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
81 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
82 $data_path, and $header_path variables; fixed documentation bugs and
85 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
86 exipick to access non-standard spools, including the "frozen" queue
89 NM/01 Bugzilla 965: Support mysql stored procedures.
90 Patch from Alain Williams
92 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
94 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
95 Patch from Andreas Metzler
97 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
98 Patch from Kirill Miazine
100 NM/05 Bugzilla 671: Added umask to procmail example.
102 JJ/03 installed exipick 20100323.0, fixing doc bug
104 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
105 directory. Notification and patch from Dan Rosenberg.
107 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
109 TK/02 Improve log output when DKIM signing operation fails.
111 MH/01 Treat the transport option dkim_domain as a colon separated
112 list, not as a single string, and sign the message with each element,
113 omitting multiple occurences of the same signer.
115 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
116 Bugzilla 985, 986. Patch by Simon Arlott
118 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
119 Patch by Simon Arlott
121 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
122 MBX locking. Notification from Dan Rosenberg.
128 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
130 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
132 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
134 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
136 NM/04 Bugzilla 915: Flag broken perl installation during build.
142 TK/01 Added patch by Johannes Berg that expands the main option
143 "spamd_address" if it starts with a dollar sign.
145 TK/02 Write list of recipients to X-Envelope-Sender header when building
146 the mbox-format spool file for content scanning (suggested by Jakob
149 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
150 (http://www.dcc-servers.net/) support via dccifd. Activated by
151 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
153 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
154 by Mark Daniel Reidel <mr@df.eu>.
156 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
157 When building exim an external PCRE library is now needed -
158 PCRE is a system library on the majority of modern systems.
159 See entry on PCRE_LIBS in EDITME file.
161 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
162 conversation. Added nologin parameter to request.
163 Patch contributed by Kirill Miazine.
165 TF/01 Do not log submission mode rewrites if they do not change the address.
167 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
169 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
170 log files in place. Contributed by Roberto Lima.
172 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
174 TF/03 Bugzilla 615: When checking the local_parts router precondition
175 after a local_part_suffix or local_part_prefix option, Exim now
176 does not use the address's named list lookup cache, since this
177 contains cached lookups for the whole local part.
179 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
180 Robert Millan. Documentation is in experimental-spec.txt.
182 TF/04 Bugzilla 668: Fix parallel build (make -j).
184 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
186 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
187 Patch provided by Jan Srzednicki.
189 TF/05 Leading white space used to be stripped from $spam_report which
190 wrecked the formatting. Now it is preserved.
192 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
193 that they are available at delivery time.
195 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
197 TF/08 TLS error reporting now respects the incoming_interface and
198 incoming_port log selectors.
200 TF/09 Produce a more useful error message if an SMTP transport's hosts
201 setting expands to an empty string.
203 NM/06 Bugzilla 744: EXPN did not work under TLS.
204 Patch provided by Phil Pennock.
206 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
207 Patch provided by Richard Godbee.
209 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
210 acl_smtp_notquit, added index entry.
212 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
213 Patch provided by Eugene Bujak.
215 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
216 accept(). Patch provided by Maxim Dounin.
218 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
219 Patch provided by Phil Pennock.
221 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
223 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
224 Patch provided by Brad "anomie" Jorsch.
226 NM/14 Bugzilla 622: Added timeout setting to transport filter.
227 Patch provided by Dean Brooks.
229 TK/05 Add native DKIM support (does not depend on external libraries).
231 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
232 Patch provided by Graeme Fowler.
234 NM/16 Bugzilla 851: Documentation example syntax fix.
236 NM/17 Changed NOTICE file to remove references to embedded PCRE.
238 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
241 NM/19 Bugzilla 745: TLS version reporting.
242 Patch provided by Phil Pennock.
244 NM/20 Bugzilla 167: bool: condition support.
245 Patch provided by Phil Pennock.
247 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
248 clients. Patch provided by Phil Pennock.
250 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
251 Patch provided by Brad "anomie" Jorsch.
253 NM/23 Bugzilla 687: Fix misparses in eximstats.
254 Patch provided by Heiko Schlittermann.
256 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
257 Patch provided by Heiko Schlittermann.
259 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
260 plus update to original patch.
262 NM/26 Bugzilla 799: Documentation correction for ratelimit.
264 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
265 Patch provided by David Brownlee.
267 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
269 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
271 NM/30 Bugzilla 888: TLS documentation bugfixes.
273 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
275 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
276 Unlike the original bugzilla I have changed all shell scripts in src tree.
278 NM/33 Bugzilla 898: Transport filter timeout fix.
279 Patch by Todd Rinaldo.
281 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
282 Patch by Serge Demonchaux.
284 NM/35 Bugzilla 39: Base64 decode bug fixes.
285 Patch by Jakob Hirsch.
287 NM/36 Bugzilla 909: Correct connect() call in dcc code.
289 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
291 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
293 NM/39 Bugzilla 911: Fixed MakeLinks build script.
299 TK/01 Add preliminary DKIM support. Currently requires a forked version of
300 ALT-N's libdkim that I have put here:
301 http://duncanthrax.net/exim-experimental/
303 Note to Michael Haardt: I had to rename some vars in sieve.c. They
304 were called 'true' and it seems that C99 defines that as a reserved
305 keyword to be used with 'bool' variable types. That means you could
306 not include C99-style headers which use bools without triggering
307 build errors in sieve.c.
309 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
310 as mailq or other aliases. Changed the --help handling significantly
311 to do whats expected. exim_usage() emits usage/help information.
313 SC/01 Added the -bylocaldomain option to eximstats.
315 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
317 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
319 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
325 PH/01 Another patch from the Sieve maintainer.
327 PH/02 When an IPv6 address is converted to a string for single-key lookup
328 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
329 dots are used instead of colons so that keys in lsearch files need not
330 contain colons. This was done some time before quoting was made available
331 in lsearch files. However, iplsearch files do require colons in IPv6 keys
332 (notated using the quote facility) so as to distinguish them from IPv4
333 keys. This meant that lookups for IP addresses in host lists did not work
334 for iplsearch lookups.
336 This has been fixed by arranging for IPv6 addresses to be expressed with
337 colons if the lookup type is iplsearch. This is not incompatible, because
338 previously such lookups could never work.
340 The situation is now rather anomolous, since one *can* have colons in
341 ordinary lsearch keys. However, making the change in all cases is
342 incompatible and would probably break a number of configurations.
344 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
347 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
348 conversion specification without a maximum field width, thereby enabling
349 a rogue spamd server to cause a buffer overflow. While nobody in their
350 right mind would setup Exim to query an untrusted spamd server, an
351 attacker that gains access to a server running spamd could potentially
352 exploit this vulnerability to run arbitrary code as the Exim user.
354 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
355 $primary_hostname instead of what libspf2 thinks the hosts name is.
357 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
358 a directory entry by the name of the lookup key. Previously, if a
359 symlink pointed to a non-existing file or a file in a directory that
360 Exim lacked permissions to read, a lookup for a key matching that
361 symlink would fail. Now it is enough that a matching directory entry
362 exists, symlink or not. (Bugzilla 503.)
364 PH/03 The body_linecount and body_zerocount variables are now exported in the
367 PH/04 Added the $dnslist_matched variable.
369 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
370 This means they are set thereafter only if the connection becomes
373 PH/06 Added the client_condition to authenticators so that some can be skipped
374 by clients under certain conditions.
376 PH/07 The error message for a badly-placed control=no_multiline_responses left
377 "_responses" off the end of the name.
379 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
381 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
382 (without spaces) instead of just copying the configuration text.
384 PH/10 Added the /noupdate option to the ratelimit ACL condition.
386 PH/11 Added $max_received_linelength.
388 PH/12 Added +ignore_defer and +include_defer to host lists.
390 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
391 way in which PCRE > 7.0 is built.
393 PH/14 Implemented queue_only_load_latch.
395 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
396 MAIL command. The effect was to mangle the value on 64-bit systems.
398 PH/16 Another patch from the Sieve maintainer.
400 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
402 PH/18 If a system quota error occurred while trying to create the file for
403 a maildir delivery, the message "Mailbox is full" was not appended to the
404 bounce if the delivery eventually timed out. Change 4.67/27 below applied
405 only to a quota excession during the actual writing of the file.
407 PH/19 It seems that peer DN values may contain newlines (and other non-printing
408 characters?) which causes problems in log lines. The DN values are now
409 passed through string_printing() before being added to log lines.
411 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
412 and InterBase are left for another time.)
414 PH/21 Added message_body_newlines option.
416 PH/22 Guard against possible overflow in moan_check_errorcopy().
418 PH/23 POSIX allows open() to be a macro; guard against that.
420 PH/24 If the recipient of an error message contained an @ in the local part
421 (suitably quoted, of course), incorrect values were put in $domain and
422 $local_part during the evaluation of errors_copy.
428 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
429 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
432 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
433 issue a MAIL command.
435 PH/02 In an ACL statement such as
437 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
439 if a client was not listed at all, or was listed with a value other than
440 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
441 the condition was not true (as it should be), so access was not denied.
442 The bug was that the ! inversion was incorrectly passed on to the second
443 item. This has been fixed.
445 PH/03 Added additional dnslists conditions == and =& which are different from
446 = and & when the dns lookup returns more than one IP address.
448 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
449 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
451 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
452 FSYNC, which compiles an option called disable_fsync that allows for
453 bypassing fsync(). The documentation is heavily laced with warnings.
455 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
457 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
458 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
459 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
460 including adding "make clean"; (3) Added -fPIC when compiling the test
461 dynamically loaded module, to get rid of a warning.
463 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
464 message fails, move_frozen_messages = true and ignore_bounce_errors_after
465 = 0s. The bug is otherwise harmless.
467 PH/07 There was a bug in the dovecot authenticator such that the value of
468 $auth1 could be overwritten, and so not correctly preserved, after a
469 successful authentication. This usually meant that the value preserved by
470 the server_setid option was incorrect.
472 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
474 PH/09 Installed PCRE release 7.0.
476 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
477 run for batched SMTP input. It is now run at the start of every message
478 in the batch. While fixing this I discovered that the process information
479 (output by running exiwhat) was not always getting set for -bs and -bS
480 input. This is fixed, and it now also says "batched" for BSMTP.
482 PH/11 Added control=no_pipelining.
484 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
485 patch, slightly modified), and move the expansion of helo_data till after
486 the connection is made in the smtp transport (so it can use these
489 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
491 PH/14 Added log_selector = +pid.
493 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
495 PH/16 Add ${if forany and ${if forall.
497 PH/17 Added dsn_from option to vary the From: line in DSNs.
499 PH/18 Flush SMTP output before performing a callout, unless control =
500 no_callout_flush is set.
502 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
503 was true (the default) a successful delivery failed to delete the retry
504 item, thus causing premature timeout of the address. The bug is now
507 PH/20 Added hosts_avoid_pipelining to the smtp transport.
509 PH/21 Long custom messages for fakedefer and fakereject are now split up
510 into multiline reponses in the same way that messages for "deny" and
511 other ACL rejections are.
513 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
514 with slight modification.
516 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
517 draft, changing the syntax and factoring some duplicate code".
519 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
520 for deliveries of the second and subsequent messages over the same SMTP
523 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
524 ${reduce, with only minor "tidies".
526 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
528 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
529 expansion side effects.
531 PH/27 When a message times out after an over-quota error from an Exim-imposed
532 quota, the bounce message says "mailbox is full". This message was not
533 being given when it was a system quota that was exceeded. It now should
536 MH/03 Made $recipients available in local_scan(). local_scan() already has
537 better access to the recipient list through recipients_list[], but
538 $recipients can be useful in postmaster-provided expansion strings.
540 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
541 in the case of a MAIL command with additional options following the
542 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
543 were accidentally chopped off.
545 PH/29 SMTP synchronization checks are implemented when a command is read -
546 there is a check that no more input is waiting when there shouldn't be
547 any. However, for some commands, a delay in an ACL can mean that it is
548 some time before the response is written. In this time, more input might
549 arrive, invalidly. So now there are extra checks after an ACL has run for
550 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
551 pipelining has not been advertised.
553 PH/30 MH's patch to allow iscntrl() characters to be list separators.
555 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
556 returned in the SMTP response when smtp_return_error_details was false.
559 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
560 instead of the C I/O that was originally supplied, because problems were
563 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
564 Exim which did not show up earlier: it was assuming that a call to
565 SSL_CTX_set_info_callback() might give an error value. In fact, there is
566 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
567 was a macro that became an assignment, so it seemed to work. This has
568 changed to a proper function call with a void return, hence the compile
569 error. Exim's code has been fixed.
571 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
574 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
576 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
578 PH/37 If a message is not accepted after it has had an id assigned (e.g.
579 because it turns out to be too big or there is a timeout) there is no
580 "Completed" line in the log. When some messages of this type were
581 selected by exigrep, they were listed as "not completed". Others were
582 picked up by some special patterns. I have improved the selection
583 criteria to be more general.
585 PH/38 The host_find_failed option in the manualroute router can now be set
586 to "ignore", to completely ignore a host whose IP address cannot be
587 found. If all hosts are ignored, the behaviour is controlled by the new
588 host_all_ignored option.
590 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
591 homing or because of multiple MX records with /mx) generated more than
592 one IP address, and the following item turned out to be the local host,
593 all the secondary addresses of the first item were incorrectly removed
594 from the list, along with the local host and any following hosts (which
595 is what is supposed to happen).
597 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
598 whoever called Exim into the -H file. In the case of the daemon it was
599 behaving confusingly. When first started, it used values for whoever
600 started the daemon, but after a SIGHUP it used the Exim user (because it
601 calls itself on a restart). I have changed the code so that it now always
604 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
605 message are rejected with the same error (e.g. no authentication or bad
606 sender address), and a DATA command is nevertheless sent (as can happen
607 with PIPELINING or a stupid MUA), the error message that was given to the
608 RCPT commands is included in the rejection of the DATA command. This is
609 intended to be helpful for MUAs that show only the final error to their
612 PH/42 Another patch from the Sieve maintainer.
614 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
615 Eximstats - Fixed some broken HTML links and added missing column headers
617 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
618 columns for Rejects, Temp Rejects, Ham, and Spam rows.
620 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
622 PH/43 Yet another patch from the Sieve maintainer.
624 PH/44 I found a way to check for a TCP/IP connection going away before sending
625 the response to the final '.' that terminates a message, but only in the
626 case where the client has not sent further data following the '.'
627 (unfortunately, this is allowed). However, in many cases there won't be
628 any further data because there won't be any more messages to send. A call
629 to select() can be used: if it shows that the input is "ready", there is
630 either input waiting, or the socket has been closed. An attempt to read
631 the next input character can distinguish the two cases. Previously, Exim
632 would have sent an OK response which the client would never have see.
633 This could lead to message repetition. This fix should cure that, at
634 least in a lot of common cases.
636 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
637 advertised in response to EHLO.
643 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
644 fixed by 4.65/MH/01 (is this a record?) are fixed:
646 (i) An empty string was always treated as zero by the numeric comparison
647 operators. This behaviour has been restored.
649 (ii) It is documented that the numeric comparison operators always treat
650 their arguments as decimal numbers. This was broken in that numbers
651 starting with 0 were being interpreted as octal.
653 While fixing these problems I realized that there was another issue that
654 hadn't been noticed. Values of message_size_limit (both the global option
655 and the transport option) were treated as octal if they started with 0.
656 The documentation was vague. These values are now always treated as
657 decimal, and I will make that clear in the documentation.
663 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
664 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
667 MH/01 Don't check that the operands of numeric comparison operators are
668 integers when their expansion is in "skipping" mode (fixes bug
669 introduced by 4.64-PH/07).
671 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
672 child addresses, Exim now panics and dies. Previously, because the count
673 is held in a short int, deliveries were likely to be lost. As such a
674 large number of recipients for a single message is ridiculous
675 (performance will be very, very poor), I have chosen to impose a limit
676 rather than extend the field.
682 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
683 leftover -K file (the existence of which was triggered by #402).
684 While we were at it, introduced process PID as part of the -K
685 filename. This should rule out race conditions when creating
688 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
689 processing considerably. Previous code took too long for large mails,
690 triggering a timeout which in turn triggers #401.
692 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
693 in the DK code in transports.c. sendfile() is not really portable,
694 hence the _LINUX specificness.
696 TF/01 In the add_headers option to the mail command in an Exim filter,
697 there was a bug that Exim would claim a syntax error in any
698 header after the first one which had an odd number of characters
701 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
702 callout verification, Exim cached a "reject" for the entire domain. This
703 is correct for most verifications, but it is not correct for a recipient
704 verification with use_sender or use_postmaster set, because in that case
705 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
706 case of MAIL FROM:<> rejection from other early rejections (e.g.
707 rejection of HELO). When verifying a recipient using a non-null MAIL
708 address, the cache is ignored if it shows MAIL FROM:<> rejection.
709 Whatever the result of the callout, the value of the domain cache is
710 left unchanged (for any other kind of callout, getting as far as trying
711 RCPT means that the domain itself is ok).
713 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
716 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
717 manifest itself as EPIPE rather than ECONNECT. When tidying away a
718 session, the daemon ignores ECONNECT errors and logs others; it now
719 ignores EPIPE as well.
721 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
722 (quoted-printable decoding).
724 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
725 later the small subsequent patch to fix an introduced bug.
727 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
729 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
731 PH/08 An error is now given if message_size_limit is specified negative.
733 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
734 to be given (somewhat) arbitrary names.
736 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
739 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
740 miscellaneous code fixes
742 PH/10 Added the log_reject_target ACL modifier to specify where to log
745 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
746 hostname. This is wrong, because it relates to the incoming message (and
747 probably the interface on which it is arriving) and not to the outgoing
748 callout (which could be using a different interface). This has been
749 changed to use the value of the helo_data option from the smtp transport
750 instead - this is what is used when a message is actually being sent. If
751 there is no remote transport (possible with a router that sets up host
752 addresses), $smtp_active_hostname is used.
754 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
755 tweaks were necessary in order to get it to work (see also 21 below):
756 (a) The code assumed that strncpy() returns a negative number on buffer
757 overflow, which isn't the case. Replaced with Exim's string_format()
759 (b) There were several signed/unsigned issues. I just did the minimum
760 hacking in of casts. There is scope for a larger refactoring.
761 (c) The code used strcasecmp() which is not a standard C function.
762 Replaced with Exim's strcmpic() function.
763 (d) The code set only $1; it now sets $auth1 as well.
764 (e) A simple test gave the error "authentication client didn't specify
765 service in request". It would seem that Dovecot has changed its
766 interface. Fortunately there's a specification; I followed it and
767 changed what the client sends and it appears to be working now.
769 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
772 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
773 address A is aliased to B and C, where B exists and C does not. Without
774 -v the output is "A verified" because verification stops after a
775 successful redirection if more than one address is generated. However,
776 with -v the child addresses are also verified. Exim was outputting "A
777 failed to verify" and then showing the successful verification for C,
778 with its parentage. It now outputs "B failed to verify", showing B's
779 parentage before showing the successful verification of C.
781 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
782 look up a TXT record in a specific list after matching in a combined
785 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
786 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
787 they consult the DNS. I had assumed they would set it the way they
788 wanted; and indeed my experiments on Linux seem to show that in some
789 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
790 To be on the safe side, however, I have now made the interface to
791 host_find_byname() similar to host_find_bydns(), with an argument
792 containing the DNS resolver options. The host_find_byname() function now
793 sets these options at its start, just as host_find_bydns() does. The smtp
794 transport options dns_qualify_single and dns_search_parents are passed to
795 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
796 of host_find_byname() use the default settings of RES_DEFNAMES
797 (qualify_single) but not RES_DNSRCH (search_parents).
799 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
800 spool_read_header() do less string testing, by means of a preliminary
801 switch on the second character of optional "-foo" lines. (This is
802 overdue, caused by the large number of possibilities that now exist.
803 Originally there were few.) While I was there, I also converted the
804 str(n)cmp tests so they don't re-test the leading "-" and the first
805 character, in the hope this might squeeze out yet more improvement.
807 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
808 flag allowing group syntax was set by the header_syntax check but not
809 turned off, possible causing trouble later; (2) The flag was not being
810 set at all for the header_verify test, causing "group"-style headers to
811 be rejected. I have now set it in this case, and also caused header_
812 verify to ignore an empty address taken from a group. While doing this, I
813 came across some other cases where the code for allowing group syntax
814 while scanning a header line wasn't quite right (mostly, not resetting
815 the flag correctly in the right place). These bugs could have caused
816 trouble for malformed header lines. I hope it is now all correct.
818 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
819 with the "reply" argument non-NULL. The code, however (which originally
820 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
821 but it didn't always do it. This confused somebody who was copying the
822 code for some other use. I have removed all the tests.
824 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
825 feature that was used to support insecure browsers during the U.S. crypto
826 embargo. It requires special client support, and Exim is probably the
827 only MTA that supported it -- and would never use it because real RSA is
828 always available. This code has been removed, because it had the bad
829 effect of slowing Exim down by computing (never used) parameters for the
830 RSA_EXPORT functionality.
832 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
833 authenticator to fail if there's a tab character in the incoming data
834 (there should never be unless someone is messing about, as it's supposed
835 to be base64-encoded). Also added, on Timo's advice, the "secured" option
836 if the connection is using TLS or if the remote IP is the same as the
837 local IP, and the "valid-client-cert option" if a client certificate has
840 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
841 authenticators. This can be used for authorization after authentication
842 succeeds. (In the case of plaintext, it servers for both authentication
845 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
846 if any retry times were supplied.
848 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
849 connection, where there is no client IP address to check. In this
850 situation, the verify now always succeeds.
852 PH/25 Applied John Jetmore's -Mset patch.
854 PH/26 Added -bem to be like -Mset, but loading a message from a file.
856 PH/27 In a string expansion for a processed (not raw) header when multiple
857 headers of the same name were present, leading whitespace was being
858 removed from all of them, but trailing whitespace was being removed only
859 from the last one. Now trailing whitespace is removed from each header
860 before concatenation. Completely empty headers in a concatenation (as
863 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
864 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
866 PH/29 [Removed. This was a change that I later backed out, and forgot to
867 correct the ChangeLog entry (that I had efficiently created) before
868 committing the later change.]
870 PH/30 Exim was sometimes attempting to deliver messages that had suffered
871 address errors (4xx response to RCPT) over the same connection as other
872 messages routed to the same hosts. Such deliveries are always "forced",
873 so retry times are not inspected. This resulted in far too many retries
874 for the affected addresses. The effect occurred only when there were more
875 hosts than the hosts_max_try setting in the smtp transport when it had
876 the 4xx errors. Those hosts that it had tried were not added to the list
877 of hosts for which the message was waiting, so if all were tried, there
878 was no problem. Two fixes have been applied:
880 (i) If there are any address or message errors in an SMTP delivery, none
881 of the hosts (tried or untried) are now added to the list of hosts
882 for which the message is waiting, so the message should not be a
883 candidate for sending over the same connection that was used for a
884 successful delivery of some other message. This seems entirely
885 reasonable: after all the message is NOT "waiting for some host".
886 This is so "obvious" that I'm not sure why it wasn't done
887 previously. Hope I haven't missed anything, but it can't do any
888 harm, as the worst effect is to miss an optimization.
890 (ii) If, despite (i), such a delivery is accidentally attempted, the
891 routing retry time is respected, so at least it doesn't keep
892 hammering the server.
894 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
895 in ${readsocket because some servers need this prod.
897 PH/32 Added some extra debug output when updating a wait-xxx database.
899 PH/33 The hint "could be header name not terminated by colon", which has been
900 given for certain expansion errors for a long time, was not being given
901 for the ${if def:h_colon_omitted{... case.
903 PH/34 The spec says: "With one important exception, whenever a domain list is
904 being scanned, $domain contains the subject domain." There was at least
905 one case where this was not true.
907 PH/35 The error "getsockname() failed: connection reset by peer" was being
908 written to the panic log as well as the main log, but it isn't really
909 panic-worthy as it just means the connection died rather early on. I have
910 removed the panic log writing for the ECONNRESET error when getsockname()
913 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
914 runs only) independently of the message's sender address. This meant
915 that, if the 4xx error was in fact related to the sender, a different
916 message to the same recipient with a different sender could confuse
917 things. In particualar, this can happen when sending to a greylisting
918 server, but other circumstances could also provoke similar problems.
919 I have changed the default so that the retry time for these errors is now
920 based a combination of the sender and recipient addresses. This change
921 can be overridden by setting address_retry_include_sender=false in the
924 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
925 remote server are returned as part of bounce messages. This was not
926 happening for LMTP over a pipe (the lmtp transport), but now it is the
927 same for both kinds of LMTP.
929 PH/38 Despite being documented as not happening, Exim was rewriting addresses
930 in header lines that were in fact CNAMEs. This is no longer the case.
932 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
933 and queue runs started by the daemon processed all messages. This has
934 been fixed so that -R and -S can now usefully be given with -q<time>.
936 PH/40 Import PCRE release 6.7 (fixes some bugs).
938 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
940 PH/42 Give an error if -q is specified more than once.
942 PH/43 Renamed the variables $interface_address and $interface_port as
943 $received_ip_address and $received_port, to make it clear that these
944 values apply to message reception, and not to the outgoing interface when
945 a message is delivered. (The old names remain recognized, of course.)
947 PH/44 There was no timeout on the connect() call when using a Unix domain
948 socket in the ${readsocket expansion. There now is.
950 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
951 be meaningful with "accept".
953 SC/01 Eximstats V1.43
954 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
956 SC/02 Eximstats V1.44
957 Use a glob alias rather than an array ref in the generated
958 parser. This improves both readability and performance.
960 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
961 Collect SpamAssassin and rejection statistics.
962 Don't display local sender or destination tables unless
963 there is data to show.
964 Added average volumes into the top table text output.
966 SC/04 Eximstats V1.46
967 Collect data on the number of addresses (recipients)
968 as well as the number of messages.
970 SC/05 Eximstats V1.47
971 Added 'Message too big' to the list of mail rejection
972 reasons (thanks to Marco Gaiarin).
974 SC/06 Eximstats V1.48
975 Mainlog lines which have GMT offsets and are too short to
976 have a flag are now skipped.
978 SC/07 Eximstats V1.49 (Alain Williams)
979 Added the -emptyok flag.
981 SC/08 Eximstats V1.50
982 Fixes for obtaining the IP address from reject messages.
984 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
985 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
986 whitesspace changes from 4.64-PH/27
988 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
991 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
992 are found, allow negative numbers in numeric criteria)
994 JJ/06 exipick.20061117.2, added new $message_body_missing variable
996 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
997 to match changes made in 4.64-PH/43
999 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1001 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1002 "domain" is actually an IP address, and force a failure. This locks out
1003 those revolvers/nameservers that support "A-for-A" lookups, in
1004 contravention of the specifications.
1006 PH/48 When a host name was looked up from an IP address, and the subsequent
1007 forward lookup of the name timed out, the host name was left in
1008 $sender_host_name, contrary to the specification.
1010 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1011 restricted to single-key lookups, Exim was not diagnosing an error if
1012 * or *@ was used with a query-style lookup.
1014 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1016 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1017 long ago, but noone interested enough thought of it. Let's just say that
1018 the "1.1" means that there are some new functions that weren't there at
1019 some point in the past.
1021 PH/51 Error processing for expansion failure of helo_data from an smtp
1022 transport during callout processing was broken.
1024 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1025 tested/used via the -bh/-bhc/-bs options.
1027 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1028 bug, fixed in subsequent PCRE releases).
1030 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1031 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1033 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1039 SC/01 Use a glob alias rather than an array ref in eximstats generated
1040 parser. This improves both readability and performance.
1042 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1043 Don't display local sender or destination tables in eximstats unless
1044 there is data to show.
1045 Added average volumes into the eximstats top table text output.
1047 SC/03 Collect data on the number of addresses (recipients) as well
1048 as the number of messages in eximstats.
1050 TF/01 Correct an error in the documentation for the redirect router. Exim
1051 does (usually) call initgroups() when daemonizing.
1053 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1054 with consistent privilege compared to when running as a daemon.
1056 TF/03 Note in the spec that $authenticated_id is not set for local
1057 submissions from trusted users.
1059 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1060 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1062 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1063 by adding some example configuration directives to the default
1064 configuration file. A little bit of work is required to uncomment the
1065 directives and define how usernames and passwords are checked, but
1066 there is now a framework to start from.
1068 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1069 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1070 without this. I don't know how relevant this is to other LDAP libraries.
1072 PH/02 Add the verb name to the "unknown ACL verb" error.
1074 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1076 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1078 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1079 directory not expanded when it should be if an expanded home directory
1080 was set for the address (which is overridden by the transport).
1082 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1085 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1086 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1087 because it is too late at that time, and has no effect.
1089 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1090 security issue with \' (bugzilla #107). I could not use the
1091 PQescapeStringConn() function, because it needs a PGconn value as one of
1094 PH/08 When testing addresses using -bt, indicate those final addresses that
1095 are duplicates that would not cause an additional delivery. At least one
1096 person was confused, thinking that -bt output corresponded to deliveries.
1097 (Suppressing duplicates isn't a good idea as you lose the information
1098 about possibly different redirections that led to the duplicates.)
1100 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1101 systems where poll() doesn't work, in particular OS X.
1103 PH/10 Added more information to debugging output for retry time not reached.
1105 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1106 operations in malware.c.
1108 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1111 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1112 syslog_duplication set false, log lines that would normally be written
1113 both the the main log and to the reject log were not written to syslog at
1116 PH/14 In the default configuration, change the use of "message" in ACL warn
1117 statements to "add_header".
1119 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1120 not followed by a command (e.g. "seen endif").
1122 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1123 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1126 PH/17 Added extra conditions to the default value of delay_warning_condition
1130 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1131 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1132 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1135 The Auto-Submitted: and various List- headers are standardised, whereas I
1136 don't think Precedence: ever was.
1138 PH/18 Refactored debugging code in route_finduser() to show more information,
1139 in particular, the error code if getpwnam() issues one.
1141 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1142 This is apparently needed in addition to the PH/07 change above to avoid
1143 any possible encoding problems.
1145 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1146 but not after initializing Perl.
1148 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1149 output them only if debugging. By default they are written stderr,
1150 apparently, which is not desirable.
1152 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1155 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1158 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1160 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1161 authenticated or an ident call has been made. Suppress the default
1162 values for $authenticated_id and $authenticated_sender (but permit -oMai
1163 and -oMas) when testing with -bh.
1165 PH/24 Re-jigged the order of the tests in the default configuration so that the
1166 tests for valid domains and recipients precede the DNS black list and CSA
1167 tests, on the grounds that those ones are more expensive.
1169 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1170 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1171 This bug exists in every version of Exim that I still have, right back to
1174 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1175 However, an attempt to turn on case-sensitivity in a regex key by
1176 including (?-i) didn't work because the subject string was already
1177 lowercased, and the effects were non-intuitive. It turns out that a
1178 one-line patch can be used to allow (?-i) to work as expected.
1184 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1185 other effects) broke the use of negated acl sub-conditions.
1187 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1190 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1191 "Deny" causes Exim to reject the incoming connection with a 554 error.
1192 Unfortunately, if there is a major crisis, such as a disk failure,
1193 tcp-wrappers gives "deny", whereas what one would like would be some
1194 kind of temporary error. A kludge has been added to help with this.
1195 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1196 554 error is used if errno is still zero or contains ENOENT (which occurs
1197 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1200 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1202 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1203 errors. Otherwise a message that provokes a temporary error (when other
1204 messages do not) can cause a whole host to time out.
1206 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1207 addresses were routed directly to files or pipes from a redirect router.
1208 File deliveries just didn't batch; pipe deliveries might have suffered
1211 PH/06 A failure to get a lock for a hints database would erroneously always say
1212 "Failed to get write lock", even when it was really a read lock.
1214 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1215 of 0600. This has been changed to use the value of the lockfile_mode
1216 option (which defaults to 0600).
1218 PH/08 Applied small patch from the Sieve maintainer.
1220 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1221 folder from quota calculations, a direct delivery into this folder messed
1222 up the contents of the maildirsize file. This was because the regex was
1223 used only to exclude .Trash (or whatever) when the size of the mailbox
1224 was calculated. There was no check that a delivery was happening into an
1225 excluded directory. This bug has been fixed by ignoring all quota
1226 processing for deliveries into excluded directories.
1228 PH/10 Added the maildirfolder_create_regex option to appendfile.
1234 PH/01 The code for finding all the local interface addresses on a FreeBSD
1235 system running IPv6 was broken. This may well have applied to all BSD
1236 systems, as well as to others that have similar system calls. The broken
1237 code found IPv4 interfaces correctly, but gave incorrect values for the
1238 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1239 that it would not match correctly against @[] and not recognize the IPv6
1242 PH/02 The ipliteral router was not recognizing addresses of the form user@
1243 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1245 PH/03 Added disable_ipv6.
1247 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1248 decoded form, because it is most often used to construct To: headers
1249 lines in autoreplies, and the decoded form may well be syntactically
1250 invalid. However, $reply_address has leading white space removed, and all
1251 newlines turned into spaces so that the autoreply transport does not
1254 PH/05 If group was specified without a user on a router, and no group or user
1255 was specified on a transport, the group from the router was ignored.
1257 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1258 for visible compile-time settings that can be used to change these
1259 numbers, for those that want even more. Backwards compatibility with old
1260 spool files has been maintained. However, going back to a previous Exim
1261 release will lost any variables that are in spool files.
1263 PH/07 Two small changes when running in the test harness: increase delay when
1264 passing a TCP/IP connection to a new process, in case the original
1265 process has to generate a bounce, and remove special handling of
1266 127.0.0.2 (sic), which is no longer necessary.
1268 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1269 be the same on different OS.
1271 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1274 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1275 whether --show-vars was specified or not
1277 JJ/02 exipick: Added support for new ACL variable spool format introduced
1280 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1281 syntactically invalid From: or Reply-to: line, and a filter used this to
1282 generate an autoreply, and therefore failed to obtain an address for the
1283 autoreply, Exim could try to deliver to a non-existent relative file
1284 name, causing unrelated and misleading errors. What now happens is that
1285 it logs this as a hard delivery error, but does not attempt to create a
1288 PH/11 The exinext utility has a -C option for testing purposes, but although
1289 the given file was scanned by exinext itself; it wasn't being passed on
1290 when Exim was called.
1292 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1293 an end-of-file indication when reading a command response.
1295 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1296 compiled. In many other places in Exim, IPv6 addresses are always
1297 recognized, so I have changed this. It also means that IPv4 domain
1298 literals of the form [IPV4:n.n.n.n] are now always recognized.
1300 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1301 used if the router is not running as root, for example, when verifying at
1302 ACL time, or when using -bh. The debugging output from this situation was
1303 non-existent - all you got was a failure to exec. I have made two
1306 (a) Failures to set uid/gid, the current directory, or a process leader
1307 in a subprocess such as that created by queryprogram now generate
1308 suitable debugging ouput when -d is set.
1310 (b) The queryprogram router detects when it is not running as root,
1311 outputs suitable debugging information if -d is set, and then runs
1312 the subprocess without attempting to change uid/gid.
1314 PH/15 Minor change to Makefile for building test_host (undocumented testing
1317 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1318 additional section of a DNS packet that returns MX or SRV records.
1319 Instead, it always explicitly searches for A/AAAA records. This avoids
1320 major problems that occur when a DNS server includes only records of one
1321 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1322 fixed another bug: if SRV records were looked up and the corresponding
1323 address records were *not* found in the additional section, the port
1324 values from the SRV records were lost.
1326 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1327 using the correct key (the original address) when searching the retry
1328 rules in order to find which one to use for generating the retry hint.
1330 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1331 adding the default one. Similarly, if it contains a Reply-To: header, the
1332 errors_reply_to option, if set, is not used.
1334 PH/19 When calculating a retry time, Exim used to measure the "time since
1335 failure" by looking at the "first failed" field in the retry record. Now
1336 it does not use this if it is later than than the arrival time of the
1337 message. Instead it uses the arrival time. This makes for better
1338 behaviour in cases where some deliveries succeed, thus re-setting the
1339 "first failed" field. An example is a quota failure for a huge message
1340 when small messages continue to be delivered. Without this change, the
1341 "time since failure" will always be short, possible causing more frequent
1342 delivery attempts for the huge message than are intended.
1343 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1345 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1346 $1, $2, $3) because the numerical variables can be reset during some
1347 expansion items (e.g. "match"), thereby losing the authentication data.
1349 PH/21 Make -bV show the size of off_t variables so that the test suite can
1350 decide whether to run tests for quotas > 2G.
1352 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1353 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1354 filecount value is greater than 2G or if a quota value is greater than 2G
1355 on a system where the size of off_t is not greater than 4, a panic error
1358 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1359 never match. The debug and -bh output now contains an explicit error
1360 message indicating a malformed IPv4 address or mask.
1362 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1363 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1364 PH/23 above applies.
1366 PH/25 Do not write to syslog when running in the test harness. The only
1367 occasion when this arises is a failure to open the main or panic logs
1368 (for which there is an explicit test).
1370 PH/26 Added the /no_tell option to "control=freeze".
1372 PH/27 If a host name lookup failed very early in a connection, for example, if
1373 the IP address matched host_lookup and the reverse lookup yielded a name
1374 that did not have a forward lookup, an error message of the form "no IP
1375 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1376 could be logged. Now it outputs the IP address instead of "NULL".
1378 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1379 allows the sender and the authenticated sender to be set when
1380 submitting a message from within Exim. Since child_open_exim() is
1381 documented for local_scan(), the new function should be too.
1383 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1384 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1385 results in an empty string is now treated as unset.
1387 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1389 PH/31 Added sender_verify_fail logging option.
1391 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1392 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1393 tidied the source and removed it altogether.
1395 PH/33 When a queue run was abandoned because the load average was too high, a
1396 log line was always written; now it is written only if the queue_run log
1397 selector is set. In addition, the log line for abandonment now contains
1398 information about the queue run such as the pid. This is always present
1399 in "start" and "stop" lines but was omitted from the "abandon" line.
1401 PH/34 Omit spaces between a header name and the colon in the error message that
1402 is given when verify = headers_syntax fails (if there are lots of them,
1403 the message gets confusing).
1405 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1406 names, as there are now some PTR records that contain slashes. This check
1407 is only to protect against broken name servers that fall over on strange
1408 characters, so the fact that it applies to all lookups doesn't matter.
1410 PH/36 Now that the new test suite is complete, we can remove some of the
1411 special code in Exim that was needed for the old test suite. For example,
1412 sorting DNS records because real resolvers return them in an arbitrary
1413 order. The new test suite's fake resolver always returns records in the
1416 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1417 bounces) except when queue_only is set, to avoid logging races between
1418 the different processes.
1420 PH/38 Panic-die if .include specifies a non-absolute path.
1422 PH/39 A tweak to the "H" retry rule from its user.
1424 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1425 a label. They prevented compilation on older perls.
1427 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1428 a warning to be raised on newish perls.
1430 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1431 on queue. Changes to match documented behaviour of showing count of
1432 messages matching specified criteria.
1434 PH/40 Changed the default ident timeout from 30s to 5s.
1436 PH/41 Added support for the use of login_cap features, on those BSD systems
1437 that have them, for controlling the resources used by pipe deliveries.
1439 PH/42 The content-scanning code uses fopen() to create files in which to put
1440 message data. Previously it was not paying any attention to the mode of
1441 the files. Exim runs with umask(0) because the rest of the code creates
1442 files with open(), and sets the required mode explicitly. Thus, these
1443 files were ending up world-writeable. This was not a big issue, because,
1444 being within the spool directory, they were not world-accessible. I have
1445 created a function called modefopen, which takes an additional mode
1446 argument. It sets umask(777), creates the file, chmods it to the required
1447 mode, then resets the umask. All the relevant calls to fopen() in the
1448 content scanning code have been changed to use this function.
1450 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1451 to 24 hours. This avoids potential overflow problems when processing G
1452 and H retry rules. I suspect nobody ever tinkers with this value.
1454 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1456 PH/45 When the plaintext authenticator is running as a client, the server's
1457 challenges are checked to ensure they are valid base64 strings. By
1458 default, the authentication attempt is cancelled if an invalid string is
1459 received. Setting client_ignore_invalid_base64 true ignores these errors.
1460 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1461 they are received. Thus, the responses can be made to depend on the
1462 challenges. If an invalid string is ignored, an empty string is placed in
1465 PH/46 Messages that are created by the autoreply transport now contains a
1466 References: header, in accordance with RFCs 2822 and 3834.
1468 PH/47 Added authenticated_sender_force to the smtp transport.
1470 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1472 PH/49 Installed latest patch from the Sieve maintainer.
1474 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1475 was also set, the appendfile transport was unnecessarily scanning a
1476 directory of message files (e.g. for maildir delivery) to find the count
1477 of files (along with the size), even though it did not need this
1478 information. It now does the scan only if it needs to find either the
1479 size of the count of files.
1481 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1483 PH/52 Two bugs concerned with error handling when the smtp transport is
1486 (i) Exim was not creating retry information for temporary errors given
1487 for individual recipients after the DATA command when the smtp transport
1488 was used in LMTP mode. This meant that they could be retried too
1489 frequently, and not timed out correctly.
1491 (ii) Exim was setting the flag that allows error details to be returned
1492 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1493 recipients that were returned after the DATA command.
1495 PH/53 This is related to PH/52, but is more general: for any failing address,
1496 when detailed error information was permitted to be returned to the
1497 sender, but the error was temporary, then after the final timeout, only
1498 "retry timeout exceeded" was returned. Now it returns the full error as
1499 well as "retry timeout exceeded".
1501 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1502 do this, and (what is worse) MTAs that accept it.
1504 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1505 will now be deprecated.
1507 PH/56 New os.c-cygwin from the Cygwin maintainer.
1509 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1510 formats (previously only available in exim formats via -bpr, -bpru,
1511 and -bpra. Now also available in native and exiqgrep formats)
1513 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1514 with very large, slow to parse queues
1516 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1518 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1520 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1521 responses to authentication challenges, though it was showing the
1522 challenges; (ii) I've removed the CR characters from the debug output for
1525 PH/58 Allow for the insertion of a newline as well as a space when a string
1526 is turned into more than one encoded-word during RFC 2047 encoding. The
1527 Sieve code now uses this.
1529 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1530 data_4xx, lost_connection, tls_required.
1532 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1533 message was being sent as an SMTP response.
1535 PH/61 Add -l and -k options to exicyclog.
1537 PH/62 When verifying, if an address was redirected to one new address, so that
1538 verification continued, and the new address failed or deferred after
1539 having set something in $address_data, the value of $address_data was not
1540 passed back to the ACL. This was different to the case when no
1541 redirection occurred. The value is now passed back in both cases.
1543 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1544 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1545 use login_cap.h, so on its own it isn't the distinguishing feature. The
1546 new name refers directly to the setclassresources() function.
1548 PH/65 Added configuration files for NetBSD3.
1550 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1552 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1553 is preferred over IPv4.
1555 PH/68 The bounce_return_message and bounce_return_body options were not being
1556 honoured for bounces generated during the reception of non-SMTP messages.
1557 In particular, this applied to messages rejected by the ACL. This bug has
1558 been fixed. However, if bounce_return_message is true and bounce_return_
1559 body is false, the headers that are returned for a non-SMTP message
1560 include only those that have been read before the error was detected.
1561 (In the case of an ACL rejection, they have all been read.)
1563 PH/69 The HTML version of the specification is now built in a directory called
1564 spec_html instead of spec.html, because the latter looks like a path with
1565 a MIME-type, and this confuses some software.
1567 PH/70 Catch two compiler warnings in sieve.c.
1569 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1570 function verify_get_ident() calls ip_connect() to connect a socket, but
1571 if the "connect()" function timed out, ip_connect() used to close the
1572 socket. However, verify_get_ident() also closes the socket later, and in
1573 between Exim writes to the log, which may get opened at this point. When
1574 the socket was closed in ip_connect(), the log could get the same file
1575 descriptor number as the socket. This naturally causes chaos. The fix is
1576 not to close the socket in ip_connect(); the socket should be closed by
1577 the function that creates it. There was only one place in the code where
1578 this was missing, in the iplookup router, which I don't think anybody now
1579 uses, but I've fixed it anyway.
1581 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1582 well as to direct DNS lookups. Otherwise the handling of names in host
1583 lists is inconsistent and therefore confusing.
1589 PH/01 Two changes to the default runtime configuration:
1591 (1) Move the checks for relay_from_hosts and authenticated clients from
1592 after to before the (commented out) DNS black list checks.
1594 (2) Add control=submission to the relay_from_hosts and authenticated
1595 clients checks, on the grounds that messages accepted by these
1596 statements are most likely to be submissions.
1598 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1600 (1) Generate an error if the third argument for the ${prvs expansion is
1603 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1606 (3) Reset the variables that are obtained from the first argument of
1607 ${prvscheck and used in the second argument before leaving the code,
1608 because their memory is reclaimed, so using them afterwards may do
1611 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1612 one (it's much easier than Tom thought :-).
1614 (5) Because of (4), we can now allow for the use of $prvscheck_result
1615 inside the third argument.
1617 PH/03 For some reason, the default setting of PATH when running a command from
1618 a pipe transport was just "/usr/bin". I have changed it to
1621 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1622 anything to be listed in the output from -bV.
1624 PH/05 When a filter generated an autoreply, the entire To: header line was
1625 quoted in the delivery log line, like this:
1627 => >A.N.Other <ano@some.domain> <original@ddress> ...
1629 This has been changed so that it extracts the operative address. There
1630 may be more than one such address. If so, they are comma-separated, like
1633 => >ano@some.domain,ona@other.domain <original@ddress> ...
1635 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1636 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1637 not being looked up in the rDNS to get a host name, Exim was showing the
1638 IP address twice in Received: lines, even though the IP addresses were
1639 identical. For example:
1641 Received: from [1.2.3.4] (helo=[1.2.3.4])
1643 However, if the real host name was known, it was omitting the HELO data
1644 if it matched the actual IP address. This has been tidied up so that it
1645 doesn't show the same IP address twice.
1647 PH/07 When both +timestamp and +memory debugging was on, the value given by
1648 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1649 called by the debug printing, thereby overwriting the timestamp buffer.
1650 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1652 PH/08 When the original message was included in an autoreply transport, it
1653 always said "this is a copy of the message, including all the headers",
1654 even if body_only or headers_only was set. It now gives an appropriate
1657 PH/09 Applied a patch from the Sieve maintainer which:
1659 o fixes some comments
1660 o adds the (disabled) notify extension core
1661 o adds some debug output for the result of if/elsif tests
1662 o points to the current vacation draft in the documentation
1663 and documents the missing references header update
1667 o fixes a bug in processing the envelope test (when testing
1668 multiple envelope elements, the last element determinted the
1671 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1672 Electronic Mail") by including:
1674 Auto-submitted: auto-generated
1676 in the messages that it generates (bounce messages and others, such as
1677 warnings). In the case of bounce messages for non-SMTP mesages, there was
1678 also a typo: it was using "Auto_submitted" (underscore instead of
1679 hyphen). Since every message generated by Exim is necessarily in response
1680 to another message, thes have all been changed to:
1682 Auto-Submitted: auto-replied
1684 in accordance with these statements in the RFC:
1686 The auto-replied keyword:
1688 - SHOULD be used on messages sent in direct response to another
1689 message by an automatic process,
1691 - MUST NOT be used on manually-generated messages,
1693 - MAY be used on Delivery Status Notifications (DSNs) and Message
1694 Disposition Notifications (MDNs),
1696 - MUST NOT be used on messages generated by automatic or periodic
1697 processes, except for messages which are automatic responses to
1700 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1701 to the default Received: header definition.
1703 PH/12 Added log selector acl_warn_skipped (default on).
1705 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1706 variables because (a) they are in the wrong storage pool and (b) even if
1707 they were copied, it wouldn't work properly because of the caching.
1709 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1710 checking when decoding. Apparently there are clients that generate
1711 overlong encoded strings. Why am I not surprised?
1713 PH/15 If the first argument of "${if match_address" was not empty, but did not
1714 contain an "@" character, Exim crashed. Now it writes a panic log message
1715 and treats the condition as false.
1717 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1719 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1720 extension "envelope-auth". The code is finished and in agreement with
1721 other implementations, but there is no documentation so far and in fact,
1722 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1723 not changing the active code.
1725 Print executed "if" and "elsif" statements when debugging is used. This
1726 helps a great deal to understand what a filter does.
1728 Document more things not specified clearly in RFC3028. I had all this
1729 sorted out, when out of a sudden new issues came to my mind. Oops."
1731 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1734 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1735 canonical form (as documented). However, after a host name lookup from
1736 the IP address, check_host() was doing a simple string comparison with
1737 addresses acquired from the DNS when checking that the found name did
1738 have the original IP as one of its addresses. Since any found IPv6
1739 addresses are likely to be in abbreviated form, the comparison could
1740 fail. Luckily, there already exists a function for doing the comparison
1741 by converting both addresses to binary, so now that is used instead of
1742 the text comparison.
1744 PH/20 There was another similar case to PH/19, when a complete host name was
1745 given in a host list; looking up its IP address could give an abbreviated
1746 form, whereas the current host's name might or might not be abbreviated.
1747 The same fix has been applied.
1753 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1754 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1757 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1758 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1760 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1762 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1763 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1764 Also, for "Auto-Submitted", treat anything other than "no" as
1765 non-personal, in accordance with RFC 3834. (Previously it treated
1766 anything starting "auto-" as non-personal.)
1768 TF/01 The control=submission/name=... option had a problem with syntax
1769 errors if the name included a slash character. The /name= option
1770 now slurps the rest of the string, so it can include any characters
1771 but it must come last in the list of options (after /sender_retain
1774 PH/05 Some modifications to the interface to the fake nameserver for the new
1782 TK/01 Added the "success_on_redirect" address verification option. See
1783 NewStuff for rationale and an example.
1785 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1787 PH/02 Patch to exigrep to allow it to work on syslog lines.
1789 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1790 fread() to skip over the body file's header line, because in Cygwin the
1791 header line is locked and is inaccessible.
1793 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1794 co-exist for some time) to make it clear that it is the Exim ID that is
1795 referenced, not the Message-ID: header line.
1797 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1798 string_format() function, because snprintf() does not exist on all
1801 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1802 ${stat: expansion item.
1804 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1805 protocol synchronization error", to keep the pedants happy.
1807 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1808 well as for IRIX systems, when gcc is being used. See the host.c source
1811 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1813 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1816 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1817 transport and to the smtp transport in LMTP mode.
1819 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1821 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1822 run a filter in a subprocess. This could lead to confusion in subsequent
1823 lookups in the parent process. There should also be a search_tidyup() at
1824 the end of the subprocess.
1826 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1827 only if the host matched helo_try_verify_hosts, which caused the
1828 verification to occur when the EHLO/HELO command was issued. The ACL just
1829 tested the remembered result. Now, if a previous verification attempt has
1830 not happened, "verify = helo" does it there and then.
1832 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1834 TK/03 Fix log output including CR from clamd.
1836 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1837 exist provoked a memory error which could cause a segfault.
1839 PH/15 Installed PCRE 6.2
1841 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1843 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1844 of the problem. Specifically, suggested +O2 rather than +O1 for the
1847 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1849 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1850 record, the port was not correctly logged when the outgoing_port log
1851 selector was set (it logged the transort's default port).
1853 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1854 fallback_hosts, and "hosts" in the smtp transport.
1856 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1857 host errors such as "Connection refused".
1859 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1860 authentication with radiusclient 0.4.9:
1862 - Error returned from rc_read_config was caught wrongly
1863 - Username/password not passed on to radius server due to wrong length.
1865 The presumption is that some radiusclient API changes for 4.51/PH/17
1866 were not taken care of correctly. The code is still untested by me (my
1867 Linux distribution still has 0.3.2 of radiusclient), but it was
1868 contributed by a Radius user.
1870 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1871 expanding the "port" option of the smtp transport.
1873 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1874 while reading a MIME header. Thanks to Tom Hughes for a patch.
1876 PH/24 Include config.h inside local_scan.h so that configuration settings are
1879 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1880 that in an ACL for RCPT (for example), you can examine exactly what was
1883 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1884 commands, but it was not correctly comparing the address with the actual
1885 client host address. Thus, it would show the EHLO address in Received:
1886 header lines when this was not necessary.
1888 PH/27 Added the % operator to ${eval:}.
1890 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1891 it should be ignoring failures (because with -C, for example, it has lost
1892 privilege). It wasn't ignoring creation failures other than "already
1895 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1898 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1899 because an input line was too long, either on its own, or by virtue of
1900 too many continuations, the temporary file was not being removed, and the
1901 return code was incorrect.
1903 PH/31 Missing "BOOL" in function definition in filtertest.c.
1905 PH/32 Applied Sieve patches from the maintainer.
1907 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1909 PH/33 Added "verify = not_blind".
1911 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1912 Local/Makefile (with some defaults set). These are used in built scripts
1913 such as exicyclog, but they have never been used in the exim_install
1914 script (though there are many overriding facilities there). I have
1915 arranged that the exim_install script now takes note of these two
1918 PH/35 Installed configuration files for Dragonfly.
1920 PH/36 When a locally submitted message by a trusted user did not contain a
1921 From: header, and the sender address was obtained from -f or from an SMTP
1922 MAIL command, and the trusted user did not use -F to supply a sender
1923 name, $originator_name was incorrectly used when constructing a From:
1924 header. Furthermore, $originator_name was used for submission mode
1925 messages from external hosts without From: headers in a similar way,
1926 which is clearly wrong.
1928 PH/37 Added control=suppress_local_fixups.
1930 PH/38 When log_selector = +received_sender was set, and the addition of the
1931 sender made the log line's construction buffer exactly full, or one byte
1932 less than full, an overflow happened when the terminating "\n" was
1935 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1936 when the result of a list match is failure because a DNS lookup failed.
1938 PH/40 RM_COMMAND is now used in the building process.
1940 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1941 the "build-* directories that it finds.
1943 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1944 address in a domain literal was a prefix of an interface address.
1946 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1947 when verifying a sender address, unless rewrite_headers is false.
1949 PH/44 Wrote a long comment about why errors_to addresses are verified as
1950 recipients, not senders.
1952 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1953 the ratelimit ACL was added.
1955 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1957 PH/46 Added extra information about PostgreSQL errors to the error string.
1959 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1960 avoiding the need to install special zones in a real server. This is
1961 backwards compatible; if it can't find the fake resolver, it drops back.
1962 Thus, both old and new test suites can be run.
1964 TF/02 Added util/ratelimit.pl
1966 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1967 clock is set back in time.
1969 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1970 Candler <B.Candler@pobox.com>.
1972 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1973 OK for addresses that are the result of redirections.
1975 PH/48 A number of further additions for the benefit of the new test suite,
1976 including a fake gethostbyname() that interfaces to the fake DNS resolver
1979 TF/06 The fix for widen_domains has also been applied to qualify_single and
1980 search_parents which are the other dnslookup options that can cause
1983 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1986 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1988 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1989 service" if it encountered an MX record with an empty target hostname.
1990 The message is now "an MX or SRV record indicated no SMTP service".
1992 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1993 if the DNS of the sending site is misconfigured. This is quite a
1994 common situation. This change restores the behaviour of treating a
1995 helo verification defer as a failure.
1997 PH/51 If self=fail was set on a router, the bounce message did not include the
1998 actual error message.
2004 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2006 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2007 command itself ended in error, the underlying message about the transport
2008 filter timeout was being overwritten with the pipe command error. Now the
2009 underlying error message should be appended to the second error message.
2011 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2013 PH/02 Reduce the amount of output that "make" produces by default. Full output
2014 can still be requested.
2016 PH/03 The warning log line about a condition test deferring for a "warn" verb
2017 was being output only once per connection, rather than after each
2018 occurrence (because it was using the same function as for successful
2019 "warn" verbs). This seems wrong, so I have changed it.
2021 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2022 it should not have, which might have caused a crash in the right
2023 circumstances, but probably never did.
2025 PH/04 Installed a modified version of Tony Finch's patch to make submission
2026 mode fix the return path as well as the Sender: header line, and to
2027 add a /name= option so that you can make the user's friendly name appear
2030 TF/03 Added the control = fakedefer ACL modifier.
2032 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2033 Mark Lowes for thorough testing.
2035 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2037 TK/03 Merged latest SRS patch from Miles Wilton.
2039 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2040 with the definition in sysexits.h (which is #included earlier).
2041 Fortunately, Exim does not actually use EX_OK. The code used to try to
2042 preserve the sysexits.h value, by assumimg that macro definitions were
2043 scanned for macro replacements. I have been disabused of this notion,
2044 so now the code just undefines EX_OK before #including unistd.h.
2046 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2047 in the smtp transport. When a block could not be written in a single
2048 write() function, the timeout was being re-applied to each part-write.
2049 This seems wrong - if the receiver was accepting one byte at a time it
2050 would take for ever. The timeout is now adjusted when this happens. It
2051 doesn't have to be particularly precise.
2053 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2054 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2056 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2057 without a domain if the check to <postmaster@domain> fails.
2059 SC/01 Eximstats: added -xls and the ability to specify output files
2060 (patch written by Frank Heydlauf).
2062 SC/02 Eximstats: use FileHandles for outputing results.
2064 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2066 SC/04 Eximstats: fixed display of large numbers with -nvr option
2068 SC/05 Eximstats: fixed merging of reports with empty tables.
2070 SC/06 Eximstats: added the -include_original_destination flag
2072 SC/07 Eximstats: removed tabs and trailing whitespace.
2074 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2076 TK/06 MBOX spool code: Add real "From " MBOX separator line
2077 so the .eml file is really in mbox format (even though
2078 most programs do not really care). Patch from Alex Miller.
2080 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2081 The latter is generated from $received_to and is only set if the
2082 message has one envelope recipient. SA can use these headers,
2083 obviously out-of-the-box. Patch from Alex Miller.
2085 PH/08 The ${def test on a variable was returning false if the variable's
2086 value was "0", contrary to what the specification has always said!
2087 The result should be true unless the variable is empty.
2089 PH/09 The syntax error of a character other than { following "${if
2090 def:variable_name" (after optional whitespace) was not being diagnosed.
2091 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2092 accidental colon was present, for example, could give incorrect results.
2094 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2095 result is used (not including appendfile, where other changes are about
2098 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2099 This involved changing a lot of size variables from int to off_t. It
2100 should work with maildirs and everything.
2102 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2103 spamd dying while we are connected to it.
2105 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2108 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2109 function for BDB 4.3.
2111 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2113 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2114 the sources for PCRE have been re-arranged and more files are now
2117 PH/15 The code I had for printing potentially long long variables in PH/11
2118 above was not the best (it lost precision). The length of off_t variables
2119 is now inspected at build time, and an appropriate printing format (%ld
2120 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2121 to be "long long int" or "long int". This is needed for the internal
2122 formatting function string_vformat().
2124 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2125 the configuration file to be ":syslog", then the script "guesses" where
2126 the logs files are, rather than using the compiled in default. In our
2127 case the guess is not the same as the compiled default, so the script
2128 suddenly stopped working when I started to use syslog. The patch checks
2129 to see if log_file_path is "". If so, it attempts to read it from exim
2130 with no configuration file to get the compiled in version, before it
2131 falls back to the previous guessing code."
2133 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2134 implementing BATV in an Exim configuration. See NewStuff for the gory
2137 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2138 Makefile that are specific to HP-UX.
2140 PH/18 If the "use_postmaster" option was set for a recipient callout together
2141 with the "random" option, the postmaster address was used as the MAIL
2142 FROM address for the random test, but not for the subsequent recipient
2143 test. It is now used for both.
2145 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2146 patch removes a few documentation additions to RFC 3028, because the
2147 latest draft now contains them. It adds the new en;ascii-case comparator
2148 and a new error check for 8bit text in MIME parts. Comparator and
2149 require names are now matched exactly. I enabled the subaddress
2150 extension, but it is not well tested yet (read: it works for me)."
2152 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2153 rework some of the code of TK/09 above to avoid the hardwired use of
2154 "%lld" and "long long". Replaced the call to snprintf() with a call to
2157 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2158 records point to non-existent hosts", "retry timeout exceeded", and
2159 "retry time not reached for any host after a long failure period".
2161 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2162 experimental DomainKeys support:
2164 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2165 (2) On an error such as an illegally used "control", the wrong name for
2166 the control was given.
2168 These problems did NOT occur unless DomainKeys support was compiled.
2170 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2172 PH/24 Added ${if match_ip condition.
2174 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2175 databases so that it will be absolutely obvious if a crash occurs in the
2176 DB library. This is a regular occurrence (often caused by mis-matched
2179 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2180 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2181 picked up on a user's system that detects such things. There doesn't seem
2182 to be a gcc warning option for this - only an attribute that has to be
2183 put on the function's prototype. It seems that in Fedora Core 4 they have
2184 set this on a number of new functions. No doubt there will be more in due
2187 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2188 specify a transport. However, if an address that was verified by such a
2189 router was the subject of a callout, Exim crashed because it tried to
2190 read the rcpt_include_affixes from the non-existent transport. Now it
2191 just assumes that the setting of that option is false. This bug was
2192 introduced by 4.51/PH/31.
2194 PH/28 Changed -d+all to exclude +memory, because that information is very
2195 rarely of interest, but it makes the output a lot bigger. People tend to
2196 do -d+all out of habit.
2198 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2199 code in os-type was giving problems when libc.so lives in lib64, like on
2202 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2203 aren't the modern standard, and it seems that some systems' include files
2204 don't always have them. Exim was already checking for some of the newer
2205 ones like T_AAAA, and defining it itself. I've added checks for all the
2206 record types that Exim uses.
2208 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2209 not automatically generating a new one, as it is supposed to. This
2210 prevented TLS from working. If the file did exist, but contained invalid
2211 data, a new version was generated, as expected. It was only the case of a
2212 non-existent file that was broken.
2214 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2215 with a change in libdomainkeys > 0.64.
2217 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2218 from DNS. If the selector record carries the flag, it now has
2219 precedence over the domain-wide flag.
2221 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2223 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2224 the use of a port name that isn't defined in /etc/services) occurred, the
2225 message was deferred as in a normal delivery, and thus remained on the
2226 spool, instead of being failed because of the mua_wrapper setting. This
2227 is now fixed, and I tidied up some of the mua_wrapper messages at the
2230 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2231 the messages in a hash of arrays rather than using individual hashes.
2232 This is a bit cleaner and results in dramatic memory savings, albeit
2233 at a slight CPU cost.
2235 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2236 as requested by Marc Sherman.
2238 SC/10 Eximstats: added histograms for user specified patterns as requested
2241 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2243 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2244 fopen() in the content-scanning modules that did not already have it.
2250 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2251 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2253 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2255 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2257 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2258 to test Sieve filters that use "vacation".
2260 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2261 that changes the way the GnuTLS parameters are stored in the cache file.
2262 The new format can be generated externally. For backward compatibility,
2263 if the data in the cache doesn't make sense, Exim assumes it has read an
2264 old-format file, and it generates new data and writes a new file. This
2265 means that you can't go back to an older release without removing the
2268 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2269 work if there are any delivery delays because "one_time" forces the
2270 parent to be marked "delivered", so its unseen clone is never tried
2271 again. For this reason, Exim now forbids the simultaneous setting of
2274 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2275 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2276 entry if you want to know the details. The fix, however, neglected to
2277 consider the case when local delivery batching is involved. The test for
2278 "previously delivered" was not happening when checking to see if an
2279 address could be batched with a previous (undelivered) one; under
2280 certain circumstances this could lead to multiple deliveries to the same
2283 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2284 in its include files, and this causes problems building Exim.
2286 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2287 header_syntax) but Exim was just ignoring anything given after a slash.
2288 In particular, this caused confusion with an attempt to use "verify =
2289 reverse_host_lookup/defer_ok". An error is now given when options are
2290 supplied for verify items that do not have them. (Maybe reverse_host_
2291 lookup should have a defer_ok option, but that's a different point.)
2293 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2294 defined by RFC 821) to 2048, because there were problems with some AUTH
2295 commands, and RFC 1869 says the size should be increased for extended
2296 SMTP commands that take arguments.
2298 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2301 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2302 "unknown" error; now it says that the functionality isn't in the binary.
2304 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2305 an address' error message when a string expansion fails (syntax or
2306 whatever). Otherwise the password may appear in the log. Following change
2307 PH/42 below, there is no longer a chance of it appearing in a bounce
2310 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2312 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2313 of its addresses was ever tried. (Bugzilla bug #2.)
2315 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2316 the result incorrectly in the debug output. (It correctly added a newline
2317 to what was transported.)
2319 TF/01 Added $received_time.
2321 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2322 commented out examples of how to interface to a virus scanner and to
2323 SpamAssassin. Also added commented examples of av_scanner and
2324 spamd_address settings.
2326 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2327 and controls are allowed in which ACLs. There were a couple of minor
2328 errors. Some of the entries in the conditions table (which is a table of
2329 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2330 negation of where the condition IS allowed.
2332 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2334 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2335 header file does not have a version number, so I've had to invent a new
2336 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2337 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2338 radiusclient), but it was contributed by a Radius user.
2340 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2341 files or directories, for OpenSSL.
2343 PH/19 When an Exim process that is running as root has to create an Exim log
2344 file, it does so in a subprocess that runs as exim:exim so as to get the
2345 ownership right at creation (otherwise, other Exim processes might see
2346 the file with the wrong ownership). There was no test for failure of this
2347 fork() call, which would lead to the process getting stuck as it waited
2348 for a non-existent subprocess. Forks do occasionally fail when resources
2349 run out. I reviewed all the other calls to fork(); they all seem to check
2352 PH/20 When checking for unexpected SMTP input at connect time (before writing
2353 the banner), Exim was not dealing correctly with a non-positive return
2354 from the read() function. If the client had disconnected by this time,
2355 the result was a log entry for a synchronization error with an empty
2356 string after "input=" when read() returned zero. If read() returned -1
2357 (an event I could not check), uninitialized data bytes were printed.
2358 There were reports of junk text (parts of files, etc) appearing after
2361 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2363 PH/22 Added support for macro redefinition, and (re)definition in between
2364 driver and ACL definitions.
2366 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2367 forgetting to use the resulting value; it was using the unexpanded value.
2369 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2370 hadn't been configured. The fix is from Juergen Kreileder, who
2371 understands it better than I do:
2373 "Here's what I see happening with three configured cyrus_sasl
2374 authenticators configured (plain, login, cram-md5):
2376 On startup auth_cyrus_sasl_init() gets called for each of these.
2377 This means three calls to sasl_listmech() without a specified mech_list.
2378 => SASL tests which mechs of all available mechs actually work
2379 => three warnings about OTP not working
2380 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2382 With the patch, sasl_listmech() also gets called three times. But now
2383 SASL's mech_list option is set to the server_mech specified in the the
2384 authenticator. Or in other words, the answer from sasl_listmech()
2385 gets limited to just the mech you're testing for (which is different
2387 => the return list contains just 'plain' or 'login', 'cram-md5' or
2388 nothing depending on the value of ob->server_mech.
2390 I've just tested the patch: Authentication still works fine,
2391 unavailable mechs specified in the exim configuration are still
2392 caught, and the auth.log warnings about OTP are gone."
2394 PH/25 When debugging is enabled, the contents of the command line are added
2395 to the debugging output, even when log_selector=+arguments is not
2398 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2399 answer is "GNU", and only if the return is "GNU/something" is the answer
2402 PH/27 $acl_verify_message is now set immediately after the failure of a
2403 verification in an ACL, and so is available in subsequent modifiers. In
2404 particular, the message can be preserved by coding like this:
2406 warn !verify = sender
2407 set acl_m0 = $acl_verify_message
2409 Previously, $acl_verify_message was set only while expanding "message"
2410 and "log_message" when a very denied access.
2412 PH/28 Modified OS/os.c-Linux with
2414 -#ifndef OS_LOAD_AVERAGE
2415 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2417 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2418 nomenclature these days.)
2420 PH/29 Installed patch from the Sieve maintainer that adds the options
2421 sieve_useraddress and sieve_subaddress to the redirect router.
2423 PH/30 In these circumstances:
2424 . Two addresses routed to the same list of hosts;
2425 . First host does not offer TLS;
2426 . First host accepts first address;
2427 . First host gives temporary error to second address;
2428 . Second host offers TLS and a TLS session is established;
2429 . Second host accepts second address.
2430 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2431 and peerdn, if requested) that were in fact used only for the second
2434 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2435 attention to any local part prefix or suffix that was matched by the
2436 router that accepted the address. It now behaves in the same way as it
2437 does for delivery: the affixes are removed from the local part unless
2438 rcpt_include_affixes is set on the transport.
2440 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2441 timeout during the DATA phase of an incoming message.
2443 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2444 applied a patch sanctioned by the Sieve maintainer.
2446 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2447 the uid or gid is negative. A case of a negative gid caused this to be
2448 noticed. The fix allows for either to be negative.
2450 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2451 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2454 PH/36 The change PH/12 above was broken. Fixed it.
2456 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2457 the grounds that routing the same address twice would always produce the
2458 same answer. This might have been true once, but it is certainly no
2459 longer true now. Routing a child address may depend on the previous
2460 routing that produced that child. Some complicated redirection strategies
2461 went wrong when messages had multiple recipients, and made Exim's
2462 behaviour dependent on the order in which the addresses were given.
2464 I have moved the duplicate checking until after the routing is complete.
2465 Exim scans the addresses that are assigned to local and remote
2466 transports, and removes any duplicates. This means that more work will be
2467 done, as duplicates will always all be routed, but duplicates are
2468 presumably rare, so I don't expect this is of any significance.
2470 For deliveries to pipes, files, and autoreplies, the duplicate checking
2471 still happens during the routing process, since they are not going to be
2474 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2475 It corrects a timeout issue with spamd. This is Ian's comment: "The
2476 background is that sometimes spamd either never reads data from a
2477 connection it has accepted, or it never writes response data. The exiscan
2478 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2479 blindly assumes that writes won't block so it may never time out."
2481 PH/39 Allow G after quota size as well as K and M.
2483 PH/40 The value set for $authenticated_id in an authenticator may not contain
2484 binary zeroes or newlines because the value is written to log lines and
2485 to spool files. There was no check on this. Now the value is run through
2486 the string_printing() function so that such characters are converted to
2487 printable escape sequences.
2489 PH/41 $message_linecount is a new variable that contains the total number of
2490 lines in the message. Compare $body_linecount, which is the count for the
2493 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2494 bounce and delay warning messages, except in certain special cases, which
2497 (a) An SMTP error message from a remote host;
2498 (b) A message specified in a :fail: redirection;
2499 (c) A message specified in a "fail" command in a system filter;
2500 (d) A message specified in a FAIL return from the queryprogram router;
2501 (e) A message specified by the cannot_route_message router option.
2503 In these cases only, Exim does include the error details in bounce and
2504 warning messages. There are also a few cases where bland messages such
2505 as "unrouteable address" or "local delivery error" are given.
2507 PH/43 $value is now also set for the "else" part of a ${run expansion.
2509 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2510 being worked on, but at least Exim now implements the latest version to
2513 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2514 process to complete was treated as a delivery failure, a timeout while
2515 writing the message to the pipe was logged, but erroneously treated as a
2516 successful delivery. Such timeouts include transport filter timeouts. For
2517 consistency with the overall process timeout, these timeouts are now
2518 treated as errors, giving rise to delivery failures by default. However,
2519 there is now a new Boolean option for the pipe transport called
2520 timeout_defer, which, if set TRUE, converts the failures into defers for
2521 both kinds of timeout. A transport filter timeout is now identified in
2524 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2525 systems where "make" and "gmake" are different, calling "gmake" at top
2526 level broke things. I've arranged for the value of $(MAKE) to be passed
2527 from the Makefile to this script so that it can call the same version of
2531 A note about Exim versions 4.44 and 4.50
2532 ----------------------------------------
2534 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2535 changes of various kinds. As a consequence, a big documentation update was
2536 needed. This delayed the release for rather longer than seemed good, especially
2537 in the light of a couple of (minor) security issues. Therefore, the changes
2538 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2539 release. So 4.44 and 4.50 are in effect two different branches that both start
2542 I have left the 4.50 change log unchanged; it contains all the changes since
2543 4.43. The change log for 4.44 is below; many of its items are identical to
2544 those for 4.50. This seems to be the most sensible way to preserve the
2545 historical information.
2551 1. Minor wording change to the doc/README.SIEVE file.
2553 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2554 computation of the current number of files was incorrect.
2556 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2557 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2560 4. Give more explanation in the error message when the command for a transport
2561 filter fails to execute.
2563 5. There are several places where Exim runs a non-Exim command in a
2564 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2565 was being done only for the command run by the queryprogram router. It is
2566 now done for all such subprocesses. The other cases are: ${run, transport
2567 filters, and the commands run by the lmtp and pipe transports.
2569 6. Added CONFIGURE_GROUP build-time option.
2571 7. Some older OS have a limit of 256 on the maximum number of file
2572 descriptors. Exim was using setrlimit() to set 1000 as a large value
2573 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2574 systems. I've change it so that if it can't get 1000, it tries for 256.
2576 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2577 was an oversight, and furthermore, ever since the addition of extra
2578 controls (e.g. 4.43/32), the checks on when to allow different forms of
2579 "control" were broken. There should now be diagnostics for all cases when a
2580 control that does not make sense is encountered.
2582 9. Added the /retain_sender option to "control=submission".
2584 10. $recipients is now available in the predata ACL (oversight).
2586 11. Tidy the search cache before the fork to do a delivery from a message
2587 received from the command line. Otherwise the child will trigger a lookup
2588 failure and thereby defer the delivery if it tries to use (for example) a
2589 cached ldap connection that the parent has called unbind on.
2591 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2592 of $address_data from the recipient verification was clobbered by the
2593 sender verification.
2595 13. The value of address_data from a sender verification is now available in
2596 $sender_address_data in subsequent conditions in the ACL statement.
2598 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2600 15. Added a new option "connect=<time>" to callout options, to set a different
2603 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2604 was its contents. (It was OK if the option was not defined at all.)
2606 17. A "Completed" log line is now written for messages that are removed from
2607 the spool by the -Mrm option.
2609 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2610 information about exactly what failed.
2612 19. Added -dd to debug only the daemon process.
2614 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2615 handles timeouts, both on the server side and network timeouts. Renamed the
2616 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2618 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2619 It is now set to "smtps".
2621 22. $host_address is now set to the target address during the checking of
2622 ignore_target_hosts.
2624 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2625 being passed; this would have caused $sender_host_name to have been used if
2626 matching the list had actually called for a host name (not very likely,
2627 since this list is usually IP addresses). A host name is now passed as
2630 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2631 code that specifies a non-restarting handler (typically sigaction() in
2632 modern systems) in an attempt to fix a rare and obscure crash bug.
2634 25. Narrowed the window for a race in the daemon that could cause it to ignore
2635 SIGCHLD signals. This is not a major problem, because they are used only to
2636 wake it up if nothing else does.
2638 26. A malformed maildirsize file could cause Exim to calculate negative values
2639 for the mailbox size or file count. Odd effects could occur as a result.
2640 The maildirsize information is now recalculated if the size or filecount
2643 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2644 support for a long time. Removed HAVE_SYS_VFS_H.
2646 28. Installed the latest version of exipick from John Jetmore.
2648 29. In an address list, if the pattern was not a regular expression, an empty
2649 subject address (from a bounce message) matched only if the pattern was an
2650 empty string. Non-empty patterns were not even tested. This was the wrong
2651 because it is perfectly reasonable to use an empty address as part of a
2652 database query. An empty address is now tested by patterns that are
2653 lookups. However, all the other forms of pattern expect the subject to
2654 contain a local part and a domain, and therefore, for them, an empty
2655 address still always fails if the pattern is not itself empty.
2657 30. Exim went into a mad DNS loop when attempting to do a callout where the
2658 host was specified on an smtp transport, and looking it up yielded more
2659 than one IP address.
2661 31. Re-factored the code for checking spool and log partition space into a
2662 function that finds that data and another that does the check. The former
2663 is then used to implement four new variables: $spool_space, $log_space,
2664 $spool_inodes, and $log_inodes.
2666 32. The RFC2047 encoding function was originally intended for short strings
2667 such as real names; it was not keeping to the 75-character limit for
2668 encoded words that the RFC imposes. It now respects the limit, and
2669 generates multiple encoded words if necessary. To be on the safe side, I
2670 have increased the buffer size for the ${rfc2047: expansion operator from
2673 33. It is now permitted to omit both strings after an "if" condition; if the
2674 condition is true, the result is "true". As before, when the second string
2675 is omitted, a false condition yields an empty string. This makes it less
2676 cumbersome to write custom ACL and router conditions.
2678 34. Failure to deliver a bounce message always caused it to be frozen, even if
2679 there was an errors_to setting on the router. The errors_to setting is now
2682 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2683 canonical form (fully expanded) before being placed in
2684 $sender_host_address.
2686 36. The table in the code that translates DNS record types into text (T_A to
2687 "A" for instance) was missing entries for NS and CNAME. It is just possible
2688 that this could have caused confusion if both these types were looked up
2689 for the same domain, because the text type is used as part of Exim's
2690 per-process caching. But the chance of anyone hitting this buglet seems
2693 37. The dnsdb lookup has been extended in a number of ways.
2695 (1) There is a new type, "zns", which walks up the domain tree until it
2696 finds some nameserver records. It should be used with care.
2698 (2) There is a new type, "mxh", which is like "mx" except that it returns
2699 just the host names, not the priorities.
2701 (3) It is now possible to give a list of domains (or IP addresses) to be
2702 looked up. The behaviour when one of the lookups defers can be
2703 controlled by a keyword.
2705 (4) It is now possible to specify the separator character for use when
2706 multiple records are returned.
2708 38. The dnslists ACL condition has been extended: it is now possible to supply
2709 a list of IP addresses and/or domains to be looked up in a particular DNS
2712 39. Added log_selector=+queue_time_overall.
2714 40. When running the queue in the test harness, wait just a tad after forking a
2715 delivery process, to get repeatability of debugging output.
2717 41. Include certificate and key file names in error message when GnuTLS fails
2718 to set them up, because the GnuTLS error message doesn't include the name
2719 of the failing file when there is a problem reading it.
2721 42. Allow both -bf and -bF in the same test run.
2723 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2725 44. The "Exiscan patch" is now merged into the mainline Exim source.
2727 45. Sometimes the final signoff response after QUIT could fail to get
2728 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2729 < 0 before doing a fflush(). This bug looks as though it goes back to the
2730 introduction of TLS in release 3.20, but "sometimes" must have been rare
2731 because the tests only now provoked it.
2733 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2734 (this can affect the format of dates).
2736 47. exim_tidydb, when checking for the continued existence of a message for
2737 which it has found a message-specific retry record, was not finding
2738 messages that were in split spool directories. Consequently, it was
2739 deleting retry records that should have stayed in existence.
2741 48. Steve fixed some bugs in eximstats.
2743 49. The SPA authentication driver was not abandoning authentication and moving
2744 on to the next authenticator when an expansion was forced to fail,
2745 contradicting the general specification for all authenticators. Instead it
2746 was generating a temporary error. It now behaves as specified.
2748 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2749 (the order specifies the preference for clients). The order is now AES256,
2750 AES128, 3DES, ARCFOUR128.
2752 51. Small patch to Sieve code - explicitly set From: when generating an
2755 52. Exim crashed if a remote delivery caused a very long error message to be
2756 recorded - for instance if somebody sent an entire SpamAssassin report back
2757 as a large number of 550 error lines. This bug was coincidentally fixed by
2758 increasing the size of one of Exim's internal buffers (big_buffer) that
2759 happened as part of the Exiscan merge. However, to be on the safe side, I
2760 have made the code more robust (and fixed the comments that describe what
2763 53. Now that there can be additional text after "Completed" in log lines (if
2764 the queue_time_overall log selector is set), a one-byte patch to exigrep
2765 was needed to allow it to recognize "Completed" as not the last thing in
2768 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2769 patch that reportedly fixes this has been added. I am not expert enough to
2770 create a test for it. This is what the patch creator wrote:
2772 "I found a little strange behaviour of ldap code when working with
2773 Windows 2003 AD Domain, where users was placed in more than one
2774 Organization Units. When I tried to give exim partial DN, the exit code
2775 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2776 But simultaneously result of request was absolutely normal ldap result,
2777 so I produce this patch..."
2779 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2780 REFERENCE, so I have modified the code to exclude the patch when that macro
2783 55. Some experimental protocols are using DNS PTR records for new purposes. The
2784 keys for these records are domain names, not reversed IP addresses. The
2785 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2786 leaves it alone. Component reversal etc. now happens only for IP addresses.
2789 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2791 57. Double the size of the debug message buffer (to 2048) so that more of very
2792 long debug lines gets shown.
2794 58. The exicyclog utility now does better if the number of log files to keep
2795 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2797 59. Two changes related to the smtp_active_hostname option:
2799 (1) $smtp_active_hostname is now available as a variable.
2800 (2) The default for smtp_banner uses $smtp_active_hostname instead
2801 of $primary_hostname.
2803 60. The host_aton() function is supposed to be passed a string that is known
2804 to be a valid IP address. However, in the case of IPv6 addresses, it was
2805 not checking this. This is a hostage to fortune. Exim now panics and dies
2806 if the condition is not met. A case was found where this could be provoked
2807 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2808 components; fortuitously, this particular loophole had already been fixed
2809 by change 4.50/55 above.
2811 If there are any other similar loopholes, the new check in host_aton()
2812 itself should stop them being exploited. The report I received stated that
2813 data on the command line could provoke the exploit when Exim was running as
2814 exim, but did not say which command line option was involved. All I could
2815 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2816 running as the user.
2819 61. There was a buffer overflow vulnerability in the SPA authentication code
2820 (which came originally from the Samba project). I have added a test to the
2821 spa_base64_to_bits() function which I hope fixes it.
2824 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2825 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2827 63. The daemon start-up calls getloadavg() while still root for those OS that
2828 need the first call to be done as root, but it missed one case: when
2829 deliver_queue_load_max is set with deliver_drop_privilege. This is
2830 necessary for the benefit of the queue runner, because there is no re-exec
2831 when deliver_drop_privilege is set.
2833 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2834 This has been fixed.
2836 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2837 was in use, was not putting the data itself into the right store pool;
2838 consequently, it could be overwritten for a subsequent message in the same
2839 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2842 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2844 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2845 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2846 the function were treating the return as a boolean value, which happened to
2847 work because 0=false and not-0=true, but is not correct code.
2849 68. The host_aton() function was not handling scoped IPv6 addresses (those
2850 with, for example, "%eth0" on the end) correctly.
2852 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2853 negated items (that is, ~something) in unsigned ints. Some compilers
2854 apparently mutter when there is no cast.
2856 70. If an address verification called from an ACL failed, and did not produce a
2857 user-specific message (i.e. there was only a "system" message), nothing was
2858 put in $acl_verify_message. In this situation, it now puts the system
2861 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2862 session; change 4.31/43 added the unwanted input to the log line - except
2863 that it did not do this in the start of session case. It now does.
2865 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2866 This is wrong and can cause the other end to generate a synchronization
2867 error if it is another Exim or anything else that does the synchronization
2868 check. A QUIT command is no longer sent after a timeout.
2870 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2871 during host lookups.
2873 74. The defer_ok option of callout verification was not working if it was used
2874 when verifying addresses in header lines, that is, for this case:
2876 verify = header_sender/callout=defer_ok
2878 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2879 those file descriptors could be used for SMTP connections. If anything
2880 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2881 could be sent to the SMTP client, causing chaos. The daemon now opens
2882 stdin, stdout, and stderr to /dev/null when it puts itself into the
2885 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2886 log by default. The user can override this with suitable Perl magic.
2888 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2889 the log message when discard triggers, was not working for the DATA ACL or
2890 for the non-SMTP ACL.
2892 78. Error message wording change in sieve.c.
2894 79. If smtp_accept_max_per_host was set, the number of connections could be
2895 restricted to fewer than expected, because the daemon was trying to set up
2896 a new connection before checking whether the processes handling previous
2897 connections had finished. The check for completed processes is now done
2898 earlier. On busy systems, this bug wouldn't be noticed because something
2899 else would have woken the daemon, and it would have reaped the completed
2902 80. If a message was submitted locally by a user whose login name contained one
2903 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2904 It caused a spool format error. I have fixed the spool reading code. A
2905 related problem was that the "from" clause in the Received: line became
2906 illegal because of the space(s). It is now covered by ${quote_local_part.
2908 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2911 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2913 83. Patch from Sieve maintainer to fix unterminated string problem in
2914 "vacation" handling.
2916 84. Some minor changes to the Linux configuration files to help with other
2917 OS variants using glibc.
2919 85. One more patch for Sieve to update vacation handling to latest spec.
2922 ----------------------------------------------------
2923 See the note above about the 4.44 and 4.50 releases.
2924 ----------------------------------------------------
2930 1. Change 4.43/35 introduced a bug that caused file counts to be
2931 incorrectly computed when quota_filecount was set in an appendfile
2934 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2935 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2938 3. Give more explanation in the error message when the command for a transport
2939 filter fails to execute.
2941 4. There are several places where Exim runs a non-Exim command in a
2942 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2943 was being done only for the command run by the queryprogram router. It is
2944 now done for all such subprocesses. The other cases are: ${run, transport
2945 filters, and the commands run by the lmtp and pipe transports.
2947 5. Some older OS have a limit of 256 on the maximum number of file
2948 descriptors. Exim was using setrlimit() to set 1000 as a large value
2949 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2950 systems. I've change it so that if it can't get 1000, it tries for 256.
2952 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2953 was an oversight, and furthermore, ever since the addition of extra
2954 controls (e.g. 4.43/32), the checks on when to allow different forms of
2955 "control" were broken. There should now be diagnostics for all cases when a
2956 control that does not make sense is encountered.
2958 7. $recipients is now available in the predata ACL (oversight).
2960 8. Tidy the search cache before the fork to do a delivery from a message
2961 received from the command line. Otherwise the child will trigger a lookup
2962 failure and thereby defer the delivery if it tries to use (for example) a
2963 cached ldap connection that the parent has called unbind on.
2965 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2966 of $address_data from the recipient verification was clobbered by the
2967 sender verification.
2969 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2970 was its contents. (It was OK if the option was not defined at all.)
2972 11. A "Completed" log line is now written for messages that are removed from
2973 the spool by the -Mrm option.
2975 12. $host_address is now set to the target address during the checking of
2976 ignore_target_hosts.
2978 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2979 being passed; this would have caused $sender_host_name to have been used if
2980 matching the list had actually called for a host name (not very likely,
2981 since this list is usually IP addresses). A host name is now passed as
2984 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2985 code that specifies a non-restarting handler (typically sigaction() in
2986 modern systems) in an attempt to fix a rare and obscure crash bug.
2988 15. Narrowed the window for a race in the daemon that could cause it to ignore
2989 SIGCHLD signals. This is not a major problem, because they are used only to
2990 wake it up if nothing else does.
2992 16. A malformed maildirsize file could cause Exim to calculate negative values
2993 for the mailbox size or file count. Odd effects could occur as a result.
2994 The maildirsize information is now recalculated if the size or filecount
2997 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2998 support for a long time. Removed HAVE_SYS_VFS_H.
3000 18. Updated exipick to current release from John Jetmore.
3002 19. Allow an empty sender to be matched against a lookup in an address list.
3003 Previously the only cases considered were a regular expression, or an
3006 20. Exim went into a mad DNS lookup loop when doing a callout where the
3007 host was specified on the transport, if the DNS lookup yielded more than
3010 21. The RFC2047 encoding function was originally intended for short strings
3011 such as real names; it was not keeping to the 75-character limit for
3012 encoded words that the RFC imposes. It now respects the limit, and
3013 generates multiple encoded words if necessary. To be on the safe side, I
3014 have increased the buffer size for the ${rfc2047: expansion operator from
3017 22. Failure to deliver a bounce message always caused it to be frozen, even if
3018 there was an errors_to setting on the router. The errors_to setting is now
3021 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3022 canonical form (fully expanded) before being placed in
3023 $sender_host_address.
3025 24. Updated eximstats to version 1.33
3027 25. Include certificate and key file names in error message when GnuTLS fails
3028 to set them up, because the GnuTLS error message doesn't include the name
3029 of the failing file when there is a problem reading it.
3031 26. Expand error message when OpenSSL has problems setting up cert/key files.
3034 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3035 (this can affect the format of dates).
3037 28. exim_tidydb, when checking for the continued existence of a message for
3038 which it has found a message-specific retry record, was not finding
3039 messages that were in split spool directories. Consequently, it was
3040 deleting retry records that should have stayed in existence.
3042 29. eximstats updated to version 1.35
3043 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3044 1.35 - bugfix such that pie charts by volume are generated correctly
3046 30. The SPA authentication driver was not abandoning authentication and moving
3047 on to the next authenticator when an expansion was forced to fail,
3048 contradicting the general specification for all authenticators. Instead it
3049 was generating a temporary error. It now behaves as specified.
3051 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3052 (the order specifies the preference for clients). The order is now AES256,
3053 AES128, 3DES, ARCFOUR128.
3055 31. Small patch to Sieve code - explicitly set From: when generating an
3058 32. Exim crashed if a remote delivery caused a very long error message to be
3059 recorded - for instance if somebody sent an entire SpamAssassin report back
3060 as a large number of 550 error lines. This bug was coincidentally fixed by
3061 increasing the size of one of Exim's internal buffers (big_buffer) that
3062 happened as part of the Exiscan merge. However, to be on the safe side, I
3063 have made the code more robust (and fixed the comments that describe what
3066 33. Some experimental protocols are using DNS PTR records for new purposes. The
3067 keys for these records are domain names, not reversed IP addresses. The
3068 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3069 leaves it alone. Component reversal etc. now happens only for IP addresses.
3072 34. The host_aton() function is supposed to be passed a string that is known
3073 to be a valid IP address. However, in the case of IPv6 addresses, it was
3074 not checking this. This is a hostage to fortune. Exim now panics and dies
3075 if the condition is not met. A case was found where this could be provoked
3076 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3077 components; fortuitously, this particular loophole had already been fixed
3078 by change 4.50/55 or 4.44/33 above.
3080 If there are any other similar loopholes, the new check in host_aton()
3081 itself should stop them being exploited. The report I received stated that
3082 data on the command line could provoke the exploit when Exim was running as
3083 exim, but did not say which command line option was involved. All I could
3084 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3085 running as the user.
3088 35. There was a buffer overflow vulnerability in the SPA authentication code
3089 (which came originally from the Samba project). I have added a test to the
3090 spa_base64_to_bits() function which I hope fixes it.
3093 36. The daemon start-up calls getloadavg() while still root for those OS that
3094 need the first call to be done as root, but it missed one case: when
3095 deliver_queue_load_max is set with deliver_drop_privilege. This is
3096 necessary for the benefit of the queue runner, because there is no re-exec
3097 when deliver_drop_privilege is set.
3099 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3100 was in use, was not putting the data itself into the right store pool;
3101 consequently, it could be overwritten for a subsequent message in the same
3102 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3105 38. Sometimes the final signoff response after QUIT could fail to get
3106 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3107 < 0 before doing a fflush(). This bug looks as though it goes back to the
3108 introduction of TLS in release 3.20, but "sometimes" must have been rare
3109 because the tests only now provoked it.
3115 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3116 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3117 Now it returns an integer. A number of calls were still expecting a T/F
3118 return. Fortuitously, in all cases, the tests worked in OK situations,
3119 which is the norm. However, things would have gone wrong on any write
3120 failures on the smtp file descriptor. This function is used when sending
3121 messages over SMTP and also when doing verify callouts.
3123 2. When Exim is called to do synchronous delivery of a locally submitted
3124 message (the -odf or -odi options), it no longer closes stderr before doing
3127 3. Implemented the mua_wrapper option.
3129 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3131 5. Implemented the functions header_remove(), header_testname(),
3132 header_add_at_position(), and receive_remove_recipient(), and exported them
3135 6. If an ACL "warn" statement specified the addition of headers, Exim already
3136 inserted X-ACL-Warn: at the start if there was no header name. However, it
3137 was not making this test for the second and subsequent header lines if
3138 there were newlines in the string. This meant that an invalid header could
3139 be inserted if Exim was badly configured.
3141 7. Allow an ACL "warn" statement to add header lines at the start or after all
3142 the Received: headers, as well as at the end.
3144 8. Added the rcpt_4xx retry error code.
3146 9. Added postmaster_mailfrom=xxx to callout verification option.
3148 10. Added mailfrom=xxxx to the callout verification option, for verify=
3151 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3152 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3154 12. Inserted some casts to stop certain compilers warning when using pointer
3155 differences as field lengths or precisions in printf-type calls (mostly
3156 affecting debugging statements).
3158 13. Added optional readline() support for -be (dynamically loaded).
3160 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3161 same clock tick as a message's arrival, so that its received time was the
3162 same as the "first fail" time on the retry record, and that message
3163 remained on the queue past the ultimate address timeout, every queue runner
3164 would try a delivery (because it was past the ultimate address timeout) but
3165 after another failure, the ultimate address timeout, which should have then
3166 bounced the address, did not kick in. This was a "< instead of <=" error;
3167 in most cases the first failure would have been in the next clock tick
3168 after the received time, and all would be well.
3170 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3171 being recognized when the domain list was tested by the match_domain
3172 condition in an expansion string.
3174 16. Added the ${str2b64: operator.
3176 17. Exim was always calling setrlimit() to set a large limit for the number of
3177 processes, without checking whether the existing limit was already
3178 adequate. (It did check for the limit on file descriptors.) Furthermore,
3179 errors from getrlimit() and setrlimit() were being ignored. Now they are
3180 logged to the main and panic logs, but Exim does carry on, to try to do its
3181 job under whatever limits there are.
3183 18. Imported PCRE 5.0.
3185 19. Trivial typo in log message " temporarily refused connection" (the leading
3188 20. If the log selector return_path_on_delivery was set and an address was
3189 redirected to /dev/null, the delivery process crashed because it assumed
3190 that a return path would always be set for a "successful" delivery. In this
3191 case, the whole delivery is bypassed as an optimization, and therefore no
3194 21. Internal re-arrangement: the function for sending a challenge and reading
3195 a response while authentication was assuming a zero-terminated challenge
3196 string. It's now changed to take a pointer and a length, to allow for
3197 binary data in such strings.
3199 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3201 23. Exim was not respecting finduser_retries when seeking the login of the
3202 uid under which it was called; it was always trying 10 times. (The default
3203 setting of finduser_retries is zero.) Also, it was sleeping after the final
3204 failure, which is pointless.
3206 24. Implemented tls_on_connect_ports.
3208 25. Implemented acl_smtp_predata.
3210 26. If the domain in control=submission is set empty, Exim assumes that the
3211 authenticated id is a complete email address when it generates From: or
3212 Sender: header lines.
3214 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3215 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3216 chown and chgrp in /bin and hostname in /usr/bin.
3218 28. Exim was keeping the "process log" file open after each use, just as it
3219 does for the main log. This opens the possibility of it remaining open for
3220 long periods when the USR1 signal hits a daemon. Occasional processlog
3221 errors were reported, that could have been caused by this. Anyway, it seems
3222 much more sensible not to leave this file open at all, so that is what now
3225 29. The long-running daemon process does not normally write to the log once it
3226 has entered its main loop, and it closes the log before doing so. This is
3227 so that log files can straightforwardly be renamed and moved. However,
3228 there are a couple of unusual error situations where the daemon does write
3229 log entries, and I had neglected to close the log afterwards.
3231 30. The text of an SMTP error response that was received during a remote
3232 delivery was being truncated at 512 bytes. This is too short for some of
3233 the long messages that one sometimes sees. I've increased the limit to
3236 31. It is now possible to make retry rules that apply only when a message has a
3237 specific sender, in particular, an empty sender.
3239 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3240 it possible to be selective about when SMTP synchronization is enforced.
3242 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3244 32. Implemented hosts_connection_nolog.
3246 33. Added an ACL for QUIT.
3248 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3251 35. Added mailbox_size and mailbox_filecount to appendfile.
3253 36. Added control = no_multiline_responses to ACLs.
3255 37. There was a bug in the logic of the code that waits for the clock to tick
3256 in the case where the clock went backwards by a substantial amount such
3257 that the microsecond fraction of "now" was more than the microsecond
3258 fraction of "then" (but the whole seconds number was less).
3260 38. Added support for the libradius Radius client library this is found on
3261 FreeBSD (previously only the radiusclient library was supported).
3267 1. When certain lookups returned multiple values in the form name=value, the
3268 quoting of the values was not always being done properly. Specifically:
3269 (a) If the value started with a double quote, but contained no whitespace,
3271 (b) If the value contained whitespace other than a space character (i.e.
3272 tabs or newlines or carriage returns) it was not quoted.
3273 This fix has been applied to the mysql and pgsql lookups by writing a
3274 separate quoting function and calling it from the lookup code. The fix
3275 should probably also be applied to nisplus, ibase and oracle lookups, but
3276 since I cannot test any of those, I have not disturbed their existing code.
3278 2. A hit in the callout cache for a specific address caused a log line with no
3279 reason for rejecting RCPT. Now it says "Previous (cached) callout
3280 verification failure".
3282 3. There was an off-by-one bug in the queryprogram router. An over-long
3283 return line was truncated at 256 instead of 255 characters, thereby
3284 overflowing its buffer with the terminating zero. As well as fixing this, I
3285 have increased the buffer size to 1024 (and made a note to document this).
3287 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3288 when Exim is waiting for an SMTP response from a remote server, Exim
3289 restarts its select() call on the socket, thereby resetting its timeout.
3290 This is not a problem when such interrupts are rare. Somebody set up a cron
3291 job to run exiwhat every 2 minutes, which is less than the normal select()
3292 timeout (5 or 10 minutes). This meant that the select() timeout never
3293 kicked in because it was always reset. I have fixed this by comparing the
3294 time when an interrupt arrives with the time at the start of the first call
3295 to select(). If more time than the timeout has elapsed, the interrupt is
3296 treated as a timeout.
3298 5. Some internal re-factoring in preparation for the addition of Sieve
3299 extensions (by MH). In particular, the "personal" test is moved to a
3300 separate function, and given an option for scanning Cc: and Bcc: (which is
3301 not set for Exim filters).
3303 6. When Exim created an email address using the login of the caller as the
3304 local part (e.g. when creating a From: or Sender: header line), it was not
3305 quoting the local part when it contained special characters such as @.
3307 7. Installed new OpenBSD configuration files.
3309 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3310 try to make them clearer.
3312 9. Callout options, other than the timeout value, were being ignored when
3313 verifying sender addresses in header lines. For example, when using
3315 verify = header_sender/callout=no_cache
3317 the cache was (incorrectly) being used.
3319 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3320 only the Cygwin environment.
3322 11. When return_path_on_delivery was set as a log selector, if different remote
3323 addresses in the same message used different return paths and parallel
3324 remote delivery occurred, the wrong values would sometimes be logged.
3325 (Whenever a remote delivery process finished, the return path value from
3326 the most recently started remote delivery process was logged.)
3328 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3329 lines when AUTH and/or TLS are in use. This is the "received protocol"
3330 field. Exim used to use "asmtp" for authenticated SMTP, without any
3331 indication (in the protocol name) for TLS use. Now it follows the RFC and
3332 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3333 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3334 names appear in log lines as well as in Received: header lines.
3336 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3337 extensions, and comparison tests, and to fix some bugs.
3339 14. Changes to the "personal" filter test:
3341 (1) The test was buggy in that it was just doing the equivalent of
3342 "contains" tests on header lines. For example, if a user's address was
3343 anne@some.where, the "personal" test would incorrectly be true for
3345 To: susanne@some.where
3347 This test is now done by extracting each address from the header in turn,
3348 and checking the entire address. Other tests that are part of "personal"
3349 are now done using regular expressions (for example, to check local parts
3350 of addresses in From: header lines).
3352 (2) The list of non-personal local parts in From: addresses has been
3353 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3354 taken from the Sieve specification recommendations.
3356 (3) If the message contains any header line starting with "List-" it is
3357 treated as non-personal.
3359 (4) The test for "circular" in the Subject: header line has been removed
3360 because it now seems ill-conceived.
3362 15. Minor typos in src/EDITME comments corrected.
3364 16. Installed latest exipick from John Jetmore.
3366 17. If headers_add on a router specified a text string that was too long for
3367 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3368 of string_sprintf() is now avoided.
3370 18. $message_body_size was not set (it was always zero) when running the DATA
3371 ACL and the local_scan() function.
3373 19. For the "mail" command in an Exim filter, no default was being set for
3374 the once_repeat time, causing a random time value to be used if "once" was
3375 specified. (If the value happened to be <= 0, no repeat happened.) The
3376 default is now 0s, meaning "never repeat". The "vacation" command was OK
3377 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3378 (I found it when inspecting the code).
3380 20. There is now an overall timeout for performing a callout verification. It
3381 defaults to 4 times the callout timeout, which applies to individual SMTP
3382 commands during the callout. The overall timeout applies when there is more
3383 than one host that can be tried. The timeout is checked before trying the
3384 next host. This prevents very long delays if there are a large number of
3385 hosts and all are timing out (e.g. when the network connections are timing
3386 out). The value of the overall timeout can be changed by specifying an
3387 additional sub-option for "callout", called "maxwait". For example:
3389 verify = sender/callout=5s,maxwait=20s
3391 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3392 to the end before writing, but this should make it even safer).
3394 22. Exim was forgetting that it had advertised PIPELINING for the second and
3395 subsequent messages on an SMTP connection. It was also not resetting its
3396 memory on STARTTLS and an internal HELO.
3398 23. When Exim logs an SMTP synchronization error within a session, it now
3399 records whether PIPELINING has been advertised or not.
3401 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3402 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3403 rather than long int.
3405 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3407 26. Added the never_mail option to autoreply.
3413 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3414 crash if the getsockname() call failed; this can happen if a connection is
3415 closed very soon after it is established. The problem was simply in the
3416 order in which certain operations were done, causing Exim to try to write
3417 to the SMTP stream before it had set up the file descriptor. The bug has
3418 been fixed by making things happen in the correct order.
3424 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3425 before the connection was closed, thus losing the rejection response.
3427 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3428 some early Solaris releases, but causes trouble in current releases where
3429 socklen_t is defined.
3431 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3434 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3437 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3438 that releasing the top of it at the end releases what was used for sub-
3439 expansions (unless the block got too big). However, discard this block if
3440 the first thing is a variable or header, so that we can use its block when
3441 it is dynamic (useful for very large $message_headers, for example).
3443 6. Lookups now cache *every* query, not just the most recent. A new, separate
3444 store pool is used for this. It can be recovered when all lookup caches are
3445 flushed. Lookups now release memory at the end of their result strings.
3446 This has involved some general refactoring of the lookup sources.
3448 7. Some code has been added to the store_xxx() functions to reduce the amount
3449 of flapping under certain conditions.
3451 8. log_incoming_interface used to affect only the <= reception log lines. Now
3452 it causes the local interface and port to be added to several more SMTP log
3453 lines, for example "SMTP connection from", and rejection lines.
3455 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3457 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3459 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3460 could be overwritten at the end of the current message (or the start of a
3461 new message if it was set in a HELO ACL). The value is now preserved for
3462 the duration of the SMTP connection.
3464 12. If a transport had a headers_rewrite setting, and a matching header line
3465 contained an unqualified address, that address was qualified, even if it
3466 did not match any rewriting rules. The underlying bug was that the values
3467 of the flags that permit the existence of unqualified sender and recipient
3468 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3469 non-local messages, and by -bnq for local messages) were not being
3470 preserved with the message after it was received.
3472 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3473 "next input=" as part of the text comprising the host identity instead of
3474 the correct text. The code was using the same buffer for two different
3475 strings. However, depending on which order the printing function evaluated
3476 its arguments, the bug did not always show up. Under Linux, for example, my
3477 test suite worked just fine.
3479 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3480 doesn't work with some older versions of Perl. It has been changed to "my",
3481 which in any case is probably the better facility to use.
3483 15. A really picky compiler found some instances of statements for creating
3484 error messages that either had too many or two few arguments for the format
3487 16. The size of the buffer for calls to the DNS resolver has been increased
3488 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3489 for addresses that have a lot of PTR records. This alleviates a problem; it
3490 does not fully solve it.
3492 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3493 buffer now truncates the list and logs the incident, which is the same
3494 action as happens when Exim is looking up a host name and its aliases.
3495 Previously in this situation something unpredictable would happen;
projects / exim.git / blob