.2 (sic), which is no longer necessary.
1302 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1303 be the same on different OS.
1305 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1308 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1309 whether --show-vars was specified or not
1311 JJ/02 exipick: Added support for new ACL variable spool format introduced
1314 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1315 syntactically invalid From: or Reply-to: line, and a filter used this to
1316 generate an autoreply, and therefore failed to obtain an address for the
1317 autoreply, Exim could try to deliver to a non-existent relative file
1318 name, causing unrelated and misleading errors. What now happens is that
1319 it logs this as a hard delivery error, but does not attempt to create a
1322 PH/11 The exinext utility has a -C option for testing purposes, but although
1323 the given file was scanned by exinext itself; it wasn't being passed on
1324 when Exim was called.
1326 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1327 an end-of-file indication when reading a command response.
1329 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1330 compiled. In many other places in Exim, IPv6 addresses are always
1331 recognized, so I have changed this. It also means that IPv4 domain
1332 literals of the form [IPV4:n.n.n.n] are now always recognized.
1334 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1335 used if the router is not running as root, for example, when verifying at
1336 ACL time, or when using -bh. The debugging output from this situation was
1337 non-existent - all you got was a failure to exec. I have made two
1340 (a) Failures to set uid/gid, the current directory, or a process leader
1341 in a subprocess such as that created by queryprogram now generate
1342 suitable debugging ouput when -d is set.
1344 (b) The queryprogram router detects when it is not running as root,
1345 outputs suitable debugging information if -d is set, and then runs
1346 the subprocess without attempting to change uid/gid.
1348 PH/15 Minor change to Makefile for building test_host (undocumented testing
1351 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1352 additional section of a DNS packet that returns MX or SRV records.
1353 Instead, it always explicitly searches for A/AAAA records. This avoids
1354 major problems that occur when a DNS server includes only records of one
1355 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1356 fixed another bug: if SRV records were looked up and the corresponding
1357 address records were *not* found in the additional section, the port
1358 values from the SRV records were lost.
1360 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1361 using the correct key (the original address) when searching the retry
1362 rules in order to find which one to use for generating the retry hint.
1364 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1365 adding the default one. Similarly, if it contains a Reply-To: header, the
1366 errors_reply_to option, if set, is not used.
1368 PH/19 When calculating a retry time, Exim used to measure the "time since
1369 failure" by looking at the "first failed" field in the retry record. Now
1370 it does not use this if it is later than than the arrival time of the
1371 message. Instead it uses the arrival time. This makes for better
1372 behaviour in cases where some deliveries succeed, thus re-setting the
1373 "first failed" field. An example is a quota failure for a huge message
1374 when small messages continue to be delivered. Without this change, the
1375 "time since failure" will always be short, possible causing more frequent
1376 delivery attempts for the huge message than are intended.
1377 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1379 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1380 $1, $2, $3) because the numerical variables can be reset during some
1381 expansion items (e.g. "match"), thereby losing the authentication data.
1383 PH/21 Make -bV show the size of off_t variables so that the test suite can
1384 decide whether to run tests for quotas > 2G.
1386 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1387 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1388 filecount value is greater than 2G or if a quota value is greater than 2G
1389 on a system where the size of off_t is not greater than 4, a panic error
1392 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1393 never match. The debug and -bh output now contains an explicit error
1394 message indicating a malformed IPv4 address or mask.
1396 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1397 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1398 PH/23 above applies.
1400 PH/25 Do not write to syslog when running in the test harness. The only
1401 occasion when this arises is a failure to open the main or panic logs
1402 (for which there is an explicit test).
1404 PH/26 Added the /no_tell option to "control=freeze".
1406 PH/27 If a host name lookup failed very early in a connection, for example, if
1407 the IP address matched host_lookup and the reverse lookup yielded a name
1408 that did not have a forward lookup, an error message of the form "no IP
1409 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1410 could be logged. Now it outputs the IP address instead of "NULL".
1412 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1413 allows the sender and the authenticated sender to be set when
1414 submitting a message from within Exim. Since child_open_exim() is
1415 documented for local_scan(), the new function should be too.
1417 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1418 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1419 results in an empty string is now treated as unset.
1421 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1423 PH/31 Added sender_verify_fail logging option.
1425 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1426 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1427 tidied the source and removed it altogether.
1429 PH/33 When a queue run was abandoned because the load average was too high, a
1430 log line was always written; now it is written only if the queue_run log
1431 selector is set. In addition, the log line for abandonment now contains
1432 information about the queue run such as the pid. This is always present
1433 in "start" and "stop" lines but was omitted from the "abandon" line.
1435 PH/34 Omit spaces between a header name and the colon in the error message that
1436 is given when verify = headers_syntax fails (if there are lots of them,
1437 the message gets confusing).
1439 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1440 names, as there are now some PTR records that contain slashes. This check
1441 is only to protect against broken name servers that fall over on strange
1442 characters, so the fact that it applies to all lookups doesn't matter.
1444 PH/36 Now that the new test suite is complete, we can remove some of the
1445 special code in Exim that was needed for the old test suite. For example,
1446 sorting DNS records because real resolvers return them in an arbitrary
1447 order. The new test suite's fake resolver always returns records in the
1450 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1451 bounces) except when queue_only is set, to avoid logging races between
1452 the different processes.
1454 PH/38 Panic-die if .include specifies a non-absolute path.
1456 PH/39 A tweak to the "H" retry rule from its user.
1458 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1459 a label. They prevented compilation on older perls.
1461 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1462 a warning to be raised on newish perls.
1464 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1465 on queue. Changes to match documented behaviour of showing count of
1466 messages matching specified criteria.
1468 PH/40 Changed the default ident timeout from 30s to 5s.
1470 PH/41 Added support for the use of login_cap features, on those BSD systems
1471 that have them, for controlling the resources used by pipe deliveries.
1473 PH/42 The content-scanning code uses fopen() to create files in which to put
1474 message data. Previously it was not paying any attention to the mode of
1475 the files. Exim runs with umask(0) because the rest of the code creates
1476 files with open(), and sets the required mode explicitly. Thus, these
1477 files were ending up world-writeable. This was not a big issue, because,
1478 being within the spool directory, they were not world-accessible. I have
1479 created a function called modefopen, which takes an additional mode
1480 argument. It sets umask(777), creates the file, chmods it to the required
1481 mode, then resets the umask. All the relevant calls to fopen() in the
1482 content scanning code have been changed to use this function.
1484 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1485 to 24 hours. This avoids potential overflow problems when processing G
1486 and H retry rules. I suspect nobody ever tinkers with this value.
1488 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1490 PH/45 When the plaintext authenticator is running as a client, the server's
1491 challenges are checked to ensure they are valid base64 strings. By
1492 default, the authentication attempt is cancelled if an invalid string is
1493 received. Setting client_ignore_invalid_base64 true ignores these errors.
1494 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1495 they are received. Thus, the responses can be made to depend on the
1496 challenges. If an invalid string is ignored, an empty string is placed in
1499 PH/46 Messages that are created by the autoreply transport now contains a
1500 References: header, in accordance with RFCs 2822 and 3834.
1502 PH/47 Added authenticated_sender_force to the smtp transport.
1504 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1506 PH/49 Installed latest patch from the Sieve maintainer.
1508 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1509 was also set, the appendfile transport was unnecessarily scanning a
1510 directory of message files (e.g. for maildir delivery) to find the count
1511 of files (along with the size), even though it did not need this
1512 information. It now does the scan only if it needs to find either the
1513 size of the count of files.
1515 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1517 PH/52 Two bugs concerned with error handling when the smtp transport is
1520 (i) Exim was not creating retry information for temporary errors given
1521 for individual recipients after the DATA command when the smtp transport
1522 was used in LMTP mode. This meant that they could be retried too
1523 frequently, and not timed out correctly.
1525 (ii) Exim was setting the flag that allows error details to be returned
1526 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1527 recipients that were returned after the DATA command.
1529 PH/53 This is related to PH/52, but is more general: for any failing address,
1530 when detailed error information was permitted to be returned to the
1531 sender, but the error was temporary, then after the final timeout, only
1532 "retry timeout exceeded" was returned. Now it returns the full error as
1533 well as "retry timeout exceeded".
1535 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1536 do this, and (what is worse) MTAs that accept it.
1538 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1539 will now be deprecated.
1541 PH/56 New os.c-cygwin from the Cygwin maintainer.
1543 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1544 formats (previously only available in exim formats via -bpr, -bpru,
1545 and -bpra. Now also available in native and exiqgrep formats)
1547 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1548 with very large, slow to parse queues
1550 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1552 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1554 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1555 responses to authentication challenges, though it was showing the
1556 challenges; (ii) I've removed the CR characters from the debug output for
1559 PH/58 Allow for the insertion of a newline as well as a space when a string
1560 is turned into more than one encoded-word during RFC 2047 encoding. The
1561 Sieve code now uses this.
1563 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1564 data_4xx, lost_connection, tls_required.
1566 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1567 message was being sent as an SMTP response.
1569 PH/61 Add -l and -k options to exicyclog.
1571 PH/62 When verifying, if an address was redirected to one new address, so that
1572 verification continued, and the new address failed or deferred after
1573 having set something in $address_data, the value of $address_data was not
1574 passed back to the ACL. This was different to the case when no
1575 redirection occurred. The value is now passed back in both cases.
1577 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1578 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1579 use login_cap.h, so on its own it isn't the distinguishing feature. The
1580 new name refers directly to the setclassresources() function.
1582 PH/65 Added configuration files for NetBSD3.
1584 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1586 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1587 is preferred over IPv4.
1589 PH/68 The bounce_return_message and bounce_return_body options were not being
1590 honoured for bounces generated during the reception of non-SMTP messages.
1591 In particular, this applied to messages rejected by the ACL. This bug has
1592 been fixed. However, if bounce_return_message is true and bounce_return_
1593 body is false, the headers that are returned for a non-SMTP message
1594 include only those that have been read before the error was detected.
1595 (In the case of an ACL rejection, they have all been read.)
1597 PH/69 The HTML version of the specification is now built in a directory called
1598 spec_html instead of spec.html, because the latter looks like a path with
1599 a MIME-type, and this confuses some software.
1601 PH/70 Catch two compiler warnings in sieve.c.
1603 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1604 function verify_get_ident() calls ip_connect() to connect a socket, but
1605 if the "connect()" function timed out, ip_connect() used to close the
1606 socket. However, verify_get_ident() also closes the socket later, and in
1607 between Exim writes to the log, which may get opened at this point. When
1608 the socket was closed in ip_connect(), the log could get the same file
1609 descriptor number as the socket. This naturally causes chaos. The fix is
1610 not to close the socket in ip_connect(); the socket should be closed by
1611 the function that creates it. There was only one place in the code where
1612 this was missing, in the iplookup router, which I don't think anybody now
1613 uses, but I've fixed it anyway.
1615 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1616 well as to direct DNS lookups. Otherwise the handling of names in host
1617 lists is inconsistent and therefore confusing.
1623 PH/01 Two changes to the default runtime configuration:
1625 (1) Move the checks for relay_from_hosts and authenticated clients from
1626 after to before the (commented out) DNS black list checks.
1628 (2) Add control=submission to the relay_from_hosts and authenticated
1629 clients checks, on the grounds that messages accepted by these
1630 statements are most likely to be submissions.
1632 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1634 (1) Generate an error if the third argument for the ${prvs expansion is
1637 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1640 (3) Reset the variables that are obtained from the first argument of
1641 ${prvscheck and used in the second argument before leaving the code,
1642 because their memory is reclaimed, so using them afterwards may do
1645 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1646 one (it's much easier than Tom thought :-).
1648 (5) Because of (4), we can now allow for the use of $prvscheck_result
1649 inside the third argument.
1651 PH/03 For some reason, the default setting of PATH when running a command from
1652 a pipe transport was just "/usr/bin". I have changed it to
1655 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1656 anything to be listed in the output from -bV.
1658 PH/05 When a filter generated an autoreply, the entire To: header line was
1659 quoted in the delivery log line, like this:
1661 => >A.N.Other <ano@some.domain> <original@ddress> ...
1663 This has been changed so that it extracts the operative address. There
1664 may be more than one such address. If so, they are comma-separated, like
1667 => >ano@some.domain,ona@other.domain <original@ddress> ...
1669 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1670 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1671 not being looked up in the rDNS to get a host name, Exim was showing the
1672 IP address twice in Received: lines, even though the IP addresses were
1673 identical. For example:
1675 Received: from [1.2.3.4] (helo=[1.2.3.4])
1677 However, if the real host name was known, it was omitting the HELO data
1678 if it matched the actual IP address. This has been tidied up so that it
1679 doesn't show the same IP address twice.
1681 PH/07 When both +timestamp and +memory debugging was on, the value given by
1682 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1683 called by the debug printing, thereby overwriting the timestamp buffer.
1684 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1686 PH/08 When the original message was included in an autoreply transport, it
1687 always said "this is a copy of the message, including all the headers",
1688 even if body_only or headers_only was set. It now gives an appropriate
1691 PH/09 Applied a patch from the Sieve maintainer which:
1693 o fixes some comments
1694 o adds the (disabled) notify extension core
1695 o adds some debug output for the result of if/elsif tests
1696 o points to the current vacation draft in the documentation
1697 and documents the missing references header update
1701 o fixes a bug in processing the envelope test (when testing
1702 multiple envelope elements, the last element determinted the
1705 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1706 Electronic Mail") by including:
1708 Auto-submitted: auto-generated
1710 in the messages that it generates (bounce messages and others, such as
1711 warnings). In the case of bounce messages for non-SMTP mesages, there was
1712 also a typo: it was using "Auto_submitted" (underscore instead of
1713 hyphen). Since every message generated by Exim is necessarily in response
1714 to another message, thes have all been changed to:
1716 Auto-Submitted: auto-replied
1718 in accordance with these statements in the RFC:
1720 The auto-replied keyword:
1722 - SHOULD be used on messages sent in direct response to another
1723 message by an automatic process,
1725 - MUST NOT be used on manually-generated messages,
1727 - MAY be used on Delivery Status Notifications (DSNs) and Message
1728 Disposition Notifications (MDNs),
1730 - MUST NOT be used on messages generated by automatic or periodic
1731 processes, except for messages which are automatic responses to
1734 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1735 to the default Received: header definition.
1737 PH/12 Added log selector acl_warn_skipped (default on).
1739 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1740 variables because (a) they are in the wrong storage pool and (b) even if
1741 they were copied, it wouldn't work properly because of the caching.
1743 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1744 checking when decoding. Apparently there are clients that generate
1745 overlong encoded strings. Why am I not surprised?
1747 PH/15 If the first argument of "${if match_address" was not empty, but did not
1748 contain an "@" character, Exim crashed. Now it writes a panic log message
1749 and treats the condition as false.
1751 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1753 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1754 extension "envelope-auth". The code is finished and in agreement with
1755 other implementations, but there is no documentation so far and in fact,
1756 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1757 not changing the active code.
1759 Print executed "if" and "elsif" statements when debugging is used. This
1760 helps a great deal to understand what a filter does.
1762 Document more things not specified clearly in RFC3028. I had all this
1763 sorted out, when out of a sudden new issues came to my mind. Oops."
1765 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1768 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1769 canonical form (as documented). However, after a host name lookup from
1770 the IP address, check_host() was doing a simple string comparison with
1771 addresses acquired from the DNS when checking that the found name did
1772 have the original IP as one of its addresses. Since any found IPv6
1773 addresses are likely to be in abbreviated form, the comparison could
1774 fail. Luckily, there already exists a function for doing the comparison
1775 by converting both addresses to binary, so now that is used instead of
1776 the text comparison.
1778 PH/20 There was another similar case to PH/19, when a complete host name was
1779 given in a host list; looking up its IP address could give an abbreviated
1780 form, whereas the current host's name might or might not be abbreviated.
1781 The same fix has been applied.
1787 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1788 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1791 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1792 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1794 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1796 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1797 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1798 Also, for "Auto-Submitted", treat anything other than "no" as
1799 non-personal, in accordance with RFC 3834. (Previously it treated
1800 anything starting "auto-" as non-personal.)
1802 TF/01 The control=submission/name=... option had a problem with syntax
1803 errors if the name included a slash character. The /name= option
1804 now slurps the rest of the string, so it can include any characters
1805 but it must come last in the list of options (after /sender_retain
1808 PH/05 Some modifications to the interface to the fake nameserver for the new
1816 TK/01 Added the "success_on_redirect" address verification option. See
1817 NewStuff for rationale and an example.
1819 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1821 PH/02 Patch to exigrep to allow it to work on syslog lines.
1823 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1824 fread() to skip over the body file's header line, because in Cygwin the
1825 header line is locked and is inaccessible.
1827 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1828 co-exist for some time) to make it clear that it is the Exim ID that is
1829 referenced, not the Message-ID: header line.
1831 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1832 string_format() function, because snprintf() does not exist on all
1835 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1836 ${stat: expansion item.
1838 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1839 protocol synchronization error", to keep the pedants happy.
1841 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1842 well as for IRIX systems, when gcc is being used. See the host.c source
1845 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1847 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1850 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1851 transport and to the smtp transport in LMTP mode.
1853 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1855 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1856 run a filter in a subprocess. This could lead to confusion in subsequent
1857 lookups in the parent process. There should also be a search_tidyup() at
1858 the end of the subprocess.
1860 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1861 only if the host matched helo_try_verify_hosts, which caused the
1862 verification to occur when the EHLO/HELO command was issued. The ACL just
1863 tested the remembered result. Now, if a previous verification attempt has
1864 not happened, "verify = helo" does it there and then.
1866 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1868 TK/03 Fix log output including CR from clamd.
1870 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1871 exist provoked a memory error which could cause a segfault.
1873 PH/15 Installed PCRE 6.2
1875 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1877 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1878 of the problem. Specifically, suggested +O2 rather than +O1 for the
1881 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1883 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1884 record, the port was not correctly logged when the outgoing_port log
1885 selector was set (it logged the transort's default port).
1887 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1888 fallback_hosts, and "hosts" in the smtp transport.
1890 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1891 host errors such as "Connection refused".
1893 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1894 authentication with radiusclient 0.4.9:
1896 - Error returned from rc_read_config was caught wrongly
1897 - Username/password not passed on to radius server due to wrong length.
1899 The presumption is that some radiusclient API changes for 4.51/PH/17
1900 were not taken care of correctly. The code is still untested by me (my
1901 Linux distribution still has 0.3.2 of radiusclient), but it was
1902 contributed by a Radius user.
1904 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1905 expanding the "port" option of the smtp transport.
1907 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1908 while reading a MIME header. Thanks to Tom Hughes for a patch.
1910 PH/24 Include config.h inside local_scan.h so that configuration settings are
1913 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1914 that in an ACL for RCPT (for example), you can examine exactly what was
1917 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1918 commands, but it was not correctly comparing the address with the actual
1919 client host address. Thus, it would show the EHLO address in Received:
1920 header lines when this was not necessary.
1922 PH/27 Added the % operator to ${eval:}.
1924 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1925 it should be ignoring failures (because with -C, for example, it has lost
1926 privilege). It wasn't ignoring creation failures other than "already
1929 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1932 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1933 because an input line was too long, either on its own, or by virtue of
1934 too many continuations, the temporary file was not being removed, and the
1935 return code was incorrect.
1937 PH/31 Missing "BOOL" in function definition in filtertest.c.
1939 PH/32 Applied Sieve patches from the maintainer.
1941 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1943 PH/33 Added "verify = not_blind".
1945 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1946 Local/Makefile (with some defaults set). These are used in built scripts
1947 such as exicyclog, but they have never been used in the exim_install
1948 script (though there are many overriding facilities there). I have
1949 arranged that the exim_install script now takes note of these two
1952 PH/35 Installed configuration files for Dragonfly.
1954 PH/36 When a locally submitted message by a trusted user did not contain a
1955 From: header, and the sender address was obtained from -f or from an SMTP
1956 MAIL command, and the trusted user did not use -F to supply a sender
1957 name, $originator_name was incorrectly used when constructing a From:
1958 header. Furthermore, $originator_name was used for submission mode
1959 messages from external hosts without From: headers in a similar way,
1960 which is clearly wrong.
1962 PH/37 Added control=suppress_local_fixups.
1964 PH/38 When log_selector = +received_sender was set, and the addition of the
1965 sender made the log line's construction buffer exactly full, or one byte
1966 less than full, an overflow happened when the terminating "\n" was
1969 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1970 when the result of a list match is failure because a DNS lookup failed.
1972 PH/40 RM_COMMAND is now used in the building process.
1974 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1975 the "build-* directories that it finds.
1977 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1978 address in a domain literal was a prefix of an interface address.
1980 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1981 when verifying a sender address, unless rewrite_headers is false.
1983 PH/44 Wrote a long comment about why errors_to addresses are verified as
1984 recipients, not senders.
1986 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1987 the ratelimit ACL was added.
1989 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1991 PH/46 Added extra information about PostgreSQL errors to the error string.
1993 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1994 avoiding the need to install special zones in a real server. This is
1995 backwards compatible; if it can't find the fake resolver, it drops back.
1996 Thus, both old and new test suites can be run.
1998 TF/02 Added util/ratelimit.pl
2000 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2001 clock is set back in time.
2003 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2004 Candler <B.Candler@pobox.com>.
2006 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2007 OK for addresses that are the result of redirections.
2009 PH/48 A number of further additions for the benefit of the new test suite,
2010 including a fake gethostbyname() that interfaces to the fake DNS resolver
2013 TF/06 The fix for widen_domains has also been applied to qualify_single and
2014 search_parents which are the other dnslookup options that can cause
2017 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2020 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2022 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2023 service" if it encountered an MX record with an empty target hostname.
2024 The message is now "an MX or SRV record indicated no SMTP service".
2026 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2027 if the DNS of the sending site is misconfigured. This is quite a
2028 common situation. This change restores the behaviour of treating a
2029 helo verification defer as a failure.
2031 PH/51 If self=fail was set on a router, the bounce message did not include the
2032 actual error message.
2038 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2040 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2041 command itself ended in error, the underlying message about the transport
2042 filter timeout was being overwritten with the pipe command error. Now the
2043 underlying error message should be appended to the second error message.
2045 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2047 PH/02 Reduce the amount of output that "make" produces by default. Full output
2048 can still be requested.
2050 PH/03 The warning log line about a condition test deferring for a "warn" verb
2051 was being output only once per connection, rather than after each
2052 occurrence (because it was using the same function as for successful
2053 "warn" verbs). This seems wrong, so I have changed it.
2055 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2056 it should not have, which might have caused a crash in the right
2057 circumstances, but probably never did.
2059 PH/04 Installed a modified version of Tony Finch's patch to make submission
2060 mode fix the return path as well as the Sender: header line, and to
2061 add a /name= option so that you can make the user's friendly name appear
2064 TF/03 Added the control = fakedefer ACL modifier.
2066 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2067 Mark Lowes for thorough testing.
2069 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2071 TK/03 Merged latest SRS patch from Miles Wilton.
2073 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2074 with the definition in sysexits.h (which is #included earlier).
2075 Fortunately, Exim does not actually use EX_OK. The code used to try to
2076 preserve the sysexits.h value, by assumimg that macro definitions were
2077 scanned for macro replacements. I have been disabused of this notion,
2078 so now the code just undefines EX_OK before #including unistd.h.
2080 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2081 in the smtp transport. When a block could not be written in a single
2082 write() function, the timeout was being re-applied to each part-write.
2083 This seems wrong - if the receiver was accepting one byte at a time it
2084 would take for ever. The timeout is now adjusted when this happens. It
2085 doesn't have to be particularly precise.
2087 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2088 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2090 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2091 without a domain if the check to <postmaster@domain> fails.
2093 SC/01 Eximstats: added -xls and the ability to specify output files
2094 (patch written by Frank Heydlauf).
2096 SC/02 Eximstats: use FileHandles for outputing results.
2098 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2100 SC/04 Eximstats: fixed display of large numbers with -nvr option
2102 SC/05 Eximstats: fixed merging of reports with empty tables.
2104 SC/06 Eximstats: added the -include_original_destination flag
2106 SC/07 Eximstats: removed tabs and trailing whitespace.
2108 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2110 TK/06 MBOX spool code: Add real "From " MBOX separator line
2111 so the .eml file is really in mbox format (even though
2112 most programs do not really care). Patch from Alex Miller.
2114 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2115 The latter is generated from $received_to and is only set if the
2116 message has one envelope recipient. SA can use these headers,
2117 obviously out-of-the-box. Patch from Alex Miller.
2119 PH/08 The ${def test on a variable was returning false if the variable's
2120 value was "0", contrary to what the specification has always said!
2121 The result should be true unless the variable is empty.
2123 PH/09 The syntax error of a character other than { following "${if
2124 def:variable_name" (after optional whitespace) was not being diagnosed.
2125 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2126 accidental colon was present, for example, could give incorrect results.
2128 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2129 result is used (not including appendfile, where other changes are about
2132 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2133 This involved changing a lot of size variables from int to off_t. It
2134 should work with maildirs and everything.
2136 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2137 spamd dying while we are connected to it.
2139 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2142 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2143 function for BDB 4.3.
2145 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2147 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2148 the sources for PCRE have been re-arranged and more files are now
2151 PH/15 The code I had for printing potentially long long variables in PH/11
2152 above was not the best (it lost precision). The length of off_t variables
2153 is now inspected at build time, and an appropriate printing format (%ld
2154 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2155 to be "long long int" or "long int". This is needed for the internal
2156 formatting function string_vformat().
2158 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2159 the configuration file to be ":syslog", then the script "guesses" where
2160 the logs files are, rather than using the compiled in default. In our
2161 case the guess is not the same as the compiled default, so the script
2162 suddenly stopped working when I started to use syslog. The patch checks
2163 to see if log_file_path is "". If so, it attempts to read it from exim
2164 with no configuration file to get the compiled in version, before it
2165 falls back to the previous guessing code."
2167 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2168 implementing BATV in an Exim configuration. See NewStuff for the gory
2171 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2172 Makefile that are specific to HP-UX.
2174 PH/18 If the "use_postmaster" option was set for a recipient callout together
2175 with the "random" option, the postmaster address was used as the MAIL
2176 FROM address for the random test, but not for the subsequent recipient
2177 test. It is now used for both.
2179 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2180 patch removes a few documentation additions to RFC 3028, because the
2181 latest draft now contains them. It adds the new en;ascii-case comparator
2182 and a new error check for 8bit text in MIME parts. Comparator and
2183 require names are now matched exactly. I enabled the subaddress
2184 extension, but it is not well tested yet (read: it works for me)."
2186 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2187 rework some of the code of TK/09 above to avoid the hardwired use of
2188 "%lld" and "long long". Replaced the call to snprintf() with a call to
2191 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2192 records point to non-existent hosts", "retry timeout exceeded", and
2193 "retry time not reached for any host after a long failure period".
2195 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2196 experimental DomainKeys support:
2198 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2199 (2) On an error such as an illegally used "control", the wrong name for
2200 the control was given.
2202 These problems did NOT occur unless DomainKeys support was compiled.
2204 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2206 PH/24 Added ${if match_ip condition.
2208 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2209 databases so that it will be absolutely obvious if a crash occurs in the
2210 DB library. This is a regular occurrence (often caused by mis-matched
2213 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2214 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2215 picked up on a user's system that detects such things. There doesn't seem
2216 to be a gcc warning option for this - only an attribute that has to be
2217 put on the function's prototype. It seems that in Fedora Core 4 they have
2218 set this on a number of new functions. No doubt there will be more in due
2221 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2222 specify a transport. However, if an address that was verified by such a
2223 router was the subject of a callout, Exim crashed because it tried to
2224 read the rcpt_include_affixes from the non-existent transport. Now it
2225 just assumes that the setting of that option is false. This bug was
2226 introduced by 4.51/PH/31.
2228 PH/28 Changed -d+all to exclude +memory, because that information is very
2229 rarely of interest, but it makes the output a lot bigger. People tend to
2230 do -d+all out of habit.
2232 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2233 code in os-type was giving problems when libc.so lives in lib64, like on
2236 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2237 aren't the modern standard, and it seems that some systems' include files
2238 don't always have them. Exim was already checking for some of the newer
2239 ones like T_AAAA, and defining it itself. I've added checks for all the
2240 record types that Exim uses.
2242 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2243 not automatically generating a new one, as it is supposed to. This
2244 prevented TLS from working. If the file did exist, but contained invalid
2245 data, a new version was generated, as expected. It was only the case of a
2246 non-existent file that was broken.
2248 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2249 with a change in libdomainkeys > 0.64.
2251 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2252 from DNS. If the selector record carries the flag, it now has
2253 precedence over the domain-wide flag.
2255 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2257 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2258 the use of a port name that isn't defined in /etc/services) occurred, the
2259 message was deferred as in a normal delivery, and thus remained on the
2260 spool, instead of being failed because of the mua_wrapper setting. This
2261 is now fixed, and I tidied up some of the mua_wrapper messages at the
2264 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2265 the messages in a hash of arrays rather than using individual hashes.
2266 This is a bit cleaner and results in dramatic memory savings, albeit
2267 at a slight CPU cost.
2269 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2270 as requested by Marc Sherman.
2272 SC/10 Eximstats: added histograms for user specified patterns as requested
2275 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2277 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2278 fopen() in the content-scanning modules that did not already have it.
2284 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2285 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2287 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2289 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2291 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2292 to test Sieve filters that use "vacation".
2294 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2295 that changes the way the GnuTLS parameters are stored in the cache file.
2296 The new format can be generated externally. For backward compatibility,
2297 if the data in the cache doesn't make sense, Exim assumes it has read an
2298 old-format file, and it generates new data and writes a new file. This
2299 means that you can't go back to an older release without removing the
2302 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2303 work if there are any delivery delays because "one_time" forces the
2304 parent to be marked "delivered", so its unseen clone is never tried
2305 again. For this reason, Exim now forbids the simultaneous setting of
2308 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2309 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2310 entry if you want to know the details. The fix, however, neglected to
2311 consider the case when local delivery batching is involved. The test for
2312 "previously delivered" was not happening when checking to see if an
2313 address could be batched with a previous (undelivered) one; under
2314 certain circumstances this could lead to multiple deliveries to the same
2317 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2318 in its include files, and this causes problems building Exim.
2320 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2321 header_syntax) but Exim was just ignoring anything given after a slash.
2322 In particular, this caused confusion with an attempt to use "verify =
2323 reverse_host_lookup/defer_ok". An error is now given when options are
2324 supplied for verify items that do not have them. (Maybe reverse_host_
2325 lookup should have a defer_ok option, but that's a different point.)
2327 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2328 defined by RFC 821) to 2048, because there were problems with some AUTH
2329 commands, and RFC 1869 says the size should be increased for extended
2330 SMTP commands that take arguments.
2332 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2335 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2336 "unknown" error; now it says that the functionality isn't in the binary.
2338 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2339 an address' error message when a string expansion fails (syntax or
2340 whatever). Otherwise the password may appear in the log. Following change
2341 PH/42 below, there is no longer a chance of it appearing in a bounce
2344 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2346 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2347 of its addresses was ever tried. (Bugzilla bug #2.)
2349 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2350 the result incorrectly in the debug output. (It correctly added a newline
2351 to what was transported.)
2353 TF/01 Added $received_time.
2355 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2356 commented out examples of how to interface to a virus scanner and to
2357 SpamAssassin. Also added commented examples of av_scanner and
2358 spamd_address settings.
2360 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2361 and controls are allowed in which ACLs. There were a couple of minor
2362 errors. Some of the entries in the conditions table (which is a table of
2363 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2364 negation of where the condition IS allowed.
2366 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2368 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2369 header file does not have a version number, so I've had to invent a new
2370 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2371 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2372 radiusclient), but it was contributed by a Radius user.
2374 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2375 files or directories, for OpenSSL.
2377 PH/19 When an Exim process that is running as root has to create an Exim log
2378 file, it does so in a subprocess that runs as exim:exim so as to get the
2379 ownership right at creation (otherwise, other Exim processes might see
2380 the file with the wrong ownership). There was no test for failure of this
2381 fork() call, which would lead to the process getting stuck as it waited
2382 for a non-existent subprocess. Forks do occasionally fail when resources
2383 run out. I reviewed all the other calls to fork(); they all seem to check
2386 PH/20 When checking for unexpected SMTP input at connect time (before writing
2387 the banner), Exim was not dealing correctly with a non-positive return
2388 from the read() function. If the client had disconnected by this time,
2389 the result was a log entry for a synchronization error with an empty
2390 string after "input=" when read() returned zero. If read() returned -1
2391 (an event I could not check), uninitialized data bytes were printed.
2392 There were reports of junk text (parts of files, etc) appearing after
2395 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2397 PH/22 Added support for macro redefinition, and (re)definition in between
2398 driver and ACL definitions.
2400 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2401 forgetting to use the resulting value; it was using the unexpanded value.
2403 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2404 hadn't been configured. The fix is from Juergen Kreileder, who
2405 understands it better than I do:
2407 "Here's what I see happening with three configured cyrus_sasl
2408 authenticators configured (plain, login, cram-md5):
2410 On startup auth_cyrus_sasl_init() gets called for each of these.
2411 This means three calls to sasl_listmech() without a specified mech_list.
2412 => SASL tests which mechs of all available mechs actually work
2413 => three warnings about OTP not working
2414 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2416 With the patch, sasl_listmech() also gets called three times. But now
2417 SASL's mech_list option is set to the server_mech specified in the the
2418 authenticator. Or in other words, the answer from sasl_listmech()
2419 gets limited to just the mech you're testing for (which is different
2421 => the return list contains just 'plain' or 'login', 'cram-md5' or
2422 nothing depending on the value of ob->server_mech.
2424 I've just tested the patch: Authentication still works fine,
2425 unavailable mechs specified in the exim configuration are still
2426 caught, and the auth.log warnings about OTP are gone."
2428 PH/25 When debugging is enabled, the contents of the command line are added
2429 to the debugging output, even when log_selector=+arguments is not
2432 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2433 answer is "GNU", and only if the return is "GNU/something" is the answer
2436 PH/27 $acl_verify_message is now set immediately after the failure of a
2437 verification in an ACL, and so is available in subsequent modifiers. In
2438 particular, the message can be preserved by coding like this:
2440 warn !verify = sender
2441 set acl_m0 = $acl_verify_message
2443 Previously, $acl_verify_message was set only while expanding "message"
2444 and "log_message" when a very denied access.
2446 PH/28 Modified OS/os.c-Linux with
2448 -#ifndef OS_LOAD_AVERAGE
2449 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2451 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2452 nomenclature these days.)
2454 PH/29 Installed patch from the Sieve maintainer that adds the options
2455 sieve_useraddress and sieve_subaddress to the redirect router.
2457 PH/30 In these circumstances:
2458 . Two addresses routed to the same list of hosts;
2459 . First host does not offer TLS;
2460 . First host accepts first address;
2461 . First host gives temporary error to second address;
2462 . Second host offers TLS and a TLS session is established;
2463 . Second host accepts second address.
2464 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2465 and peerdn, if requested) that were in fact used only for the second
2468 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2469 attention to any local part prefix or suffix that was matched by the
2470 router that accepted the address. It now behaves in the same way as it
2471 does for delivery: the affixes are removed from the local part unless
2472 rcpt_include_affixes is set on the transport.
2474 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2475 timeout during the DATA phase of an incoming message.
2477 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2478 applied a patch sanctioned by the Sieve maintainer.
2480 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2481 the uid or gid is negative. A case of a negative gid caused this to be
2482 noticed. The fix allows for either to be negative.
2484 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2485 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2488 PH/36 The change PH/12 above was broken. Fixed it.
2490 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2491 the grounds that routing the same address twice would always produce the
2492 same answer. This might have been true once, but it is certainly no
2493 longer true now. Routing a child address may depend on the previous
2494 routing that produced that child. Some complicated redirection strategies
2495 went wrong when messages had multiple recipients, and made Exim's
2496 behaviour dependent on the order in which the addresses were given.
2498 I have moved the duplicate checking until after the routing is complete.
2499 Exim scans the addresses that are assigned to local and remote
2500 transports, and removes any duplicates. This means that more work will be
2501 done, as duplicates will always all be routed, but duplicates are
2502 presumably rare, so I don't expect this is of any significance.
2504 For deliveries to pipes, files, and autoreplies, the duplicate checking
2505 still happens during the routing process, since they are not going to be
2508 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2509 It corrects a timeout issue with spamd. This is Ian's comment: "The
2510 background is that sometimes spamd either never reads data from a
2511 connection it has accepted, or it never writes response data. The exiscan
2512 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2513 blindly assumes that writes won't block so it may never time out."
2515 PH/39 Allow G after quota size as well as K and M.
2517 PH/40 The value set for $authenticated_id in an authenticator may not contain
2518 binary zeroes or newlines because the value is written to log lines and
2519 to spool files. There was no check on this. Now the value is run through
2520 the string_printing() function so that such characters are converted to
2521 printable escape sequences.
2523 PH/41 $message_linecount is a new variable that contains the total number of
2524 lines in the message. Compare $body_linecount, which is the count for the
2527 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2528 bounce and delay warning messages, except in certain special cases, which
2531 (a) An SMTP error message from a remote host;
2532 (b) A message specified in a :fail: redirection;
2533 (c) A message specified in a "fail" command in a system filter;
2534 (d) A message specified in a FAIL return from the queryprogram router;
2535 (e) A message specified by the cannot_route_message router option.
2537 In these cases only, Exim does include the error details in bounce and
2538 warning messages. There are also a few cases where bland messages such
2539 as "unrouteable address" or "local delivery error" are given.
2541 PH/43 $value is now also set for the "else" part of a ${run expansion.
2543 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2544 being worked on, but at least Exim now implements the latest version to
2547 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2548 process to complete was treated as a delivery failure, a timeout while
2549 writing the message to the pipe was logged, but erroneously treated as a
2550 successful delivery. Such timeouts include transport filter timeouts. For
2551 consistency with the overall process timeout, these timeouts are now
2552 treated as errors, giving rise to delivery failures by default. However,
2553 there is now a new Boolean option for the pipe transport called
2554 timeout_defer, which, if set TRUE, converts the failures into defers for
2555 both kinds of timeout. A transport filter timeout is now identified in
2558 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2559 systems where "make" and "gmake" are different, calling "gmake" at top
2560 level broke things. I've arranged for the value of $(MAKE) to be passed
2561 from the Makefile to this script so that it can call the same version of
2565 A note about Exim versions 4.44 and 4.50
2566 ----------------------------------------
2568 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2569 changes of various kinds. As a consequence, a big documentation update was
2570 needed. This delayed the release for rather longer than seemed good, especially
2571 in the light of a couple of (minor) security issues. Therefore, the changes
2572 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2573 release. So 4.44 and 4.50 are in effect two different branches that both start
2576 I have left the 4.50 change log unchanged; it contains all the changes since
2577 4.43. The change log for 4.44 is below; many of its items are identical to
2578 those for 4.50. This seems to be the most sensible way to preserve the
2579 historical information.
2585 1. Minor wording change to the doc/README.SIEVE file.
2587 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2588 computation of the current number of files was incorrect.
2590 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2591 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2594 4. Give more explanation in the error message when the command for a transport
2595 filter fails to execute.
2597 5. There are several places where Exim runs a non-Exim command in a
2598 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2599 was being done only for the command run by the queryprogram router. It is
2600 now done for all such subprocesses. The other cases are: ${run, transport
2601 filters, and the commands run by the lmtp and pipe transports.
2603 6. Added CONFIGURE_GROUP build-time option.
2605 7. Some older OS have a limit of 256 on the maximum number of file
2606 descriptors. Exim was using setrlimit() to set 1000 as a large value
2607 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2608 systems. I've change it so that if it can't get 1000, it tries for 256.
2610 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2611 was an oversight, and furthermore, ever since the addition of extra
2612 controls (e.g. 4.43/32), the checks on when to allow different forms of
2613 "control" were broken. There should now be diagnostics for all cases when a
2614 control that does not make sense is encountered.
2616 9. Added the /retain_sender option to "control=submission".
2618 10. $recipients is now available in the predata ACL (oversight).
2620 11. Tidy the search cache before the fork to do a delivery from a message
2621 received from the command line. Otherwise the child will trigger a lookup
2622 failure and thereby defer the delivery if it tries to use (for example) a
2623 cached ldap connection that the parent has called unbind on.
2625 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2626 of $address_data from the recipient verification was clobbered by the
2627 sender verification.
2629 13. The value of address_data from a sender verification is now available in
2630 $sender_address_data in subsequent conditions in the ACL statement.
2632 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2634 15. Added a new option "connect=<time>" to callout options, to set a different
2637 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2638 was its contents. (It was OK if the option was not defined at all.)
2640 17. A "Completed" log line is now written for messages that are removed from
2641 the spool by the -Mrm option.
2643 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2644 information about exactly what failed.
2646 19. Added -dd to debug only the daemon process.
2648 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2649 handles timeouts, both on the server side and network timeouts. Renamed the
2650 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2652 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2653 It is now set to "smtps".
2655 22. $host_address is now set to the target address during the checking of
2656 ignore_target_hosts.
2658 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2659 being passed; this would have caused $sender_host_name to have been used if
2660 matching the list had actually called for a host name (not very likely,
2661 since this list is usually IP addresses). A host name is now passed as
2664 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2665 code that specifies a non-restarting handler (typically sigaction() in
2666 modern systems) in an attempt to fix a rare and obscure crash bug.
2668 25. Narrowed the window for a race in the daemon that could cause it to ignore
2669 SIGCHLD signals. This is not a major problem, because they are used only to
2670 wake it up if nothing else does.
2672 26. A malformed maildirsize file could cause Exim to calculate negative values
2673 for the mailbox size or file count. Odd effects could occur as a result.
2674 The maildirsize information is now recalculated if the size or filecount
2677 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2678 support for a long time. Removed HAVE_SYS_VFS_H.
2680 28. Installed the latest version of exipick from John Jetmore.
2682 29. In an address list, if the pattern was not a regular expression, an empty
2683 subject address (from a bounce message) matched only if the pattern was an
2684 empty string. Non-empty patterns were not even tested. This was the wrong
2685 because it is perfectly reasonable to use an empty address as part of a
2686 database query. An empty address is now tested by patterns that are
2687 lookups. However, all the other forms of pattern expect the subject to
2688 contain a local part and a domain, and therefore, for them, an empty
2689 address still always fails if the pattern is not itself empty.
2691 30. Exim went into a mad DNS loop when attempting to do a callout where the
2692 host was specified on an smtp transport, and looking it up yielded more
2693 than one IP address.
2695 31. Re-factored the code for checking spool and log partition space into a
2696 function that finds that data and another that does the check. The former
2697 is then used to implement four new variables: $spool_space, $log_space,
2698 $spool_inodes, and $log_inodes.
2700 32. The RFC2047 encoding function was originally intended for short strings
2701 such as real names; it was not keeping to the 75-character limit for
2702 encoded words that the RFC imposes. It now respects the limit, and
2703 generates multiple encoded words if necessary. To be on the safe side, I
2704 have increased the buffer size for the ${rfc2047: expansion operator from
2707 33. It is now permitted to omit both strings after an "if" condition; if the
2708 condition is true, the result is "true". As before, when the second string
2709 is omitted, a false condition yields an empty string. This makes it less
2710 cumbersome to write custom ACL and router conditions.
2712 34. Failure to deliver a bounce message always caused it to be frozen, even if
2713 there was an errors_to setting on the router. The errors_to setting is now
2716 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2717 canonical form (fully expanded) before being placed in
2718 $sender_host_address.
2720 36. The table in the code that translates DNS record types into text (T_A to
2721 "A" for instance) was missing entries for NS and CNAME. It is just possible
2722 that this could have caused confusion if both these types were looked up
2723 for the same domain, because the text type is used as part of Exim's
2724 per-process caching. But the chance of anyone hitting this buglet seems
2727 37. The dnsdb lookup has been extended in a number of ways.
2729 (1) There is a new type, "zns", which walks up the domain tree until it
2730 finds some nameserver records. It should be used with care.
2732 (2) There is a new type, "mxh", which is like "mx" except that it returns
2733 just the host names, not the priorities.
2735 (3) It is now possible to give a list of domains (or IP addresses) to be
2736 looked up. The behaviour when one of the lookups defers can be
2737 controlled by a keyword.
2739 (4) It is now possible to specify the separator character for use when
2740 multiple records are returned.
2742 38. The dnslists ACL condition has been extended: it is now possible to supply
2743 a list of IP addresses and/or domains to be looked up in a particular DNS
2746 39. Added log_selector=+queue_time_overall.
2748 40. When running the queue in the test harness, wait just a tad after forking a
2749 delivery process, to get repeatability of debugging output.
2751 41. Include certificate and key file names in error message when GnuTLS fails
2752 to set them up, because the GnuTLS error message doesn't include the name
2753 of the failing file when there is a problem reading it.
2755 42. Allow both -bf and -bF in the same test run.
2757 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2759 44. The "Exiscan patch" is now merged into the mainline Exim source.
2761 45. Sometimes the final signoff response after QUIT could fail to get
2762 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2763 < 0 before doing a fflush(). This bug looks as though it goes back to the
2764 introduction of TLS in release 3.20, but "sometimes" must have been rare
2765 because the tests only now provoked it.
2767 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2768 (this can affect the format of dates).
2770 47. exim_tidydb, when checking for the continued existence of a message for
2771 which it has found a message-specific retry record, was not finding
2772 messages that were in split spool directories. Consequently, it was
2773 deleting retry records that should have stayed in existence.
2775 48. Steve fixed some bugs in eximstats.
2777 49. The SPA authentication driver was not abandoning authentication and moving
2778 on to the next authenticator when an expansion was forced to fail,
2779 contradicting the general specification for all authenticators. Instead it
2780 was generating a temporary error. It now behaves as specified.
2782 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2783 (the order specifies the preference for clients). The order is now AES256,
2784 AES128, 3DES, ARCFOUR128.
2786 51. Small patch to Sieve code - explicitly set From: when generating an
2789 52. Exim crashed if a remote delivery caused a very long error message to be
2790 recorded - for instance if somebody sent an entire SpamAssassin report back
2791 as a large number of 550 error lines. This bug was coincidentally fixed by
2792 increasing the size of one of Exim's internal buffers (big_buffer) that
2793 happened as part of the Exiscan merge. However, to be on the safe side, I
2794 have made the code more robust (and fixed the comments that describe what
2797 53. Now that there can be additional text after "Completed" in log lines (if
2798 the queue_time_overall log selector is set), a one-byte patch to exigrep
2799 was needed to allow it to recognize "Completed" as not the last thing in
2802 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2803 patch that reportedly fixes this has been added. I am not expert enough to
2804 create a test for it. This is what the patch creator wrote:
2806 "I found a little strange behaviour of ldap code when working with
2807 Windows 2003 AD Domain, where users was placed in more than one
2808 Organization Units. When I tried to give exim partial DN, the exit code
2809 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2810 But simultaneously result of request was absolutely normal ldap result,
2811 so I produce this patch..."
2813 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2814 REFERENCE, so I have modified the code to exclude the patch when that macro
2817 55. Some experimental protocols are using DNS PTR records for new purposes. The
2818 keys for these records are domain names, not reversed IP addresses. The
2819 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2820 leaves it alone. Component reversal etc. now happens only for IP addresses.
2823 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2825 57. Double the size of the debug message buffer (to 2048) so that more of very
2826 long debug lines gets shown.
2828 58. The exicyclog utility now does better if the number of log files to keep
2829 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2831 59. Two changes related to the smtp_active_hostname option:
2833 (1) $smtp_active_hostname is now available as a variable.
2834 (2) The default for smtp_banner uses $smtp_active_hostname instead
2835 of $primary_hostname.
2837 60. The host_aton() function is supposed to be passed a string that is known
2838 to be a valid IP address. However, in the case of IPv6 addresses, it was
2839 not checking this. This is a hostage to fortune. Exim now panics and dies
2840 if the condition is not met. A case was found where this could be provoked
2841 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2842 components; fortuitously, this particular loophole had already been fixed
2843 by change 4.50/55 above.
2845 If there are any other similar loopholes, the new check in host_aton()
2846 itself should stop them being exploited. The report I received stated that
2847 data on the command line could provoke the exploit when Exim was running as
2848 exim, but did not say which command line option was involved. All I could
2849 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2850 running as the user.
2853 61. There was a buffer overflow vulnerability in the SPA authentication code
2854 (which came originally from the Samba project). I have added a test to the
2855 spa_base64_to_bits() function which I hope fixes it.
2858 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2859 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2861 63. The daemon start-up calls getloadavg() while still root for those OS that
2862 need the first call to be done as root, but it missed one case: when
2863 deliver_queue_load_max is set with deliver_drop_privilege. This is
2864 necessary for the benefit of the queue runner, because there is no re-exec
2865 when deliver_drop_privilege is set.
2867 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2868 This has been fixed.
2870 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2871 was in use, was not putting the data itself into the right store pool;
2872 consequently, it could be overwritten for a subsequent message in the same
2873 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2876 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2878 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2879 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2880 the function were treating the return as a boolean value, which happened to
2881 work because 0=false and not-0=true, but is not correct code.
2883 68. The host_aton() function was not handling scoped IPv6 addresses (those
2884 with, for example, "%eth0" on the end) correctly.
2886 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2887 negated items (that is, ~something) in unsigned ints. Some compilers
2888 apparently mutter when there is no cast.
2890 70. If an address verification called from an ACL failed, and did not produce a
2891 user-specific message (i.e. there was only a "system" message), nothing was
2892 put in $acl_verify_message. In this situation, it now puts the system
2895 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2896 session; change 4.31/43 added the unwanted input to the log line - except
2897 that it did not do this in the start of session case. It now does.
2899 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2900 This is wrong and can cause the other end to generate a synchronization
2901 error if it is another Exim or anything else that does the synchronization
2902 check. A QUIT command is no longer sent after a timeout.
2904 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2905 during host lookups.
2907 74. The defer_ok option of callout verification was not working if it was used
2908 when verifying addresses in header lines, that is, for this case:
2910 verify = header_sender/callout=defer_ok
2912 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2913 those file descriptors could be used for SMTP connections. If anything
2914 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2915 could be sent to the SMTP client, causing chaos. The daemon now opens
2916 stdin, stdout, and stderr to /dev/null when it puts itself into the
2919 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2920 log by default. The user can override this with suitable Perl magic.
2922 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2923 the log message when discard triggers, was not working for the DATA ACL or
2924 for the non-SMTP ACL.
2926 78. Error message wording change in sieve.c.
2928 79. If smtp_accept_max_per_host was set, the number of connections could be
2929 restricted to fewer than expected, because the daemon was trying to set up
2930 a new connection before checking whether the processes handling previous
2931 connections had finished. The check for completed processes is now done
2932 earlier. On busy systems, this bug wouldn't be noticed because something
2933 else would have woken the daemon, and it would have reaped the completed
2936 80. If a message was submitted locally by a user whose login name contained one
2937 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2938 It caused a spool format error. I have fixed the spool reading code. A
2939 related problem was that the "from" clause in the Received: line became
2940 illegal because of the space(s). It is now covered by ${quote_local_part.
2942 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2945 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2947 83. Patch from Sieve maintainer to fix unterminated string problem in
2948 "vacation" handling.
2950 84. Some minor changes to the Linux configuration files to help with other
2951 OS variants using glibc.
2953 85. One more patch for Sieve to update vacation handling to latest spec.
2956 ----------------------------------------------------
2957 See the note above about the 4.44 and 4.50 releases.
2958 ----------------------------------------------------
2964 1. Change 4.43/35 introduced a bug that caused file counts to be
2965 incorrectly computed when quota_filecount was set in an appendfile
2968 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2969 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2972 3. Give more explanation in the error message when the command for a transport
2973 filter fails to execute.
2975 4. There are several places where Exim runs a non-Exim command in a
2976 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2977 was being done only for the command run by the queryprogram router. It is
2978 now done for all such subprocesses. The other cases are: ${run, transport
2979 filters, and the commands run by the lmtp and pipe transports.
2981 5. Some older OS have a limit of 256 on the maximum number of file
2982 descriptors. Exim was using setrlimit() to set 1000 as a large value
2983 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2984 systems. I've change it so that if it can't get 1000, it tries for 256.
2986 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2987 was an oversight, and furthermore, ever since the addition of extra
2988 controls (e.g. 4.43/32), the checks on when to allow different forms of
2989 "control" were broken. There should now be diagnostics for all cases when a
2990 control that does not make sense is encountered.
2992 7. $recipients is now available in the predata ACL (oversight).
2994 8. Tidy the search cache before the fork to do a delivery from a message
2995 received from the command line. Otherwise the child will trigger a lookup
2996 failure and thereby defer the delivery if it tries to use (for example) a
2997 cached ldap connection that the parent has called unbind on.
2999 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3000 of $address_data from the recipient verification was clobbered by the
3001 sender verification.
3003 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3004 was its contents. (It was OK if the option was not defined at all.)
3006 11. A "Completed" log line is now written for messages that are removed from
3007 the spool by the -Mrm option.
3009 12. $host_address is now set to the target address during the checking of
3010 ignore_target_hosts.
3012 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3013 being passed; this would have caused $sender_host_name to have been used if
3014 matching the list had actually called for a host name (not very likely,
3015 since this list is usually IP addresses). A host name is now passed as
3018 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3019 code that specifies a non-restarting handler (typically sigaction() in
3020 modern systems) in an attempt to fix a rare and obscure crash bug.
3022 15. Narrowed the window for a race in the daemon that could cause it to ignore
3023 SIGCHLD signals. This is not a major problem, because they are used only to
3024 wake it up if nothing else does.
3026 16. A malformed maildirsize file could cause Exim to calculate negative values
3027 for the mailbox size or file count. Odd effects could occur as a result.
3028 The maildirsize information is now recalculated if the size or filecount
3031 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3032 support for a long time. Removed HAVE_SYS_VFS_H.
3034 18. Updated exipick to current release from John Jetmore.
3036 19. Allow an empty sender to be matched against a lookup in an address list.
3037 Previously the only cases considered were a regular expression, or an
3040 20. Exim went into a mad DNS lookup loop when doing a callout where the
3041 host was specified on the transport, if the DNS lookup yielded more than
3044 21. The RFC2047 encoding function was originally intended for short strings
3045 such as real names; it was not keeping to the 75-character limit for
3046 encoded words that the RFC imposes. It now respects the limit, and
3047 generates multiple encoded words if necessary. To be on the safe side, I
3048 have increased the buffer size for the ${rfc2047: expansion operator from
3051 22. Failure to deliver a bounce message always caused it to be frozen, even if
3052 there was an errors_to setting on the router. The errors_to setting is now
3055 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3056 canonical form (fully expanded) before being placed in
3057 $sender_host_address.
3059 24. Updated eximstats to version 1.33
3061 25. Include certificate and key file names in error message when GnuTLS fails
3062 to set them up, because the GnuTLS error message doesn't include the name
3063 of the failing file when there is a problem reading it.
3065 26. Expand error message when OpenSSL has problems setting up cert/key files.
3068 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3069 (this can affect the format of dates).
3071 28. exim_tidydb, when checking for the continued existence of a message for
3072 which it has found a message-specific retry record, was not finding
3073 messages that were in split spool directories. Consequently, it was
3074 deleting retry records that should have stayed in existence.
3076 29. eximstats updated to version 1.35
3077 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3078 1.35 - bugfix such that pie charts by volume are generated correctly
3080 30. The SPA authentication driver was not abandoning authentication and moving
3081 on to the next authenticator when an expansion was forced to fail,
3082 contradicting the general specification for all authenticators. Instead it
3083 was generating a temporary error. It now behaves as specified.
3085 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3086 (the order specifies the preference for clients). The order is now AES256,
3087 AES128, 3DES, ARCFOUR128.
3089 31. Small patch to Sieve code - explicitly set From: when generating an
3092 32. Exim crashed if a remote delivery caused a very long error message to be
3093 recorded - for instance if somebody sent an entire SpamAssassin report back
3094 as a large number of 550 error lines. This bug was coincidentally fixed by
3095 increasing the size of one of Exim's internal buffers (big_buffer) that
3096 happened as part of the Exiscan merge. However, to be on the safe side, I
3097 have made the code more robust (and fixed the comments that describe what
3100 33. Some experimental protocols are using DNS PTR records for new purposes. The
3101 keys for these records are domain names, not reversed IP addresses. The
3102 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3103 leaves it alone. Component reversal etc. now happens only for IP addresses.
3106 34. The host_aton() function is supposed to be passed a string that is known
3107 to be a valid IP address. However, in the case of IPv6 addresses, it was
3108 not checking this. This is a hostage to fortune. Exim now panics and dies
3109 if the condition is not met. A case was found where this could be provoked
3110 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3111 components; fortuitously, this particular loophole had already been fixed
3112 by change 4.50/55 or 4.44/33 above.
3114 If there are any other similar loopholes, the new check in host_aton()
3115 itself should stop them being exploited. The report I received stated that
3116 data on the command line could provoke the exploit when Exim was running as
3117 exim, but did not say which command line option was involved. All I could
3118 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3119 running as the user.
3122 35. There was a buffer overflow vulnerability in the SPA authentication code
3123 (which came originally from the Samba project). I have added a test to the
3124 spa_base64_to_bits() function which I hope fixes it.
3127 36. The daemon start-up calls getloadavg() while still root for those OS that
3128 need the first call to be done as root, but it missed one case: when
3129 deliver_queue_load_max is set with deliver_drop_privilege. This is
3130 necessary for the benefit of the queue runner, because there is no re-exec
3131 when deliver_drop_privilege is set.
3133 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3134 was in use, was not putting the data itself into the right store pool;
3135 consequently, it could be overwritten for a subsequent message in the same
3136 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3139 38. Sometimes the final signoff response after QUIT could fail to get
3140 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3141 < 0 before doing a fflush(). This bug looks as though it goes back to the
3142 introduction of TLS in release 3.20, but "sometimes" must have been rare
3143 because the tests only now provoked it.
3149 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3150 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3151 Now it returns an integer. A number of calls were still expecting a T/F
3152 return. Fortuitously, in all cases, the tests worked in OK situations,
3153 which is the norm. However, things would have gone wrong on any write
3154 failures on the smtp file descriptor. This function is used when sending
3155 messages over SMTP and also when doing verify callouts.
3157 2. When Exim is called to do synchronous delivery of a locally submitted
3158 message (the -odf or -odi options), it no longer closes stderr before doing
3161 3. Implemented the mua_wrapper option.
3163 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3165 5. Implemented the functions header_remove(), header_testname(),
3166 header_add_at_position(), and receive_remove_recipient(), and exported them
3169 6. If an ACL "warn" statement specified the addition of headers, Exim already
3170 inserted X-ACL-Warn: at the start if there was no header name. However, it
3171 was not making this test for the second and subsequent header lines if
3172 there were newlines in the string. This meant that an invalid header could
3173 be inserted if Exim was badly configured.
3175 7. Allow an ACL "warn" statement to add header lines at the start or after all
3176 the Received: headers, as well as at the end.
3178 8. Added the rcpt_4xx retry error code.
3180 9. Added postmaster_mailfrom=xxx to callout verification option.
3182 10. Added mailfrom=xxxx to the callout verification option, for verify=
3185 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3186 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3188 12. Inserted some casts to stop certain compilers warning when using pointer
3189 differences as field lengths or precisions in printf-type calls (mostly
3190 affecting debugging statements).
3192 13. Added optional readline() support for -be (dynamically loaded).
3194 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3195 same clock tick as a message's arrival, so that its received time was the
3196 same as the "first fail" time on the retry record, and that message
3197 remained on the queue past the ultimate address timeout, every queue runner
3198 would try a delivery (because it was past the ultimate address timeout) but
3199 after another failure, the ultimate address timeout, which should have then
3200 bounced the address, did not kick in. This was a "< instead of <=" error;
3201 in most cases the first failure would have been in the next clock tick
3202 after the received time, and all would be well.
3204 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3205 being recognized when the domain list was tested by the match_domain
3206 condition in an expansion string.
3208 16. Added the ${str2b64: operator.
3210 17. Exim was always calling setrlimit() to set a large limit for the number of
3211 processes, without checking whether the existing limit was already
3212 adequate. (It did check for the limit on file descriptors.) Furthermore,
3213 errors from getrlimit() and setrlimit() were being ignored. Now they are
3214 logged to the main and panic logs, but Exim does carry on, to try to do its
3215 job under whatever limits there are.
3217 18. Imported PCRE 5.0.
3219 19. Trivial typo in log message " temporarily refused connection" (the leading
3222 20. If the log selector return_path_on_delivery was set and an address was
3223 redirected to /dev/null, the delivery process crashed because it assumed
3224 that a return path would always be set for a "successful" delivery. In this
3225 case, the whole delivery is bypassed as an optimization, and therefore no
3228 21. Internal re-arrangement: the function for sending a challenge and reading
3229 a response while authentication was assuming a zero-terminated challenge
3230 string. It's now changed to take a pointer and a length, to allow for
3231 binary data in such strings.
3233 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3235 23. Exim was not respecting finduser_retries when seeking the login of the
3236 uid under which it was called; it was always trying 10 times. (The default
3237 setting of finduser_retries is zero.) Also, it was sleeping after the final
3238 failure, which is pointless.
3240 24. Implemented tls_on_connect_ports.
3242 25. Implemented acl_smtp_predata.
3244 26. If the domain in control=submission is set empty, Exim assumes that the
3245 authenticated id is a complete email address when it generates From: or
3246 Sender: header lines.
3248 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3249 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3250 chown and chgrp in /bin and hostname in /usr/bin.
3252 28. Exim was keeping the "process log" file open after each use, just as it
3253 does for the main log. This opens the possibility of it remaining open for
3254 long periods when the USR1 signal hits a daemon. Occasional processlog
3255 errors were reported, that could have been caused by this. Anyway, it seems
3256 much more sensible not to leave this file open at all, so that is what now
3259 29. The long-running daemon process does not normally write to the log once it
3260 has entered its main loop, and it closes the log before doing so. This is
3261 so that log files can straightforwardly be renamed and moved. However,
3262 there are a couple of unusual error situations where the daemon does write
3263 log entries, and I had neglected to close the log afterwards.
3265 30. The text of an SMTP error response that was received during a remote
3266 delivery was being truncated at 512 bytes. This is too short for some of
3267 the long messages that one sometimes sees. I've increased the limit to
3270 31. It is now possible to make retry rules that apply only when a message has a
3271 specific sender, in particular, an empty sender.
3273 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3274 it possible to be selective about when SMTP synchronization is enforced.
3276 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3278 32. Implemented hosts_connection_nolog.
3280 33. Added an ACL for QUIT.
3282 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3285 35. Added mailbox_size and mailbox_filecount to appendfile.
3287 36. Added control = no_multiline_responses to ACLs.
3289 37. There was a bug in the logic of the code that waits for the clock to tick
3290 in the case where the clock went backwards by a substantial amount such
3291 that the microsecond fraction of "now" was more than the microsecond
3292 fraction of "then" (but the whole seconds number was less).
3294 38. Added support for the libradius Radius client library this is found on
3295 FreeBSD (previously only the radiusclient library was supported).
3301 1. When certain lookups returned multiple values in the form name=value, the
3302 quoting of the values was not always being done properly. Specifically:
3303 (a) If the value started with a double quote, but contained no whitespace,
3305 (b) If the value contained whitespace other than a space character (i.e.
3306 tabs or newlines or carriage returns) it was not quoted.
3307 This fix has been applied to the mysql and pgsql lookups by writing a
3308 separate quoting function and calling it from the lookup code. The fix
3309 should probably also be applied to nisplus, ibase and oracle lookups, but
3310 since I cannot test any of those, I have not disturbed their existing code.
3312 2. A hit in the callout cache for a specific address caused a log line with no
3313 reason for rejecting RCPT. Now it says "Previous (cached) callout
3314 verification failure".
3316 3. There was an off-by-one bug in the queryprogram router. An over-long
3317 return line was truncated at 256 instead of 255 characters, thereby
3318 overflowing its buffer with the terminating zero. As well as fixing this, I
3319 have increased the buffer size to 1024 (and made a note to document this).
3321 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3322 when Exim is waiting for an SMTP response from a remote server, Exim
3323 restarts its select() call on the socket, thereby resetting its timeout.
3324 This is not a problem when such interrupts are rare. Somebody set up a cron
3325 job to run exiwhat every 2 minutes, which is less than the normal select()
3326 timeout (5 or 10 minutes). This meant that the select() timeout never
3327 kicked in because it was always reset. I have fixed this by comparing the
3328 time when an interrupt arrives with the time at the start of the first call
3329 to select(). If more time than the timeout has elapsed, the interrupt is
3330 treated as a timeout.
3332 5. Some internal re-factoring in preparation for the addition of Sieve
3333 extensions (by MH). In particular, the "personal" test is moved to a
3334 separate function, and given an option for scanning Cc: and Bcc: (which is
3335 not set for Exim filters).
3337 6. When Exim created an email address using the login of the caller as the
3338 local part (e.g. when creating a From: or Sender: header line), it was not
3339 quoting the local part when it contained special characters such as @.
3341 7. Installed new OpenBSD configuration files.
3343 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3344 try to make them clearer.
3346 9. Callout options, other than the timeout value, were being ignored when
3347 verifying sender addresses in header lines. For example, when using
3349 verify = header_sender/callout=no_cache
3351 the cache was (incorrectly) being used.
3353 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3354 only the Cygwin environment.
3356 11. When return_path_on_delivery was set as a log selector, if different remote
3357 addresses in the same message used different return paths and parallel
3358 remote delivery occurred, the wrong values would sometimes be logged.
3359 (Whenever a remote delivery process finished, the return path value from
3360 the most recently started remote delivery process was logged.)
3362 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3363 lines when AUTH and/or TLS are in use. This is the "received protocol"
3364 field. Exim used to use "asmtp" for authenticated SMTP, without any
3365 indication (in the protocol name) for TLS use. Now it follows the RFC and
3366 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3367 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3368 names appear in log lines as well as in Received: header lines.
3370 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3371 extensions, and comparison tests, and to fix some bugs.
3373 14. Changes to the "personal" filter test:
3375 (1) The test was buggy in that it was just doing the equivalent of
3376 "contains" tests on header lines. For example, if a user's address was
3377 anne@some.where, the "personal" test would incorrectly be true for
3379 To: susanne@some.where
3381 This test is now done by extracting each address from the header in turn,
3382 and checking the entire address. Other tests that are part of "personal"
3383 are now done using regular expressions (for example, to check local parts
3384 of addresses in From: header lines).
3386 (2) The list of non-personal local parts in From: addresses has been
3387 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3388 taken from the Sieve specification recommendations.
3390 (3) If the message contains any header line starting with "List-" it is
3391 treated as non-personal.
3393 (4) The test for "circular" in the Subject: header line has been removed
3394 because it now seems ill-conceived.
3396 15. Minor typos in src/EDITME comments corrected.
3398 16. Installed latest exipick from John Jetmore.
3400 17. If headers_add on a router specified a text string that was too long for
3401 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3402 of string_sprintf() is now avoided.
3404 18. $message_body_size was not set (it was always zero) when running the DATA
3405 ACL and the local_scan() function.
3407 19. For the "mail" command in an Exim filter, no default was being set for
3408 the once_repeat time, causing a random time value to be used if "once" was
3409 specified. (If the value happened to be <= 0, no repeat happened.) The
3410 default is now 0s, meaning "never repeat". The "vacation" command was OK
3411 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3412 (I found it when inspecting the code).
3414 20. There is now an overall timeout for performing a callout verification. It
3415 defaults to 4 times the callout timeout, which applies to individual SMTP
3416 commands during the callout. The overall timeout applies when there is more
3417 than one host that can be tried. The timeout is checked before trying the
3418 next host. This prevents very long delays if there are a large number of
3419 hosts and all are timing out (e.g. when the network connections are timing
3420 out). The value of the overall timeout can be changed by specifying an
3421 additional sub-option for "callout", called "maxwait". For example:
3423 verify = sender/callout=5s,maxwait=20s
3425 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3426 to the end before writing, but this should make it even safer).
3428 22. Exim was forgetting that it had advertised PIPELINING for the second and
3429 subsequent messages on an SMTP connection. It was also not resetting its
3430 memory on STARTTLS and an internal HELO.
3432 23. When Exim logs an SMTP synchronization error within a session, it now
3433 records whether PIPELINING has been advertised or not.
3435 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3436 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3437 rather than long int.
3439 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3441 26. Added the never_mail option to autoreply.
3447 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3448 crash if the getsockname() call failed; this can happen if a connection is
3449 closed very soon after it is established. The problem was simply in the
3450 order in which certain operations were done, causing Exim to try to write
3451 to the SMTP stream before it had set up the file descriptor. The bug has
3452 been fixed by making things happen in the correct order.
3458 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3459 before the connection was closed, thus losing the rejection response.
3461 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3462 some early Solaris releases, but causes trouble in current releases where
3463 socklen_t is defined.
3465 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3468 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3471 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3472 that releasing the top of it at the end releases what was used for sub-
3473 expansions (unless the block got too big). However, discard this block if
3474 the first thing is a variable or header, so that we can use its block when
3475 it is dynamic (useful for very large $message_headers, for example).
3477 6. Lookups now cache *every* query, not just the most recent. A new, separate
3478 store pool is used for this. It can be recovered when all lookup caches are
3479 flushed. Lookups now release memory at the end of their result strings.
3480 This has involved some general refactoring of the lookup sources.
3482 7. Some code has been added to the store_xxx() functions to reduce the amount
3483 of flapping under certain conditions.
3485 8. log_incoming_interface used to affect only the <= reception log lines. Now
3486 it causes the local interface and port to be added to several more SMTP log
3487 lines, for example "SMTP connection from", and rejection lines.
3489 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3491 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3493 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3494 could be overwritten at the end of the current message (or the start of a
3495 new message if it was set in a HELO ACL). The value is now preserved for
3496 the duration of the SMTP connection.
3498 12. If a transport had a headers_rewrite setting, and a matching header line
3499 contained an unqualified address, that address was qualified, even if it
3500 did not match any rewriting rules. The underlying bug was that the values
3501 of the flags that permit the existence of unqualified sender and recipient
3502 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3503 non-local messages, and by -bnq for local messages) were not being
3504 preserved with the message after it was received.
3506 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3507 "next input=" as part of the text comprising the host identity instead of
3508 the correct text. The code was using the same buffer for two different
3509 strings. However, depending on which order the printing function evaluated
3510 its arguments, the bug did not always show up. Under Linux, for example, my
3511 test suite worked just fine.
3513 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3514 doesn't work with some older versions of Perl. It has been changed to "my",
3515 which in any case is probably the better facility to use.
3517 15. A really picky compiler found some instances of statements for creating
3518 error messages that either had too many or two few arguments for the format
3521 16. The size of the buffer for calls to the DNS resolver has been increased
3522 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3523 for addresses that have a lot of PTR records. This alleviates a problem; it
3524 does not fully solve it.
3526 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3527 buffer now truncates the list and logs the incident, which is the same
3528 action as happens when Exim is looking up a host name and its aliases.
3529 Previously in this situation something unpredictable would happen;
3530 sometimes it was "internal error: store_reset failed".
3532 18. If a server dropped the connection unexpectedly when an Exim client was
3533 using GnuTLS and trying to read a response, the client delivery process
3534 crashed while trying to generate an error log message.
3536 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3537 string, for example:
3539 warn message = H1: something\nH2: something
3541 the text was added as a single header line from Exim's point of view
3542 though it ended up OK in the delivered message. However, searching for the
3543 second and subsequent header lines using $h_h2: did not work. This has been
3544 fixed. Similarly, if a system filter added multiple headers in this way,
3545 the routers could not see them.
3547 20. Expanded the error message when iplsearch is called with an invalid key to
3548 suggest using net-iplsearch in a host list.
3550 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3551 ACLs are no longer actually imposed (and a message to that effect is
3554 22. If a "gecos" field in a passwd entry contained escaped characters, in
3555 particular, if it contained a \" sequence, Exim got it wrong when building
3556 a From: or a Sender: header from that name. A second bug also caused
3557 incorrect handling when an unquoted " was present following a character
3558 that needed quoting.
3560 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3561 was not being matched caselessly.
3563 24. Arranged for all hyphens in the exim.8 source to be escaped with
3566 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3567 properly. Recipient callout cache records were still being keyed to include
3568 the sender, even when use_sender was set false. This led to far more
3569 callouts that were necessary. The sender is no longer included in the key
3570 when use_sender is false.
3572 26. Added "control = submission" modifier to ACLs.
3574 27. Added the ${base62d: operator to decode base 62 numbers.
3576 28. dnsdb lookups can now access SRV records.
3578 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3579 the configuration file.
3581 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3582 (-v) mode. This makes the output for a verbose queue run more intelligible.
3584 31. Added a use_postmaster feature to recipient callouts.
3586 32. Added the $body_zerocount variable, containing the number of binary zero
3587 bytes in the message body.
3589 33. The time of last modification of the "new" subdirectory is now used as the
3590 "mailbox time last read" when there is a quota error for a maildir
3593 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3595 35. Added +ignore_unknown as a special item in host lists.
3597 36. Code for decoding IPv6 addresses in host lists is now included, even if
3598 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3599 address was recognized as an IP address, but was then not correctly decoded
3600 into binary, causing unexpected and incorrect effects when compared with
3607 1. Very minor rewording of debugging text in manualroute to say "list of
3608 hosts" instead of "hostlist".
3610 2. If verify=header_syntax was set, and a header line with an unqualified
3611 address (no domain) and a large number of spaces between the end of the
3612 name and the colon was received, the reception process suffered a buffer
3613 overflow, and (when I tested it) crashed. This was caused by some obsolete
3614 code that should have been removed. The fix is to remove it!
3616 3. When running in the test harness, delay a bit after writing a bounce
3617 message to get a bit more predictability in the log output.
3619 4. Added a call to search_tidyup() just before forking a reception process. In
3620 theory, someone could use a lookup in the expansion of smtp_accept_max_
3621 per_host which, without the tidyup, could leave open a database connection.
3623 5. Added the variables $recipient_data and $sender_data which get set from a
3624 lookup success in an ACL "recipients" or "senders" condition, or a router
3625 "senders" option, similar to $domain_data and $local_part_data.
3627 6. Moved the writing of debug_print from before to after the "senders" test
3630 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3631 problems for message scanning, either using a data ACL, or using
3632 local_scan() because the Received: header was not generated till after they
3633 were called (in order to set the time as the time of reception completion).
3634 I have revised the way this works. The header is now generated after the
3635 body is received, but before the ACL or local_scan() are called. After they
3636 are run, the timestamp in the header is updated.
3642 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3643 before starting a queue runner without re-exec. This happened only when
3644 deliver_drop_privilege was set or when the Exim user was set to root. The
3645 effect of the bug was that timeouts during subsequent deliveries caused
3646 crashes instead of being properly handled. The handler is now left at its
3647 default (and expected) setting.
3649 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3650 message, again when deliver_drop_privilege is set or Exim is run as root.
3651 The bug described in (1) was not present in this case, but the tidying up
3652 of the other signals was missing. I have made the two cases consistent.
3654 3. The ignore_target_hosts setting on a manualroute router was being ignored
3655 for hosts that were looked up using the /MX notation.
3657 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3660 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3661 operated on the sender address. After changing the $sender_address to <>
3662 for the sender address verify, Exim was re-instated it as the original
3663 (before rewriting) address, but remembering that it had rewritten it, so it
3664 wasn't rewriting it again. This bug also had the effect of breaking the
3665 sender address verification caching when the sender address was rewritten.
3667 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3668 This has been changed so that if the ip literal address matches
3669 ignore_target_hosts, the router declines.
3671 7. Added expansion conditions match_domain, match_address, and match_local_
3672 part (NOT match_host).
3674 8. The placeholder for the Received: header didn't have a length field set.
3676 9. Added code to Exim itself and to exim_lock to test for a specific race
3677 condition that could lead to file corruption when using MBX delivery. The
3678 issue is with the lockfile that is created in /tmp. If this file is removed
3679 after a process has opened it but before that process has acquired a lock,
3680 there is the potential for a second process to recreate the file and also
3681 acquire a lock. This could lead to two Exim processes writing to the file
3682 at the same time. The added code performs the same test as UW imapd; it
3683 checks after acquiring the lock that its file descriptor still refers to
3684 the same named file.
3686 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3687 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3690 11. Added the smtp_active_hostname option. If used, this will typically be made
3691 to depend on the incoming interface address. Because $interface_address is
3692 not set up until the daemon has forked a reception process, error responses
3693 that can happen earlier (such as "too many connections") no longer contain
3696 12. If an expansion in a condition on a "warn" statement fails because a lookup
3697 defers, the "warn" statement is abandoned, and the next ACL statement is
3698 processed. Previously this caused the whole ACL to be aborted.
3700 13. Added the iplsearch lookup type.
3702 14. Added ident_timeout as a log selector.
3704 15. Added tls_certificate_verified as a log selector.
3706 16. Added a global option tls_require_ciphers (compare the smtp transport
3707 option of the same name). This controls incoming TLS connections.
3709 17. I finally figured out how to make tls_require_ciphers do a similar thing
3710 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3711 before starting the TLS session.
3713 18. Tabs are now shown as \t in -bP output.
3715 19. If the log selector return_path_on_delivery was set, Exim crashed when
3716 bouncing a message because it had too many Received: header lines.
3718 20. If two routers both had headers_remove settings, and the first one included
3719 a superfluous trailing colon, the final name in the first list and the
3720 first name in the second list were incorrectly joined into one item (with a
3721 colon in the middle).
3727 1. Added -C and -D options to the exinext utility, mainly to make it easier
3728 to include in the automated testing, but these could be helpful when
3729 multiple configurations are in use.
3731 2. The exinext utility was not formatting the output nicely when there was
3732 an alternate port involved in the retry record key, nor when there was a
3733 message id as well (for retries that were specific to a specific message
3734 and a specific host). It was also confused by IPv6 addresses, because of
3735 the additional colons they contain. I have fixed the IPv4 problem, and
3736 patched it up to do a reasonable job for IPv6.
3738 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3739 delivery, the log line now contains "pipelined" if PIPELINING was used.
3741 4. An SMTP transport process used to panic and die if the bind() call to set
3742 an explicit outgoing interface failed. This has been changed; it is now
3743 treated in the same way as a connect() failure.
3745 5. A reference to $sender_host_name in the part of a conditional expansion
3746 that was being skipped was still causing a DNS lookup. This no longer
3749 6. The def: expansion condition was not recognizing references to header lines
3750 that used bh_ and bheader_.
3752 7. Added the _cache feature to named lists.
3754 8. The code for checking quota_filecount in the appendfile transport was
3755 allowing one more file than it should have been.
3757 9. For compatibility with Sendmail, the command line option
3765 and sets the incoming protocol and host name (for trusted callers). The
3766 host name and its colon can be omitted when only the protocol is to be set.
3767 Note the Exim already has two private options, -pd and -ps, that refer to
3768 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3769 "s", but I don't think that's a major issue.
3771 10. A number of refactoring changes to the code, none of which should affect
3774 (a) The number of logging options was getting close to filling up the
3775 32-bit word that was used as a bit map. I have split them into two classes:
3776 those that are passed in the argument to log_write(), and those that are
3777 only ever tested independently outside of that function. These are now in
3778 separate 32-bit words, so there is plenty of room for expansion again.
3779 There is no change in the user interface or the logging behaviour.
3781 (b) When building, for example, log lines, the code previously used a
3782 macro that called string_cat() twice, in order to add two strings. This is
3783 not really sufficiently general. Furthermore, there was one instance where
3784 it was actually wrong because one of the argument was used twice, and in
3785 one call a function was used. (As it happened, calling the function twice
3786 did not affect the overall behaviour.) The macro has been replaced by a
3787 function that can join an arbitrary number of extra strings onto a growing
3790 (c) The code for expansion conditions now uses a table and a binary chop
3791 instead of a serial search (which was left over from when there were very
3792 few conditions). Also, it now recognizes conditions like "pam" even when
3793 the relevant support is not compiled in: a suitably worded error message is
3794 given if an attempt is made to use such a condition.
3796 11. Added ${time_interval:xxxxx}.
3798 12. A bug was causing one of the ddress fields not to be passed back correctly
3799 from remote delivery subprocesses. The field in question was not being
3800 subsequently used, so this caused to problems in practice.
3802 13. Added new log selectors queue_time and deliver_time.
3804 14. Might have fixed a bug in maildirsizefile handling that threw up
3805 "unexpected character" debug warnings, and recalculated the data
3806 unnecessarily. In any case, I expanded the warning message to give more
3809 15. Added the message "Restricted characters in address" to the statements in
3810 the default ACL that block characters like @ and % in local parts.
3812 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3813 Three changes have been made:
3815 (a) There was a serious bug; a negative response to MAIL caused the whole
3816 recipient domain to be cached as invalid, thereby blocking all messages
3817 to all local parts at the same domain, from all senders. This bug has
3818 been fixed. The domain is no longer cached after a negative response to
3819 MAIL if the sender used is not empty.
3821 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3824 (c) A new callout option, "use_sender" has been added for people who want
3825 the modified behaviour.
3831 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3834 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3835 indeed breaks things for older releases.
3837 3. Added additional logging to the case where there is a problem reading data
3838 from a filter that is running in a subprocess using a pipe, in order to
3839 try to track down a specific problem.
3841 4. Testing facility fudge: when running in the test harness and attempting
3842 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3843 getting "No route to host". Convert this to a timeout.
3845 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3848 6. Some OS don't have socklen_t but use size_t instead. This affects the
3849 fifth argument of getsockopt() amongst other things. This is now
3850 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3851 can be set for individual OS. I have set it for SunOS5, OSF1, and
3852 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3853 some earlier ones do not.
3855 7. Change 4.30/15 was not doing the test caselessly.
3857 8. The standard form for an IPv6 address literal was being rejected by address
3858 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3859 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3860 this, as well as the form without the "IPv6" on the front (but only when
3861 address literals are enabled, of course).
3863 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3865 10. Exim crashed if a message with an empty sender address specified by -f
3866 encountered a router with an errors_to setting. This could be provoked only
3867 by a command such as
3871 where an empty string was supplied; "<>" did not hit this bug.
3873 11. Installed PCRE release 4.5.
3875 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3876 remained set. It is now erased.
3878 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3879 times from message ids (which are base 36 rather than the normal 62).
3881 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3882 were being counted as actual protocol errors, and logged if the log
3883 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3884 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3885 and DATA following a set of rejected RCPTs do not count as protocol errors.
3886 In other words, Exim assumes they were pipelined, though this may not
3887 actually be the case. Of course, in all cases the client gets an
3888 appropriate error code.
3890 15. If a lookup fails in an ACL condition, a message about the failure may
3891 be available; it is used if testing the ACL cannot continue, because most
3892 such messages specify what the cause of the deferral is. However, some
3893 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3894 that caused an old message to be retained and used if a later statement
3895 caused a defer, replacing the real cause of the deferral.
3897 16. If an IP address had so many PTR records that the DNS lookup buffer
3898 was not large enough to hold them, Exim could crash while trying to process
3899 the truncated data. It now detects and logs this case.
3901 17. Further to 4.21/58, another change has been made: if (and only if) the
3902 first line of a message (the first header line) ends with CRLF, a bare LF
3903 in a subsequent header line has a space inserted after it, so as not to
3904 terminate the header.
3906 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3907 unnecessarily, used atoi() instead of strtol(), and didn't check the
3908 termination when getting file sizes from file names by regex.
3910 19. Completely re-implemented the support for maildirsize files, in the light
3911 of a number of problems with the previous contributed implementation
3912 (4.30/29). In particular:
3914 . If the quota is zero, the maildirsize file is maintained, but no quota is
3917 . If the maildir directory does not exist, it is created before any attempt
3918 to write a maildirsize file.
3920 . The quota value in the file is just a cache; if the quota is changed in
3921 the transport, the new value overrides.
3923 . A regular expression is available for excluding directories from the
3926 20. The autoreply transport checks the characters in options that define the
3927 message's headers; it allows continued headers, but it was checking with
3928 isspace() after an embedded newline instead of explicitly looking for a
3931 21. If all the "regular" hosts to which an address was routed had passed their
3932 expiry times, and had not reached their retry times, the address was
3933 bounced, even if fallback hosts were defined. Now Exim should go on to try
3936 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3937 equivalent code in the SMTP transport. Some hosts send humungous responses
3938 to HELO/EHLO, more than 1024 it seems.
3940 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3941 gives compiler warnings in some environments. I've now done it "properly",
3944 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3945 (because of problems with the built-in one) was declared to return uschar *
3946 instead of char *, causing compiler failure.
3948 25. Fixed a file descriptor leak when processing alias/forward files.
3950 26. Fixed a minor format string issue in dbfn.c.
3952 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3954 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3955 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3958 29. When a sender address is verified, it is cached, to save repeating the test
3959 when there is more than one recipient in a message. However, when the
3960 verification involves a callout, it is possible for different callout
3961 options to be set for different recipients. It is too complicated to keep
3962 track of this in the cache, so now Exim always runs a verification when a
3963 callout is required, relying on the callout cache for the optimization.
3964 The overhead is duplication of the address routing, but this should not be
3967 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3968 to be verified with callout=postmaster, and the main callout worked but the
3969 postmaster check failed, the verification correctly failed. However, if a
3970 subsequent RCPT command asked for sender verification *without* the
3971 postmaster check, incorrect caching caused this verification also to fail,
3974 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3975 it was not caching the DNS options (qualify_single, search_parents) that
3976 were used when the lookup failed. A subsequent lookup with different
3977 options therefore always gave the same answer, though there were cases
3978 where it should not have. (Example: a "domains = !$mx_any" option on a
3979 dnslookup router: the "domains" option is always processed without any
3980 widening, but the router might have qualify_single set.) Now Exim uses the
3981 cached value only when the same options are set.
3983 32. Added John Jetmore's "exipick" utility to the distribution.
3985 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3986 than a timeout (e.g. a certificate is required, and is not provided), an
3987 Exim server now closes the connection immediately. Previously it waited for
3988 the client to close - but if the client is SSL, it seems that they each
3989 wait for each other, leading to a delay before one of them times out.
3991 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3992 maintained 0.8.x compatibility because I don't think many are using it, and
3993 it is clearly obsolete.
3995 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3998 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3999 client certificate was expired. A simple patch fixes this, though I don't
4000 understand the full logic of why the verify callback is called multiple
4003 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4004 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4005 which causes problems with some clients (such as the Certicom SSL Plus
4006 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4007 disables the coutermeasure allowing Eudora to connect."
4009 38. Exim was not checking that a write() to a log file succeeded. This could
4010 lead to Bad Things if a log got too big, in particular if it hit a file
4011 size limit. Exim now panics and dies if it cannot write to a log file, just
4012 as it does if it cannot open a log file.
4014 39. Modified OS/Makefile-Linux so that it now contains
4016 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4018 The two -D definitions ensure that Exim is compiled with large file
4019 support, which makes it possible to handle log files that are bigger than
4022 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4023 instance) a domain was checked against a named list that involved a lookup,
4024 causing $domain_data to be set, then another domain was checked against the
4025 same list, then the first domain was re-checked, the value of $domain_data
4026 after the final check could be wrong. In particular, if the second check
4027 failed, it could be set empty. This bug probably also applied to
4030 41. The strip_trailing_dot option was not being applied to the address given
4031 with the -f command-line option.
4033 42. The code for reading a message's header from the spool was incrementing
4034 $received_count, but never initializing it. This meant that the value was
4035 incorrect (doubled) while delivering a message in the same process in which
4036 it was received. In the most common configuration of Exim, this never
4037 happens - a fresh exec is done - but it can happen when
4038 deliver_drop_privilege is set.
4040 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4041 it now includes up to 150 characters of the unexpected data in the log
4044 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4045 and building data strings. The size of both of these buffers was 10 000
4046 bytes - far larger than anybody would *ever* want, thought I. Needless to
4047 say, somebody hit the limit. I have increased the maximum line length to
4048 20 000 and the maximum data length of concatenated lines to 100 000. I have
4049 also fixed two bugs, because there was no checking on these buffers. Tsk,
4050 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4051 buffer is too small.
4053 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4054 lsearch lookups. Now it does.
4056 46. When parsing a route_list item in a manualroute router, a fixed-length
4057 buffer was used for the list of hosts. I made this 1024 bytes long,
4058 thinking that nobody would ever have a list of hosts that long. Wrong.
4059 Somebody had a whole pile of complicated expansion conditions, and the
4060 string was silently truncated, leading to an expansion error. It turns out
4061 that it is easier to change to an unlimited length (owing to other changes
4062 that have happened since this code was originally written) than to build
4063 structure for giving a limitation error. The length of the item that
4064 expands into the list of hosts is now unlimited.
4066 47. The lsearch lookup could not handle data where the length of text line was
4067 more than 4095 characters. Such lines were truncated, leading to shortened
4068 data being returned. It should now handle lines of any length.
4070 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4071 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4074 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4075 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4076 no longer happens in comment lines. A list of replacements is now placed
4077 at the head of all of the source files, except those whose only change is
4078 to replace PERL_COMMAND in the very first #! line.
4080 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4081 messages on the queue, with a bottom-up merge sort, using code contributed
4082 by Michael Haardt. This should make operations like -bp somewhat faster on
4083 large queues. It won't affect queue runners, except when queue_run_in_order
4086 51. Installed eximstats 1.31 in the distribution.
4088 52. Added support for SRV lookups to the dnslookup router.
4090 53. If an ACL referred to $message_body or $message_body_end, the value was not
4091 reset for any messages that followed in the same SMTP session.
4093 54. The store-handling optimization for building very long strings was not
4094 differentiating between the different store pools. I don't think this
4095 actually made any difference in practice, but I've tidied it.
4097 55. While running the routers to verify a sender address, $sender_address
4098 was still set to the sender address. This is wrong, because when routing to
4099 send a bounce to the sender, it would be empty. Therefore, I have changed
4100 it so that, while verifying a sender address, $sender_address is set to <>.
4101 (There is no change to what happens when verifying a recipient address.)
4103 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4104 target A or AAAA records (if not already returned) without resetting the
4105 qualify_single or search_parents options of the DNS resolver. These are
4106 inappropriate in this case because the targets of MX and SRV records must
4107 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4108 target that, when qualified, matched something in the local domain. These
4109 two options are now turned off when doing these lookups.
4111 57. It seems that at least some releases of Reiserfs (which does not have the
4112 concept of a fixed number of inodes) returns zero and not -1 for the
4113 number of available inodes. This interacted badly with check_spool_inodes,
4114 which assumed that -1 was the "no such thing" setting. What I have done is
4115 to check that the total number of inodes is greater than zero before doing
4116 the test of how many are available.
4118 58. When a "warn" ACL statement has a log_message modifier, the message is
4119 remembered, and not repeated. This is to avoid a lot of repetition when a
4120 message has many recipients that cause the same warning to be written.
4121 Howewer, Exim was preserving the list of already written lines for an
4122 entire SMTP session, which doesn't seem right. The memory is now reset if a
4123 new message is started.
4125 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4126 debugging output unless log_rewrite was also set.
4128 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4129 of (int)(handle) when we know that handle contains (void *)(-1).
4131 61. The Exim daemon panic-logs an error return when it closes the incoming
4132 connection. However "connection reset by peer" seems to be common, and
4133 isn't really an error worthy of noting specially, so that particular error
4136 62. When Exim is trying to find all the local interfaces, it used to panic and
4137 die if the ioctl to get the interface flags failed. However, it seems that
4138 on at least one OS (Solaris 9) it is possible to have an interface that is
4139 included in the list of interfaces, but for which you get a failure error
4140 for this call. This happens when the interface is not "plumbed" into a
4141 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4142 failure of the "get flags" call assumes that the interface is down.
4144 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4145 makes life easier for people who are doing arithmetic on fields extracted
4146 from dates, where you often get leading zeros that should not be
4147 interpreted as octal.
4149 64. Added qualify_domain to the redirect router, to override the global
4152 65. If a pathologically long header line contained very many addresses (the
4153 report of this problem mentioned 10 000) and each of them was rewritten,
4154 Exim could use up a very large amount of memory. (It kept on making new
4155 copies of the header line as it rewrote, and never released the old ones.)
4156 At the expense of a bit more processing, the header rewriting function has
4157 been changed so that it no longer eats memory in this way.
4159 66. The generation of the Received: header has been moved from the time that a
4160 message starts to be received, to the time that it finishes. The timestamp
4161 in the Received: header should now be very close to that of the <= log
4162 line. There are two side-effects of this change:
4164 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4165 logged header lines no longer include the local Received: line, because
4166 it has not yet been created. The same applies to a copy of the message
4167 that is returned to a non-SMTP sender when a message is rejected.
4169 (b) When a filter file is tested using -bf, no additional Received: header
4170 is added to the test message. After some thought, I decided that this
4173 This change does not affect the value of $received_for. It is still set
4174 after address rewriting, but before local_scan() is called.
4176 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4178 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4179 gave an unhelpful panic error message, and a defer error. I have managed to
4180 change this behaviour so that it now rejects any supplied certificate,
4181 which seems right, as the list of acceptable certificates is empty.
4183 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4184 gave an unhelpful defer error. I have not managed to make this reject any
4185 supplied certificates, but the error message it gives is "no certificate
4186 supplied", which is not helpful.
4188 70. exigrep's output now also includes lines that are not associated with any
4189 message, but which match the given pattern. Implemented by a patch from
4190 Martin Sluka, which also tidied up the Perl a bit.
4192 71. Recipient callout verification, like sender verification, was using <> in
4193 the MAIL FROM command. This isn't really the right thing, since the actual
4194 sender may affect whether the remote host accepts the recipient or not. I
4195 have changed it to use the actual sender in the callout; this means that
4196 the cache record is now keyed on a recipient/sender pair, not just the
4197 recipient address. There doesn't seem to be a real danger of callout loops,
4198 since a callout by the remote host to check the sender would use <>.
4199 [SEE ABOVE: changed after hitting problems.]
4201 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4202 temporary errors. However, in the case of such a code being given after
4203 the end of a data transmission (i.e. after ".") Exim was failing to write
4204 a retry record for the message. (Yes, there was some broken host that was
4205 actually sending 8xx at this point.)
4207 73. An unknown lookup type in a host list could cause Exim to panic-die when
4208 the list was checked. (An example that provoked this was putting <; in the
4209 middle of a list instead of at the start.) If this happened during a DATA
4210 ACL check, a -D file could be left lying around. This kind of configuration
4211 error no longer causes Exim to die; instead it causes a defer errror. The
4212 incident is still logged to the main and panic logs.
4214 74. Buglet left over from Exim 3 conversion. The message "too many messages
4215 in one connection" was written to the rejectlog but not the mainlog, except
4216 when address rewriting (yes!) was being logged.
4218 75. Added write_rejectlog option.
4220 76. When a system filter was run not as root (that is, when system_filter_user
4221 was set), the values of the $n variables were not being returned to the
4222 main process; thus, they were not subsequently available in the $sn
4225 77. Added +return_path_on_delivery log selector.
4227 78. A connection timeout was being treated differently from recipients deferred
4228 when testing hosts_max_try with a message that was older than the host's
4229 retry timeout. (The host should not be counted, thus allowing all hosts to
4230 be tried at least once before bouncing.) This may have been the cause of an
4231 occasionally reported bug whereby a message would remain on the queue
4232 longer than the retry timeout, but would be bounced if a delivery was
4233 forced. I say "may" because I never totally pinned down the problem;
4234 setting up timeout/retry tests is difficult. See also the next item.
4236 79. The ultimate address timeout was not being applied to errors that involved
4237 a combination of host plus message (for example, a timeout on a MAIL
4238 command). When an address resolved to a number of possible hosts, and they
4239 were not all tried for each delivery (e.g. because of hosts_max_try), a
4240 message could remain on the queue longer than the retry timeout.
4242 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4245 81. Fixed an obscure SMTP outgoing bug which required at least the following
4246 conditions: (a) there was another message waiting for the same server;
4247 (b) the server returned 5xx to all RCPT commands in the first message so
4248 that the message was not completed; (c) the server dropped the connection
4249 or gave a negative response to the RSET that Exim sends to abort the
4250 transaction. The observed case was a dropped connection after DATA that had
4251 been sent in pipelining mode. That is, the server had advertised PIPELINING
4252 but was not implementing it correctly. The effect of the bug was incorrect
4253 behaviour, such as trying another host, and this could lead to a crash.
4259 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4260 and daemon.c were passed as pointers to ints; they should have been
4261 pointers to socklen_t variables (which are typically unsigned ints).
4263 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4266 3. Fixed a really odd bug that affected only the testing scheme; patching a
4267 certain fixed string in the binary changed the value of another string that
4268 happened to be identical to the end of the original first string.
4270 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4271 name", it returns that address as the IP address. On some operating
4272 systems (e.g. Solaris), it also passes back the IP address string as the
4273 "host name". However, on others (e.g. Linux), it passes back an empty
4274 string. Exim wasn't checking for this, and was changing the host name to an
4275 empty string, assuming it had been canonicized.
4277 5. Although rare, it is permitted to have more than one PTR record for a given
4278 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4279 all the names associated with an address, because they do in Solaris.
4280 However, it seems that they do not in Linux for data that comes from the
4281 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4282 I found this out when I moved to a new Linux workstation and tried to run
4283 the Exim test suite.
4285 To get round this problem I have changed the code so that it now does its
4286 own call to the DNS to look up PTR records when searching for a host name.
4287 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4288 addresses that are only in /etc/hosts are still found.
4290 This behaviour is, however, controlled by an option called host_lookup_
4291 order, which defaults to "bydns:byaddr". If people want to use the other
4292 order, or indeed, just use one or the other means of lookup, they can
4293 specify it in this variable.
4295 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4296 some operating systems, this comes back from gethostbyaddr() as an empty
4297 string, and this is what Exim used to test for. However, it seems that in
4298 other systems, "." is yielded. Exim now tests for this case too.
4300 7. The values of check_spool_space and check_log_space are now held internally
4301 as a number of kilobytes instead of an absolute number of bytes. If a
4302 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4303 kilobyte. This means that much larger values can be stored.
4305 8. Exim monitor: an attempt to get the action menu when not actually pointing
4306 at a message produces an empty menu entitled "No message selected". This
4307 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4308 no entries in it ("Shell widget menu has zero width and/or height"). So I
4309 have added a single, blank menu entry in this case.
4311 9. Added ${quote_local_part.
4313 10. MIME decoding is now applied to the contents of Subject: header lines when
4316 11. Now that a reference to $sender_host_address automatically causes a reverse
4317 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4318 host lookup before query-style lookups in lists that might use this
4319 variable. This has therefore been abolished, and the "net-" prefix is no
4320 longer necessary for query-style lookups.
4322 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4323 have been a typo for LFLAGS, so it has been changed.
4325 13. The install script calls Exim with "-C /dev/null" in order to find the
4326 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4327 to be output. Howeve, since Exim outputs its version number before the
4328 error, it didn't break the script. It just looked ugly. I fixed this by
4329 always allowing "-C /dev/null" if the caller is root.
4331 14. Ignore overlarge ACL variable number when reading spool file - insurance
4332 against a later release with more variables having written the file.
4334 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4335 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4336 this, as well as the form without the "IPv6" on the front.
4338 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4339 OS/Makefile-Darwin file.
4341 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4342 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4344 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4345 variable. However, some OS use ints for inodes. I've added cast to long int
4346 to get rid of the compiler warning.
4348 19. I had forgotten to lock out "/../" in configuration file names when
4349 ALT_CONFIG_PREFIX was set.
4351 20. Routers used for verification do not need to specify transports. However,
4352 if such a router generated a host list, and callout was configured, Exim
4353 crashed, because it could not find a port number from the (non-existent)
4354 transport. It now assumes port 25 in this circumstance.
4356 21. Added the -t option to exigrep.
4358 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4359 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4360 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4361 them is set, LOOKUP_LSEARCH is forced.
4363 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4365 24. Added sender and host information to the "rejected by local_scan()" log
4366 line; previously there was no indication of these.
4368 25. Added .include_if_exists.
4370 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4371 Linux. It turns out that not all file systems support this. Apparently some
4372 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4373 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4374 not supported on the file descriptor, is now ignored when Exim is trying to
4375 sync a directory. This applies only to Linux.
4377 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4379 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4380 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4383 29. Added support for maildirsize files from supplied patch (modified a bit).
4385 30. The use of :fail: followed by an empty string could lead Exim to respond to
4386 sender verification failures with (e.g.):
4388 550 Verification failed for <xxx>
4389 550 Sender verify failed
4391 where the first response line was missing the '-' that indicates it is not
4392 the final line of the response.
4394 31. The loop for finding the name of the user that called Exim had a hardwired
4395 limit of 10; it now uses the value of finduser_retries, which is used for
4396 all other user lookups.
4398 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4401 33. Exim was neglecting to zero errno before one call of strtol() when
4402 expanding a string and expecting an integer value. On some systems this
4403 resulted in spurious "integer overflow" errors. Also, it was casting the
4404 result into an int without checking.
4406 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4407 did not work. The code looks as if it has *never* worked, though it appears
4408 to have been documented since at least releast 1.62. I have made it work.
4410 35. The "timeout_DNS" error in retry rules, also documented since at least
4411 1.62, also never worked. As it isn't clear exactly what this means, and
4412 clearly it isn't a major issue, I have abolished the feature by treating it
4413 as "timeout", and writing a warning to the main and panic logs.
4415 36. The display of retry rules for -brt wasn't always showing the error code
4418 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4419 timeout_connect_A, timeout_connect_MX.
4421 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4422 to the empty sender.
4424 39. The daemon was not analysing the content of -oX till after it had closed
4425 stderr and disconnected from the controlling terminal. This meant that any
4426 syntax errors were only noted on the panic log, and the return code from
4427 the command was 0. By re-arranging the code a little, I've made the
4428 decoding happen first, so such errors now appear on stderr, and the return
4429 code is 1. However, the actual setting up of the sockets still happens in
4430 the disconnected process, so errors there are still only recorded on the
4433 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4434 connections (as happens on some IP stacks) was logged at start up time as
4435 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4436 it from "IPv6 and IPv4", which means that two separate sockets are being
4439 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4440 says whether AF_INET or AF_INET6 was passed as an argument.
4442 42. Exiwhat output was messed up when time zones were included in log
4445 43. Exiwhat now gives more information about the daemon's listening ports,
4446 and whether -tls-on-connect was used.
4448 44. The "port" option of the smtp transport is now expanded.
4450 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4451 silently ignored. Now an error message is written to the main and panic
4454 46. There's a new ACL modifier called "logwrite" which writes to a log file
4455 as soon as it is encountered.
4457 47. Added $local_user_uid and $local_user_gid at routing time.
4459 48. Exim crashed when trying to verify a sender address that was being
4462 49. Exim was recognizing only a space character after ".include". It now also
4463 recognizes a tab character.
4465 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4466 extracting the relevant information from the specification. The man page no
4467 longer contains scrambled data for the -d option, and I've added a section
4468 at the front about calling Exim under different names.
4470 51. Added "extra_headers" argument to the "mail" command in filter files.
4472 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4475 53. Installed eximstats 1.29.
4477 54. Added transport_filter_timeout as a generic transport option.
4479 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4480 Cc: header lines. This was required by RFC 822, but it not required by RFC
4483 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4484 incoming messages that did not have them. Now it does so only if the
4485 message originates locally, that is, if there is no associated remote host
4486 address. When Resent- header lines are present, this applies to the Resent-
4487 lines rather than the non-Resent- lines.
4489 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4490 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4492 58. Messages for configuration errors now include the name of the main
4493 configuration file - useful now that there may be more than one file in a
4494 list (.included file names were always shown).
4496 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4497 for those rare installations that do not start the daemon as root or run it
4498 setuid root. I've cut out the call to initgroups() if the daemon is not
4501 60. The Exim user and group can now be bound into the binary as text strings
4502 that are looked up at the start of Exim's processing.
4504 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4506 62. Added $mailstore_basename variable.
4508 63. Installed patch to sieve.c from Michael Haardt.
4510 64. When Exim failed to open the panic log after failing to open the main log,
4511 the original message it was trying to log was written to stderr and debug
4512 output, but if they were not available (the usual case in production), it
4513 was lost. Now it is written to syslog before the two lines that record the
4514 failures to open the logs.
4516 65. Users' Exim filters run in subprocesses under the user's uid. It is
4517 possible for a "deliver" command or an alias in a "personal" command to
4518 provoke an address rewrite. If logging of address rewriting is configured,
4519 this fails because the process is not running as root or exim. There may be
4520 a better way of dealing with this, but for the moment (because 4.30 needs
4521 to be released), I have disabled address rewrite logging when running a
4522 filter in a non-root, non-exim process.
4528 1. The buildconfig auxiliary program wasn't quoting the value set for
4529 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4530 not defined. This bug was masked in 4.22 by the effect that was fixed in
4533 2. Some messages that were rejected after a message id was allocated were
4534 shown as "incomplete" by exigrep. It no longer does this for messages that
4535 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4537 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4538 have allow_domain_literals set, the ID did not get logged in the <= line.
4539 Domain literals are now always recognized in Message-ID: header lines.
4541 4. The first argument for a ${extract expansion item is the key name or field
4542 number. Leading and trailing spaces in this item were not being ignored,
4543 causing some misleading effects.
4545 5. When deliver_drop_privilege was set, single queue runner processes started
4546 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4547 same command in the process it spins off) were not dropping privilege.
4549 6. When the daemon running as "exim" started a queue runner, it always
4550 re-executed Exim in the spun-off process. This is a waste of effort when
4551 deliver_drop_privilege is set. The new process now just calls the
4552 queue-runner function directly.
4558 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4561 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4562 ignored. Though the use of -oP was forcing the writing of a pid file, it
4563 was always written to the default place.
4565 3. If the message "no IP address found for host xxxx" is generated during
4566 incoming verification, it is now followed by identification of the incoming
4567 connection (so you can more easily find what provoked it).
4569 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4571 5. Added some features to "harden" Exim a bit more against certain attacks:
4573 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4574 be put in Local/Makefile. This is like the never_users runtime option,
4575 but it cannot be overridden. The default setting is "root".
4577 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4578 prefix string with which any file named in a -C command line option
4581 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4582 is retained for -C and -D only if the caller of Exim is root. Without
4583 it, the exim user may also use -C and -D and retain privilege.
4585 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4586 command line option is disabled.
4588 6. Macro names set by the -D option must start with an upper case letter, just
4589 like macro names defined in the configuration file.
4591 7. Added "dereference=" facility to LDAP.
4593 8. Two instances of the typo "uknown" in the source files are fixed.
4595 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4596 the Configure-Makefile script screwed up while processing it.
4598 10. Incorporated PCRE 4.4.
4600 11. The SMTP synchronization check was not operating right at the start of an
4601 SMTP session. For example, it could not catch a HELO sent before the client
4602 waited for the greeting. There is now a check for outstanding input at the
4603 point when the greeting is written. Because of the duplex, asynchronous
4604 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4605 way, but not yet received, when the check is performed.
4607 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4608 on TCP/IP sockets, because this apparently causes some broken clients to
4611 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4612 unchanged) from the Cygwin maintainer.
4614 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4615 was supported instead of testing for mbx. Effectively a typo.
4617 15. The spa authenticator server code was not checking that the input it
4618 received was valid base64.
4620 16. The debug output line for the "set" modifier in ACLs was not showing the
4621 name of the variable that was being set.
4623 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4625 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4626 lookup on its own. Something else was needed to trigger the lookup. For
4627 example, a match in host_lookup or the need for a host name in a host list.
4628 Now, if $sender_host_name is referenced and the host name has not yet been
4629 looked up, a lookup is performed. If the lookup fails, the variable remains
4630 empty, and $host_lookup_failed is set to "1".
4632 19. Added "eqi" as a case-independent comparison operator.
4634 20. The saslauthd authentication condition could segfault if neither service
4635 nor realm was specified.
4637 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4638 error message that was logged was misleading, and incoming SMTP
4639 connections were dropped. The message is now more accurate, and temporary
4640 errors are given to SMTP connections.
4642 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4643 (except RSET) with a 421 error, until QUIT is received. However, it was
4644 failing to send a response to QUIT.
4646 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4647 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4648 ACL is now run after the helo_try_verify_hosts code.
4650 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4651 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4652 case-independent, so other case variants are also recognized). Apparently
4653 some systems use these upper case variants.
4655 25. If more than two messages were waiting for the same host, and a transport
4656 filter was specified for the transport, Exim sent two messages over the
4657 same TCP/IP connection, and then failed with "socket operation on non-
4658 socket" when it tried to send the third.
4660 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4662 27. The extern definition of crypt16() in expand.c was not being excluded when
4663 the OS had its own crypt16() function.
4665 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4666 as a preferred synonym for return_size_limit, both as an option and as an
4669 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4671 30. Changed the default configuration ACL to relax the local part checking rule
4672 for addresses that are not in any local domains. For these addresses,
4673 slashes and pipe symbols are allowed within local parts, but the sequence
4674 /../ is explicitly forbidden.
4676 31. SPA server authentication was not clearing the challenge buffer before
4679 32. log_message in a "warn" ACL statement was writing to the reject log as
4680 well as to the main log, which contradicts the documentation and doesn't
4681 seem right (because no rejection is happening). So I have stopped it.
4683 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4684 However, I am unable to do any testing of this.
4686 34. Fixed an infelicity in the appendfile transport. When checking directories
4687 for a mailbox, to see if any needed to be created, it was accidentally
4688 using path names with one or more superfluous leading slashes; tracing
4689 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4691 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4692 contents are added to the log line that is written for every discarded
4693 recipient. (Previously a log_message setting was ignored.)
4695 36. The ${quote: operator now quotes the string if it is empty.
4697 37. The install script runs exim in order to find its version number. If for
4698 some reason other than non-existence or emptiness, which it checks, it
4699 could not run './exim', it was installing it with an empty version number,
4700 i.e. as "exim-". This error state is now caught, and the installation is
4703 38. An argument was missing from the function that creates an error message
4704 when Exim fails to connect to the socket for saslauthd authentication.
4705 This could cause Exim to crash, or give a corrupted message.
4707 39. Added isip, isip4, and isip6 to ${if conditions.
4709 40. The ACL variables $acl_xx are now saved with the message, and can be
4710 accessed later in routers, transports, and filters.
4712 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4713 strings in the file are not string-expanded.
4715 42. If a MAIL command specified a SIZE value that was too large to fit into an
4716 int variable, the check against message_size_limit failed. Such values are
4717 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4718 day this will have to be increased, but I don't think I want to be around
4719 when emails are that large.
4726 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4727 iconv() is not standard in FreeBSD.
4729 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4730 IPv6 enabled. The observed symptom was a segmentation fault on return from
4731 the function os_common_find_running_interfaces() in src/os.c.
4733 3. In the check_special_case() function in daemon.c I had used "errno" as an
4734 argument name, which causes warnings on some systems. This was basically a
4735 typo, since it was named "eno" in the comments!
4737 4. The code that waits for the clock to tick (at a resolution of some fraction
4738 of a second) so as to ensure message-id uniqueness was always waiting for
4739 at least one whole tick, when it could have waited for less. [This is
4740 almost certainly not relevant at current processor speeds, where it is
4741 unlikely to ever wait at all. But we try to future-proof.]
4743 5. The function that sleeps for a time interval that includes fractions of a
4744 second contained a race. It did not block SIGALRM between setting the
4745 timer, and suspending (a couple of lines later). If the interval was short
4746 and the sigsuspend() was delayed until after it had expired, the suspension
4747 never ended. On busy systems this could lead to processes getting stuck for
4750 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4751 process, before it forks any delivery processes. The open lookup caching
4752 mechanism meant that the open file or database connection was passed into
4753 the delivery process. The problem was that delivery processes always tidy
4754 up cached lookup data. This could cause a problem for the next delivery
4755 process started by the queue runner, because the external queue runner
4756 process does not know about the closure. So the next delivery process
4757 still has data in the lookup cache. In the case of a file lookup, there was
4758 no problem because closing a file descriptor in a subprocess doesn't affect
4759 the parent. However, if the lookup was caching a connection to a database,
4760 the connection was closed, and the second delivery process was likely to
4761 see errors such as "PGSQL: query failed: server closed the connection
4762 unexpectedly". The problem has been fixed by closing all cached lookups
4763 in a queue runner before running a delivery process.
4765 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4766 seem to have the "const" qualifier which it has on other OS. I've
4769 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4770 *of the same type* (client or server) with the same public name that an
4771 error should be diagnosed.
4773 9. When Exim looked up a host name for an IP address, but failed to find the
4774 original IP address when looking up the host name (a safety check), it
4775 output the message "<ip address> does not match any IP for NULL", which was
4776 confusing, to say the least. The bug was that the host name should have
4777 appeared instead of "NULL".
4779 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4780 user that is built into the binary, and the -C or -D options is used, root
4781 privilege is dropped before the configuration file is read. In addition,
4782 logging is switched to stderr instead of the normal log files. If the
4783 configuration then re-defines the Exim user, the unprivileged environment
4784 is probably not what is expected, so Exim logs a panic warning message (but
4787 However, if deliver_drop_privilege is set, the unprivileged state may well
4788 be exactly what is intended, so the warning has been cut out in that case,
4789 and Exim is allowed to try to write to its normal log files.
4795 1. smtp_return_error_details was not giving details for temporary sender
4796 or receiver verification errors.
4798 2. Diagnose a configuration error if two authenticators have the same public
4801 3. Exim used not to create the message log file for a message until the first
4802 delivery attempt. This could be confusing when incoming messages were held
4803 for policy or load reasons. The message log file is now created at the time
4804 the message is received, and an initial "Received" line is written to it.
4806 4. The automatically generated man page for command line options had a minor
4807 bug that caused no ill effects; however, a more serious problem was that
4808 the procedure for building the man page automatically didn't always
4809 operate. Consequently, release 4.20 contains an out-of-date version. This
4810 shouldn't happen again.
4812 5. When building Exim with embedded Perl support, the script that builds the
4813 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4814 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4816 6. The freeze_tell option was not being used for messages that were frozen on
4817 arrival, either by an ACL or by local_scan().
4819 7. Added the smtp_incomplete_transaction log selector.
4821 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4822 was accepting AUTH without a new EHLO.
4824 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4825 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4828 10. Logging of TCP/IP connections (when configured) now happens in the main
4829 daemon process instead of the child process, so that the TCP/IP connection
4830 count is more accurate (but it can never be perfect).
4832 11. The use of "drop" in a nested ACL was not being handled correctly in the
4833 outer ACL. Now, if condition failure induced by the nested "drop" causes
4834 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4835 or "require"), the connection is dropped.
4837 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4838 that yield "discard" can now be used with an "accept" or a "discard" verb,
4839 but an error is generated for any others (because I can't see a useful way
4840 to define what should happen).
4842 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4843 are now processed in the same way as lines in the Exim configuration file.
4844 In particular, continuation lines are supported.
4846 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4848 15. Added -ti meaning -t -i.
4850 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4851 domains, and warn by logging if others are found.
4853 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4854 returned from GIFCONF when Exim is trying to find the list of interfaces on
4855 a host. The code in os.c has been modified to copy each ifreq to an aligned
4856 structure in all cases.
4858 Also, in some cases, the returned ifreq's were being copied to a 'struct
4859 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4860 means the last couple of bytes of an IPv6 address could be chopped if the
4861 ifreq contained only a normal sockaddr (14 bytes storage).
4863 18. Named domain lists were not supported in the hosts_treat_as_local option.
4864 An entry such as +xxxx was not recognized, and was treated as a literal
4867 19. Ensure that header lines added by a DATA ACL are included in the reject log
4868 if the ACL subsequently rejects the message.
4870 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4871 MD5 (which is deprecated).
4873 21. When testing a filter file using -bf, Exim was writing a message when it
4874 took the sender from a "From " line in the message, but it was not doing so
4875 when it took $return_path from a Return-Path: header line. It now does.
4877 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4878 with a valid header line field name (a series of printing characters
4879 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4881 23. Changed "disc" in the source to "disk" to conform to the documentation and
4882 the book and for uniformity.
4884 24. Ignore Sendmail's -Ooption=value command line item.
4886 25. When execve() failed while trying to run a command in a pipe transport,
4887 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4888 could be confused with a return value of 69 from the command itself. This
4889 has been changed to 127, the value the shell returns if it is asked to run
4890 a non-existent command. The wording for the related log line suggests a
4891 non-existent command as the problem.
4893 26. If received_header_text expands to an empty string, do not add a Received:
4894 header line to the message. (Well, it adds a token one on the spool, but
4895 marks it "old" so that it doesn't get used or transmitted.)
4897 27. Installed eximstats 1.28 (addition of -nt option).
4899 28. There was no check for failure on the call to getsockname() in the daemon
4900 code. This can fail if there is a shortage of resources on the system, with
4901 ENOMEM, for example. A temporary error is now given on failure.
4903 29. Contrary to the C standard, it seems that in some environments, the
4904 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4905 program. Exim now does this explicitly; it affects the formatting of
4906 timestamps using strftime().
4908 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4909 complaints. I've now initialized all the variables, to avoid this.
4911 32. Header lines added by a system filter were not being "seen" during
4912 transport-time rewrites.
4914 33. The info_callback() function passed to OpenSSL is set up with type void
4915 (*)(SSL *, int, int), as described somewhere. However, when calling the
4916 function (actually a macro) that sets it up, the type void(*)() is
4917 expected. I've put in a cast to prevent warnings from picky compilers.
4919 34. If a DNS black list lookup found a CNAME record, but there were no A
4920 records associated with the domain it pointed at, Exim crashed.
4922 35. If a DNS black list lookup returned more than one A record, Exim ignored
4923 all but the first. It now scans all returned addresses if a particular IP
4924 value is being sought. In this situation, the contents of the
4925 $dnslist_value variable are a list of all the addresses, separated by a
4928 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4929 to accept a host name and all its aliases if the forward lookup for any of
4930 them yielded the IP address of the incoming connection. Now it accepts only
4931 those names whose forward lookup yields the correct IP address. Any other
4932 names are discarded. This closes a loophole whereby a rogue DNS
4933 administrator could create reverse DNS records to break through a
4934 wildcarded host restriction in an ACL.
4936 37. If a user filter or a system filter that ran in a subprocess used any of
4937 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4938 the wrong values were passed to the pipe command ($thisaddress had the
4939 value of $0, $0 had the value of $1, etc). This bug was introduced by
4940 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4942 38. Improved the line breaking for long SMTP error messages from ACLs.
4943 Previously, if there was no break point between 40 and 75 characters, Exim
4944 left the rest of the message alone. Two changes have been made: (a) I've
4945 reduced the minimum length to 35 characters; (b) if it can't find a break
4946 point between 35 and 75 characters, it looks ahead and uses the first one
4947 that it finds. This may give the occasional overlong line, but at least the
4948 remaining text gets split now.
4950 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4951 file descriptors might be low, and that setting 1000 would always raise it.
4952 It turns out that in some environments, the limit is already over 1000 and
4953 that lowering it causes trouble. So now Exim takes care not to decrease it.
4955 40. When delivering a message, the value of $return_path is set to $sender_
4956 address at the start of routing (routers may change the value). By an
4957 oversight, this default was not being set up when an address was tested by
4958 -bt or -bv, which affected the outcome if any router or filter referred to
4961 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4962 log only when "message" or "log_message" are set. However, if one of the
4963 conditions was an address verification, or a call to a nested ACL, the
4964 messages generated by the underlying test were being passed through. This
4965 no longer happens. The underlying message is available in $acl_verify_
4966 message for both "message" and "log_message" expansions, so it can be
4967 passed through if needed.
4969 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4970 new expansion $bh_ to give the encoded byte string without charset
4971 translation. Translation happens only if iconv() is available; HAVE_ICONV
4972 indicates this at build time. HEADERS_CHARSET gives the charset to
4973 translate to; headers_charset can change it in the configuration, and
4974 "headers charset" can change it in an individual filter file.
4976 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4977 that creates RFC 2047 encoded "words" labels them as that charset instead
4978 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4979 expansion operator; (ii) when Exim creates a From: line for a local
4980 message; (iii) when a header line is rewritten to include a "phrase" part.
4982 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4983 buggy, causing it to skip the first lines of messages whose message ID
4984 ended in 'D'. This would not have bitten before Exim release 4.14, because
4985 message IDs were unlikely to end in 'D' before then. The effect was to have
4986 incorrect size information for certain domains.
4988 45. #include "config.h" was missing at the start of the crypt16.c module. This
4989 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4992 46. If there was a timeout during a "random" callout check, Exim treated it as
4993 a failure of the random address, and carried on sending RSET and the real
4994 address. If the delay was just some slowness somewhere, the response to the
4995 original RCPT would be taken as a response to RSET and so on, causing
4996 mayhem of various kinds.
4998 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4999 when I implemented it. It didn't allow for the fact that some option values
5000 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5001 the right test for positive values.
5003 48. Domain names in DNS records are case-independent. Exim always looks them up
5004 in lower case. Some resolvers return domain names in exactly the case they
5005 appear in the zone file, that is, they may contain uppercase letters. Not
5006 all resolvers do this - some return always lower case. Exim was treating a
5007 change of case by a resolver as a change of domain, similar to a widening
5008 of a domain abbreviation. This triggered its re-routing code and so it was
5009 trying to route what was effectively the same domain again. This normally
5010 caused routing to fail (because the router wouldn't handle the domain
5011 twice). Now Exim checks for this case specially, and just changes the
5012 casing of the domain that it ultimately uses when it transmits the message
5015 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5018 50. If a filter generated a file delivery with a non-absolute name (possible if
5019 no home directory exists for the router), the forbid_file option was not
5022 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5023 the existing equality matching.
5025 52. Exim was using ints instead of ino_t variables in some places where it was
5026 dealing with inode numbers.
5028 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5029 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5030 called TMPDIR, and if it finds it is different, it changes its value.
5032 54. The smtp_printf() function is now made available to local_scan() so
5033 additional output lines can be written before returning. There is also an
5034 smtp_fflush() function to enable the detection of a dropped connection.
5035 The variables smtp_input and smtp_batched_input are exported to
5038 55. Changed the default runtime configuration: the message "Unknown user"
5039 has been removed from the ACL, and instead placed on the localuser router,
5040 using the cannot_route_message feature. This means that any verification
5041 failures that generate their own messages won't get overridden. Similarly,
5042 the "Unrouteable address" message that was in the ACL for unverifiable
5043 relay addresses has also been removed.
5045 56. Added hosts_avoid_esmtp to the smtp transport.
5047 57. The exicyclog script was not checking for the esoteric option
5048 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5049 will work only if exicyclog is run under the appropriate euid.
5051 58. Following a discussion on the list, the rules by which Exim recognises line
5052 endings on incoming messages have been changed. The -dropcr and drop_cr
5053 options are now no-ops, retained only for backwards compatibility. The
5054 following line terminators are recognized: LF CRLF CR. However, special
5055 processing applies to CR:
5057 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5058 nor a local message in the state where . is a terminator.
5060 (ii) If a bare CR is encountered in a header line, an extra space is added
5061 after the line terminator so as not to end the header. The reasoning
5062 behind this is that bare CRs in header lines are most likely either
5063 to be mistakes, or people trying to play silly games.
5065 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5066 was being incorrectly given as 18 bytes larger than it should have been.
5067 This is a VOB (very old bug).
5069 60. This may never have affected anything current, but just in case it has:
5070 When the local host is found other than at the start of a list of hosts,
5071 the local host, those with the same MX, and any that follow, are discarded.
5072 When the list in question was part of a longer list of hosts, the following
5073 hosts (not currently being processed) were also being discarded. This no
5074 longer happens. I'm not sure if this situation could ever has previously
5077 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5080 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5081 header. This is something that is recommended in a new Internet Draft, and
5082 is something that is documented as being done by Sendmail. There are two
5083 possible values. For messages generated by the autoreply transport, Exim
5086 Auto-Submitted: auto-replied
5088 whereas for all other generated messages (e.g. bounces) it adds
5090 Auto-Submitted: auto-generated
5092 63. The "personal" condition in filters now includes a test for the
5093 Auto-Submitted: header. If it contains the string "auto-" the message it
5094 not considered personal.
5096 64. Added rcpt_include_affixes as a generic transport option.
5098 65. Added queue_only_override (default true).
5100 66. Added the syslog_duplication option.
5102 67. If what should have been the first header line of a message consisted of
5103 a space followed by a colon, Exim was mis-interpreting it as a header line.
5104 It isn't of course - it is syntactically invalid and should therefore be
5105 treated as the start of the message body. The misbehaviour could have
5106 caused a number of strange effects, including loss of data in subsequent
5107 header lines, and spool format errors.
5109 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5110 client host had authenticated. This control can now be exercised by an ACL
5111 for more flexibility.
5113 69. By default, callouts do not happen when testing with -bh. There is now a
5114 variant, -bhc, which does actually run the callout code, including
5115 consulting and updating the callout cache.
5117 70. Added support for saslauthd authentication, courtesy of Alexander
5120 71. If statvfs() failed on the spool or log directories while checking their
5121 size for availability, Exim confusingly gave the error "space shortage".
5122 Furthermore, in debugging mode it crashed with a floating point exception.
5123 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5124 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5125 serious problem, Exim now writes to the main and panic logs when this
5126 happens, with details of the failure. It then refuses to accept the
5127 incoming message, giving the message "spool directory problem" or "log
5128 directory problem" with a 421 code for SMTP messages.
5130 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5131 0, 1, and 2 exist, because some OS complain for execs without them (see
5132 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5133 descriptors. However, the code omitted to check that the open succeeded,
5134 causing mysterious errors if for some reason the permissions on /dev/null
5135 got screwed. Now Exim writes a message to the main and panic logs, and
5136 bombs out if it can't open /dev/null.
5138 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5139 interact so that it is all more flexible. It is supposed to remain
5140 backwards compatible. Also added extra_local_interfaces.
5142 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5143 to bomb out with an assertion failure - to the client this appears as a
5144 connection drop. This problem occurs in the part of the code that was taken
5145 from the Samba project. Fortunately, the assertion is in a very simple
5146 function, so I have fixed this by reproducing the function inline in the
5147 one place where it is called, and arranging for authentication to fail
5148 instead of killing the process with assert().
5150 75. The SPA client code was not working when the server requested OEM rather
5151 than Unicode encoding.
5153 76. Added code to make require_files with a specific uid setting more usable in
5154 the case where statting the file as root fails - usually a non-root-mounted
5155 NFS file system. When this happens and the failure is EACCES, Exim now
5156 forks a subprocess and does the per-uid checking as the relevant uid.
5158 77. Added process_log_path.
5160 78. If log_file_path was not explicitly set, a setting of check_log_space or
5161 check_log_inodes was ignored.
5163 79. If a space check for the spool or log partitions fails, the incident is now
5164 logged. Of course, in the latter case the data may get lost...
5166 80. Added the %p formatting code to string_format() so that it can be used to
5167 print addresses in debug_print(). Adjusted all the address printing in the
5168 debugging in store.c to use %p rather than %d.
5170 81. There was a concern that a line of code in smtp_in.c could overflow a
5171 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5172 initially expressed, the concern was not well-founded, because trailing
5173 spaces are removed early. However, if the trailing spaces were followed by
5174 a NULL, they did not get removed, so the overflow was possible. Two fixes
5177 (a) I re-wrote the offending code in a cleaner fashion.
5178 (b) If an incoming SMTP command contains a NULL character, it is rejected
5181 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5182 runs initgroups(), so that if the Exim user is in any additional groups,
5183 they will be used during message reception.
5189 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob