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.
102 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
103 $data_path, and $header_path variables; fixed documentation bugs and
106 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
107 exipick to access non-standard spools, including the "frozen" queue
110 NM/01 Bugzilla 965: Support mysql stored procedures.
111 Patch from Alain Williams
113 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
115 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
116 Patch from Andreas Metzler
118 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
119 Patch from Kirill Miazine
121 NM/05 Bugzilla 671: Added umask to procmail example.
123 JJ/03 installed exipick 20100323.0, fixing doc bug
125 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
126 directory. Notification and patch from Dan Rosenberg.
128 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
130 TK/02 Improve log output when DKIM signing operation fails.
132 MH/01 Treat the transport option dkim_domain as a colon separated
133 list, not as a single string, and sign the message with each element,
134 omitting multiple occurences of the same signer.
136 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
137 Bugzilla 985, 986. Patch by Simon Arlott
139 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
140 Patch by Simon Arlott
142 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
143 MBX locking. Notification from Dan Rosenberg.
149 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
151 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
153 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
155 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
157 NM/04 Bugzilla 915: Flag broken perl installation during build.
163 TK/01 Added patch by Johannes Berg that expands the main option
164 "spamd_address" if it starts with a dollar sign.
166 TK/02 Write list of recipients to X-Envelope-Sender header when building
167 the mbox-format spool file for content scanning (suggested by Jakob
170 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
171 (http://www.dcc-servers.net/) support via dccifd. Activated by
172 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
174 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
175 by Mark Daniel Reidel <mr@df.eu>.
177 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
178 When building exim an external PCRE library is now needed -
179 PCRE is a system library on the majority of modern systems.
180 See entry on PCRE_LIBS in EDITME file.
182 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
183 conversation. Added nologin parameter to request.
184 Patch contributed by Kirill Miazine.
186 TF/01 Do not log submission mode rewrites if they do not change the address.
188 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
190 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
191 log files in place. Contributed by Roberto Lima.
193 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
195 TF/03 Bugzilla 615: When checking the local_parts router precondition
196 after a local_part_suffix or local_part_prefix option, Exim now
197 does not use the address's named list lookup cache, since this
198 contains cached lookups for the whole local part.
200 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
201 Robert Millan. Documentation is in experimental-spec.txt.
203 TF/04 Bugzilla 668: Fix parallel build (make -j).
205 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
207 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
208 Patch provided by Jan Srzednicki.
210 TF/05 Leading white space used to be stripped from $spam_report which
211 wrecked the formatting. Now it is preserved.
213 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
214 that they are available at delivery time.
216 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
218 TF/08 TLS error reporting now respects the incoming_interface and
219 incoming_port log selectors.
221 TF/09 Produce a more useful error message if an SMTP transport's hosts
222 setting expands to an empty string.
224 NM/06 Bugzilla 744: EXPN did not work under TLS.
225 Patch provided by Phil Pennock.
227 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
228 Patch provided by Richard Godbee.
230 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
231 acl_smtp_notquit, added index entry.
233 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
234 Patch provided by Eugene Bujak.
236 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
237 accept(). Patch provided by Maxim Dounin.
239 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
240 Patch provided by Phil Pennock.
242 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
244 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
245 Patch provided by Brad "anomie" Jorsch.
247 NM/14 Bugzilla 622: Added timeout setting to transport filter.
248 Patch provided by Dean Brooks.
250 TK/05 Add native DKIM support (does not depend on external libraries).
252 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
253 Patch provided by Graeme Fowler.
255 NM/16 Bugzilla 851: Documentation example syntax fix.
257 NM/17 Changed NOTICE file to remove references to embedded PCRE.
259 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
262 NM/19 Bugzilla 745: TLS version reporting.
263 Patch provided by Phil Pennock.
265 NM/20 Bugzilla 167: bool: condition support.
266 Patch provided by Phil Pennock.
268 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
269 clients. Patch provided by Phil Pennock.
271 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
272 Patch provided by Brad "anomie" Jorsch.
274 NM/23 Bugzilla 687: Fix misparses in eximstats.
275 Patch provided by Heiko Schlittermann.
277 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
278 Patch provided by Heiko Schlittermann.
280 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
281 plus update to original patch.
283 NM/26 Bugzilla 799: Documentation correction for ratelimit.
285 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
286 Patch provided by David Brownlee.
288 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
290 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
292 NM/30 Bugzilla 888: TLS documentation bugfixes.
294 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
296 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
297 Unlike the original bugzilla I have changed all shell scripts in src tree.
299 NM/33 Bugzilla 898: Transport filter timeout fix.
300 Patch by Todd Rinaldo.
302 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
303 Patch by Serge Demonchaux.
305 NM/35 Bugzilla 39: Base64 decode bug fixes.
306 Patch by Jakob Hirsch.
308 NM/36 Bugzilla 909: Correct connect() call in dcc code.
310 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
312 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
314 NM/39 Bugzilla 911: Fixed MakeLinks build script.
320 TK/01 Add preliminary DKIM support. Currently requires a forked version of
321 ALT-N's libdkim that I have put here:
322 http://duncanthrax.net/exim-experimental/
324 Note to Michael Haardt: I had to rename some vars in sieve.c. They
325 were called 'true' and it seems that C99 defines that as a reserved
326 keyword to be used with 'bool' variable types. That means you could
327 not include C99-style headers which use bools without triggering
328 build errors in sieve.c.
330 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
331 as mailq or other aliases. Changed the --help handling significantly
332 to do whats expected. exim_usage() emits usage/help information.
334 SC/01 Added the -bylocaldomain option to eximstats.
336 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
338 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
340 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
346 PH/01 Another patch from the Sieve maintainer.
348 PH/02 When an IPv6 address is converted to a string for single-key lookup
349 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
350 dots are used instead of colons so that keys in lsearch files need not
351 contain colons. This was done some time before quoting was made available
352 in lsearch files. However, iplsearch files do require colons in IPv6 keys
353 (notated using the quote facility) so as to distinguish them from IPv4
354 keys. This meant that lookups for IP addresses in host lists did not work
355 for iplsearch lookups.
357 This has been fixed by arranging for IPv6 addresses to be expressed with
358 colons if the lookup type is iplsearch. This is not incompatible, because
359 previously such lookups could never work.
361 The situation is now rather anomolous, since one *can* have colons in
362 ordinary lsearch keys. However, making the change in all cases is
363 incompatible and would probably break a number of configurations.
365 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
368 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
369 conversion specification without a maximum field width, thereby enabling
370 a rogue spamd server to cause a buffer overflow. While nobody in their
371 right mind would setup Exim to query an untrusted spamd server, an
372 attacker that gains access to a server running spamd could potentially
373 exploit this vulnerability to run arbitrary code as the Exim user.
375 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
376 $primary_hostname instead of what libspf2 thinks the hosts name is.
378 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
379 a directory entry by the name of the lookup key. Previously, if a
380 symlink pointed to a non-existing file or a file in a directory that
381 Exim lacked permissions to read, a lookup for a key matching that
382 symlink would fail. Now it is enough that a matching directory entry
383 exists, symlink or not. (Bugzilla 503.)
385 PH/03 The body_linecount and body_zerocount variables are now exported in the
388 PH/04 Added the $dnslist_matched variable.
390 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
391 This means they are set thereafter only if the connection becomes
394 PH/06 Added the client_condition to authenticators so that some can be skipped
395 by clients under certain conditions.
397 PH/07 The error message for a badly-placed control=no_multiline_responses left
398 "_responses" off the end of the name.
400 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
402 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
403 (without spaces) instead of just copying the configuration text.
405 PH/10 Added the /noupdate option to the ratelimit ACL condition.
407 PH/11 Added $max_received_linelength.
409 PH/12 Added +ignore_defer and +include_defer to host lists.
411 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
412 way in which PCRE > 7.0 is built.
414 PH/14 Implemented queue_only_load_latch.
416 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
417 MAIL command. The effect was to mangle the value on 64-bit systems.
419 PH/16 Another patch from the Sieve maintainer.
421 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
423 PH/18 If a system quota error occurred while trying to create the file for
424 a maildir delivery, the message "Mailbox is full" was not appended to the
425 bounce if the delivery eventually timed out. Change 4.67/27 below applied
426 only to a quota excession during the actual writing of the file.
428 PH/19 It seems that peer DN values may contain newlines (and other non-printing
429 characters?) which causes problems in log lines. The DN values are now
430 passed through string_printing() before being added to log lines.
432 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
433 and InterBase are left for another time.)
435 PH/21 Added message_body_newlines option.
437 PH/22 Guard against possible overflow in moan_check_errorcopy().
439 PH/23 POSIX allows open() to be a macro; guard against that.
441 PH/24 If the recipient of an error message contained an @ in the local part
442 (suitably quoted, of course), incorrect values were put in $domain and
443 $local_part during the evaluation of errors_copy.
449 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
450 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
453 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
454 issue a MAIL command.
456 PH/02 In an ACL statement such as
458 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
460 if a client was not listed at all, or was listed with a value other than
461 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
462 the condition was not true (as it should be), so access was not denied.
463 The bug was that the ! inversion was incorrectly passed on to the second
464 item. This has been fixed.
466 PH/03 Added additional dnslists conditions == and =& which are different from
467 = and & when the dns lookup returns more than one IP address.
469 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
470 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
472 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
473 FSYNC, which compiles an option called disable_fsync that allows for
474 bypassing fsync(). The documentation is heavily laced with warnings.
476 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
478 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
479 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
480 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
481 including adding "make clean"; (3) Added -fPIC when compiling the test
482 dynamically loaded module, to get rid of a warning.
484 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
485 message fails, move_frozen_messages = true and ignore_bounce_errors_after
486 = 0s. The bug is otherwise harmless.
488 PH/07 There was a bug in the dovecot authenticator such that the value of
489 $auth1 could be overwritten, and so not correctly preserved, after a
490 successful authentication. This usually meant that the value preserved by
491 the server_setid option was incorrect.
493 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
495 PH/09 Installed PCRE release 7.0.
497 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
498 run for batched SMTP input. It is now run at the start of every message
499 in the batch. While fixing this I discovered that the process information
500 (output by running exiwhat) was not always getting set for -bs and -bS
501 input. This is fixed, and it now also says "batched" for BSMTP.
503 PH/11 Added control=no_pipelining.
505 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
506 patch, slightly modified), and move the expansion of helo_data till after
507 the connection is made in the smtp transport (so it can use these
510 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
512 PH/14 Added log_selector = +pid.
514 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
516 PH/16 Add ${if forany and ${if forall.
518 PH/17 Added dsn_from option to vary the From: line in DSNs.
520 PH/18 Flush SMTP output before performing a callout, unless control =
521 no_callout_flush is set.
523 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
524 was true (the default) a successful delivery failed to delete the retry
525 item, thus causing premature timeout of the address. The bug is now
528 PH/20 Added hosts_avoid_pipelining to the smtp transport.
530 PH/21 Long custom messages for fakedefer and fakereject are now split up
531 into multiline reponses in the same way that messages for "deny" and
532 other ACL rejections are.
534 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
535 with slight modification.
537 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
538 draft, changing the syntax and factoring some duplicate code".
540 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
541 for deliveries of the second and subsequent messages over the same SMTP
544 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
545 ${reduce, with only minor "tidies".
547 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
549 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
550 expansion side effects.
552 PH/27 When a message times out after an over-quota error from an Exim-imposed
553 quota, the bounce message says "mailbox is full". This message was not
554 being given when it was a system quota that was exceeded. It now should
557 MH/03 Made $recipients available in local_scan(). local_scan() already has
558 better access to the recipient list through recipients_list[], but
559 $recipients can be useful in postmaster-provided expansion strings.
561 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
562 in the case of a MAIL command with additional options following the
563 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
564 were accidentally chopped off.
566 PH/29 SMTP synchronization checks are implemented when a command is read -
567 there is a check that no more input is waiting when there shouldn't be
568 any. However, for some commands, a delay in an ACL can mean that it is
569 some time before the response is written. In this time, more input might
570 arrive, invalidly. So now there are extra checks after an ACL has run for
571 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
572 pipelining has not been advertised.
574 PH/30 MH's patch to allow iscntrl() characters to be list separators.
576 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
577 returned in the SMTP response when smtp_return_error_details was false.
580 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
581 instead of the C I/O that was originally supplied, because problems were
584 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
585 Exim which did not show up earlier: it was assuming that a call to
586 SSL_CTX_set_info_callback() might give an error value. In fact, there is
587 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
588 was a macro that became an assignment, so it seemed to work. This has
589 changed to a proper function call with a void return, hence the compile
590 error. Exim's code has been fixed.
592 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
595 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
597 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
599 PH/37 If a message is not accepted after it has had an id assigned (e.g.
600 because it turns out to be too big or there is a timeout) there is no
601 "Completed" line in the log. When some messages of this type were
602 selected by exigrep, they were listed as "not completed". Others were
603 picked up by some special patterns. I have improved the selection
604 criteria to be more general.
606 PH/38 The host_find_failed option in the manualroute router can now be set
607 to "ignore", to completely ignore a host whose IP address cannot be
608 found. If all hosts are ignored, the behaviour is controlled by the new
609 host_all_ignored option.
611 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
612 homing or because of multiple MX records with /mx) generated more than
613 one IP address, and the following item turned out to be the local host,
614 all the secondary addresses of the first item were incorrectly removed
615 from the list, along with the local host and any following hosts (which
616 is what is supposed to happen).
618 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
619 whoever called Exim into the -H file. In the case of the daemon it was
620 behaving confusingly. When first started, it used values for whoever
621 started the daemon, but after a SIGHUP it used the Exim user (because it
622 calls itself on a restart). I have changed the code so that it now always
625 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
626 message are rejected with the same error (e.g. no authentication or bad
627 sender address), and a DATA command is nevertheless sent (as can happen
628 with PIPELINING or a stupid MUA), the error message that was given to the
629 RCPT commands is included in the rejection of the DATA command. This is
630 intended to be helpful for MUAs that show only the final error to their
633 PH/42 Another patch from the Sieve maintainer.
635 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
636 Eximstats - Fixed some broken HTML links and added missing column headers
638 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
639 columns for Rejects, Temp Rejects, Ham, and Spam rows.
641 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
643 PH/43 Yet another patch from the Sieve maintainer.
645 PH/44 I found a way to check for a TCP/IP connection going away before sending
646 the response to the final '.' that terminates a message, but only in the
647 case where the client has not sent further data following the '.'
648 (unfortunately, this is allowed). However, in many cases there won't be
649 any further data because there won't be any more messages to send. A call
650 to select() can be used: if it shows that the input is "ready", there is
651 either input waiting, or the socket has been closed. An attempt to read
652 the next input character can distinguish the two cases. Previously, Exim
653 would have sent an OK response which the client would never have see.
654 This could lead to message repetition. This fix should cure that, at
655 least in a lot of common cases.
657 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
658 advertised in response to EHLO.
664 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
665 fixed by 4.65/MH/01 (is this a record?) are fixed:
667 (i) An empty string was always treated as zero by the numeric comparison
668 operators. This behaviour has been restored.
670 (ii) It is documented that the numeric comparison operators always treat
671 their arguments as decimal numbers. This was broken in that numbers
672 starting with 0 were being interpreted as octal.
674 While fixing these problems I realized that there was another issue that
675 hadn't been noticed. Values of message_size_limit (both the global option
676 and the transport option) were treated as octal if they started with 0.
677 The documentation was vague. These values are now always treated as
678 decimal, and I will make that clear in the documentation.
684 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
685 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
688 MH/01 Don't check that the operands of numeric comparison operators are
689 integers when their expansion is in "skipping" mode (fixes bug
690 introduced by 4.64-PH/07).
692 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
693 child addresses, Exim now panics and dies. Previously, because the count
694 is held in a short int, deliveries were likely to be lost. As such a
695 large number of recipients for a single message is ridiculous
696 (performance will be very, very poor), I have chosen to impose a limit
697 rather than extend the field.
703 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
704 leftover -K file (the existence of which was triggered by #402).
705 While we were at it, introduced process PID as part of the -K
706 filename. This should rule out race conditions when creating
709 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
710 processing considerably. Previous code took too long for large mails,
711 triggering a timeout which in turn triggers #401.
713 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
714 in the DK code in transports.c. sendfile() is not really portable,
715 hence the _LINUX specificness.
717 TF/01 In the add_headers option to the mail command in an Exim filter,
718 there was a bug that Exim would claim a syntax error in any
719 header after the first one which had an odd number of characters
722 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
723 callout verification, Exim cached a "reject" for the entire domain. This
724 is correct for most verifications, but it is not correct for a recipient
725 verification with use_sender or use_postmaster set, because in that case
726 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
727 case of MAIL FROM:<> rejection from other early rejections (e.g.
728 rejection of HELO). When verifying a recipient using a non-null MAIL
729 address, the cache is ignored if it shows MAIL FROM:<> rejection.
730 Whatever the result of the callout, the value of the domain cache is
731 left unchanged (for any other kind of callout, getting as far as trying
732 RCPT means that the domain itself is ok).
734 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
737 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
738 manifest itself as EPIPE rather than ECONNECT. When tidying away a
739 session, the daemon ignores ECONNECT errors and logs others; it now
740 ignores EPIPE as well.
742 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
743 (quoted-printable decoding).
745 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
746 later the small subsequent patch to fix an introduced bug.
748 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
750 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
752 PH/08 An error is now given if message_size_limit is specified negative.
754 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
755 to be given (somewhat) arbitrary names.
757 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
760 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
761 miscellaneous code fixes
763 PH/10 Added the log_reject_target ACL modifier to specify where to log
766 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
767 hostname. This is wrong, because it relates to the incoming message (and
768 probably the interface on which it is arriving) and not to the outgoing
769 callout (which could be using a different interface). This has been
770 changed to use the value of the helo_data option from the smtp transport
771 instead - this is what is used when a message is actually being sent. If
772 there is no remote transport (possible with a router that sets up host
773 addresses), $smtp_active_hostname is used.
775 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
776 tweaks were necessary in order to get it to work (see also 21 below):
777 (a) The code assumed that strncpy() returns a negative number on buffer
778 overflow, which isn't the case. Replaced with Exim's string_format()
780 (b) There were several signed/unsigned issues. I just did the minimum
781 hacking in of casts. There is scope for a larger refactoring.
782 (c) The code used strcasecmp() which is not a standard C function.
783 Replaced with Exim's strcmpic() function.
784 (d) The code set only $1; it now sets $auth1 as well.
785 (e) A simple test gave the error "authentication client didn't specify
786 service in request". It would seem that Dovecot has changed its
787 interface. Fortunately there's a specification; I followed it and
788 changed what the client sends and it appears to be working now.
790 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
793 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
794 address A is aliased to B and C, where B exists and C does not. Without
795 -v the output is "A verified" because verification stops after a
796 successful redirection if more than one address is generated. However,
797 with -v the child addresses are also verified. Exim was outputting "A
798 failed to verify" and then showing the successful verification for C,
799 with its parentage. It now outputs "B failed to verify", showing B's
800 parentage before showing the successful verification of C.
802 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
803 look up a TXT record in a specific list after matching in a combined
806 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
807 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
808 they consult the DNS. I had assumed they would set it the way they
809 wanted; and indeed my experiments on Linux seem to show that in some
810 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
811 To be on the safe side, however, I have now made the interface to
812 host_find_byname() similar to host_find_bydns(), with an argument
813 containing the DNS resolver options. The host_find_byname() function now
814 sets these options at its start, just as host_find_bydns() does. The smtp
815 transport options dns_qualify_single and dns_search_parents are passed to
816 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
817 of host_find_byname() use the default settings of RES_DEFNAMES
818 (qualify_single) but not RES_DNSRCH (search_parents).
820 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
821 spool_read_header() do less string testing, by means of a preliminary
822 switch on the second character of optional "-foo" lines. (This is
823 overdue, caused by the large number of possibilities that now exist.
824 Originally there were few.) While I was there, I also converted the
825 str(n)cmp tests so they don't re-test the leading "-" and the first
826 character, in the hope this might squeeze out yet more improvement.
828 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
829 flag allowing group syntax was set by the header_syntax check but not
830 turned off, possible causing trouble later; (2) The flag was not being
831 set at all for the header_verify test, causing "group"-style headers to
832 be rejected. I have now set it in this case, and also caused header_
833 verify to ignore an empty address taken from a group. While doing this, I
834 came across some other cases where the code for allowing group syntax
835 while scanning a header line wasn't quite right (mostly, not resetting
836 the flag correctly in the right place). These bugs could have caused
837 trouble for malformed header lines. I hope it is now all correct.
839 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
840 with the "reply" argument non-NULL. The code, however (which originally
841 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
842 but it didn't always do it. This confused somebody who was copying the
843 code for some other use. I have removed all the tests.
845 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
846 feature that was used to support insecure browsers during the U.S. crypto
847 embargo. It requires special client support, and Exim is probably the
848 only MTA that supported it -- and would never use it because real RSA is
849 always available. This code has been removed, because it had the bad
850 effect of slowing Exim down by computing (never used) parameters for the
851 RSA_EXPORT functionality.
853 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
854 authenticator to fail if there's a tab character in the incoming data
855 (there should never be unless someone is messing about, as it's supposed
856 to be base64-encoded). Also added, on Timo's advice, the "secured" option
857 if the connection is using TLS or if the remote IP is the same as the
858 local IP, and the "valid-client-cert option" if a client certificate has
861 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
862 authenticators. This can be used for authorization after authentication
863 succeeds. (In the case of plaintext, it servers for both authentication
866 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
867 if any retry times were supplied.
869 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
870 connection, where there is no client IP address to check. In this
871 situation, the verify now always succeeds.
873 PH/25 Applied John Jetmore's -Mset patch.
875 PH/26 Added -bem to be like -Mset, but loading a message from a file.
877 PH/27 In a string expansion for a processed (not raw) header when multiple
878 headers of the same name were present, leading whitespace was being
879 removed from all of them, but trailing whitespace was being removed only
880 from the last one. Now trailing whitespace is removed from each header
881 before concatenation. Completely empty headers in a concatenation (as
884 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
885 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
887 PH/29 [Removed. This was a change that I later backed out, and forgot to
888 correct the ChangeLog entry (that I had efficiently created) before
889 committing the later change.]
891 PH/30 Exim was sometimes attempting to deliver messages that had suffered
892 address errors (4xx response to RCPT) over the same connection as other
893 messages routed to the same hosts. Such deliveries are always "forced",
894 so retry times are not inspected. This resulted in far too many retries
895 for the affected addresses. The effect occurred only when there were more
896 hosts than the hosts_max_try setting in the smtp transport when it had
897 the 4xx errors. Those hosts that it had tried were not added to the list
898 of hosts for which the message was waiting, so if all were tried, there
899 was no problem. Two fixes have been applied:
901 (i) If there are any address or message errors in an SMTP delivery, none
902 of the hosts (tried or untried) are now added to the list of hosts
903 for which the message is waiting, so the message should not be a
904 candidate for sending over the same connection that was used for a
905 successful delivery of some other message. This seems entirely
906 reasonable: after all the message is NOT "waiting for some host".
907 This is so "obvious" that I'm not sure why it wasn't done
908 previously. Hope I haven't missed anything, but it can't do any
909 harm, as the worst effect is to miss an optimization.
911 (ii) If, despite (i), such a delivery is accidentally attempted, the
912 routing retry time is respected, so at least it doesn't keep
913 hammering the server.
915 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
916 in ${readsocket because some servers need this prod.
918 PH/32 Added some extra debug output when updating a wait-xxx database.
920 PH/33 The hint "could be header name not terminated by colon", which has been
921 given for certain expansion errors for a long time, was not being given
922 for the ${if def:h_colon_omitted{... case.
924 PH/34 The spec says: "With one important exception, whenever a domain list is
925 being scanned, $domain contains the subject domain." There was at least
926 one case where this was not true.
928 PH/35 The error "getsockname() failed: connection reset by peer" was being
929 written to the panic log as well as the main log, but it isn't really
930 panic-worthy as it just means the connection died rather early on. I have
931 removed the panic log writing for the ECONNRESET error when getsockname()
934 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
935 runs only) independently of the message's sender address. This meant
936 that, if the 4xx error was in fact related to the sender, a different
937 message to the same recipient with a different sender could confuse
938 things. In particualar, this can happen when sending to a greylisting
939 server, but other circumstances could also provoke similar problems.
940 I have changed the default so that the retry time for these errors is now
941 based a combination of the sender and recipient addresses. This change
942 can be overridden by setting address_retry_include_sender=false in the
945 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
946 remote server are returned as part of bounce messages. This was not
947 happening for LMTP over a pipe (the lmtp transport), but now it is the
948 same for both kinds of LMTP.
950 PH/38 Despite being documented as not happening, Exim was rewriting addresses
951 in header lines that were in fact CNAMEs. This is no longer the case.
953 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
954 and queue runs started by the daemon processed all messages. This has
955 been fixed so that -R and -S can now usefully be given with -q<time>.
957 PH/40 Import PCRE release 6.7 (fixes some bugs).
959 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
961 PH/42 Give an error if -q is specified more than once.
963 PH/43 Renamed the variables $interface_address and $interface_port as
964 $received_ip_address and $received_port, to make it clear that these
965 values apply to message reception, and not to the outgoing interface when
966 a message is delivered. (The old names remain recognized, of course.)
968 PH/44 There was no timeout on the connect() call when using a Unix domain
969 socket in the ${readsocket expansion. There now is.
971 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
972 be meaningful with "accept".
974 SC/01 Eximstats V1.43
975 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
977 SC/02 Eximstats V1.44
978 Use a glob alias rather than an array ref in the generated
979 parser. This improves both readability and performance.
981 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
982 Collect SpamAssassin and rejection statistics.
983 Don't display local sender or destination tables unless
984 there is data to show.
985 Added average volumes into the top table text output.
987 SC/04 Eximstats V1.46
988 Collect data on the number of addresses (recipients)
989 as well as the number of messages.
991 SC/05 Eximstats V1.47
992 Added 'Message too big' to the list of mail rejection
993 reasons (thanks to Marco Gaiarin).
995 SC/06 Eximstats V1.48
996 Mainlog lines which have GMT offsets and are too short to
997 have a flag are now skipped.
999 SC/07 Eximstats V1.49 (Alain Williams)
1000 Added the -emptyok flag.
1002 SC/08 Eximstats V1.50
1003 Fixes for obtaining the IP address from reject messages.
1005 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1006 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1007 whitesspace changes from 4.64-PH/27
1009 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1012 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1013 are found, allow negative numbers in numeric criteria)
1015 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1017 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1018 to match changes made in 4.64-PH/43
1020 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1022 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1023 "domain" is actually an IP address, and force a failure. This locks out
1024 those revolvers/nameservers that support "A-for-A" lookups, in
1025 contravention of the specifications.
1027 PH/48 When a host name was looked up from an IP address, and the subsequent
1028 forward lookup of the name timed out, the host name was left in
1029 $sender_host_name, contrary to the specification.
1031 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1032 restricted to single-key lookups, Exim was not diagnosing an error if
1033 * or *@ was used with a query-style lookup.
1035 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1037 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1038 long ago, but noone interested enough thought of it. Let's just say that
1039 the "1.1" means that there are some new functions that weren't there at
1040 some point in the past.
1042 PH/51 Error processing for expansion failure of helo_data from an smtp
1043 transport during callout processing was broken.
1045 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1046 tested/used via the -bh/-bhc/-bs options.
1048 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1049 bug, fixed in subsequent PCRE releases).
1051 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1052 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1054 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1060 SC/01 Use a glob alias rather than an array ref in eximstats generated
1061 parser. This improves both readability and performance.
1063 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1064 Don't display local sender or destination tables in eximstats unless
1065 there is data to show.
1066 Added average volumes into the eximstats top table text output.
1068 SC/03 Collect data on the number of addresses (recipients) as well
1069 as the number of messages in eximstats.
1071 TF/01 Correct an error in the documentation for the redirect router. Exim
1072 does (usually) call initgroups() when daemonizing.
1074 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1075 with consistent privilege compared to when running as a daemon.
1077 TF/03 Note in the spec that $authenticated_id is not set for local
1078 submissions from trusted users.
1080 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1081 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1083 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1084 by adding some example configuration directives to the default
1085 configuration file. A little bit of work is required to uncomment the
1086 directives and define how usernames and passwords are checked, but
1087 there is now a framework to start from.
1089 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1090 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1091 without this. I don't know how relevant this is to other LDAP libraries.
1093 PH/02 Add the verb name to the "unknown ACL verb" error.
1095 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1097 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1099 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1100 directory not expanded when it should be if an expanded home directory
1101 was set for the address (which is overridden by the transport).
1103 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1106 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1107 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1108 because it is too late at that time, and has no effect.
1110 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1111 security issue with \' (bugzilla #107). I could not use the
1112 PQescapeStringConn() function, because it needs a PGconn value as one of
1115 PH/08 When testing addresses using -bt, indicate those final addresses that
1116 are duplicates that would not cause an additional delivery. At least one
1117 person was confused, thinking that -bt output corresponded to deliveries.
1118 (Suppressing duplicates isn't a good idea as you lose the information
1119 about possibly different redirections that led to the duplicates.)
1121 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1122 systems where poll() doesn't work, in particular OS X.
1124 PH/10 Added more information to debugging output for retry time not reached.
1126 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1127 operations in malware.c.
1129 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1132 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1133 syslog_duplication set false, log lines that would normally be written
1134 both the the main log and to the reject log were not written to syslog at
1137 PH/14 In the default configuration, change the use of "message" in ACL warn
1138 statements to "add_header".
1140 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1141 not followed by a command (e.g. "seen endif").
1143 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1144 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1147 PH/17 Added extra conditions to the default value of delay_warning_condition
1151 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1152 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1153 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1156 The Auto-Submitted: and various List- headers are standardised, whereas I
1157 don't think Precedence: ever was.
1159 PH/18 Refactored debugging code in route_finduser() to show more information,
1160 in particular, the error code if getpwnam() issues one.
1162 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1163 This is apparently needed in addition to the PH/07 change above to avoid
1164 any possible encoding problems.
1166 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1167 but not after initializing Perl.
1169 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1170 output them only if debugging. By default they are written stderr,
1171 apparently, which is not desirable.
1173 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1176 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1179 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1181 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1182 authenticated or an ident call has been made. Suppress the default
1183 values for $authenticated_id and $authenticated_sender (but permit -oMai
1184 and -oMas) when testing with -bh.
1186 PH/24 Re-jigged the order of the tests in the default configuration so that the
1187 tests for valid domains and recipients precede the DNS black list and CSA
1188 tests, on the grounds that those ones are more expensive.
1190 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1191 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1192 This bug exists in every version of Exim that I still have, right back to
1195 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1196 However, an attempt to turn on case-sensitivity in a regex key by
1197 including (?-i) didn't work because the subject string was already
1198 lowercased, and the effects were non-intuitive. It turns out that a
1199 one-line patch can be used to allow (?-i) to work as expected.
1205 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1206 other effects) broke the use of negated acl sub-conditions.
1208 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1211 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1212 "Deny" causes Exim to reject the incoming connection with a 554 error.
1213 Unfortunately, if there is a major crisis, such as a disk failure,
1214 tcp-wrappers gives "deny", whereas what one would like would be some
1215 kind of temporary error. A kludge has been added to help with this.
1216 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1217 554 error is used if errno is still zero or contains ENOENT (which occurs
1218 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1221 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1223 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1224 errors. Otherwise a message that provokes a temporary error (when other
1225 messages do not) can cause a whole host to time out.
1227 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1228 addresses were routed directly to files or pipes from a redirect router.
1229 File deliveries just didn't batch; pipe deliveries might have suffered
1232 PH/06 A failure to get a lock for a hints database would erroneously always say
1233 "Failed to get write lock", even when it was really a read lock.
1235 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1236 of 0600. This has been changed to use the value of the lockfile_mode
1237 option (which defaults to 0600).
1239 PH/08 Applied small patch from the Sieve maintainer.
1241 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1242 folder from quota calculations, a direct delivery into this folder messed
1243 up the contents of the maildirsize file. This was because the regex was
1244 used only to exclude .Trash (or whatever) when the size of the mailbox
1245 was calculated. There was no check that a delivery was happening into an
1246 excluded directory. This bug has been fixed by ignoring all quota
1247 processing for deliveries into excluded directories.
1249 PH/10 Added the maildirfolder_create_regex option to appendfile.
1255 PH/01 The code for finding all the local interface addresses on a FreeBSD
1256 system running IPv6 was broken. This may well have applied to all BSD
1257 systems, as well as to others that have similar system calls. The broken
1258 code found IPv4 interfaces correctly, but gave incorrect values for the
1259 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1260 that it would not match correctly against @[] and not recognize the IPv6
1263 PH/02 The ipliteral router was not recognizing addresses of the form user@
1264 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1266 PH/03 Added disable_ipv6.
1268 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1269 decoded form, because it is most often used to construct To: headers
1270 lines in autoreplies, and the decoded form may well be syntactically
1271 invalid. However, $reply_address has leading white space removed, and all
1272 newlines turned into spaces so that the autoreply transport does not
1275 PH/05 If group was specified without a user on a router, and no group or user
1276 was specified on a transport, the group from the router was ignored.
1278 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1279 for visible compile-time settings that can be used to change these
1280 numbers, for those that want even more. Backwards compatibility with old
1281 spool files has been maintained. However, going back to a previous Exim
1282 release will lost any variables that are in spool files.
1284 PH/07 Two small changes when running in the test harness: increase delay when
1285 passing a TCP/IP connection to a new process, in case the original
1286 process has to generate a bounce, and remove special handling of
1287 127.0.0.2 (sic), which is no longer necessary.
1289 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1290 be the same on different OS.
1292 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1295 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1296 whether --show-vars was specified or not
1298 JJ/02 exipick: Added support for new ACL variable spool format introduced
1301 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1302 syntactically invalid From: or Reply-to: line, and a filter used this to
1303 generate an autoreply, and therefore failed to obtain an address for the
1304 autoreply, Exim could try to deliver to a non-existent relative file
1305 name, causing unrelated and misleading errors. What now happens is that
1306 it logs this as a hard delivery error, but does not attempt to create a
1309 PH/11 The exinext utility has a -C option for testing purposes, but although
1310 the given file was scanned by exinext itself; it wasn't being passed on
1311 when Exim was called.
1313 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1314 an end-of-file indication when reading a command response.
1316 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1317 compiled. In many other places in Exim, IPv6 addresses are always
1318 recognized, so I have changed this. It also means that IPv4 domain
1319 literals of the form [IPV4:n.n.n.n] are now always recognized.
1321 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1322 used if the router is not running as root, for example, when verifying at
1323 ACL time, or when using -bh. The debugging output from this situation was
1324 non-existent - all you got was a failure to exec. I have made two
1327 (a) Failures to set uid/gid, the current directory, or a process leader
1328 in a subprocess such as that created by queryprogram now generate
1329 suitable debugging ouput when -d is set.
1331 (b) The queryprogram router detects when it is not running as root,
1332 outputs suitable debugging information if -d is set, and then runs
1333 the subprocess without attempting to change uid/gid.
1335 PH/15 Minor change to Makefile for building test_host (undocumented testing
1338 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1339 additional section of a DNS packet that returns MX or SRV records.
1340 Instead, it always explicitly searches for A/AAAA records. This avoids
1341 major problems that occur when a DNS server includes only records of one
1342 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1343 fixed another bug: if SRV records were looked up and the corresponding
1344 address records were *not* found in the additional section, the port
1345 values from the SRV records were lost.
1347 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1348 using the correct key (the original address) when searching the retry
1349 rules in order to find which one to use for generating the retry hint.
1351 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1352 adding the default one. Similarly, if it contains a Reply-To: header, the
1353 errors_reply_to option, if set, is not used.
1355 PH/19 When calculating a retry time, Exim used to measure the "time since
1356 failure" by looking at the "first failed" field in the retry record. Now
1357 it does not use this if it is later than than the arrival time of the
1358 message. Instead it uses the arrival time. This makes for better
1359 behaviour in cases where some deliveries succeed, thus re-setting the
1360 "first failed" field. An example is a quota failure for a huge message
1361 when small messages continue to be delivered. Without this change, the
1362 "time since failure" will always be short, possible causing more frequent
1363 delivery attempts for the huge message than are intended.
1364 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1366 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1367 $1, $2, $3) because the numerical variables can be reset during some
1368 expansion items (e.g. "match"), thereby losing the authentication data.
1370 PH/21 Make -bV show the size of off_t variables so that the test suite can
1371 decide whether to run tests for quotas > 2G.
1373 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1374 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1375 filecount value is greater than 2G or if a quota value is greater than 2G
1376 on a system where the size of off_t is not greater than 4, a panic error
1379 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1380 never match. The debug and -bh output now contains an explicit error
1381 message indicating a malformed IPv4 address or mask.
1383 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1384 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1385 PH/23 above applies.
1387 PH/25 Do not write to syslog when running in the test harness. The only
1388 occasion when this arises is a failure to open the main or panic logs
1389 (for which there is an explicit test).
1391 PH/26 Added the /no_tell option to "control=freeze".
1393 PH/27 If a host name lookup failed very early in a connection, for example, if
1394 the IP address matched host_lookup and the reverse lookup yielded a name
1395 that did not have a forward lookup, an error message of the form "no IP
1396 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1397 could be logged. Now it outputs the IP address instead of "NULL".
1399 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1400 allows the sender and the authenticated sender to be set when
1401 submitting a message from within Exim. Since child_open_exim() is
1402 documented for local_scan(), the new function should be too.
1404 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1405 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1406 results in an empty string is now treated as unset.
1408 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1410 PH/31 Added sender_verify_fail logging option.
1412 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1413 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1414 tidied the source and removed it altogether.
1416 PH/33 When a queue run was abandoned because the load average was too high, a
1417 log line was always written; now it is written only if the queue_run log
1418 selector is set. In addition, the log line for abandonment now contains
1419 information about the queue run such as the pid. This is always present
1420 in "start" and "stop" lines but was omitted from the "abandon" line.
1422 PH/34 Omit spaces between a header name and the colon in the error message that
1423 is given when verify = headers_syntax fails (if there are lots of them,
1424 the message gets confusing).
1426 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1427 names, as there are now some PTR records that contain slashes. This check
1428 is only to protect against broken name servers that fall over on strange
1429 characters, so the fact that it applies to all lookups doesn't matter.
1431 PH/36 Now that the new test suite is complete, we can remove some of the
1432 special code in Exim that was needed for the old test suite. For example,
1433 sorting DNS records because real resolvers return them in an arbitrary
1434 order. The new test suite's fake resolver always returns records in the
1437 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1438 bounces) except when queue_only is set, to avoid logging races between
1439 the different processes.
1441 PH/38 Panic-die if .include specifies a non-absolute path.
1443 PH/39 A tweak to the "H" retry rule from its user.
1445 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1446 a label. They prevented compilation on older perls.
1448 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1449 a warning to be raised on newish perls.
1451 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1452 on queue. Changes to match documented behaviour of showing count of
1453 messages matching specified criteria.
1455 PH/40 Changed the default ident timeout from 30s to 5s.
1457 PH/41 Added support for the use of login_cap features, on those BSD systems
1458 that have them, for controlling the resources used by pipe deliveries.
1460 PH/42 The content-scanning code uses fopen() to create files in which to put
1461 message data. Previously it was not paying any attention to the mode of
1462 the files. Exim runs with umask(0) because the rest of the code creates
1463 files with open(), and sets the required mode explicitly. Thus, these
1464 files were ending up world-writeable. This was not a big issue, because,
1465 being within the spool directory, they were not world-accessible. I have
1466 created a function called modefopen, which takes an additional mode
1467 argument. It sets umask(777), creates the file, chmods it to the required
1468 mode, then resets the umask. All the relevant calls to fopen() in the
1469 content scanning code have been changed to use this function.
1471 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1472 to 24 hours. This avoids potential overflow problems when processing G
1473 and H retry rules. I suspect nobody ever tinkers with this value.
1475 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1477 PH/45 When the plaintext authenticator is running as a client, the server's
1478 challenges are checked to ensure they are valid base64 strings. By
1479 default, the authentication attempt is cancelled if an invalid string is
1480 received. Setting client_ignore_invalid_base64 true ignores these errors.
1481 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1482 they are received. Thus, the responses can be made to depend on the
1483 challenges. If an invalid string is ignored, an empty string is placed in
1486 PH/46 Messages that are created by the autoreply transport now contains a
1487 References: header, in accordance with RFCs 2822 and 3834.
1489 PH/47 Added authenticated_sender_force to the smtp transport.
1491 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1493 PH/49 Installed latest patch from the Sieve maintainer.
1495 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1496 was also set, the appendfile transport was unnecessarily scanning a
1497 directory of message files (e.g. for maildir delivery) to find the count
1498 of files (along with the size), even though it did not need this
1499 information. It now does the scan only if it needs to find either the
1500 size of the count of files.
1502 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1504 PH/52 Two bugs concerned with error handling when the smtp transport is
1507 (i) Exim was not creating retry information for temporary errors given
1508 for individual recipients after the DATA command when the smtp transport
1509 was used in LMTP mode. This meant that they could be retried too
1510 frequently, and not timed out correctly.
1512 (ii) Exim was setting the flag that allows error details to be returned
1513 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1514 recipients that were returned after the DATA command.
1516 PH/53 This is related to PH/52, but is more general: for any failing address,
1517 when detailed error information was permitted to be returned to the
1518 sender, but the error was temporary, then after the final timeout, only
1519 "retry timeout exceeded" was returned. Now it returns the full error as
1520 well as "retry timeout exceeded".
1522 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1523 do this, and (what is worse) MTAs that accept it.
1525 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1526 will now be deprecated.
1528 PH/56 New os.c-cygwin from the Cygwin maintainer.
1530 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1531 formats (previously only available in exim formats via -bpr, -bpru,
1532 and -bpra. Now also available in native and exiqgrep formats)
1534 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1535 with very large, slow to parse queues
1537 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1539 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1541 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1542 responses to authentication challenges, though it was showing the
1543 challenges; (ii) I've removed the CR characters from the debug output for
1546 PH/58 Allow for the insertion of a newline as well as a space when a string
1547 is turned into more than one encoded-word during RFC 2047 encoding. The
1548 Sieve code now uses this.
1550 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1551 data_4xx, lost_connection, tls_required.
1553 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1554 message was being sent as an SMTP response.
1556 PH/61 Add -l and -k options to exicyclog.
1558 PH/62 When verifying, if an address was redirected to one new address, so that
1559 verification continued, and the new address failed or deferred after
1560 having set something in $address_data, the value of $address_data was not
1561 passed back to the ACL. This was different to the case when no
1562 redirection occurred. The value is now passed back in both cases.
1564 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1565 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1566 use login_cap.h, so on its own it isn't the distinguishing feature. The
1567 new name refers directly to the setclassresources() function.
1569 PH/65 Added configuration files for NetBSD3.
1571 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1573 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1574 is preferred over IPv4.
1576 PH/68 The bounce_return_message and bounce_return_body options were not being
1577 honoured for bounces generated during the reception of non-SMTP messages.
1578 In particular, this applied to messages rejected by the ACL. This bug has
1579 been fixed. However, if bounce_return_message is true and bounce_return_
1580 body is false, the headers that are returned for a non-SMTP message
1581 include only those that have been read before the error was detected.
1582 (In the case of an ACL rejection, they have all been read.)
1584 PH/69 The HTML version of the specification is now built in a directory called
1585 spec_html instead of spec.html, because the latter looks like a path with
1586 a MIME-type, and this confuses some software.
1588 PH/70 Catch two compiler warnings in sieve.c.
1590 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1591 function verify_get_ident() calls ip_connect() to connect a socket, but
1592 if the "connect()" function timed out, ip_connect() used to close the
1593 socket. However, verify_get_ident() also closes the socket later, and in
1594 between Exim writes to the log, which may get opened at this point. When
1595 the socket was closed in ip_connect(), the log could get the same file
1596 descriptor number as the socket. This naturally causes chaos. The fix is
1597 not to close the socket in ip_connect(); the socket should be closed by
1598 the function that creates it. There was only one place in the code where
1599 this was missing, in the iplookup router, which I don't think anybody now
1600 uses, but I've fixed it anyway.
1602 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1603 well as to direct DNS lookups. Otherwise the handling of names in host
1604 lists is inconsistent and therefore confusing.
1610 PH/01 Two changes to the default runtime configuration:
1612 (1) Move the checks for relay_from_hosts and authenticated clients from
1613 after to before the (commented out) DNS black list checks.
1615 (2) Add control=submission to the relay_from_hosts and authenticated
1616 clients checks, on the grounds that messages accepted by these
1617 statements are most likely to be submissions.
1619 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1621 (1) Generate an error if the third argument for the ${prvs expansion is
1624 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1627 (3) Reset the variables that are obtained from the first argument of
1628 ${prvscheck and used in the second argument before leaving the code,
1629 because their memory is reclaimed, so using them afterwards may do
1632 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1633 one (it's much easier than Tom thought :-).
1635 (5) Because of (4), we can now allow for the use of $prvscheck_result
1636 inside the third argument.
1638 PH/03 For some reason, the default setting of PATH when running a command from
1639 a pipe transport was just "/usr/bin". I have changed it to
1642 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1643 anything to be listed in the output from -bV.
1645 PH/05 When a filter generated an autoreply, the entire To: header line was
1646 quoted in the delivery log line, like this:
1648 => >A.N.Other <ano@some.domain> <original@ddress> ...
1650 This has been changed so that it extracts the operative address. There
1651 may be more than one such address. If so, they are comma-separated, like
1654 => >ano@some.domain,ona@other.domain <original@ddress> ...
1656 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1657 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1658 not being looked up in the rDNS to get a host name, Exim was showing the
1659 IP address twice in Received: lines, even though the IP addresses were
1660 identical. For example:
1662 Received: from [1.2.3.4] (helo=[1.2.3.4])
1664 However, if the real host name was known, it was omitting the HELO data
1665 if it matched the actual IP address. This has been tidied up so that it
1666 doesn't show the same IP address twice.
1668 PH/07 When both +timestamp and +memory debugging was on, the value given by
1669 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1670 called by the debug printing, thereby overwriting the timestamp buffer.
1671 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1673 PH/08 When the original message was included in an autoreply transport, it
1674 always said "this is a copy of the message, including all the headers",
1675 even if body_only or headers_only was set. It now gives an appropriate
1678 PH/09 Applied a patch from the Sieve maintainer which:
1680 o fixes some comments
1681 o adds the (disabled) notify extension core
1682 o adds some debug output for the result of if/elsif tests
1683 o points to the current vacation draft in the documentation
1684 and documents the missing references header update
1688 o fixes a bug in processing the envelope test (when testing
1689 multiple envelope elements, the last element determinted the
1692 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1693 Electronic Mail") by including:
1695 Auto-submitted: auto-generated
1697 in the messages that it generates (bounce messages and others, such as
1698 warnings). In the case of bounce messages for non-SMTP mesages, there was
1699 also a typo: it was using "Auto_submitted" (underscore instead of
1700 hyphen). Since every message generated by Exim is necessarily in response
1701 to another message, thes have all been changed to:
1703 Auto-Submitted: auto-replied
1705 in accordance with these statements in the RFC:
1707 The auto-replied keyword:
1709 - SHOULD be used on messages sent in direct response to another
1710 message by an automatic process,
1712 - MUST NOT be used on manually-generated messages,
1714 - MAY be used on Delivery Status Notifications (DSNs) and Message
1715 Disposition Notifications (MDNs),
1717 - MUST NOT be used on messages generated by automatic or periodic
1718 processes, except for messages which are automatic responses to
1721 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1722 to the default Received: header definition.
1724 PH/12 Added log selector acl_warn_skipped (default on).
1726 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1727 variables because (a) they are in the wrong storage pool and (b) even if
1728 they were copied, it wouldn't work properly because of the caching.
1730 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1731 checking when decoding. Apparently there are clients that generate
1732 overlong encoded strings. Why am I not surprised?
1734 PH/15 If the first argument of "${if match_address" was not empty, but did not
1735 contain an "@" character, Exim crashed. Now it writes a panic log message
1736 and treats the condition as false.
1738 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1740 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1741 extension "envelope-auth". The code is finished and in agreement with
1742 other implementations, but there is no documentation so far and in fact,
1743 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1744 not changing the active code.
1746 Print executed "if" and "elsif" statements when debugging is used. This
1747 helps a great deal to understand what a filter does.
1749 Document more things not specified clearly in RFC3028. I had all this
1750 sorted out, when out of a sudden new issues came to my mind. Oops."
1752 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1755 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1756 canonical form (as documented). However, after a host name lookup from
1757 the IP address, check_host() was doing a simple string comparison with
1758 addresses acquired from the DNS when checking that the found name did
1759 have the original IP as one of its addresses. Since any found IPv6
1760 addresses are likely to be in abbreviated form, the comparison could
1761 fail. Luckily, there already exists a function for doing the comparison
1762 by converting both addresses to binary, so now that is used instead of
1763 the text comparison.
1765 PH/20 There was another similar case to PH/19, when a complete host name was
1766 given in a host list; looking up its IP address could give an abbreviated
1767 form, whereas the current host's name might or might not be abbreviated.
1768 The same fix has been applied.
1774 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1775 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1778 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1779 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1781 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1783 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1784 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1785 Also, for "Auto-Submitted", treat anything other than "no" as
1786 non-personal, in accordance with RFC 3834. (Previously it treated
1787 anything starting "auto-" as non-personal.)
1789 TF/01 The control=submission/name=... option had a problem with syntax
1790 errors if the name included a slash character. The /name= option
1791 now slurps the rest of the string, so it can include any characters
1792 but it must come last in the list of options (after /sender_retain
1795 PH/05 Some modifications to the interface to the fake nameserver for the new
1803 TK/01 Added the "success_on_redirect" address verification option. See
1804 NewStuff for rationale and an example.
1806 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1808 PH/02 Patch to exigrep to allow it to work on syslog lines.
1810 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1811 fread() to skip over the body file's header line, because in Cygwin the
1812 header line is locked and is inaccessible.
1814 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1815 co-exist for some time) to make it clear that it is the Exim ID that is
1816 referenced, not the Message-ID: header line.
1818 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1819 string_format() function, because snprintf() does not exist on all
1822 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1823 ${stat: expansion item.
1825 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1826 protocol synchronization error", to keep the pedants happy.
1828 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1829 well as for IRIX systems, when gcc is being used. See the host.c source
1832 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1834 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1837 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1838 transport and to the smtp transport in LMTP mode.
1840 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1842 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1843 run a filter in a subprocess. This could lead to confusion in subsequent
1844 lookups in the parent process. There should also be a search_tidyup() at
1845 the end of the subprocess.
1847 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1848 only if the host matched helo_try_verify_hosts, which caused the
1849 verification to occur when the EHLO/HELO command was issued. The ACL just
1850 tested the remembered result. Now, if a previous verification attempt has
1851 not happened, "verify = helo" does it there and then.
1853 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1855 TK/03 Fix log output including CR from clamd.
1857 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1858 exist provoked a memory error which could cause a segfault.
1860 PH/15 Installed PCRE 6.2
1862 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1864 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1865 of the problem. Specifically, suggested +O2 rather than +O1 for the
1868 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1870 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1871 record, the port was not correctly logged when the outgoing_port log
1872 selector was set (it logged the transort's default port).
1874 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1875 fallback_hosts, and "hosts" in the smtp transport.
1877 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1878 host errors such as "Connection refused".
1880 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1881 authentication with radiusclient 0.4.9:
1883 - Error returned from rc_read_config was caught wrongly
1884 - Username/password not passed on to radius server due to wrong length.
1886 The presumption is that some radiusclient API changes for 4.51/PH/17
1887 were not taken care of correctly. The code is still untested by me (my
1888 Linux distribution still has 0.3.2 of radiusclient), but it was
1889 contributed by a Radius user.
1891 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1892 expanding the "port" option of the smtp transport.
1894 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1895 while reading a MIME header. Thanks to Tom Hughes for a patch.
1897 PH/24 Include config.h inside local_scan.h so that configuration settings are
1900 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1901 that in an ACL for RCPT (for example), you can examine exactly what was
1904 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1905 commands, but it was not correctly comparing the address with the actual
1906 client host address. Thus, it would show the EHLO address in Received:
1907 header lines when this was not necessary.
1909 PH/27 Added the % operator to ${eval:}.
1911 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1912 it should be ignoring failures (because with -C, for example, it has lost
1913 privilege). It wasn't ignoring creation failures other than "already
1916 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1919 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1920 because an input line was too long, either on its own, or by virtue of
1921 too many continuations, the temporary file was not being removed, and the
1922 return code was incorrect.
1924 PH/31 Missing "BOOL" in function definition in filtertest.c.
1926 PH/32 Applied Sieve patches from the maintainer.
1928 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1930 PH/33 Added "verify = not_blind".
1932 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1933 Local/Makefile (with some defaults set). These are used in built scripts
1934 such as exicyclog, but they have never been used in the exim_install
1935 script (though there are many overriding facilities there). I have
1936 arranged that the exim_install script now takes note of these two
1939 PH/35 Installed configuration files for Dragonfly.
1941 PH/36 When a locally submitted message by a trusted user did not contain a
1942 From: header, and the sender address was obtained from -f or from an SMTP
1943 MAIL command, and the trusted user did not use -F to supply a sender
1944 name, $originator_name was incorrectly used when constructing a From:
1945 header. Furthermore, $originator_name was used for submission mode
1946 messages from external hosts without From: headers in a similar way,
1947 which is clearly wrong.
1949 PH/37 Added control=suppress_local_fixups.
1951 PH/38 When log_selector = +received_sender was set, and the addition of the
1952 sender made the log line's construction buffer exactly full, or one byte
1953 less than full, an overflow happened when the terminating "\n" was
1956 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1957 when the result of a list match is failure because a DNS lookup failed.
1959 PH/40 RM_COMMAND is now used in the building process.
1961 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1962 the "build-* directories that it finds.
1964 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1965 address in a domain literal was a prefix of an interface address.
1967 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1968 when verifying a sender address, unless rewrite_headers is false.
1970 PH/44 Wrote a long comment about why errors_to addresses are verified as
1971 recipients, not senders.
1973 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1974 the ratelimit ACL was added.
1976 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1978 PH/46 Added extra information about PostgreSQL errors to the error string.
1980 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1981 avoiding the need to install special zones in a real server. This is
1982 backwards compatible; if it can't find the fake resolver, it drops back.
1983 Thus, both old and new test suites can be run.
1985 TF/02 Added util/ratelimit.pl
1987 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1988 clock is set back in time.
1990 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1991 Candler <B.Candler@pobox.com>.
1993 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1994 OK for addresses that are the result of redirections.
1996 PH/48 A number of further additions for the benefit of the new test suite,
1997 including a fake gethostbyname() that interfaces to the fake DNS resolver
2000 TF/06 The fix for widen_domains has also been applied to qualify_single and
2001 search_parents which are the other dnslookup options that can cause
2004 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2007 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2009 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2010 service" if it encountered an MX record with an empty target hostname.
2011 The message is now "an MX or SRV record indicated no SMTP service".
2013 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2014 if the DNS of the sending site is misconfigured. This is quite a
2015 common situation. This change restores the behaviour of treating a
2016 helo verification defer as a failure.
2018 PH/51 If self=fail was set on a router, the bounce message did not include the
2019 actual error message.
2025 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2027 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2028 command itself ended in error, the underlying message about the transport
2029 filter timeout was being overwritten with the pipe command error. Now the
2030 underlying error message should be appended to the second error message.
2032 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2034 PH/02 Reduce the amount of output that "make" produces by default. Full output
2035 can still be requested.
2037 PH/03 The warning log line about a condition test deferring for a "warn" verb
2038 was being output only once per connection, rather than after each
2039 occurrence (because it was using the same function as for successful
2040 "warn" verbs). This seems wrong, so I have changed it.
2042 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2043 it should not have, which might have caused a crash in the right
2044 circumstances, but probably never did.
2046 PH/04 Installed a modified version of Tony Finch's patch to make submission
2047 mode fix the return path as well as the Sender: header line, and to
2048 add a /name= option so that you can make the user's friendly name appear
2051 TF/03 Added the control = fakedefer ACL modifier.
2053 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2054 Mark Lowes for thorough testing.
2056 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2058 TK/03 Merged latest SRS patch from Miles Wilton.
2060 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2061 with the definition in sysexits.h (which is #included earlier).
2062 Fortunately, Exim does not actually use EX_OK. The code used to try to
2063 preserve the sysexits.h value, by assumimg that macro definitions were
2064 scanned for macro replacements. I have been disabused of this notion,
2065 so now the code just undefines EX_OK before #including unistd.h.
2067 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2068 in the smtp transport. When a block could not be written in a single
2069 write() function, the timeout was being re-applied to each part-write.
2070 This seems wrong - if the receiver was accepting one byte at a time it
2071 would take for ever. The timeout is now adjusted when this happens. It
2072 doesn't have to be particularly precise.
2074 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2075 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2077 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2078 without a domain if the check to <postmaster@domain> fails.
2080 SC/01 Eximstats: added -xls and the ability to specify output files
2081 (patch written by Frank Heydlauf).
2083 SC/02 Eximstats: use FileHandles for outputing results.
2085 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2087 SC/04 Eximstats: fixed display of large numbers with -nvr option
2089 SC/05 Eximstats: fixed merging of reports with empty tables.
2091 SC/06 Eximstats: added the -include_original_destination flag
2093 SC/07 Eximstats: removed tabs and trailing whitespace.
2095 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2097 TK/06 MBOX spool code: Add real "From " MBOX separator line
2098 so the .eml file is really in mbox format (even though
2099 most programs do not really care). Patch from Alex Miller.
2101 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2102 The latter is generated from $received_to and is only set if the
2103 message has one envelope recipient. SA can use these headers,
2104 obviously out-of-the-box. Patch from Alex Miller.
2106 PH/08 The ${def test on a variable was returning false if the variable's
2107 value was "0", contrary to what the specification has always said!
2108 The result should be true unless the variable is empty.
2110 PH/09 The syntax error of a character other than { following "${if
2111 def:variable_name" (after optional whitespace) was not being diagnosed.
2112 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2113 accidental colon was present, for example, could give incorrect results.
2115 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2116 result is used (not including appendfile, where other changes are about
2119 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2120 This involved changing a lot of size variables from int to off_t. It
2121 should work with maildirs and everything.
2123 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2124 spamd dying while we are connected to it.
2126 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2129 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2130 function for BDB 4.3.
2132 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2134 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2135 the sources for PCRE have been re-arranged and more files are now
2138 PH/15 The code I had for printing potentially long long variables in PH/11
2139 above was not the best (it lost precision). The length of off_t variables
2140 is now inspected at build time, and an appropriate printing format (%ld
2141 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2142 to be "long long int" or "long int". This is needed for the internal
2143 formatting function string_vformat().
2145 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2146 the configuration file to be ":syslog", then the script "guesses" where
2147 the logs files are, rather than using the compiled in default. In our
2148 case the guess is not the same as the compiled default, so the script
2149 suddenly stopped working when I started to use syslog. The patch checks
2150 to see if log_file_path is "". If so, it attempts to read it from exim
2151 with no configuration file to get the compiled in version, before it
2152 falls back to the previous guessing code."
2154 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2155 implementing BATV in an Exim configuration. See NewStuff for the gory
2158 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2159 Makefile that are specific to HP-UX.
2161 PH/18 If the "use_postmaster" option was set for a recipient callout together
2162 with the "random" option, the postmaster address was used as the MAIL
2163 FROM address for the random test, but not for the subsequent recipient
2164 test. It is now used for both.
2166 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2167 patch removes a few documentation additions to RFC 3028, because the
2168 latest draft now contains them. It adds the new en;ascii-case comparator
2169 and a new error check for 8bit text in MIME parts. Comparator and
2170 require names are now matched exactly. I enabled the subaddress
2171 extension, but it is not well tested yet (read: it works for me)."
2173 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2174 rework some of the code of TK/09 above to avoid the hardwired use of
2175 "%lld" and "long long". Replaced the call to snprintf() with a call to
2178 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2179 records point to non-existent hosts", "retry timeout exceeded", and
2180 "retry time not reached for any host after a long failure period".
2182 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2183 experimental DomainKeys support:
2185 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2186 (2) On an error such as an illegally used "control", the wrong name for
2187 the control was given.
2189 These problems did NOT occur unless DomainKeys support was compiled.
2191 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2193 PH/24 Added ${if match_ip condition.
2195 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2196 databases so that it will be absolutely obvious if a crash occurs in the
2197 DB library. This is a regular occurrence (often caused by mis-matched
2200 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2201 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2202 picked up on a user's system that detects such things. There doesn't seem
2203 to be a gcc warning option for this - only an attribute that has to be
2204 put on the function's prototype. It seems that in Fedora Core 4 they have
2205 set this on a number of new functions. No doubt there will be more in due
2208 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2209 specify a transport. However, if an address that was verified by such a
2210 router was the subject of a callout, Exim crashed because it tried to
2211 read the rcpt_include_affixes from the non-existent transport. Now it
2212 just assumes that the setting of that option is false. This bug was
2213 introduced by 4.51/PH/31.
2215 PH/28 Changed -d+all to exclude +memory, because that information is very
2216 rarely of interest, but it makes the output a lot bigger. People tend to
2217 do -d+all out of habit.
2219 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2220 code in os-type was giving problems when libc.so lives in lib64, like on
2223 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2224 aren't the modern standard, and it seems that some systems' include files
2225 don't always have them. Exim was already checking for some of the newer
2226 ones like T_AAAA, and defining it itself. I've added checks for all the
2227 record types that Exim uses.
2229 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2230 not automatically generating a new one, as it is supposed to. This
2231 prevented TLS from working. If the file did exist, but contained invalid
2232 data, a new version was generated, as expected. It was only the case of a
2233 non-existent file that was broken.
2235 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2236 with a change in libdomainkeys > 0.64.
2238 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2239 from DNS. If the selector record carries the flag, it now has
2240 precedence over the domain-wide flag.
2242 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2244 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2245 the use of a port name that isn't defined in /etc/services) occurred, the
2246 message was deferred as in a normal delivery, and thus remained on the
2247 spool, instead of being failed because of the mua_wrapper setting. This
2248 is now fixed, and I tidied up some of the mua_wrapper messages at the
2251 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2252 the messages in a hash of arrays rather than using individual hashes.
2253 This is a bit cleaner and results in dramatic memory savings, albeit
2254 at a slight CPU cost.
2256 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2257 as requested by Marc Sherman.
2259 SC/10 Eximstats: added histograms for user specified patterns as requested
2262 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2264 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2265 fopen() in the content-scanning modules that did not already have it.
2271 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2272 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2274 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2276 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2278 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2279 to test Sieve filters that use "vacation".
2281 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2282 that changes the way the GnuTLS parameters are stored in the cache file.
2283 The new format can be generated externally. For backward compatibility,
2284 if the data in the cache doesn't make sense, Exim assumes it has read an
2285 old-format file, and it generates new data and writes a new file. This
2286 means that you can't go back to an older release without removing the
2289 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2290 work if there are any delivery delays because "one_time" forces the
2291 parent to be marked "delivered", so its unseen clone is never tried
2292 again. For this reason, Exim now forbids the simultaneous setting of
2295 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2296 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2297 entry if you want to know the details. The fix, however, neglected to
2298 consider the case when local delivery batching is involved. The test for
2299 "previously delivered" was not happening when checking to see if an
2300 address could be batched with a previous (undelivered) one; under
2301 certain circumstances this could lead to multiple deliveries to the same
2304 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2305 in its include files, and this causes problems building Exim.
2307 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2308 header_syntax) but Exim was just ignoring anything given after a slash.
2309 In particular, this caused confusion with an attempt to use "verify =
2310 reverse_host_lookup/defer_ok". An error is now given when options are
2311 supplied for verify items that do not have them. (Maybe reverse_host_
2312 lookup should have a defer_ok option, but that's a different point.)
2314 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2315 defined by RFC 821) to 2048, because there were problems with some AUTH
2316 commands, and RFC 1869 says the size should be increased for extended
2317 SMTP commands that take arguments.
2319 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2322 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2323 "unknown" error; now it says that the functionality isn't in the binary.
2325 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2326 an address' error message when a string expansion fails (syntax or
2327 whatever). Otherwise the password may appear in the log. Following change
2328 PH/42 below, there is no longer a chance of it appearing in a bounce
2331 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2333 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2334 of its addresses was ever tried. (Bugzilla bug #2.)
2336 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2337 the result incorrectly in the debug output. (It correctly added a newline
2338 to what was transported.)
2340 TF/01 Added $received_time.
2342 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2343 commented out examples of how to interface to a virus scanner and to
2344 SpamAssassin. Also added commented examples of av_scanner and
2345 spamd_address settings.
2347 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2348 and controls are allowed in which ACLs. There were a couple of minor
2349 errors. Some of the entries in the conditions table (which is a table of
2350 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2351 negation of where the condition IS allowed.
2353 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2355 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2356 header file does not have a version number, so I've had to invent a new
2357 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2358 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2359 radiusclient), but it was contributed by a Radius user.
2361 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2362 files or directories, for OpenSSL.
2364 PH/19 When an Exim process that is running as root has to create an Exim log
2365 file, it does so in a subprocess that runs as exim:exim so as to get the
2366 ownership right at creation (otherwise, other Exim processes might see
2367 the file with the wrong ownership). There was no test for failure of this
2368 fork() call, which would lead to the process getting stuck as it waited
2369 for a non-existent subprocess. Forks do occasionally fail when resources
2370 run out. I reviewed all the other calls to fork(); they all seem to check
2373 PH/20 When checking for unexpected SMTP input at connect time (before writing
2374 the banner), Exim was not dealing correctly with a non-positive return
2375 from the read() function. If the client had disconnected by this time,
2376 the result was a log entry for a synchronization error with an empty
2377 string after "input=" when read() returned zero. If read() returned -1
2378 (an event I could not check), uninitialized data bytes were printed.
2379 There were reports of junk text (parts of files, etc) appearing after
2382 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2384 PH/22 Added support for macro redefinition, and (re)definition in between
2385 driver and ACL definitions.
2387 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2388 forgetting to use the resulting value; it was using the unexpanded value.
2390 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2391 hadn't been configured. The fix is from Juergen Kreileder, who
2392 understands it better than I do:
2394 "Here's what I see happening with three configured cyrus_sasl
2395 authenticators configured (plain, login, cram-md5):
2397 On startup auth_cyrus_sasl_init() gets called for each of these.
2398 This means three calls to sasl_listmech() without a specified mech_list.
2399 => SASL tests which mechs of all available mechs actually work
2400 => three warnings about OTP not working
2401 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2403 With the patch, sasl_listmech() also gets called three times. But now
2404 SASL's mech_list option is set to the server_mech specified in the the
2405 authenticator. Or in other words, the answer from sasl_listmech()
2406 gets limited to just the mech you're testing for (which is different
2408 => the return list contains just 'plain' or 'login', 'cram-md5' or
2409 nothing depending on the value of ob->server_mech.
2411 I've just tested the patch: Authentication still works fine,
2412 unavailable mechs specified in the exim configuration are still
2413 caught, and the auth.log warnings about OTP are gone."
2415 PH/25 When debugging is enabled, the contents of the command line are added
2416 to the debugging output, even when log_selector=+arguments is not
2419 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2420 answer is "GNU", and only if the return is "GNU/something" is the answer
2423 PH/27 $acl_verify_message is now set immediately after the failure of a
2424 verification in an ACL, and so is available in subsequent modifiers. In
2425 particular, the message can be preserved by coding like this:
2427 warn !verify = sender
2428 set acl_m0 = $acl_verify_message
2430 Previously, $acl_verify_message was set only while expanding "message"
2431 and "log_message" when a very denied access.
2433 PH/28 Modified OS/os.c-Linux with
2435 -#ifndef OS_LOAD_AVERAGE
2436 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2438 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2439 nomenclature these days.)
2441 PH/29 Installed patch from the Sieve maintainer that adds the options
2442 sieve_useraddress and sieve_subaddress to the redirect router.
2444 PH/30 In these circumstances:
2445 . Two addresses routed to the same list of hosts;
2446 . First host does not offer TLS;
2447 . First host accepts first address;
2448 . First host gives temporary error to second address;
2449 . Second host offers TLS and a TLS session is established;
2450 . Second host accepts second address.
2451 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2452 and peerdn, if requested) that were in fact used only for the second
2455 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2456 attention to any local part prefix or suffix that was matched by the
2457 router that accepted the address. It now behaves in the same way as it
2458 does for delivery: the affixes are removed from the local part unless
2459 rcpt_include_affixes is set on the transport.
2461 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2462 timeout during the DATA phase of an incoming message.
2464 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2465 applied a patch sanctioned by the Sieve maintainer.
2467 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2468 the uid or gid is negative. A case of a negative gid caused this to be
2469 noticed. The fix allows for either to be negative.
2471 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2472 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2475 PH/36 The change PH/12 above was broken. Fixed it.
2477 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2478 the grounds that routing the same address twice would always produce the
2479 same answer. This might have been true once, but it is certainly no
2480 longer true now. Routing a child address may depend on the previous
2481 routing that produced that child. Some complicated redirection strategies
2482 went wrong when messages had multiple recipients, and made Exim's
2483 behaviour dependent on the order in which the addresses were given.
2485 I have moved the duplicate checking until after the routing is complete.
2486 Exim scans the addresses that are assigned to local and remote
2487 transports, and removes any duplicates. This means that more work will be
2488 done, as duplicates will always all be routed, but duplicates are
2489 presumably rare, so I don't expect this is of any significance.
2491 For deliveries to pipes, files, and autoreplies, the duplicate checking
2492 still happens during the routing process, since they are not going to be
2495 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2496 It corrects a timeout issue with spamd. This is Ian's comment: "The
2497 background is that sometimes spamd either never reads data from a
2498 connection it has accepted, or it never writes response data. The exiscan
2499 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2500 blindly assumes that writes won't block so it may never time out."
2502 PH/39 Allow G after quota size as well as K and M.
2504 PH/40 The value set for $authenticated_id in an authenticator may not contain
2505 binary zeroes or newlines because the value is written to log lines and
2506 to spool files. There was no check on this. Now the value is run through
2507 the string_printing() function so that such characters are converted to
2508 printable escape sequences.
2510 PH/41 $message_linecount is a new variable that contains the total number of
2511 lines in the message. Compare $body_linecount, which is the count for the
2514 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2515 bounce and delay warning messages, except in certain special cases, which
2518 (a) An SMTP error message from a remote host;
2519 (b) A message specified in a :fail: redirection;
2520 (c) A message specified in a "fail" command in a system filter;
2521 (d) A message specified in a FAIL return from the queryprogram router;
2522 (e) A message specified by the cannot_route_message router option.
2524 In these cases only, Exim does include the error details in bounce and
2525 warning messages. There are also a few cases where bland messages such
2526 as "unrouteable address" or "local delivery error" are given.
2528 PH/43 $value is now also set for the "else" part of a ${run expansion.
2530 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2531 being worked on, but at least Exim now implements the latest version to
2534 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2535 process to complete was treated as a delivery failure, a timeout while
2536 writing the message to the pipe was logged, but erroneously treated as a
2537 successful delivery. Such timeouts include transport filter timeouts. For
2538 consistency with the overall process timeout, these timeouts are now
2539 treated as errors, giving rise to delivery failures by default. However,
2540 there is now a new Boolean option for the pipe transport called
2541 timeout_defer, which, if set TRUE, converts the failures into defers for
2542 both kinds of timeout. A transport filter timeout is now identified in
2545 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2546 systems where "make" and "gmake" are different, calling "gmake" at top
2547 level broke things. I've arranged for the value of $(MAKE) to be passed
2548 from the Makefile to this script so that it can call the same version of
2552 A note about Exim versions 4.44 and 4.50
2553 ----------------------------------------
2555 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2556 changes of various kinds. As a consequence, a big documentation update was
2557 needed. This delayed the release for rather longer than seemed good, especially
2558 in the light of a couple of (minor) security issues. Therefore, the changes
2559 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2560 release. So 4.44 and 4.50 are in effect two different branches that both start
2563 I have left the 4.50 change log unchanged; it contains all the changes since
2564 4.43. The change log for 4.44 is below; many of its items are identical to
2565 those for 4.50. This seems to be the most sensible way to preserve the
2566 historical information.
2572 1. Minor wording change to the doc/README.SIEVE file.
2574 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2575 computation of the current number of files was incorrect.
2577 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2578 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2581 4. Give more explanation in the error message when the command for a transport
2582 filter fails to execute.
2584 5. There are several places where Exim runs a non-Exim command in a
2585 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2586 was being done only for the command run by the queryprogram router. It is
2587 now done for all such subprocesses. The other cases are: ${run, transport
2588 filters, and the commands run by the lmtp and pipe transports.
2590 6. Added CONFIGURE_GROUP build-time option.
2592 7. Some older OS have a limit of 256 on the maximum number of file
2593 descriptors. Exim was using setrlimit() to set 1000 as a large value
2594 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2595 systems. I've change it so that if it can't get 1000, it tries for 256.
2597 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2598 was an oversight, and furthermore, ever since the addition of extra
2599 controls (e.g. 4.43/32), the checks on when to allow different forms of
2600 "control" were broken. There should now be diagnostics for all cases when a
2601 control that does not make sense is encountered.
2603 9. Added the /retain_sender option to "control=submission".
2605 10. $recipients is now available in the predata ACL (oversight).
2607 11. Tidy the search cache before the fork to do a delivery from a message
2608 received from the command line. Otherwise the child will trigger a lookup
2609 failure and thereby defer the delivery if it tries to use (for example) a
2610 cached ldap connection that the parent has called unbind on.
2612 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2613 of $address_data from the recipient verification was clobbered by the
2614 sender verification.
2616 13. The value of address_data from a sender verification is now available in
2617 $sender_address_data in subsequent conditions in the ACL statement.
2619 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2621 15. Added a new option "connect=<time>" to callout options, to set a different
2624 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2625 was its contents. (It was OK if the option was not defined at all.)
2627 17. A "Completed" log line is now written for messages that are removed from
2628 the spool by the -Mrm option.
2630 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2631 information about exactly what failed.
2633 19. Added -dd to debug only the daemon process.
2635 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2636 handles timeouts, both on the server side and network timeouts. Renamed the
2637 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2639 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2640 It is now set to "smtps".
2642 22. $host_address is now set to the target address during the checking of
2643 ignore_target_hosts.
2645 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2646 being passed; this would have caused $sender_host_name to have been used if
2647 matching the list had actually called for a host name (not very likely,
2648 since this list is usually IP addresses). A host name is now passed as
2651 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2652 code that specifies a non-restarting handler (typically sigaction() in
2653 modern systems) in an attempt to fix a rare and obscure crash bug.
2655 25. Narrowed the window for a race in the daemon that could cause it to ignore
2656 SIGCHLD signals. This is not a major problem, because they are used only to
2657 wake it up if nothing else does.
2659 26. A malformed maildirsize file could cause Exim to calculate negative values
2660 for the mailbox size or file count. Odd effects could occur as a result.
2661 The maildirsize information is now recalculated if the size or filecount
2664 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2665 support for a long time. Removed HAVE_SYS_VFS_H.
2667 28. Installed the latest version of exipick from John Jetmore.
2669 29. In an address list, if the pattern was not a regular expression, an empty
2670 subject address (from a bounce message) matched only if the pattern was an
2671 empty string. Non-empty patterns were not even tested. This was the wrong
2672 because it is perfectly reasonable to use an empty address as part of a
2673 database query. An empty address is now tested by patterns that are
2674 lookups. However, all the other forms of pattern expect the subject to
2675 contain a local part and a domain, and therefore, for them, an empty
2676 address still always fails if the pattern is not itself empty.
2678 30. Exim went into a mad DNS loop when attempting to do a callout where the
2679 host was specified on an smtp transport, and looking it up yielded more
2680 than one IP address.
2682 31. Re-factored the code for checking spool and log partition space into a
2683 function that finds that data and another that does the check. The former
2684 is then used to implement four new variables: $spool_space, $log_space,
2685 $spool_inodes, and $log_inodes.
2687 32. The RFC2047 encoding function was originally intended for short strings
2688 such as real names; it was not keeping to the 75-character limit for
2689 encoded words that the RFC imposes. It now respects the limit, and
2690 generates multiple encoded words if necessary. To be on the safe side, I
2691 have increased the buffer size for the ${rfc2047: expansion operator from
2694 33. It is now permitted to omit both strings after an "if" condition; if the
2695 condition is true, the result is "true". As before, when the second string
2696 is omitted, a false condition yields an empty string. This makes it less
2697 cumbersome to write custom ACL and router conditions.
2699 34. Failure to deliver a bounce message always caused it to be frozen, even if
2700 there was an errors_to setting on the router. The errors_to setting is now
2703 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2704 canonical form (fully expanded) before being placed in
2705 $sender_host_address.
2707 36. The table in the code that translates DNS record types into text (T_A to
2708 "A" for instance) was missing entries for NS and CNAME. It is just possible
2709 that this could have caused confusion if both these types were looked up
2710 for the same domain, because the text type is used as part of Exim's
2711 per-process caching. But the chance of anyone hitting this buglet seems
2714 37. The dnsdb lookup has been extended in a number of ways.
2716 (1) There is a new type, "zns", which walks up the domain tree until it
2717 finds some nameserver records. It should be used with care.
2719 (2) There is a new type, "mxh", which is like "mx" except that it returns
2720 just the host names, not the priorities.
2722 (3) It is now possible to give a list of domains (or IP addresses) to be
2723 looked up. The behaviour when one of the lookups defers can be
2724 controlled by a keyword.
2726 (4) It is now possible to specify the separator character for use when
2727 multiple records are returned.
2729 38. The dnslists ACL condition has been extended: it is now possible to supply
2730 a list of IP addresses and/or domains to be looked up in a particular DNS
2733 39. Added log_selector=+queue_time_overall.
2735 40. When running the queue in the test harness, wait just a tad after forking a
2736 delivery process, to get repeatability of debugging output.
2738 41. Include certificate and key file names in error message when GnuTLS fails
2739 to set them up, because the GnuTLS error message doesn't include the name
2740 of the failing file when there is a problem reading it.
2742 42. Allow both -bf and -bF in the same test run.
2744 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2746 44. The "Exiscan patch" is now merged into the mainline Exim source.
2748 45. Sometimes the final signoff response after QUIT could fail to get
2749 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2750 < 0 before doing a fflush(). This bug looks as though it goes back to the
2751 introduction of TLS in release 3.20, but "sometimes" must have been rare
2752 because the tests only now provoked it.
2754 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2755 (this can affect the format of dates).
2757 47. exim_tidydb, when checking for the continued existence of a message for
2758 which it has found a message-specific retry record, was not finding
2759 messages that were in split spool directories. Consequently, it was
2760 deleting retry records that should have stayed in existence.
2762 48. Steve fixed some bugs in eximstats.
2764 49. The SPA authentication driver was not abandoning authentication and moving
2765 on to the next authenticator when an expansion was forced to fail,
2766 contradicting the general specification for all authenticators. Instead it
2767 was generating a temporary error. It now behaves as specified.
2769 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2770 (the order specifies the preference for clients). The order is now AES256,
2771 AES128, 3DES, ARCFOUR128.
2773 51. Small patch to Sieve code - explicitly set From: when generating an
2776 52. Exim crashed if a remote delivery caused a very long error message to be
2777 recorded - for instance if somebody sent an entire SpamAssassin report back
2778 as a large number of 550 error lines. This bug was coincidentally fixed by
2779 increasing the size of one of Exim's internal buffers (big_buffer) that
2780 happened as part of the Exiscan merge. However, to be on the safe side, I
2781 have made the code more robust (and fixed the comments that describe what
2784 53. Now that there can be additional text after "Completed" in log lines (if
2785 the queue_time_overall log selector is set), a one-byte patch to exigrep
2786 was needed to allow it to recognize "Completed" as not the last thing in
2789 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2790 patch that reportedly fixes this has been added. I am not expert enough to
2791 create a test for it. This is what the patch creator wrote:
2793 "I found a little strange behaviour of ldap code when working with
2794 Windows 2003 AD Domain, where users was placed in more than one
2795 Organization Units. When I tried to give exim partial DN, the exit code
2796 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2797 But simultaneously result of request was absolutely normal ldap result,
2798 so I produce this patch..."
2800 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2801 REFERENCE, so I have modified the code to exclude the patch when that macro
2804 55. Some experimental protocols are using DNS PTR records for new purposes. The
2805 keys for these records are domain names, not reversed IP addresses. The
2806 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2807 leaves it alone. Component reversal etc. now happens only for IP addresses.
2810 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2812 57. Double the size of the debug message buffer (to 2048) so that more of very
2813 long debug lines gets shown.
2815 58. The exicyclog utility now does better if the number of log files to keep
2816 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2818 59. Two changes related to the smtp_active_hostname option:
2820 (1) $smtp_active_hostname is now available as a variable.
2821 (2) The default for smtp_banner uses $smtp_active_hostname instead
2822 of $primary_hostname.
2824 60. The host_aton() function is supposed to be passed a string that is known
2825 to be a valid IP address. However, in the case of IPv6 addresses, it was
2826 not checking this. This is a hostage to fortune. Exim now panics and dies
2827 if the condition is not met. A case was found where this could be provoked
2828 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2829 components; fortuitously, this particular loophole had already been fixed
2830 by change 4.50/55 above.
2832 If there are any other similar loopholes, the new check in host_aton()
2833 itself should stop them being exploited. The report I received stated that
2834 data on the command line could provoke the exploit when Exim was running as
2835 exim, but did not say which command line option was involved. All I could
2836 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2837 running as the user.
2840 61. There was a buffer overflow vulnerability in the SPA authentication code
2841 (which came originally from the Samba project). I have added a test to the
2842 spa_base64_to_bits() function which I hope fixes it.
2845 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2846 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2848 63. The daemon start-up calls getloadavg() while still root for those OS that
2849 need the first call to be done as root, but it missed one case: when
2850 deliver_queue_load_max is set with deliver_drop_privilege. This is
2851 necessary for the benefit of the queue runner, because there is no re-exec
2852 when deliver_drop_privilege is set.
2854 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2855 This has been fixed.
2857 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2858 was in use, was not putting the data itself into the right store pool;
2859 consequently, it could be overwritten for a subsequent message in the same
2860 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2863 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2865 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2866 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2867 the function were treating the return as a boolean value, which happened to
2868 work because 0=false and not-0=true, but is not correct code.
2870 68. The host_aton() function was not handling scoped IPv6 addresses (those
2871 with, for example, "%eth0" on the end) correctly.
2873 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2874 negated items (that is, ~something) in unsigned ints. Some compilers
2875 apparently mutter when there is no cast.
2877 70. If an address verification called from an ACL failed, and did not produce a
2878 user-specific message (i.e. there was only a "system" message), nothing was
2879 put in $acl_verify_message. In this situation, it now puts the system
2882 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2883 session; change 4.31/43 added the unwanted input to the log line - except
2884 that it did not do this in the start of session case. It now does.
2886 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2887 This is wrong and can cause the other end to generate a synchronization
2888 error if it is another Exim or anything else that does the synchronization
2889 check. A QUIT command is no longer sent after a timeout.
2891 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2892 during host lookups.
2894 74. The defer_ok option of callout verification was not working if it was used
2895 when verifying addresses in header lines, that is, for this case:
2897 verify = header_sender/callout=defer_ok
2899 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2900 those file descriptors could be used for SMTP connections. If anything
2901 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2902 could be sent to the SMTP client, causing chaos. The daemon now opens
2903 stdin, stdout, and stderr to /dev/null when it puts itself into the
2906 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2907 log by default. The user can override this with suitable Perl magic.
2909 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2910 the log message when discard triggers, was not working for the DATA ACL or
2911 for the non-SMTP ACL.
2913 78. Error message wording change in sieve.c.
2915 79. If smtp_accept_max_per_host was set, the number of connections could be
2916 restricted to fewer than expected, because the daemon was trying to set up
2917 a new connection before checking whether the processes handling previous
2918 connections had finished. The check for completed processes is now done
2919 earlier. On busy systems, this bug wouldn't be noticed because something
2920 else would have woken the daemon, and it would have reaped the completed
2923 80. If a message was submitted locally by a user whose login name contained one
2924 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2925 It caused a spool format error. I have fixed the spool reading code. A
2926 related problem was that the "from" clause in the Received: line became
2927 illegal because of the space(s). It is now covered by ${quote_local_part.
2929 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2932 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2934 83. Patch from Sieve maintainer to fix unterminated string problem in
2935 "vacation" handling.
2937 84. Some minor changes to the Linux configuration files to help with other
2938 OS variants using glibc.
2940 85. One more patch for Sieve to update vacation handling to latest spec.
2943 ----------------------------------------------------
2944 See the note above about the 4.44 and 4.50 releases.
2945 ----------------------------------------------------
2951 1. Change 4.43/35 introduced a bug that caused file counts to be
2952 incorrectly computed when quota_filecount was set in an appendfile
2955 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2956 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2959 3. Give more explanation in the error message when the command for a transport
2960 filter fails to execute.
2962 4. There are several places where Exim runs a non-Exim command in a
2963 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2964 was being done only for the command run by the queryprogram router. It is
2965 now done for all such subprocesses. The other cases are: ${run, transport
2966 filters, and the commands run by the lmtp and pipe transports.
2968 5. Some older OS have a limit of 256 on the maximum number of file
2969 descriptors. Exim was using setrlimit() to set 1000 as a large value
2970 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2971 systems. I've change it so that if it can't get 1000, it tries for 256.
2973 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2974 was an oversight, and furthermore, ever since the addition of extra
2975 controls (e.g. 4.43/32), the checks on when to allow different forms of
2976 "control" were broken. There should now be diagnostics for all cases when a
2977 control that does not make sense is encountered.
2979 7. $recipients is now available in the predata ACL (oversight).
2981 8. Tidy the search cache before the fork to do a delivery from a message
2982 received from the command line. Otherwise the child will trigger a lookup
2983 failure and thereby defer the delivery if it tries to use (for example) a
2984 cached ldap connection that the parent has called unbind on.
2986 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2987 of $address_data from the recipient verification was clobbered by the
2988 sender verification.
2990 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2991 was its contents. (It was OK if the option was not defined at all.)
2993 11. A "Completed" log line is now written for messages that are removed from
2994 the spool by the -Mrm option.
2996 12. $host_address is now set to the target address during the checking of
2997 ignore_target_hosts.
2999 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3000 being passed; this would have caused $sender_host_name to have been used if
3001 matching the list had actually called for a host name (not very likely,
3002 since this list is usually IP addresses). A host name is now passed as
3005 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3006 code that specifies a non-restarting handler (typically sigaction() in
3007 modern systems) in an attempt to fix a rare and obscure crash bug.
3009 15. Narrowed the window for a race in the daemon that could cause it to ignore
3010 SIGCHLD signals. This is not a major problem, because they are used only to
3011 wake it up if nothing else does.
3013 16. A malformed maildirsize file could cause Exim to calculate negative values
3014 for the mailbox size or file count. Odd effects could occur as a result.
3015 The maildirsize information is now recalculated if the size or filecount
3018 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3019 support for a long time. Removed HAVE_SYS_VFS_H.
3021 18. Updated exipick to current release from John Jetmore.
3023 19. Allow an empty sender to be matched against a lookup in an address list.
3024 Previously the only cases considered were a regular expression, or an
3027 20. Exim went into a mad DNS lookup loop when doing a callout where the
3028 host was specified on the transport, if the DNS lookup yielded more than
3031 21. The RFC2047 encoding function was originally intended for short strings
3032 such as real names; it was not keeping to the 75-character limit for
3033 encoded words that the RFC imposes. It now respects the limit, and
3034 generates multiple encoded words if necessary. To be on the safe side, I
3035 have increased the buffer size for the ${rfc2047: expansion operator from
3038 22. Failure to deliver a bounce message always caused it to be frozen, even if
3039 there was an errors_to setting on the router. The errors_to setting is now
3042 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3043 canonical form (fully expanded) before being placed in
3044 $sender_host_address.
3046 24. Updated eximstats to version 1.33
3048 25. Include certificate and key file names in error message when GnuTLS fails
3049 to set them up, because the GnuTLS error message doesn't include the name
3050 of the failing file when there is a problem reading it.
3052 26. Expand error message when OpenSSL has problems setting up cert/key files.
3055 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3056 (this can affect the format of dates).
3058 28. exim_tidydb, when checking for the continued existence of a message for
3059 which it has found a message-specific retry record, was not finding
3060 messages that were in split spool directories. Consequently, it was
3061 deleting retry records that should have stayed in existence.
3063 29. eximstats updated to version 1.35
3064 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3065 1.35 - bugfix such that pie charts by volume are generated correctly
3067 30. The SPA authentication driver was not abandoning authentication and moving
3068 on to the next authenticator when an expansion was forced to fail,
3069 contradicting the general specification for all authenticators. Instead it
3070 was generating a temporary error. It now behaves as specified.
3072 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3073 (the order specifies the preference for clients). The order is now AES256,
3074 AES128, 3DES, ARCFOUR128.
3076 31. Small patch to Sieve code - explicitly set From: when generating an
3079 32. Exim crashed if a remote delivery caused a very long error message to be
3080 recorded - for instance if somebody sent an entire SpamAssassin report back
3081 as a large number of 550 error lines. This bug was coincidentally fixed by
3082 increasing the size of one of Exim's internal buffers (big_buffer) that
3083 happened as part of the Exiscan merge. However, to be on the safe side, I
3084 have made the code more robust (and fixed the comments that describe what
3087 33. Some experimental protocols are using DNS PTR records for new purposes. The
3088 keys for these records are domain names, not reversed IP addresses. The
3089 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3090 leaves it alone. Component reversal etc. now happens only for IP addresses.
3093 34. The host_aton() function is supposed to be passed a string that is known
3094 to be a valid IP address. However, in the case of IPv6 addresses, it was
3095 not checking this. This is a hostage to fortune. Exim now panics and dies
3096 if the condition is not met. A case was found where this could be provoked
3097 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3098 components; fortuitously, this particular loophole had already been fixed
3099 by change 4.50/55 or 4.44/33 above.
3101 If there are any other similar loopholes, the new check in host_aton()
3102 itself should stop them being exploited. The report I received stated that
3103 data on the command line could provoke the exploit when Exim was running as
3104 exim, but did not say which command line option was involved. All I could
3105 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3106 running as the user.
3109 35. There was a buffer overflow vulnerability in the SPA authentication code
3110 (which came originally from the Samba project). I have added a test to the
3111 spa_base64_to_bits() function which I hope fixes it.
3114 36. The daemon start-up calls getloadavg() while still root for those OS that
3115 need the first call to be done as root, but it missed one case: when
3116 deliver_queue_load_max is set with deliver_drop_privilege. This is
3117 necessary for the benefit of the queue runner, because there is no re-exec
3118 when deliver_drop_privilege is set.
3120 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3121 was in use, was not putting the data itself into the right store pool;
3122 consequently, it could be overwritten for a subsequent message in the same
3123 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3126 38. Sometimes the final signoff response after QUIT could fail to get
3127 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3128 < 0 before doing a fflush(). This bug looks as though it goes back to the
3129 introduction of TLS in release 3.20, but "sometimes" must have been rare
3130 because the tests only now provoked it.
3136 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3137 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3138 Now it returns an integer. A number of calls were still expecting a T/F
3139 return. Fortuitously, in all cases, the tests worked in OK situations,
3140 which is the norm. However, things would have gone wrong on any write
3141 failures on the smtp file descriptor. This function is used when sending
3142 messages over SMTP and also when doing verify callouts.
3144 2. When Exim is called to do synchronous delivery of a locally submitted
3145 message (the -odf or -odi options), it no longer closes stderr before doing
3148 3. Implemented the mua_wrapper option.
3150 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3152 5. Implemented the functions header_remove(), header_testname(),
3153 header_add_at_position(), and receive_remove_recipient(), and exported them
3156 6. If an ACL "warn" statement specified the addition of headers, Exim already
3157 inserted X-ACL-Warn: at the start if there was no header name. However, it
3158 was not making this test for the second and subsequent header lines if
3159 there were newlines in the string. This meant that an invalid header could
3160 be inserted if Exim was badly configured.
3162 7. Allow an ACL "warn" statement to add header lines at the start or after all
3163 the Received: headers, as well as at the end.
3165 8. Added the rcpt_4xx retry error code.
3167 9. Added postmaster_mailfrom=xxx to callout verification option.
3169 10. Added mailfrom=xxxx to the callout verification option, for verify=
3172 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3173 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3175 12. Inserted some casts to stop certain compilers warning when using pointer
3176 differences as field lengths or precisions in printf-type calls (mostly
3177 affecting debugging statements).
3179 13. Added optional readline() support for -be (dynamically loaded).
3181 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3182 same clock tick as a message's arrival, so that its received time was the
3183 same as the "first fail" time on the retry record, and that message
3184 remained on the queue past the ultimate address timeout, every queue runner
3185 would try a delivery (because it was past the ultimate address timeout) but
3186 after another failure, the ultimate address timeout, which should have then
3187 bounced the address, did not kick in. This was a "< instead of <=" error;
3188 in most cases the first failure would have been in the next clock tick
3189 after the received time, and all would be well.
3191 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3192 being recognized when the domain list was tested by the match_domain
3193 condition in an expansion string.
3195 16. Added the ${str2b64: operator.
3197 17. Exim was always calling setrlimit() to set a large limit for the number of
3198 processes, without checking whether the existing limit was already
3199 adequate. (It did check for the limit on file descriptors.) Furthermore,
3200 errors from getrlimit() and setrlimit() were being ignored. Now they are
3201 logged to the main and panic logs, but Exim does carry on, to try to do its
3202 job under whatever limits there are.
3204 18. Imported PCRE 5.0.
3206 19. Trivial typo in log message " temporarily refused connection" (the leading
3209 20. If the log selector return_path_on_delivery was set and an address was
3210 redirected to /dev/null, the delivery process crashed because it assumed
3211 that a return path would always be set for a "successful" delivery. In this
3212 case, the whole delivery is bypassed as an optimization, and therefore no
3215 21. Internal re-arrangement: the function for sending a challenge and reading
3216 a response while authentication was assuming a zero-terminated challenge
3217 string. It's now changed to take a pointer and a length, to allow for
3218 binary data in such strings.
3220 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3222 23. Exim was not respecting finduser_retries when seeking the login of the
3223 uid under which it was called; it was always trying 10 times. (The default
3224 setting of finduser_retries is zero.) Also, it was sleeping after the final
3225 failure, which is pointless.
3227 24. Implemented tls_on_connect_ports.
3229 25. Implemented acl_smtp_predata.
3231 26. If the domain in control=submission is set empty, Exim assumes that the
3232 authenticated id is a complete email address when it generates From: or
3233 Sender: header lines.
3235 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3236 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3237 chown and chgrp in /bin and hostname in /usr/bin.
3239 28. Exim was keeping the "process log" file open after each use, just as it
3240 does for the main log. This opens the possibility of it remaining open for
3241 long periods when the USR1 signal hits a daemon. Occasional processlog
3242 errors were reported, that could have been caused by this. Anyway, it seems
3243 much more sensible not to leave this file open at all, so that is what now
3246 29. The long-running daemon process does not normally write to the log once it
3247 has entered its main loop, and it closes the log before doing so. This is
3248 so that log files can straightforwardly be renamed and moved. However,
3249 there are a couple of unusual error situations where the daemon does write
3250 log entries, and I had neglected to close the log afterwards.
3252 30. The text of an SMTP error response that was received during a remote
3253 delivery was being truncated at 512 bytes. This is too short for some of
3254 the long messages that one sometimes sees. I've increased the limit to
3257 31. It is now possible to make retry rules that apply only when a message has a
3258 specific sender, in particular, an empty sender.
3260 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3261 it possible to be selective about when SMTP synchronization is enforced.
3263 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3265 32. Implemented hosts_connection_nolog.
3267 33. Added an ACL for QUIT.
3269 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3272 35. Added mailbox_size and mailbox_filecount to appendfile.
3274 36. Added control = no_multiline_responses to ACLs.
3276 37. There was a bug in the logic of the code that waits for the clock to tick
3277 in the case where the clock went backwards by a substantial amount such
3278 that the microsecond fraction of "now" was more than the microsecond
3279 fraction of "then" (but the whole seconds number was less).
3281 38. Added support for the libradius Radius client library this is found on
3282 FreeBSD (previously only the radiusclient library was supported).
3288 1. When certain lookups returned multiple values in the form name=value, the
3289 quoting of the values was not always being done properly. Specifically:
3290 (a) If the value started with a double quote, but contained no whitespace,
3292 (b) If the value contained whitespace other than a space character (i.e.
3293 tabs or newlines or carriage returns) it was not quoted.
3294 This fix has been applied to the mysql and pgsql lookups by writing a
3295 separate quoting function and calling it from the lookup code. The fix
3296 should probably also be applied to nisplus, ibase and oracle lookups, but
3297 since I cannot test any of those, I have not disturbed their existing code.
3299 2. A hit in the callout cache for a specific address caused a log line with no
3300 reason for rejecting RCPT. Now it says "Previous (cached) callout
3301 verification failure".
3303 3. There was an off-by-one bug in the queryprogram router. An over-long
3304 return line was truncated at 256 instead of 255 characters, thereby
3305 overflowing its buffer with the terminating zero. As well as fixing this, I
3306 have increased the buffer size to 1024 (and made a note to document this).
3308 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3309 when Exim is waiting for an SMTP response from a remote server, Exim
3310 restarts its select() call on the socket, thereby resetting its timeout.
3311 This is not a problem when such interrupts are rare. Somebody set up a cron
3312 job to run exiwhat every 2 minutes, which is less than the normal select()
3313 timeout (5 or 10 minutes). This meant that the select() timeout never
3314 kicked in because it was always reset. I have fixed this by comparing the
3315 time when an interrupt arrives with the time at the start of the first call
3316 to select(). If more time than the timeout has elapsed, the interrupt is
3317 treated as a timeout.
3319 5. Some internal re-factoring in preparation for the addition of Sieve
3320 extensions (by MH). In particular, the "personal" test is moved to a
3321 separate function, and given an option for scanning Cc: and Bcc: (which is
3322 not set for Exim filters).
3324 6. When Exim created an email address using the login of the caller as the
3325 local part (e.g. when creating a From: or Sender: header line), it was not
3326 quoting the local part when it contained special characters such as @.
3328 7. Installed new OpenBSD configuration files.
3330 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3331 try to make them clearer.
3333 9. Callout options, other than the timeout value, were being ignored when
3334 verifying sender addresses in header lines. For example, when using
3336 verify = header_sender/callout=no_cache
3338 the cache was (incorrectly) being used.
3340 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3341 only the Cygwin environment.
3343 11. When return_path_on_delivery was set as a log selector, if different remote
3344 addresses in the same message used different return paths and parallel
3345 remote delivery occurred, the wrong values would sometimes be logged.
3346 (Whenever a remote delivery process finished, the return path value from
3347 the most recently started remote delivery process was logged.)
3349 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3350 lines when AUTH and/or TLS are in use. This is the "received protocol"
3351 field. Exim used to use "asmtp" for authenticated SMTP, without any
3352 indication (in the protocol name) for TLS use. Now it follows the RFC and
3353 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3354 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3355 names appear in log lines as well as in Received: header lines.
3357 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3358 extensions, and comparison tests, and to fix some bugs.
3360 14. Changes to the "personal" filter test:
3362 (1) The test was buggy in that it was just doing the equivalent of
3363 "contains" tests on header lines. For example, if a user's address was
3364 anne@some.where, the "personal" test would incorrectly be true for
3366 To: susanne@some.where
3368 This test is now done by extracting each address from the header in turn,
3369 and checking the entire address. Other tests that are part of "personal"
3370 are now done using regular expressions (for example, to check local parts
3371 of addresses in From: header lines).
3373 (2) The list of non-personal local parts in From: addresses has been
3374 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3375 taken from the Sieve specification recommendations.
3377 (3) If the message contains any header line starting with "List-" it is
3378 treated as non-personal.
3380 (4) The test for "circular" in the Subject: header line has been removed
3381 because it now seems ill-conceived.
3383 15. Minor typos in src/EDITME comments corrected.
3385 16. Installed latest exipick from John Jetmore.
3387 17. If headers_add on a router specified a text string that was too long for
3388 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3389 of string_sprintf() is now avoided.
3391 18. $message_body_size was not set (it was always zero) when running the DATA
3392 ACL and the local_scan() function.
3394 19. For the "mail" command in an Exim filter, no default was being set for
3395 the once_repeat time, causing a random time value to be used if "once" was
3396 specified. (If the value happened to be <= 0, no repeat happened.) The
3397 default is now 0s, meaning "never repeat". The "vacation" command was OK
3398 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3399 (I found it when inspecting the code).
3401 20. There is now an overall timeout for performing a callout verification. It
3402 defaults to 4 times the callout timeout, which applies to individual SMTP
3403 commands during the callout. The overall timeout applies when there is more
3404 than one host that can be tried. The timeout is checked before trying the
3405 next host. This prevents very long delays if there are a large number of
3406 hosts and all are timing out (e.g. when the network connections are timing
3407 out). The value of the overall timeout can be changed by specifying an
3408 additional sub-option for "callout", called "maxwait". For example:
3410 verify = sender/callout=5s,maxwait=20s
3412 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3413 to the end before writing, but this should make it even safer).
3415 22. Exim was forgetting that it had advertised PIPELINING for the second and
3416 subsequent messages on an SMTP connection. It was also not resetting its
3417 memory on STARTTLS and an internal HELO.
3419 23. When Exim logs an SMTP synchronization error within a session, it now
3420 records whether PIPELINING has been advertised or not.
3422 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3423 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3424 rather than long int.
3426 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3428 26. Added the never_mail option to autoreply.
3434 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3435 crash if the getsockname() call failed; this can happen if a connection is
3436 closed very soon after it is established. The problem was simply in the
3437 order in which certain operations were done, causing Exim to try to write
3438 to the SMTP stream before it had set up the file descriptor. The bug has
3439 been fixed by making things happen in the correct order.
3445 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3446 before the connection was closed, thus losing the rejection response.
3448 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3449 some early Solaris releases, but causes trouble in current releases where
3450 socklen_t is defined.
3452 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3455 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3458 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3459 that releasing the top of it at the end releases what was used for sub-
3460 expansions (unless the block got too big). However, discard this block if
3461 the first thing is a variable or header, so that we can use its block when
3462 it is dynamic (useful for very large $message_headers, for example).
3464 6. Lookups now cache *every* query, not just the most recent. A new, separate
3465 store pool is used for this. It can be recovered when all lookup caches are
3466 flushed. Lookups now release memory at the end of their result strings.
3467 This has involved some general refactoring of the lookup sources.
3469 7. Some code has been added to the store_xxx() functions to reduce the amount
3470 of flapping under certain conditions.
3472 8. log_incoming_interface used to affect only the <= reception log lines. Now
3473 it causes the local interface and port to be added to several more SMTP log
3474 lines, for example "SMTP connection from", and rejection lines.
3476 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3478 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3480 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3481 could be overwritten at the end of the current message (or the start of a
3482 new message if it was set in a HELO ACL). The value is now preserved for
3483 the duration of the SMTP connection.
3485 12. If a transport had a headers_rewrite setting, and a matching header line
3486 contained an unqualified address, that address was qualified, even if it
3487 did not match any rewriting rules. The underlying bug was that the values
3488 of the flags that permit the existence of unqualified sender and recipient
3489 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3490 non-local messages, and by -bnq for local messages) were not being
3491 preserved with the message after it was received.
3493 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3494 "next input=" as part of the text comprising the host identity instead of
3495 the correct text. The code was using the same buffer for two different
3496 strings. However, depending on which order the printing function evaluated
3497 its arguments, the bug did not always show up. Under Linux, for example, my
3498 test suite worked just fine.
3500 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3501 doesn't work with some older versions of Perl. It has been changed to "my",
3502 which in any case is probably the better facility to use.
3504 15. A really picky compiler found some instances of statements for creating
3505 error messages that either had too many or two few arguments for the format
3508 16. The size of the buffer for calls to the DNS resolver has been increased
3509 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3510 for addresses that have a lot of PTR records. This alleviates a problem; it
3511 does not fully solve it.
3513 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3514 buffer now truncates the list and logs the incident, which is the same
3515 action as happens when Exim is looking up a host name and its aliases.
3516 Previously in this situation something unpredictable would happen;
3517 sometimes it was "internal error: store_reset failed".
3519 18. If a server dropped the connection unexpectedly when an Exim client was
3520 using GnuTLS and trying to read a response, the client delivery process
3521 crashed while trying to generate an error log message.
3523 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3524 string, for example:
3526 warn message = H1: something\nH2: something
3528 the text was added as a single header line from Exim's point of view
3529 though it ended up OK in the delivered message. However, searching for the
3530 second and subsequent header lines using $h_h2: did not work. This has been
3531 fixed. Similarly, if a system filter added multiple headers in this way,
3532 the routers could not see them.
3534 20. Expanded the error message when iplsearch is called with an invalid key to
3535 suggest using net-iplsearch in a host list.
3537 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3538 ACLs are no longer actually imposed (and a message to that effect is
3541 22. If a "gecos" field in a passwd entry contained escaped characters, in
3542 particular, if it contained a \" sequence, Exim got it wrong when building
3543 a From: or a Sender: header from that name. A second bug also caused
3544 incorrect handling when an unquoted " was present following a character
3545 that needed quoting.
3547 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3548 was not being matched caselessly.
3550 24. Arranged for all hyphens in the exim.8 source to be escaped with
3553 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3554 properly. Recipient callout cache records were still being keyed to include
3555 the sender, even when use_sender was set false. This led to far more
3556 callouts that were necessary. The sender is no longer included in the key
3557 when use_sender is false.
3559 26. Added "control = submission" modifier to ACLs.
3561 27. Added the ${base62d: operator to decode base 62 numbers.
3563 28. dnsdb lookups can now access SRV records.
3565 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3566 the configuration file.
3568 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3569 (-v) mode. This makes the output for a verbose queue run more intelligible.
3571 31. Added a use_postmaster feature to recipient callouts.
3573 32. Added the $body_zerocount variable, containing the number of binary zero
3574 bytes in the message body.
3576 33. The time of last modification of the "new" subdirectory is now used as the
3577 "mailbox time last read" when there is a quota error for a maildir
3580 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3582 35. Added +ignore_unknown as a special item in host lists.
3584 36. Code for decoding IPv6 addresses in host lists is now included, even if
3585 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3586 address was recognized as an IP address, but was then not correctly decoded
3587 into binary, causing unexpected and incorrect effects when compared with
3594 1. Very minor rewording of debugging text in manualroute to say "list of
3595 hosts" instead of "hostlist".
3597 2. If verify=header_syntax was set, and a header line with an unqualified
3598 address (no domain) and a large number of spaces between the end of the
3599 name and the colon was received, the reception process suffered a buffer
3600 overflow, and (when I tested it) crashed. This was caused by some obsolete
3601 code that should have been removed. The fix is to remove it!
3603 3. When running in the test harness, delay a bit after writing a bounce
3604 message to get a bit more predictability in the log output.
3606 4. Added a call to search_tidyup() just before forking a reception process. In
3607 theory, someone could use a lookup in the expansion of smtp_accept_max_
3608 per_host which, without the tidyup, could leave open a database connection.
3610 5. Added the variables $recipient_data and $sender_data which get set from a
3611 lookup success in an ACL "recipients" or "senders" condition, or a router
3612 "senders" option, similar to $domain_data and $local_part_data.
3614 6. Moved the writing of debug_print from before to after the "senders" test
3617 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3618 problems for message scanning, either using a data ACL, or using
3619 local_scan() because the Received: header was not generated till after they
3620 were called (in order to set the time as the time of reception completion).
3621 I have revised the way this works. The header is now generated after the
3622 body is received, but before the ACL or local_scan() are called. After they
3623 are run, the timestamp in the header is updated.
3629 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3630 before starting a queue runner without re-exec. This happened only when
3631 deliver_drop_privilege was set or when the Exim user was set to root. The
3632 effect of the bug was that timeouts during subsequent deliveries caused
3633 crashes instead of being properly handled. The handler is now left at its
3634 default (and expected) setting.
3636 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3637 message, again when deliver_drop_privilege is set or Exim is run as root.
3638 The bug described in (1) was not present in this case, but the tidying up
3639 of the other signals was missing. I have made the two cases consistent.
3641 3. The ignore_target_hosts setting on a manualroute router was being ignored
3642 for hosts that were looked up using the /MX notation.
3644 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3647 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3648 operated on the sender address. After changing the $sender_address to <>
3649 for the sender address verify, Exim was re-instated it as the original
3650 (before rewriting) address, but remembering that it had rewritten it, so it
3651 wasn't rewriting it again. This bug also had the effect of breaking the
3652 sender address verification caching when the sender address was rewritten.
3654 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3655 This has been changed so that if the ip literal address matches
3656 ignore_target_hosts, the router declines.
3658 7. Added expansion conditions match_domain, match_address, and match_local_
3659 part (NOT match_host).
3661 8. The placeholder for the Received: header didn't have a length field set.
3663 9. Added code to Exim itself and to exim_lock to test for a specific race
3664 condition that could lead to file corruption when using MBX delivery. The
3665 issue is with the lockfile that is created in /tmp. If this file is removed
3666 after a process has opened it but before that process has acquired a lock,
3667 there is the potential for a second process to recreate the file and also
3668 acquire a lock. This could lead to two Exim processes writing to the file
3669 at the same time. The added code performs the same test as UW imapd; it
3670 checks after acquiring the lock that its file descriptor still refers to
3671 the same named file.
3673 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3674 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3677 11. Added the smtp_active_hostname option. If used, this will typically be made
3678 to depend on the incoming interface address. Because $interface_address is
3679 not set up until the daemon has forked a reception process, error responses
3680 that can happen earlier (such as "too many connections") no longer contain
3683 12. If an expansion in a condition on a "warn" statement fails because a lookup
3684 defers, the "warn" statement is abandoned, and the next ACL statement is
3685 processed. Previously this caused the whole ACL to be aborted.
3687 13. Added the iplsearch lookup type.
3689 14. Added ident_timeout as a log selector.
3691 15. Added tls_certificate_verified as a log selector.
3693 16. Added a global option tls_require_ciphers (compare the smtp transport
3694 option of the same name). This controls incoming TLS connections.
3696 17. I finally figured out how to make tls_require_ciphers do a similar thing
3697 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3698 before starting the TLS session.
3700 18. Tabs are now shown as \t in -bP output.
3702 19. If the log selector return_path_on_delivery was set, Exim crashed when
3703 bouncing a message because it had too many Received: header lines.
3705 20. If two routers both had headers_remove settings, and the first one included
3706 a superfluous trailing colon, the final name in the first list and the
3707 first name in the second list were incorrectly joined into one item (with a
3708 colon in the middle).
3714 1. Added -C and -D options to the exinext utility, mainly to make it easier
3715 to include in the automated testing, but these could be helpful when
3716 multiple configurations are in use.
3718 2. The exinext utility was not formatting the output nicely when there was
3719 an alternate port involved in the retry record key, nor when there was a
3720 message id as well (for retries that were specific to a specific message
3721 and a specific host). It was also confused by IPv6 addresses, because of
3722 the additional colons they contain. I have fixed the IPv4 problem, and
3723 patched it up to do a reasonable job for IPv6.
3725 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3726 delivery, the log line now contains "pipelined" if PIPELINING was used.
3728 4. An SMTP transport process used to panic and die if the bind() call to set
3729 an explicit outgoing interface failed. This has been changed; it is now
3730 treated in the same way as a connect() failure.
3732 5. A reference to $sender_host_name in the part of a conditional expansion
3733 that was being skipped was still causing a DNS lookup. This no longer
3736 6. The def: expansion condition was not recognizing references to header lines
3737 that used bh_ and bheader_.
3739 7. Added the _cache feature to named lists.
3741 8. The code for checking quota_filecount in the appendfile transport was
3742 allowing one more file than it should have been.
3744 9. For compatibility with Sendmail, the command line option
3752 and sets the incoming protocol and host name (for trusted callers). The
3753 host name and its colon can be omitted when only the protocol is to be set.
3754 Note the Exim already has two private options, -pd and -ps, that refer to
3755 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3756 "s", but I don't think that's a major issue.
3758 10. A number of refactoring changes to the code, none of which should affect
3761 (a) The number of logging options was getting close to filling up the
3762 32-bit word that was used as a bit map. I have split them into two classes:
3763 those that are passed in the argument to log_write(), and those that are
3764 only ever tested independently outside of that function. These are now in
3765 separate 32-bit words, so there is plenty of room for expansion again.
3766 There is no change in the user interface or the logging behaviour.
3768 (b) When building, for example, log lines, the code previously used a
3769 macro that called string_cat() twice, in order to add two strings. This is
3770 not really sufficiently general. Furthermore, there was one instance where
3771 it was actually wrong because one of the argument was used twice, and in
3772 one call a function was used. (As it happened, calling the function twice
3773 did not affect the overall behaviour.) The macro has been replaced by a
3774 function that can join an arbitrary number of extra strings onto a growing
3777 (c) The code for expansion conditions now uses a table and a binary chop
3778 instead of a serial search (which was left over from when there were very
3779 few conditions). Also, it now recognizes conditions like "pam" even when
3780 the relevant support is not compiled in: a suitably worded error message is
3781 given if an attempt is made to use such a condition.
3783 11. Added ${time_interval:xxxxx}.
3785 12. A bug was causing one of the ddress fields not to be passed back correctly
3786 from remote delivery subprocesses. The field in question was not being
3787 subsequently used, so this caused to problems in practice.
3789 13. Added new log selectors queue_time and deliver_time.
3791 14. Might have fixed a bug in maildirsizefile handling that threw up
3792 "unexpected character" debug warnings, and recalculated the data
3793 unnecessarily. In any case, I expanded the warning message to give more
3796 15. Added the message "Restricted characters in address" to the statements in
3797 the default ACL that block characters like @ and % in local parts.
3799 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3800 Three changes have been made:
3802 (a) There was a serious bug; a negative response to MAIL caused the whole
3803 recipient domain to be cached as invalid, thereby blocking all messages
3804 to all local parts at the same domain, from all senders. This bug has
3805 been fixed. The domain is no longer cached after a negative response to
3806 MAIL if the sender used is not empty.
3808 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3811 (c) A new callout option, "use_sender" has been added for people who want
3812 the modified behaviour.
3818 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3821 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3822 indeed breaks things for older releases.
3824 3. Added additional logging to the case where there is a problem reading data
3825 from a filter that is running in a subprocess using a pipe, in order to
3826 try to track down a specific problem.
3828 4. Testing facility fudge: when running in the test harness and attempting
3829 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3830 getting "No route to host". Convert this to a timeout.
3832 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3835 6. Some OS don't have socklen_t but use size_t instead. This affects the
3836 fifth argument of getsockopt() amongst other things. This is now
3837 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3838 can be set for individual OS. I have set it for SunOS5, OSF1, and
3839 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3840 some earlier ones do not.
3842 7. Change 4.30/15 was not doing the test caselessly.
3844 8. The standard form for an IPv6 address literal was being rejected by address
3845 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3846 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3847 this, as well as the form without the "IPv6" on the front (but only when
3848 address literals are enabled, of course).
3850 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3852 10. Exim crashed if a message with an empty sender address specified by -f
3853 encountered a router with an errors_to setting. This could be provoked only
3854 by a command such as
3858 where an empty string was supplied; "<>" did not hit this bug.
3860 11. Installed PCRE release 4.5.
3862 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3863 remained set. It is now erased.
3865 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3866 times from message ids (which are base 36 rather than the normal 62).
3868 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3869 were being counted as actual protocol errors, and logged if the log
3870 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3871 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3872 and DATA following a set of rejected RCPTs do not count as protocol errors.
3873 In other words, Exim assumes they were pipelined, though this may not
3874 actually be the case. Of course, in all cases the client gets an
3875 appropriate error code.
3877 15. If a lookup fails in an ACL condition, a message about the failure may
3878 be available; it is used if testing the ACL cannot continue, because most
3879 such messages specify what the cause of the deferral is. However, some
3880 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3881 that caused an old message to be retained and used if a later statement
3882 caused a defer, replacing the real cause of the deferral.
3884 16. If an IP address had so many PTR records that the DNS lookup buffer
3885 was not large enough to hold them, Exim could crash while trying to process
3886 the truncated data. It now detects and logs this case.
3888 17. Further to 4.21/58, another change has been made: if (and only if) the
3889 first line of a message (the first header line) ends with CRLF, a bare LF
3890 in a subsequent header line has a space inserted after it, so as not to
3891 terminate the header.
3893 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3894 unnecessarily, used atoi() instead of strtol(), and didn't check the
3895 termination when getting file sizes from file names by regex.
3897 19. Completely re-implemented the support for maildirsize files, in the light
3898 of a number of problems with the previous contributed implementation
3899 (4.30/29). In particular:
3901 . If the quota is zero, the maildirsize file is maintained, but no quota is
3904 . If the maildir directory does not exist, it is created before any attempt
3905 to write a maildirsize file.
3907 . The quota value in the file is just a cache; if the quota is changed in
3908 the transport, the new value overrides.
3910 . A regular expression is available for excluding directories from the
3913 20. The autoreply transport checks the characters in options that define the
3914 message's headers; it allows continued headers, but it was checking with
3915 isspace() after an embedded newline instead of explicitly looking for a
3918 21. If all the "regular" hosts to which an address was routed had passed their
3919 expiry times, and had not reached their retry times, the address was
3920 bounced, even if fallback hosts were defined. Now Exim should go on to try
3923 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3924 equivalent code in the SMTP transport. Some hosts send humungous responses
3925 to HELO/EHLO, more than 1024 it seems.
3927 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3928 gives compiler warnings in some environments. I've now done it "properly",
3931 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3932 (because of problems with the built-in one) was declared to return uschar *
3933 instead of char *, causing compiler failure.
3935 25. Fixed a file descriptor leak when processing alias/forward files.
3937 26. Fixed a minor format string issue in dbfn.c.
3939 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3941 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3942 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3945 29. When a sender address is verified, it is cached, to save repeating the test
3946 when there is more than one recipient in a message. However, when the
3947 verification involves a callout, it is possible for different callout
3948 options to be set for different recipients. It is too complicated to keep
3949 track of this in the cache, so now Exim always runs a verification when a
3950 callout is required, relying on the callout cache for the optimization.
3951 The overhead is duplication of the address routing, but this should not be
3954 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3955 to be verified with callout=postmaster, and the main callout worked but the
3956 postmaster check failed, the verification correctly failed. However, if a
3957 subsequent RCPT command asked for sender verification *without* the
3958 postmaster check, incorrect caching caused this verification also to fail,
3961 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3962 it was not caching the DNS options (qualify_single, search_parents) that
3963 were used when the lookup failed. A subsequent lookup with different
3964 options therefore always gave the same answer, though there were cases
3965 where it should not have. (Example: a "domains = !$mx_any" option on a
3966 dnslookup router: the "domains" option is always processed without any
3967 widening, but the router might have qualify_single set.) Now Exim uses the
3968 cached value only when the same options are set.
3970 32. Added John Jetmore's "exipick" utility to the distribution.
3972 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3973 than a timeout (e.g. a certificate is required, and is not provided), an
3974 Exim server now closes the connection immediately. Previously it waited for
3975 the client to close - but if the client is SSL, it seems that they each
3976 wait for each other, leading to a delay before one of them times out.
3978 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3979 maintained 0.8.x compatibility because I don't think many are using it, and
3980 it is clearly obsolete.
3982 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3985 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3986 client certificate was expired. A simple patch fixes this, though I don't
3987 understand the full logic of why the verify callback is called multiple
3990 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3991 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3992 which causes problems with some clients (such as the Certicom SSL Plus
3993 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3994 disables the coutermeasure allowing Eudora to connect."
3996 38. Exim was not checking that a write() to a log file succeeded. This could
3997 lead to Bad Things if a log got too big, in particular if it hit a file
3998 size limit. Exim now panics and dies if it cannot write to a log file, just
3999 as it does if it cannot open a log file.
4001 39. Modified OS/Makefile-Linux so that it now contains
4003 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4005 The two -D definitions ensure that Exim is compiled with large file
4006 support, which makes it possible to handle log files that are bigger than
4009 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4010 instance) a domain was checked against a named list that involved a lookup,
4011 causing $domain_data to be set, then another domain was checked against the
4012 same list, then the first domain was re-checked, the value of $domain_data
4013 after the final check could be wrong. In particular, if the second check
4014 failed, it could be set empty. This bug probably also applied to
4017 41. The strip_trailing_dot option was not being applied to the address given
4018 with the -f command-line option.
4020 42. The code for reading a message's header from the spool was incrementing
4021 $received_count, but never initializing it. This meant that the value was
4022 incorrect (doubled) while delivering a message in the same process in which
4023 it was received. In the most common configuration of Exim, this never
4024 happens - a fresh exec is done - but it can happen when
4025 deliver_drop_privilege is set.
4027 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4028 it now includes up to 150 characters of the unexpected data in the log
4031 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4032 and building data strings. The size of both of these buffers was 10 000
4033 bytes - far larger than anybody would *ever* want, thought I. Needless to
4034 say, somebody hit the limit. I have increased the maximum line length to
4035 20 000 and the maximum data length of concatenated lines to 100 000. I have
4036 also fixed two bugs, because there was no checking on these buffers. Tsk,
4037 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4038 buffer is too small.
4040 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4041 lsearch lookups. Now it does.
4043 46. When parsing a route_list item in a manualroute router, a fixed-length
4044 buffer was used for the list of hosts. I made this 1024 bytes long,
4045 thinking that nobody would ever have a list of hosts that long. Wrong.
4046 Somebody had a whole pile of complicated expansion conditions, and the
4047 string was silently truncated, leading to an expansion error. It turns out
4048 that it is easier to change to an unlimited length (owing to other changes
4049 that have happened since this code was originally written) than to build
4050 structure for giving a limitation error. The length of the item that
4051 expands into the list of hosts is now unlimited.
4053 47. The lsearch lookup could not handle data where the length of text line was
4054 more than 4095 characters. Such lines were truncated, leading to shortened
4055 data being returned. It should now handle lines of any length.
4057 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4058 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4061 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4062 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4063 no longer happens in comment lines. A list of replacements is now placed
4064 at the head of all of the source files, except those whose only change is
4065 to replace PERL_COMMAND in the very first #! line.
4067 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4068 messages on the queue, with a bottom-up merge sort, using code contributed
4069 by Michael Haardt. This should make operations like -bp somewhat faster on
4070 large queues. It won't affect queue runners, except when queue_run_in_order
4073 51. Installed eximstats 1.31 in the distribution.
4075 52. Added support for SRV lookups to the dnslookup router.
4077 53. If an ACL referred to $message_body or $message_body_end, the value was not
4078 reset for any messages that followed in the same SMTP session.
4080 54. The store-handling optimization for building very long strings was not
4081 differentiating between the different store pools. I don't think this
4082 actually made any difference in practice, but I've tidied it.
4084 55. While running the routers to verify a sender address, $sender_address
4085 was still set to the sender address. This is wrong, because when routing to
4086 send a bounce to the sender, it would be empty. Therefore, I have changed
4087 it so that, while verifying a sender address, $sender_address is set to <>.
4088 (There is no change to what happens when verifying a recipient address.)
4090 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4091 target A or AAAA records (if not already returned) without resetting the
4092 qualify_single or search_parents options of the DNS resolver. These are
4093 inappropriate in this case because the targets of MX and SRV records must
4094 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4095 target that, when qualified, matched something in the local domain. These
4096 two options are now turned off when doing these lookups.
4098 57. It seems that at least some releases of Reiserfs (which does not have the
4099 concept of a fixed number of inodes) returns zero and not -1 for the
4100 number of available inodes. This interacted badly with check_spool_inodes,
4101 which assumed that -1 was the "no such thing" setting. What I have done is
4102 to check that the total number of inodes is greater than zero before doing
4103 the test of how many are available.
4105 58. When a "warn" ACL statement has a log_message modifier, the message is
4106 remembered, and not repeated. This is to avoid a lot of repetition when a
4107 message has many recipients that cause the same warning to be written.
4108 Howewer, Exim was preserving the list of already written lines for an
4109 entire SMTP session, which doesn't seem right. The memory is now reset if a
4110 new message is started.
4112 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4113 debugging output unless log_rewrite was also set.
4115 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4116 of (int)(handle) when we know that handle contains (void *)(-1).
4118 61. The Exim daemon panic-logs an error return when it closes the incoming
4119 connection. However "connection reset by peer" seems to be common, and
4120 isn't really an error worthy of noting specially, so that particular error
4123 62. When Exim is trying to find all the local interfaces, it used to panic and
4124 die if the ioctl to get the interface flags failed. However, it seems that
4125 on at least one OS (Solaris 9) it is possible to have an interface that is
4126 included in the list of interfaces, but for which you get a failure error
4127 for this call. This happens when the interface is not "plumbed" into a
4128 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4129 failure of the "get flags" call assumes that the interface is down.
4131 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4132 makes life easier for people who are doing arithmetic on fields extracted
4133 from dates, where you often get leading zeros that should not be
4134 interpreted as octal.
4136 64. Added qualify_domain to the redirect router, to override the global
4139 65. If a pathologically long header line contained very many addresses (the
4140 report of this problem mentioned 10 000) and each of them was rewritten,
4141 Exim could use up a very large amount of memory. (It kept on making new
4142 copies of the header line as it rewrote, and never released the old ones.)
4143 At the expense of a bit more processing, the header rewriting function has
4144 been changed so that it no longer eats memory in this way.
4146 66. The generation of the Received: header has been moved from the time that a
4147 message starts to be received, to the time that it finishes. The timestamp
4148 in the Received: header should now be very close to that of the <= log
4149 line. There are two side-effects of this change:
4151 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4152 logged header lines no longer include the local Received: line, because
4153 it has not yet been created. The same applies to a copy of the message
4154 that is returned to a non-SMTP sender when a message is rejected.
4156 (b) When a filter file is tested using -bf, no additional Received: header
4157 is added to the test message. After some thought, I decided that this
4160 This change does not affect the value of $received_for. It is still set
4161 after address rewriting, but before local_scan() is called.
4163 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4165 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4166 gave an unhelpful panic error message, and a defer error. I have managed to
4167 change this behaviour so that it now rejects any supplied certificate,
4168 which seems right, as the list of acceptable certificates is empty.
4170 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4171 gave an unhelpful defer error. I have not managed to make this reject any
4172 supplied certificates, but the error message it gives is "no certificate
4173 supplied", which is not helpful.
4175 70. exigrep's output now also includes lines that are not associated with any
4176 message, but which match the given pattern. Implemented by a patch from
4177 Martin Sluka, which also tidied up the Perl a bit.
4179 71. Recipient callout verification, like sender verification, was using <> in
4180 the MAIL FROM command. This isn't really the right thing, since the actual
4181 sender may affect whether the remote host accepts the recipient or not. I
4182 have changed it to use the actual sender in the callout; this means that
4183 the cache record is now keyed on a recipient/sender pair, not just the
4184 recipient address. There doesn't seem to be a real danger of callout loops,
4185 since a callout by the remote host to check the sender would use <>.
4186 [SEE ABOVE: changed after hitting problems.]
4188 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4189 temporary errors. However, in the case of such a code being given after
4190 the end of a data transmission (i.e. after ".") Exim was failing to write
4191 a retry record for the message. (Yes, there was some broken host that was
4192 actually sending 8xx at this point.)
4194 73. An unknown lookup type in a host list could cause Exim to panic-die when
4195 the list was checked. (An example that provoked this was putting <; in the
4196 middle of a list instead of at the start.) If this happened during a DATA
4197 ACL check, a -D file could be left lying around. This kind of configuration
4198 error no longer causes Exim to die; instead it causes a defer errror. The
4199 incident is still logged to the main and panic logs.
4201 74. Buglet left over from Exim 3 conversion. The message "too many messages
4202 in one connection" was written to the rejectlog but not the mainlog, except
4203 when address rewriting (yes!) was being logged.
4205 75. Added write_rejectlog option.
4207 76. When a system filter was run not as root (that is, when system_filter_user
4208 was set), the values of the $n variables were not being returned to the
4209 main process; thus, they were not subsequently available in the $sn
4212 77. Added +return_path_on_delivery log selector.
4214 78. A connection timeout was being treated differently from recipients deferred
4215 when testing hosts_max_try with a message that was older than the host's
4216 retry timeout. (The host should not be counted, thus allowing all hosts to
4217 be tried at least once before bouncing.) This may have been the cause of an
4218 occasionally reported bug whereby a message would remain on the queue
4219 longer than the retry timeout, but would be bounced if a delivery was
4220 forced. I say "may" because I never totally pinned down the problem;
4221 setting up timeout/retry tests is difficult. See also the next item.
4223 79. The ultimate address timeout was not being applied to errors that involved
4224 a combination of host plus message (for example, a timeout on a MAIL
4225 command). When an address resolved to a number of possible hosts, and they
4226 were not all tried for each delivery (e.g. because of hosts_max_try), a
4227 message could remain on the queue longer than the retry timeout.
4229 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4232 81. Fixed an obscure SMTP outgoing bug which required at least the following
4233 conditions: (a) there was another message waiting for the same server;
4234 (b) the server returned 5xx to all RCPT commands in the first message so
4235 that the message was not completed; (c) the server dropped the connection
4236 or gave a negative response to the RSET that Exim sends to abort the
4237 transaction. The observed case was a dropped connection after DATA that had
4238 been sent in pipelining mode. That is, the server had advertised PIPELINING
4239 but was not implementing it correctly. The effect of the bug was incorrect
4240 behaviour, such as trying another host, and this could lead to a crash.
4246 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4247 and daemon.c were passed as pointers to ints; they should have been
4248 pointers to socklen_t variables (which are typically unsigned ints).
4250 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4253 3. Fixed a really odd bug that affected only the testing scheme; patching a
4254 certain fixed string in the binary changed the value of another string that
4255 happened to be identical to the end of the original first string.
4257 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4258 name", it returns that address as the IP address. On some operating
4259 systems (e.g. Solaris), it also passes back the IP address string as the
4260 "host name". However, on others (e.g. Linux), it passes back an empty
4261 string. Exim wasn't checking for this, and was changing the host name to an
4262 empty string, assuming it had been canonicized.
4264 5. Although rare, it is permitted to have more than one PTR record for a given
4265 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4266 all the names associated with an address, because they do in Solaris.
4267 However, it seems that they do not in Linux for data that comes from the
4268 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4269 I found this out when I moved to a new Linux workstation and tried to run
4270 the Exim test suite.
4272 To get round this problem I have changed the code so that it now does its
4273 own call to the DNS to look up PTR records when searching for a host name.
4274 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4275 addresses that are only in /etc/hosts are still found.
4277 This behaviour is, however, controlled by an option called host_lookup_
4278 order, which defaults to "bydns:byaddr". If people want to use the other
4279 order, or indeed, just use one or the other means of lookup, they can
4280 specify it in this variable.
4282 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4283 some operating systems, this comes back from gethostbyaddr() as an empty
4284 string, and this is what Exim used to test for. However, it seems that in
4285 other systems, "." is yielded. Exim now tests for this case too.
4287 7. The values of check_spool_space and check_log_space are now held internally
4288 as a number of kilobytes instead of an absolute number of bytes. If a
4289 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4290 kilobyte. This means that much larger values can be stored.
4292 8. Exim monitor: an attempt to get the action menu when not actually pointing
4293 at a message produces an empty menu entitled "No message selected". This
4294 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4295 no entries in it ("Shell widget menu has zero width and/or height"). So I
4296 have added a single, blank menu entry in this case.
4298 9. Added ${quote_local_part.
4300 10. MIME decoding is now applied to the contents of Subject: header lines when
4303 11. Now that a reference to $sender_host_address automatically causes a reverse
4304 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4305 host lookup before query-style lookups in lists that might use this
4306 variable. This has therefore been abolished, and the "net-" prefix is no
4307 longer necessary for query-style lookups.
4309 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4310 have been a typo for LFLAGS, so it has been changed.
4312 13. The install script calls Exim with "-C /dev/null" in order to find the
4313 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4314 to be output. Howeve, since Exim outputs its version number before the
4315 error, it didn't break the script. It just looked ugly. I fixed this by
4316 always allowing "-C /dev/null" if the caller is root.
4318 14. Ignore overlarge ACL variable number when reading spool file - insurance
4319 against a later release with more variables having written the file.
4321 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4322 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4323 this, as well as the form without the "IPv6" on the front.
4325 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4326 OS/Makefile-Darwin file.
4328 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4329 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4331 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4332 variable. However, some OS use ints for inodes. I've added cast to long int
4333 to get rid of the compiler warning.
4335 19. I had forgotten to lock out "/../" in configuration file names when
4336 ALT_CONFIG_PREFIX was set.
4338 20. Routers used for verification do not need to specify transports. However,
4339 if such a router generated a host list, and callout was configured, Exim
4340 crashed, because it could not find a port number from the (non-existent)
4341 transport. It now assumes port 25 in this circumstance.
4343 21. Added the -t option to exigrep.
4345 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4346 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4347 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4348 them is set, LOOKUP_LSEARCH is forced.
4350 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4352 24. Added sender and host information to the "rejected by local_scan()" log
4353 line; previously there was no indication of these.
4355 25. Added .include_if_exists.
4357 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4358 Linux. It turns out that not all file systems support this. Apparently some
4359 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4360 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4361 not supported on the file descriptor, is now ignored when Exim is trying to
4362 sync a directory. This applies only to Linux.
4364 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4366 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4367 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4370 29. Added support for maildirsize files from supplied patch (modified a bit).
4372 30. The use of :fail: followed by an empty string could lead Exim to respond to
4373 sender verification failures with (e.g.):
4375 550 Verification failed for <xxx>
4376 550 Sender verify failed
4378 where the first response line was missing the '-' that indicates it is not
4379 the final line of the response.
4381 31. The loop for finding the name of the user that called Exim had a hardwired
4382 limit of 10; it now uses the value of finduser_retries, which is used for
4383 all other user lookups.
4385 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4388 33. Exim was neglecting to zero errno before one call of strtol() when
4389 expanding a string and expecting an integer value. On some systems this
4390 resulted in spurious "integer overflow" errors. Also, it was casting the
4391 result into an int without checking.
4393 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4394 did not work. The code looks as if it has *never* worked, though it appears
4395 to have been documented since at least releast 1.62. I have made it work.
4397 35. The "timeout_DNS" error in retry rules, also documented since at least
4398 1.62, also never worked. As it isn't clear exactly what this means, and
4399 clearly it isn't a major issue, I have abolished the feature by treating it
4400 as "timeout", and writing a warning to the main and panic logs.
4402 36. The display of retry rules for -brt wasn't always showing the error code
4405 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4406 timeout_connect_A, timeout_connect_MX.
4408 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4409 to the empty sender.
4411 39. The daemon was not analysing the content of -oX till after it had closed
4412 stderr and disconnected from the controlling terminal. This meant that any
4413 syntax errors were only noted on the panic log, and the return code from
4414 the command was 0. By re-arranging the code a little, I've made the
4415 decoding happen first, so such errors now appear on stderr, and the return
4416 code is 1. However, the actual setting up of the sockets still happens in
4417 the disconnected process, so errors there are still only recorded on the
4420 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4421 connections (as happens on some IP stacks) was logged at start up time as
4422 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4423 it from "IPv6 and IPv4", which means that two separate sockets are being
4426 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4427 says whether AF_INET or AF_INET6 was passed as an argument.
4429 42. Exiwhat output was messed up when time zones were included in log
4432 43. Exiwhat now gives more information about the daemon's listening ports,
4433 and whether -tls-on-connect was used.
4435 44. The "port" option of the smtp transport is now expanded.
4437 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4438 silently ignored. Now an error message is written to the main and panic
4441 46. There's a new ACL modifier called "logwrite" which writes to a log file
4442 as soon as it is encountered.
4444 47. Added $local_user_uid and $local_user_gid at routing time.
4446 48. Exim crashed when trying to verify a sender address that was being
4449 49. Exim was recognizing only a space character after ".include". It now also
4450 recognizes a tab character.
4452 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4453 extracting the relevant information from the specification. The man page no
4454 longer contains scrambled data for the -d option, and I've added a section
4455 at the front about calling Exim under different names.
4457 51. Added "extra_headers" argument to the "mail" command in filter files.
4459 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4462 53. Installed eximstats 1.29.
4464 54. Added transport_filter_timeout as a generic transport option.
4466 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4467 Cc: header lines. This was required by RFC 822, but it not required by RFC
4470 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4471 incoming messages that did not have them. Now it does so only if the
4472 message originates locally, that is, if there is no associated remote host
4473 address. When Resent- header lines are present, this applies to the Resent-
4474 lines rather than the non-Resent- lines.
4476 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4477 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4479 58. Messages for configuration errors now include the name of the main
4480 configuration file - useful now that there may be more than one file in a
4481 list (.included file names were always shown).
4483 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4484 for those rare installations that do not start the daemon as root or run it
4485 setuid root. I've cut out the call to initgroups() if the daemon is not
4488 60. The Exim user and group can now be bound into the binary as text strings
4489 that are looked up at the start of Exim's processing.
4491 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4493 62. Added $mailstore_basename variable.
4495 63. Installed patch to sieve.c from Michael Haardt.
4497 64. When Exim failed to open the panic log after failing to open the main log,
4498 the original message it was trying to log was written to stderr and debug
4499 output, but if they were not available (the usual case in production), it
4500 was lost. Now it is written to syslog before the two lines that record the
4501 failures to open the logs.
4503 65. Users' Exim filters run in subprocesses under the user's uid. It is
4504 possible for a "deliver" command or an alias in a "personal" command to
4505 provoke an address rewrite. If logging of address rewriting is configured,
4506 this fails because the process is not running as root or exim. There may be
4507 a better way of dealing with this, but for the moment (because 4.30 needs
4508 to be released), I have disabled address rewrite logging when running a
4509 filter in a non-root, non-exim process.
4515 1. The buildconfig auxiliary program wasn't quoting the value set for
4516 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4517 not defined. This bug was masked in 4.22 by the effect that was fixed in
4520 2. Some messages that were rejected after a message id was allocated were
4521 shown as "incomplete" by exigrep. It no longer does this for messages that
4522 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4524 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4525 have allow_domain_literals set, the ID did not get logged in the <= line.
4526 Domain literals are now always recognized in Message-ID: header lines.
4528 4. The first argument for a ${extract expansion item is the key name or field
4529 number. Leading and trailing spaces in this item were not being ignored,
4530 causing some misleading effects.
4532 5. When deliver_drop_privilege was set, single queue runner processes started
4533 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4534 same command in the process it spins off) were not dropping privilege.
4536 6. When the daemon running as "exim" started a queue runner, it always
4537 re-executed Exim in the spun-off process. This is a waste of effort when
4538 deliver_drop_privilege is set. The new process now just calls the
4539 queue-runner function directly.
4545 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4548 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4549 ignored. Though the use of -oP was forcing the writing of a pid file, it
4550 was always written to the default place.
4552 3. If the message "no IP address found for host xxxx" is generated during
4553 incoming verification, it is now followed by identification of the incoming
4554 connection (so you can more easily find what provoked it).
4556 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4558 5. Added some features to "harden" Exim a bit more against certain attacks:
4560 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4561 be put in Local/Makefile. This is like the never_users runtime option,
4562 but it cannot be overridden. The default setting is "root".
4564 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4565 prefix string with which any file named in a -C command line option
4568 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4569 is retained for -C and -D only if the caller of Exim is root. Without
4570 it, the exim user may also use -C and -D and retain privilege.
4572 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4573 command line option is disabled.
4575 6. Macro names set by the -D option must start with an upper case letter, just
4576 like macro names defined in the configuration file.
4578 7. Added "dereference=" facility to LDAP.
4580 8. Two instances of the typo "uknown" in the source files are fixed.
4582 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4583 the Configure-Makefile script screwed up while processing it.
4585 10. Incorporated PCRE 4.4.
4587 11. The SMTP synchronization check was not operating right at the start of an
4588 SMTP session. For example, it could not catch a HELO sent before the client
4589 waited for the greeting. There is now a check for outstanding input at the
4590 point when the greeting is written. Because of the duplex, asynchronous
4591 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4592 way, but not yet received, when the check is performed.
4594 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4595 on TCP/IP sockets, because this apparently causes some broken clients to
4598 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4599 unchanged) from the Cygwin maintainer.
4601 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4602 was supported instead of testing for mbx. Effectively a typo.
4604 15. The spa authenticator server code was not checking that the input it
4605 received was valid base64.
4607 16. The debug output line for the "set" modifier in ACLs was not showing the
4608 name of the variable that was being set.
4610 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4612 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4613 lookup on its own. Something else was needed to trigger the lookup. For
4614 example, a match in host_lookup or the need for a host name in a host list.
4615 Now, if $sender_host_name is referenced and the host name has not yet been
4616 looked up, a lookup is performed. If the lookup fails, the variable remains
4617 empty, and $host_lookup_failed is set to "1".
4619 19. Added "eqi" as a case-independent comparison operator.
4621 20. The saslauthd authentication condition could segfault if neither service
4622 nor realm was specified.
4624 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4625 error message that was logged was misleading, and incoming SMTP
4626 connections were dropped. The message is now more accurate, and temporary
4627 errors are given to SMTP connections.
4629 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4630 (except RSET) with a 421 error, until QUIT is received. However, it was
4631 failing to send a response to QUIT.
4633 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4634 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4635 ACL is now run after the helo_try_verify_hosts code.
4637 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4638 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4639 case-independent, so other case variants are also recognized). Apparently
4640 some systems use these upper case variants.
4642 25. If more than two messages were waiting for the same host, and a transport
4643 filter was specified for the transport, Exim sent two messages over the
4644 same TCP/IP connection, and then failed with "socket operation on non-
4645 socket" when it tried to send the third.
4647 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4649 27. The extern definition of crypt16() in expand.c was not being excluded when
4650 the OS had its own crypt16() function.
4652 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4653 as a preferred synonym for return_size_limit, both as an option and as an
4656 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4658 30. Changed the default configuration ACL to relax the local part checking rule
4659 for addresses that are not in any local domains. For these addresses,
4660 slashes and pipe symbols are allowed within local parts, but the sequence
4661 /../ is explicitly forbidden.
4663 31. SPA server authentication was not clearing the challenge buffer before
4666 32. log_message in a "warn" ACL statement was writing to the reject log as
4667 well as to the main log, which contradicts the documentation and doesn't
4668 seem right (because no rejection is happening). So I have stopped it.
4670 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4671 However, I am unable to do any testing of this.
4673 34. Fixed an infelicity in the appendfile transport. When checking directories
4674 for a mailbox, to see if any needed to be created, it was accidentally
4675 using path names with one or more superfluous leading slashes; tracing
4676 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4678 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4679 contents are added to the log line that is written for every discarded
4680 recipient. (Previously a log_message setting was ignored.)
4682 36. The ${quote: operator now quotes the string if it is empty.
4684 37. The install script runs exim in order to find its version number. If for
4685 some reason other than non-existence or emptiness, which it checks, it
4686 could not run './exim', it was installing it with an empty version number,
4687 i.e. as "exim-". This error state is now caught, and the installation is
4690 38. An argument was missing from the function that creates an error message
4691 when Exim fails to connect to the socket for saslauthd authentication.
4692 This could cause Exim to crash, or give a corrupted message.
4694 39. Added isip, isip4, and isip6 to ${if conditions.
4696 40. The ACL variables $acl_xx are now saved with the message, and can be
4697 accessed later in routers, transports, and filters.
4699 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4700 strings in the file are not string-expanded.
4702 42. If a MAIL command specified a SIZE value that was too large to fit into an
4703 int variable, the check against message_size_limit failed. Such values are
4704 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4705 day this will have to be increased, but I don't think I want to be around
4706 when emails are that large.
4713 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4714 iconv() is not standard in FreeBSD.
4716 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4717 IPv6 enabled. The observed symptom was a segmentation fault on return from
4718 the function os_common_find_running_interfaces() in src/os.c.
4720 3. In the check_special_case() function in daemon.c I had used "errno" as an
4721 argument name, which causes warnings on some systems. This was basically a
4722 typo, since it was named "eno" in the comments!
4724 4. The code that waits for the clock to tick (at a resolution of some fraction
4725 of a second) so as to ensure message-id uniqueness was always waiting for
4726 at least one whole tick, when it could have waited for less. [This is
4727 almost certainly not relevant at current processor speeds, where it is
4728 unlikely to ever wait at all. But we try to future-proof.]
4730 5. The function that sleeps for a time interval that includes fractions of a
4731 second contained a race. It did not block SIGALRM between setting the
4732 timer, and suspending (a couple of lines later). If the interval was short
4733 and the sigsuspend() was delayed until after it had expired, the suspension
4734 never ended. On busy systems this could lead to processes getting stuck for
4737 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4738 process, before it forks any delivery processes. The open lookup caching
4739 mechanism meant that the open file or database connection was passed into
4740 the delivery process. The problem was that delivery processes always tidy
4741 up cached lookup data. This could cause a problem for the next delivery
4742 process started by the queue runner, because the external queue runner
4743 process does not know about the closure. So the next delivery process
4744 still has data in the lookup cache. In the case of a file lookup, there was
4745 no problem because closing a file descriptor in a subprocess doesn't affect
4746 the parent. However, if the lookup was caching a connection to a database,
4747 the connection was closed, and the second delivery process was likely to
4748 see errors such as "PGSQL: query failed: server closed the connection
4749 unexpectedly". The problem has been fixed by closing all cached lookups
4750 in a queue runner before running a delivery process.
4752 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4753 seem to have the "const" qualifier which it has on other OS. I've
4756 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4757 *of the same type* (client or server) with the same public name that an
4758 error should be diagnosed.
4760 9. When Exim looked up a host name for an IP address, but failed to find the
4761 original IP address when looking up the host name (a safety check), it
4762 output the message "<ip address> does not match any IP for NULL", which was
4763 confusing, to say the least. The bug was that the host name should have
4764 appeared instead of "NULL".
4766 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4767 user that is built into the binary, and the -C or -D options is used, root
4768 privilege is dropped before the configuration file is read. In addition,
4769 logging is switched to stderr instead of the normal log files. If the
4770 configuration then re-defines the Exim user, the unprivileged environment
4771 is probably not what is expected, so Exim logs a panic warning message (but
4774 However, if deliver_drop_privilege is set, the unprivileged state may well
4775 be exactly what is intended, so the warning has been cut out in that case,
4776 and Exim is allowed to try to write to its normal log files.
4782 1. smtp_return_error_details was not giving details for temporary sender
4783 or receiver verification errors.
4785 2. Diagnose a configuration error if two authenticators have the same public
4788 3. Exim used not to create the message log file for a message until the first
4789 delivery attempt. This could be confusing when incoming messages were held
4790 for policy or load reasons. The message log file is now created at the time
4791 the message is received, and an initial "Received" line is written to it.
4793 4. The automatically generated man page for command line options had a minor
4794 bug that caused no ill effects; however, a more serious problem was that
4795 the procedure for building the man page automatically didn't always
4796 operate. Consequently, release 4.20 contains an out-of-date version. This
4797 shouldn't happen again.
4799 5. When building Exim with embedded Perl support, the script that builds the
4800 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4801 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4803 6. The freeze_tell option was not being used for messages that were frozen on
4804 arrival, either by an ACL or by local_scan().
4806 7. Added the smtp_incomplete_transaction log selector.
4808 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4809 was accepting AUTH without a new EHLO.
4811 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4812 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4815 10. Logging of TCP/IP connections (when configured) now happens in the main
4816 daemon process instead of the child process, so that the TCP/IP connection
4817 count is more accurate (but it can never be perfect).
4819 11. The use of "drop" in a nested ACL was not being handled correctly in the
4820 outer ACL. Now, if condition failure induced by the nested "drop" causes
4821 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4822 or "require"), the connection is dropped.
4824 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4825 that yield "discard" can now be used with an "accept" or a "discard" verb,
4826 but an error is generated for any others (because I can't see a useful way
4827 to define what should happen).
4829 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4830 are now processed in the same way as lines in the Exim configuration file.
4831 In particular, continuation lines are supported.
4833 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4835 15. Added -ti meaning -t -i.
4837 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4838 domains, and warn by logging if others are found.
4840 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4841 returned from GIFCONF when Exim is trying to find the list of interfaces on
4842 a host. The code in os.c has been modified to copy each ifreq to an aligned
4843 structure in all cases.
4845 Also, in some cases, the returned ifreq's were being copied to a 'struct
4846 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4847 means the last couple of bytes of an IPv6 address could be chopped if the
4848 ifreq contained only a normal sockaddr (14 bytes storage).
4850 18. Named domain lists were not supported in the hosts_treat_as_local option.
4851 An entry such as +xxxx was not recognized, and was treated as a literal
4854 19. Ensure that header lines added by a DATA ACL are included in the reject log
4855 if the ACL subsequently rejects the message.
4857 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4858 MD5 (which is deprecated).
4860 21. When testing a filter file using -bf, Exim was writing a message when it
4861 took the sender from a "From " line in the message, but it was not doing so
4862 when it took $return_path from a Return-Path: header line. It now does.
4864 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4865 with a valid header line field name (a series of printing characters
4866 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4868 23. Changed "disc" in the source to "disk" to conform to the documentation and
4869 the book and for uniformity.
4871 24. Ignore Sendmail's -Ooption=value command line item.
4873 25. When execve() failed while trying to run a command in a pipe transport,
4874 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4875 could be confused with a return value of 69 from the command itself. This
4876 has been changed to 127, the value the shell returns if it is asked to run
4877 a non-existent command. The wording for the related log line suggests a
4878 non-existent command as the problem.
4880 26. If received_header_text expands to an empty string, do not add a Received:
4881 header line to the message. (Well, it adds a token one on the spool, but
4882 marks it "old" so that it doesn't get used or transmitted.)
4884 27. Installed eximstats 1.28 (addition of -nt option).
4886 28. There was no check for failure on the call to getsockname() in the daemon
4887 code. This can fail if there is a shortage of resources on the system, with
4888 ENOMEM, for example. A temporary error is now given on failure.
4890 29. Contrary to the C standard, it seems that in some environments, the
4891 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4892 program. Exim now does this explicitly; it affects the formatting of
4893 timestamps using strftime().
4895 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4896 complaints. I've now initialized all the variables, to avoid this.
4898 32. Header lines added by a system filter were not being "seen" during
4899 transport-time rewrites.
4901 33. The info_callback() function passed to OpenSSL is set up with type void
4902 (*)(SSL *, int, int), as described somewhere. However, when calling the
4903 function (actually a macro) that sets it up, the type void(*)() is
4904 expected. I've put in a cast to prevent warnings from picky compilers.
4906 34. If a DNS black list lookup found a CNAME record, but there were no A
4907 records associated with the domain it pointed at, Exim crashed.
4909 35. If a DNS black list lookup returned more than one A record, Exim ignored
4910 all but the first. It now scans all returned addresses if a particular IP
4911 value is being sought. In this situation, the contents of the
4912 $dnslist_value variable are a list of all the addresses, separated by a
4915 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4916 to accept a host name and all its aliases if the forward lookup for any of
4917 them yielded the IP address of the incoming connection. Now it accepts only
4918 those names whose forward lookup yields the correct IP address. Any other
4919 names are discarded. This closes a loophole whereby a rogue DNS
4920 administrator could create reverse DNS records to break through a
4921 wildcarded host restriction in an ACL.
4923 37. If a user filter or a system filter that ran in a subprocess used any of
4924 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4925 the wrong values were passed to the pipe command ($thisaddress had the
4926 value of $0, $0 had the value of $1, etc). This bug was introduced by
4927 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4929 38. Improved the line breaking for long SMTP error messages from ACLs.
4930 Previously, if there was no break point between 40 and 75 characters, Exim
4931 left the rest of the message alone. Two changes have been made: (a) I've
4932 reduced the minimum length to 35 characters; (b) if it can't find a break
4933 point between 35 and 75 characters, it looks ahead and uses the first one
4934 that it finds. This may give the occasional overlong line, but at least the
4935 remaining text gets split now.
4937 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4938 file descriptors might be low, and that setting 1000 would always raise it.
4939 It turns out that in some environments, the limit is already over 1000 and
4940 that lowering it causes trouble. So now Exim takes care not to decrease it.
4942 40. When delivering a message, the value of $return_path is set to $sender_
4943 address at the start of routing (routers may change the value). By an
4944 oversight, this default was not being set up when an address was tested by
4945 -bt or -bv, which affected the outcome if any router or filter referred to
4948 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4949 log only when "message" or "log_message" are set. However, if one of the
4950 conditions was an address verification, or a call to a nested ACL, the
4951 messages generated by the underlying test were being passed through. This
4952 no longer happens. The underlying message is available in $acl_verify_
4953 message for both "message" and "log_message" expansions, so it can be
4954 passed through if needed.
4956 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4957 new expansion $bh_ to give the encoded byte string without charset
4958 translation. Translation happens only if iconv() is available; HAVE_ICONV
4959 indicates this at build time. HEADERS_CHARSET gives the charset to
4960 translate to; headers_charset can change it in the configuration, and
4961 "headers charset" can change it in an individual filter file.
4963 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4964 that creates RFC 2047 encoded "words" labels them as that charset instead
4965 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4966 expansion operator; (ii) when Exim creates a From: line for a local
4967 message; (iii) when a header line is rewritten to include a "phrase" part.
4969 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4970 buggy, causing it to skip the first lines of messages whose message ID
4971 ended in 'D'. This would not have bitten before Exim release 4.14, because
4972 message IDs were unlikely to end in 'D' before then. The effect was to have
4973 incorrect size information for certain domains.
4975 45. #include "config.h" was missing at the start of the crypt16.c module. This
4976 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4979 46. If there was a timeout during a "random" callout check, Exim treated it as
4980 a failure of the random address, and carried on sending RSET and the real
4981 address. If the delay was just some slowness somewhere, the response to the
4982 original RCPT would be taken as a response to RSET and so on, causing
4983 mayhem of various kinds.
4985 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4986 when I implemented it. It didn't allow for the fact that some option values
4987 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4988 the right test for positive values.
4990 48. Domain names in DNS records are case-independent. Exim always looks them up
4991 in lower case. Some resolvers return domain names in exactly the case they
4992 appear in the zone file, that is, they may contain uppercase letters. Not
4993 all resolvers do this - some return always lower case. Exim was treating a
4994 change of case by a resolver as a change of domain, similar to a widening
4995 of a domain abbreviation. This triggered its re-routing code and so it was
4996 trying to route what was effectively the same domain again. This normally
4997 caused routing to fail (because the router wouldn't handle the domain
4998 twice). Now Exim checks for this case specially, and just changes the
4999 casing of the domain that it ultimately uses when it transmits the message
5002 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5005 50. If a filter generated a file delivery with a non-absolute name (possible if
5006 no home directory exists for the router), the forbid_file option was not
5009 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5010 the existing equality matching.
5012 52. Exim was using ints instead of ino_t variables in some places where it was
5013 dealing with inode numbers.
5015 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5016 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5017 called TMPDIR, and if it finds it is different, it changes its value.
5019 54. The smtp_printf() function is now made available to local_scan() so
5020 additional output lines can be written before returning. There is also an
5021 smtp_fflush() function to enable the detection of a dropped connection.
5022 The variables smtp_input and smtp_batched_input are exported to
5025 55. Changed the default runtime configuration: the message "Unknown user"
5026 has been removed from the ACL, and instead placed on the localuser router,
5027 using the cannot_route_message feature. This means that any verification
5028 failures that generate their own messages won't get overridden. Similarly,
5029 the "Unrouteable address" message that was in the ACL for unverifiable
5030 relay addresses has also been removed.
5032 56. Added hosts_avoid_esmtp to the smtp transport.
5034 57. The exicyclog script was not checking for the esoteric option
5035 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5036 will work only if exicyclog is run under the appropriate euid.
5038 58. Following a discussion on the list, the rules by which Exim recognises line
5039 endings on incoming messages have been changed. The -dropcr and drop_cr
5040 options are now no-ops, retained only for backwards compatibility. The
5041 following line terminators are recognized: LF CRLF CR. However, special
5042 processing applies to CR:
5044 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5045 nor a local message in the state where . is a terminator.
5047 (ii) If a bare CR is encountered in a header line, an extra space is added
5048 after the line terminator so as not to end the header. The reasoning
5049 behind this is that bare CRs in header lines are most likely either
5050 to be mistakes, or people trying to play silly games.
5052 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5053 was being incorrectly given as 18 bytes larger than it should have been.
5054 This is a VOB (very old bug).
5056 60. This may never have affected anything current, but just in case it has:
5057 When the local host is found other than at the start of a list of hosts,
5058 the local host, those with the same MX, and any that follow, are discarded.
5059 When the list in question was part of a longer list of hosts, the following
5060 hosts (not currently being processed) were also being discarded. This no
5061 longer happens. I'm not sure if this situation could ever has previously
5064 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5067 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5068 header. This is something that is recommended in a new Internet Draft, and
5069 is something that is documented as being done by Sendmail. There are two
5070 possible values. For messages generated by the autoreply transport, Exim
5073 Auto-Submitted: auto-replied
5075 whereas for all other generated messages (e.g. bounces) it adds
5077 Auto-Submitted: auto-generated
5079 63. The "personal" condition in filters now includes a test for the
5080 Auto-Submitted: header. If it contains the string "auto-" the message it
5081 not considered personal.
5083 64. Added rcpt_include_affixes as a generic transport option.
5085 65. Added queue_only_override (default true).
5087 66. Added the syslog_duplication option.
5089 67. If what should have been the first header line of a message consisted of
5090 a space followed by a colon, Exim was mis-interpreting it as a header line.
5091 It isn't of course - it is syntactically invalid and should therefore be
5092 treated as the start of the message body. The misbehaviour could have
5093 caused a number of strange effects, including loss of data in subsequent
5094 header lines, and spool format errors.
5096 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5097 client host had authenticated. This control can now be exercised by an ACL
5098 for more flexibility.
5100 69. By default, callouts do not happen when testing with -bh. There is now a
5101 variant, -bhc, which does actually run the callout code, including
5102 consulting and updating the callout cache.
5104 70. Added support for saslauthd authentication, courtesy of Alexander
5107 71. If statvfs() failed on the spool or log directories while checking their
5108 size for availability, Exim confusingly gave the error "space shortage".
5109 Furthermore, in debugging mode it crashed with a floating point exception.
5110 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5111 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5112 serious problem, Exim now writes to the main and panic logs when this
5113 happens, with details of the failure. It then refuses to accept the
5114 incoming message, giving the message "spool directory problem" or "log
5115 directory problem" with a 421 code for SMTP messages.
5117 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5118 0, 1, and 2 exist, because some OS complain for execs without them (see
5119 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5120 descriptors. However, the code omitted to check that the open succeeded,
5121 causing mysterious errors if for some reason the permissions on /dev/null
5122 got screwed. Now Exim writes a message to the main and panic logs, and
5123 bombs out if it can't open /dev/null.
5125 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5126 interact so that it is all more flexible. It is supposed to remain
5127 backwards compatible. Also added extra_local_interfaces.
5129 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5130 to bomb out with an assertion failure - to the client this appears as a
5131 connection drop. This problem occurs in the part of the code that was taken
5132 from the Samba project. Fortunately, the assertion is in a very simple
5133 function, so I have fixed this by reproducing the function inline in the
5134 one place where it is called, and arranging for authentication to fail
5135 instead of killing the process with assert().
5137 75. The SPA client code was not working when the server requested OEM rather
5138 than Unicode encoding.
5140 76. Added code to make require_files with a specific uid setting more usable in
5141 the case where statting the file as root fails - usually a non-root-mounted
5142 NFS file system. When this happens and the failure is EACCES, Exim now
5143 forks a subprocess and does the per-uid checking as the relevant uid.
5145 77. Added process_log_path.
5147 78. If log_file_path was not explicitly set, a setting of check_log_space or
5148 check_log_inodes was ignored.
5150 79. If a space check for the spool or log partitions fails, the incident is now
5151 logged. Of course, in the latter case the data may get lost...
5153 80. Added the %p formatting code to string_format() so that it can be used to
5154 print addresses in debug_print(). Adjusted all the address printing in the
5155 debugging in store.c to use %p rather than %d.
5157 81. There was a concern that a line of code in smtp_in.c could overflow a
5158 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5159 initially expressed, the concern was not well-founded, because trailing
5160 spaces are removed early. However, if the trailing spaces were followed by
5161 a NULL, they did not get removed, so the overflow was possible. Two fixes
5164 (a) I re-wrote the offending code in a cleaner fashion.
5165 (b) If an incoming SMTP command contains a NULL character, it is rejected
5168 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5169 runs initgroups(), so that if the Exim user is in any additional groups,
5170 they will be used during message reception.
5176 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob