1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 PP/01 Date: & Message-Id: revert to normally being appended to a message,
10 only prepend for the Resent-* case. Fixes regression introduced in
11 Exim 4.70 by NM/22 for Bugzilla 607.
13 PP/02 Include check_rfc2047_length in configure.default because we're seeing
14 increasing numbers of administrators be bitten by this.
16 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
18 PP/03 Bugzilla 994: added openssl_options main configuration option.
20 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
22 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
24 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
26 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
27 without a peer certificate, leading to a segfault because of an
28 assumption that peers always have certificates. Be a little more
29 paranoid. Problem reported by Martin Tscholak.
31 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
32 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
33 NB: ClamAV planning to remove STREAM in "middle of 2010".
34 CL also introduces -bmalware, various -d+acl logging additions and
35 more caution in buffer sizes.
37 PP/09 Implemented reverse_ip expansion operator.
39 PP/10 Bugzilla 937: provide a "debug" ACL control.
41 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
43 PP/12 Bugzilla 973: Implement --version.
45 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
47 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
49 PP/15 Bugzilla 816: support multiple condition rules on Routers.
51 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
52 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
53 ignore trailing whitespace.
55 JJ/02 prevent non-panic DKIM error from being sent to paniclog
57 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
60 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
61 Notification from Dr Andrew Aitchison.
63 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
64 ExtendedDetectionInfo response format.
65 Notification from John Horne.
67 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
70 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
71 XSL and documented dependency on system catalogs, with examples of how
74 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
77 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
78 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
79 configuration file which is writeable by the Exim user or group.
81 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
82 of configuration files to cover files specified with the -C option if
83 they are going to be used with root privileges, not just the default
86 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
87 option (effectively making it always true).
89 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
90 files to be used while preserving root privileges.
92 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
93 that rogue child processes cannot use them.
95 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
96 run-time user, instead of root.
98 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
99 Exim run-time user without dropping privileges.
106 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
107 $data_path, and $header_path variables; fixed documentation bugs and
110 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
111 exipick to access non-standard spools, including the "frozen" queue
114 NM/01 Bugzilla 965: Support mysql stored procedures.
115 Patch from Alain Williams
117 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
119 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
120 Patch from Andreas Metzler
122 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
123 Patch from Kirill Miazine
125 NM/05 Bugzilla 671: Added umask to procmail example.
127 JJ/03 installed exipick 20100323.0, fixing doc bug
129 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
130 directory. Notification and patch from Dan Rosenberg.
132 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
134 TK/02 Improve log output when DKIM signing operation fails.
136 MH/01 Treat the transport option dkim_domain as a colon separated
137 list, not as a single string, and sign the message with each element,
138 omitting multiple occurences of the same signer.
140 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
141 Bugzilla 985, 986. Patch by Simon Arlott
143 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
144 Patch by Simon Arlott
146 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
147 MBX locking. Notification from Dan Rosenberg.
153 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
155 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
157 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
159 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
161 NM/04 Bugzilla 915: Flag broken perl installation during build.
167 TK/01 Added patch by Johannes Berg that expands the main option
168 "spamd_address" if it starts with a dollar sign.
170 TK/02 Write list of recipients to X-Envelope-Sender header when building
171 the mbox-format spool file for content scanning (suggested by Jakob
174 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
175 (http://www.dcc-servers.net/) support via dccifd. Activated by
176 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
178 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
179 by Mark Daniel Reidel <mr@df.eu>.
181 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
182 When building exim an external PCRE library is now needed -
183 PCRE is a system library on the majority of modern systems.
184 See entry on PCRE_LIBS in EDITME file.
186 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
187 conversation. Added nologin parameter to request.
188 Patch contributed by Kirill Miazine.
190 TF/01 Do not log submission mode rewrites if they do not change the address.
192 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
194 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
195 log files in place. Contributed by Roberto Lima.
197 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
199 TF/03 Bugzilla 615: When checking the local_parts router precondition
200 after a local_part_suffix or local_part_prefix option, Exim now
201 does not use the address's named list lookup cache, since this
202 contains cached lookups for the whole local part.
204 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
205 Robert Millan. Documentation is in experimental-spec.txt.
207 TF/04 Bugzilla 668: Fix parallel build (make -j).
209 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
211 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
212 Patch provided by Jan Srzednicki.
214 TF/05 Leading white space used to be stripped from $spam_report which
215 wrecked the formatting. Now it is preserved.
217 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
218 that they are available at delivery time.
220 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
222 TF/08 TLS error reporting now respects the incoming_interface and
223 incoming_port log selectors.
225 TF/09 Produce a more useful error message if an SMTP transport's hosts
226 setting expands to an empty string.
228 NM/06 Bugzilla 744: EXPN did not work under TLS.
229 Patch provided by Phil Pennock.
231 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
232 Patch provided by Richard Godbee.
234 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
235 acl_smtp_notquit, added index entry.
237 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
238 Patch provided by Eugene Bujak.
240 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
241 accept(). Patch provided by Maxim Dounin.
243 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
244 Patch provided by Phil Pennock.
246 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
248 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
249 Patch provided by Brad "anomie" Jorsch.
251 NM/14 Bugzilla 622: Added timeout setting to transport filter.
252 Patch provided by Dean Brooks.
254 TK/05 Add native DKIM support (does not depend on external libraries).
256 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
257 Patch provided by Graeme Fowler.
259 NM/16 Bugzilla 851: Documentation example syntax fix.
261 NM/17 Changed NOTICE file to remove references to embedded PCRE.
263 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
266 NM/19 Bugzilla 745: TLS version reporting.
267 Patch provided by Phil Pennock.
269 NM/20 Bugzilla 167: bool: condition support.
270 Patch provided by Phil Pennock.
272 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
273 clients. Patch provided by Phil Pennock.
275 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
276 Patch provided by Brad "anomie" Jorsch.
278 NM/23 Bugzilla 687: Fix misparses in eximstats.
279 Patch provided by Heiko Schlittermann.
281 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
282 Patch provided by Heiko Schlittermann.
284 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
285 plus update to original patch.
287 NM/26 Bugzilla 799: Documentation correction for ratelimit.
289 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
290 Patch provided by David Brownlee.
292 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
294 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
296 NM/30 Bugzilla 888: TLS documentation bugfixes.
298 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
300 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
301 Unlike the original bugzilla I have changed all shell scripts in src tree.
303 NM/33 Bugzilla 898: Transport filter timeout fix.
304 Patch by Todd Rinaldo.
306 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
307 Patch by Serge Demonchaux.
309 NM/35 Bugzilla 39: Base64 decode bug fixes.
310 Patch by Jakob Hirsch.
312 NM/36 Bugzilla 909: Correct connect() call in dcc code.
314 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
316 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
318 NM/39 Bugzilla 911: Fixed MakeLinks build script.
324 TK/01 Add preliminary DKIM support. Currently requires a forked version of
325 ALT-N's libdkim that I have put here:
326 http://duncanthrax.net/exim-experimental/
328 Note to Michael Haardt: I had to rename some vars in sieve.c. They
329 were called 'true' and it seems that C99 defines that as a reserved
330 keyword to be used with 'bool' variable types. That means you could
331 not include C99-style headers which use bools without triggering
332 build errors in sieve.c.
334 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
335 as mailq or other aliases. Changed the --help handling significantly
336 to do whats expected. exim_usage() emits usage/help information.
338 SC/01 Added the -bylocaldomain option to eximstats.
340 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
342 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
344 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
350 PH/01 Another patch from the Sieve maintainer.
352 PH/02 When an IPv6 address is converted to a string for single-key lookup
353 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
354 dots are used instead of colons so that keys in lsearch files need not
355 contain colons. This was done some time before quoting was made available
356 in lsearch files. However, iplsearch files do require colons in IPv6 keys
357 (notated using the quote facility) so as to distinguish them from IPv4
358 keys. This meant that lookups for IP addresses in host lists did not work
359 for iplsearch lookups.
361 This has been fixed by arranging for IPv6 addresses to be expressed with
362 colons if the lookup type is iplsearch. This is not incompatible, because
363 previously such lookups could never work.
365 The situation is now rather anomolous, since one *can* have colons in
366 ordinary lsearch keys. However, making the change in all cases is
367 incompatible and would probably break a number of configurations.
369 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
372 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
373 conversion specification without a maximum field width, thereby enabling
374 a rogue spamd server to cause a buffer overflow. While nobody in their
375 right mind would setup Exim to query an untrusted spamd server, an
376 attacker that gains access to a server running spamd could potentially
377 exploit this vulnerability to run arbitrary code as the Exim user.
379 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
380 $primary_hostname instead of what libspf2 thinks the hosts name is.
382 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
383 a directory entry by the name of the lookup key. Previously, if a
384 symlink pointed to a non-existing file or a file in a directory that
385 Exim lacked permissions to read, a lookup for a key matching that
386 symlink would fail. Now it is enough that a matching directory entry
387 exists, symlink or not. (Bugzilla 503.)
389 PH/03 The body_linecount and body_zerocount variables are now exported in the
392 PH/04 Added the $dnslist_matched variable.
394 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
395 This means they are set thereafter only if the connection becomes
398 PH/06 Added the client_condition to authenticators so that some can be skipped
399 by clients under certain conditions.
401 PH/07 The error message for a badly-placed control=no_multiline_responses left
402 "_responses" off the end of the name.
404 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
406 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
407 (without spaces) instead of just copying the configuration text.
409 PH/10 Added the /noupdate option to the ratelimit ACL condition.
411 PH/11 Added $max_received_linelength.
413 PH/12 Added +ignore_defer and +include_defer to host lists.
415 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
416 way in which PCRE > 7.0 is built.
418 PH/14 Implemented queue_only_load_latch.
420 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
421 MAIL command. The effect was to mangle the value on 64-bit systems.
423 PH/16 Another patch from the Sieve maintainer.
425 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
427 PH/18 If a system quota error occurred while trying to create the file for
428 a maildir delivery, the message "Mailbox is full" was not appended to the
429 bounce if the delivery eventually timed out. Change 4.67/27 below applied
430 only to a quota excession during the actual writing of the file.
432 PH/19 It seems that peer DN values may contain newlines (and other non-printing
433 characters?) which causes problems in log lines. The DN values are now
434 passed through string_printing() before being added to log lines.
436 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
437 and InterBase are left for another time.)
439 PH/21 Added message_body_newlines option.
441 PH/22 Guard against possible overflow in moan_check_errorcopy().
443 PH/23 POSIX allows open() to be a macro; guard against that.
445 PH/24 If the recipient of an error message contained an @ in the local part
446 (suitably quoted, of course), incorrect values were put in $domain and
447 $local_part during the evaluation of errors_copy.
453 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
454 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
457 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
458 issue a MAIL command.
460 PH/02 In an ACL statement such as
462 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
464 if a client was not listed at all, or was listed with a value other than
465 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
466 the condition was not true (as it should be), so access was not denied.
467 The bug was that the ! inversion was incorrectly passed on to the second
468 item. This has been fixed.
470 PH/03 Added additional dnslists conditions == and =& which are different from
471 = and & when the dns lookup returns more than one IP address.
473 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
474 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
476 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
477 FSYNC, which compiles an option called disable_fsync that allows for
478 bypassing fsync(). The documentation is heavily laced with warnings.
480 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
482 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
483 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
484 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
485 including adding "make clean"; (3) Added -fPIC when compiling the test
486 dynamically loaded module, to get rid of a warning.
488 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
489 message fails, move_frozen_messages = true and ignore_bounce_errors_after
490 = 0s. The bug is otherwise harmless.
492 PH/07 There was a bug in the dovecot authenticator such that the value of
493 $auth1 could be overwritten, and so not correctly preserved, after a
494 successful authentication. This usually meant that the value preserved by
495 the server_setid option was incorrect.
497 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
499 PH/09 Installed PCRE release 7.0.
501 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
502 run for batched SMTP input. It is now run at the start of every message
503 in the batch. While fixing this I discovered that the process information
504 (output by running exiwhat) was not always getting set for -bs and -bS
505 input. This is fixed, and it now also says "batched" for BSMTP.
507 PH/11 Added control=no_pipelining.
509 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
510 patch, slightly modified), and move the expansion of helo_data till after
511 the connection is made in the smtp transport (so it can use these
514 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
516 PH/14 Added log_selector = +pid.
518 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
520 PH/16 Add ${if forany and ${if forall.
522 PH/17 Added dsn_from option to vary the From: line in DSNs.
524 PH/18 Flush SMTP output before performing a callout, unless control =
525 no_callout_flush is set.
527 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
528 was true (the default) a successful delivery failed to delete the retry
529 item, thus causing premature timeout of the address. The bug is now
532 PH/20 Added hosts_avoid_pipelining to the smtp transport.
534 PH/21 Long custom messages for fakedefer and fakereject are now split up
535 into multiline reponses in the same way that messages for "deny" and
536 other ACL rejections are.
538 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
539 with slight modification.
541 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
542 draft, changing the syntax and factoring some duplicate code".
544 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
545 for deliveries of the second and subsequent messages over the same SMTP
548 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
549 ${reduce, with only minor "tidies".
551 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
553 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
554 expansion side effects.
556 PH/27 When a message times out after an over-quota error from an Exim-imposed
557 quota, the bounce message says "mailbox is full". This message was not
558 being given when it was a system quota that was exceeded. It now should
561 MH/03 Made $recipients available in local_scan(). local_scan() already has
562 better access to the recipient list through recipients_list[], but
563 $recipients can be useful in postmaster-provided expansion strings.
565 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
566 in the case of a MAIL command with additional options following the
567 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
568 were accidentally chopped off.
570 PH/29 SMTP synchronization checks are implemented when a command is read -
571 there is a check that no more input is waiting when there shouldn't be
572 any. However, for some commands, a delay in an ACL can mean that it is
573 some time before the response is written. In this time, more input might
574 arrive, invalidly. So now there are extra checks after an ACL has run for
575 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
576 pipelining has not been advertised.
578 PH/30 MH's patch to allow iscntrl() characters to be list separators.
580 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
581 returned in the SMTP response when smtp_return_error_details was false.
584 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
585 instead of the C I/O that was originally supplied, because problems were
588 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
589 Exim which did not show up earlier: it was assuming that a call to
590 SSL_CTX_set_info_callback() might give an error value. In fact, there is
591 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
592 was a macro that became an assignment, so it seemed to work. This has
593 changed to a proper function call with a void return, hence the compile
594 error. Exim's code has been fixed.
596 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
599 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
601 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
603 PH/37 If a message is not accepted after it has had an id assigned (e.g.
604 because it turns out to be too big or there is a timeout) there is no
605 "Completed" line in the log. When some messages of this type were
606 selected by exigrep, they were listed as "not completed". Others were
607 picked up by some special patterns. I have improved the selection
608 criteria to be more general.
610 PH/38 The host_find_failed option in the manualroute router can now be set
611 to "ignore", to completely ignore a host whose IP address cannot be
612 found. If all hosts are ignored, the behaviour is controlled by the new
613 host_all_ignored option.
615 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
616 homing or because of multiple MX records with /mx) generated more than
617 one IP address, and the following item turned out to be the local host,
618 all the secondary addresses of the first item were incorrectly removed
619 from the list, along with the local host and any following hosts (which
620 is what is supposed to happen).
622 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
623 whoever called Exim into the -H file. In the case of the daemon it was
624 behaving confusingly. When first started, it used values for whoever
625 started the daemon, but after a SIGHUP it used the Exim user (because it
626 calls itself on a restart). I have changed the code so that it now always
629 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
630 message are rejected with the same error (e.g. no authentication or bad
631 sender address), and a DATA command is nevertheless sent (as can happen
632 with PIPELINING or a stupid MUA), the error message that was given to the
633 RCPT commands is included in the rejection of the DATA command. This is
634 intended to be helpful for MUAs that show only the final error to their
637 PH/42 Another patch from the Sieve maintainer.
639 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
640 Eximstats - Fixed some broken HTML links and added missing column headers
642 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
643 columns for Rejects, Temp Rejects, Ham, and Spam rows.
645 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
647 PH/43 Yet another patch from the Sieve maintainer.
649 PH/44 I found a way to check for a TCP/IP connection going away before sending
650 the response to the final '.' that terminates a message, but only in the
651 case where the client has not sent further data following the '.'
652 (unfortunately, this is allowed). However, in many cases there won't be
653 any further data because there won't be any more messages to send. A call
654 to select() can be used: if it shows that the input is "ready", there is
655 either input waiting, or the socket has been closed. An attempt to read
656 the next input character can distinguish the two cases. Previously, Exim
657 would have sent an OK response which the client would never have see.
658 This could lead to message repetition. This fix should cure that, at
659 least in a lot of common cases.
661 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
662 advertised in response to EHLO.
668 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
669 fixed by 4.65/MH/01 (is this a record?) are fixed:
671 (i) An empty string was always treated as zero by the numeric comparison
672 operators. This behaviour has been restored.
674 (ii) It is documented that the numeric comparison operators always treat
675 their arguments as decimal numbers. This was broken in that numbers
676 starting with 0 were being interpreted as octal.
678 While fixing these problems I realized that there was another issue that
679 hadn't been noticed. Values of message_size_limit (both the global option
680 and the transport option) were treated as octal if they started with 0.
681 The documentation was vague. These values are now always treated as
682 decimal, and I will make that clear in the documentation.
688 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
689 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
692 MH/01 Don't check that the operands of numeric comparison operators are
693 integers when their expansion is in "skipping" mode (fixes bug
694 introduced by 4.64-PH/07).
696 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
697 child addresses, Exim now panics and dies. Previously, because the count
698 is held in a short int, deliveries were likely to be lost. As such a
699 large number of recipients for a single message is ridiculous
700 (performance will be very, very poor), I have chosen to impose a limit
701 rather than extend the field.
707 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
708 leftover -K file (the existence of which was triggered by #402).
709 While we were at it, introduced process PID as part of the -K
710 filename. This should rule out race conditions when creating
713 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
714 processing considerably. Previous code took too long for large mails,
715 triggering a timeout which in turn triggers #401.
717 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
718 in the DK code in transports.c. sendfile() is not really portable,
719 hence the _LINUX specificness.
721 TF/01 In the add_headers option to the mail command in an Exim filter,
722 there was a bug that Exim would claim a syntax error in any
723 header after the first one which had an odd number of characters
726 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
727 callout verification, Exim cached a "reject" for the entire domain. This
728 is correct for most verifications, but it is not correct for a recipient
729 verification with use_sender or use_postmaster set, because in that case
730 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
731 case of MAIL FROM:<> rejection from other early rejections (e.g.
732 rejection of HELO). When verifying a recipient using a non-null MAIL
733 address, the cache is ignored if it shows MAIL FROM:<> rejection.
734 Whatever the result of the callout, the value of the domain cache is
735 left unchanged (for any other kind of callout, getting as far as trying
736 RCPT means that the domain itself is ok).
738 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
741 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
742 manifest itself as EPIPE rather than ECONNECT. When tidying away a
743 session, the daemon ignores ECONNECT errors and logs others; it now
744 ignores EPIPE as well.
746 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
747 (quoted-printable decoding).
749 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
750 later the small subsequent patch to fix an introduced bug.
752 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
754 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
756 PH/08 An error is now given if message_size_limit is specified negative.
758 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
759 to be given (somewhat) arbitrary names.
761 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
764 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
765 miscellaneous code fixes
767 PH/10 Added the log_reject_target ACL modifier to specify where to log
770 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
771 hostname. This is wrong, because it relates to the incoming message (and
772 probably the interface on which it is arriving) and not to the outgoing
773 callout (which could be using a different interface). This has been
774 changed to use the value of the helo_data option from the smtp transport
775 instead - this is what is used when a message is actually being sent. If
776 there is no remote transport (possible with a router that sets up host
777 addresses), $smtp_active_hostname is used.
779 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
780 tweaks were necessary in order to get it to work (see also 21 below):
781 (a) The code assumed that strncpy() returns a negative number on buffer
782 overflow, which isn't the case. Replaced with Exim's string_format()
784 (b) There were several signed/unsigned issues. I just did the minimum
785 hacking in of casts. There is scope for a larger refactoring.
786 (c) The code used strcasecmp() which is not a standard C function.
787 Replaced with Exim's strcmpic() function.
788 (d) The code set only $1; it now sets $auth1 as well.
789 (e) A simple test gave the error "authentication client didn't specify
790 service in request". It would seem that Dovecot has changed its
791 interface. Fortunately there's a specification; I followed it and
792 changed what the client sends and it appears to be working now.
794 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
797 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
798 address A is aliased to B and C, where B exists and C does not. Without
799 -v the output is "A verified" because verification stops after a
800 successful redirection if more than one address is generated. However,
801 with -v the child addresses are also verified. Exim was outputting "A
802 failed to verify" and then showing the successful verification for C,
803 with its parentage. It now outputs "B failed to verify", showing B's
804 parentage before showing the successful verification of C.
806 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
807 look up a TXT record in a specific list after matching in a combined
810 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
811 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
812 they consult the DNS. I had assumed they would set it the way they
813 wanted; and indeed my experiments on Linux seem to show that in some
814 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
815 To be on the safe side, however, I have now made the interface to
816 host_find_byname() similar to host_find_bydns(), with an argument
817 containing the DNS resolver options. The host_find_byname() function now
818 sets these options at its start, just as host_find_bydns() does. The smtp
819 transport options dns_qualify_single and dns_search_parents are passed to
820 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
821 of host_find_byname() use the default settings of RES_DEFNAMES
822 (qualify_single) but not RES_DNSRCH (search_parents).
824 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
825 spool_read_header() do less string testing, by means of a preliminary
826 switch on the second character of optional "-foo" lines. (This is
827 overdue, caused by the large number of possibilities that now exist.
828 Originally there were few.) While I was there, I also converted the
829 str(n)cmp tests so they don't re-test the leading "-" and the first
830 character, in the hope this might squeeze out yet more improvement.
832 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
833 flag allowing group syntax was set by the header_syntax check but not
834 turned off, possible causing trouble later; (2) The flag was not being
835 set at all for the header_verify test, causing "group"-style headers to
836 be rejected. I have now set it in this case, and also caused header_
837 verify to ignore an empty address taken from a group. While doing this, I
838 came across some other cases where the code for allowing group syntax
839 while scanning a header line wasn't quite right (mostly, not resetting
840 the flag correctly in the right place). These bugs could have caused
841 trouble for malformed header lines. I hope it is now all correct.
843 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
844 with the "reply" argument non-NULL. The code, however (which originally
845 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
846 but it didn't always do it. This confused somebody who was copying the
847 code for some other use. I have removed all the tests.
849 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
850 feature that was used to support insecure browsers during the U.S. crypto
851 embargo. It requires special client support, and Exim is probably the
852 only MTA that supported it -- and would never use it because real RSA is
853 always available. This code has been removed, because it had the bad
854 effect of slowing Exim down by computing (never used) parameters for the
855 RSA_EXPORT functionality.
857 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
858 authenticator to fail if there's a tab character in the incoming data
859 (there should never be unless someone is messing about, as it's supposed
860 to be base64-encoded). Also added, on Timo's advice, the "secured" option
861 if the connection is using TLS or if the remote IP is the same as the
862 local IP, and the "valid-client-cert option" if a client certificate has
865 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
866 authenticators. This can be used for authorization after authentication
867 succeeds. (In the case of plaintext, it servers for both authentication
870 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
871 if any retry times were supplied.
873 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
874 connection, where there is no client IP address to check. In this
875 situation, the verify now always succeeds.
877 PH/25 Applied John Jetmore's -Mset patch.
879 PH/26 Added -bem to be like -Mset, but loading a message from a file.
881 PH/27 In a string expansion for a processed (not raw) header when multiple
882 headers of the same name were present, leading whitespace was being
883 removed from all of them, but trailing whitespace was being removed only
884 from the last one. Now trailing whitespace is removed from each header
885 before concatenation. Completely empty headers in a concatenation (as
888 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
889 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
891 PH/29 [Removed. This was a change that I later backed out, and forgot to
892 correct the ChangeLog entry (that I had efficiently created) before
893 committing the later change.]
895 PH/30 Exim was sometimes attempting to deliver messages that had suffered
896 address errors (4xx response to RCPT) over the same connection as other
897 messages routed to the same hosts. Such deliveries are always "forced",
898 so retry times are not inspected. This resulted in far too many retries
899 for the affected addresses. The effect occurred only when there were more
900 hosts than the hosts_max_try setting in the smtp transport when it had
901 the 4xx errors. Those hosts that it had tried were not added to the list
902 of hosts for which the message was waiting, so if all were tried, there
903 was no problem. Two fixes have been applied:
905 (i) If there are any address or message errors in an SMTP delivery, none
906 of the hosts (tried or untried) are now added to the list of hosts
907 for which the message is waiting, so the message should not be a
908 candidate for sending over the same connection that was used for a
909 successful delivery of some other message. This seems entirely
910 reasonable: after all the message is NOT "waiting for some host".
911 This is so "obvious" that I'm not sure why it wasn't done
912 previously. Hope I haven't missed anything, but it can't do any
913 harm, as the worst effect is to miss an optimization.
915 (ii) If, despite (i), such a delivery is accidentally attempted, the
916 routing retry time is respected, so at least it doesn't keep
917 hammering the server.
919 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
920 in ${readsocket because some servers need this prod.
922 PH/32 Added some extra debug output when updating a wait-xxx database.
924 PH/33 The hint "could be header name not terminated by colon", which has been
925 given for certain expansion errors for a long time, was not being given
926 for the ${if def:h_colon_omitted{... case.
928 PH/34 The spec says: "With one important exception, whenever a domain list is
929 being scanned, $domain contains the subject domain." There was at least
930 one case where this was not true.
932 PH/35 The error "getsockname() failed: connection reset by peer" was being
933 written to the panic log as well as the main log, but it isn't really
934 panic-worthy as it just means the connection died rather early on. I have
935 removed the panic log writing for the ECONNRESET error when getsockname()
938 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
939 runs only) independently of the message's sender address. This meant
940 that, if the 4xx error was in fact related to the sender, a different
941 message to the same recipient with a different sender could confuse
942 things. In particualar, this can happen when sending to a greylisting
943 server, but other circumstances could also provoke similar problems.
944 I have changed the default so that the retry time for these errors is now
945 based a combination of the sender and recipient addresses. This change
946 can be overridden by setting address_retry_include_sender=false in the
949 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
950 remote server are returned as part of bounce messages. This was not
951 happening for LMTP over a pipe (the lmtp transport), but now it is the
952 same for both kinds of LMTP.
954 PH/38 Despite being documented as not happening, Exim was rewriting addresses
955 in header lines that were in fact CNAMEs. This is no longer the case.
957 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
958 and queue runs started by the daemon processed all messages. This has
959 been fixed so that -R and -S can now usefully be given with -q<time>.
961 PH/40 Import PCRE release 6.7 (fixes some bugs).
963 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
965 PH/42 Give an error if -q is specified more than once.
967 PH/43 Renamed the variables $interface_address and $interface_port as
968 $received_ip_address and $received_port, to make it clear that these
969 values apply to message reception, and not to the outgoing interface when
970 a message is delivered. (The old names remain recognized, of course.)
972 PH/44 There was no timeout on the connect() call when using a Unix domain
973 socket in the ${readsocket expansion. There now is.
975 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
976 be meaningful with "accept".
978 SC/01 Eximstats V1.43
979 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
981 SC/02 Eximstats V1.44
982 Use a glob alias rather than an array ref in the generated
983 parser. This improves both readability and performance.
985 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
986 Collect SpamAssassin and rejection statistics.
987 Don't display local sender or destination tables unless
988 there is data to show.
989 Added average volumes into the top table text output.
991 SC/04 Eximstats V1.46
992 Collect data on the number of addresses (recipients)
993 as well as the number of messages.
995 SC/05 Eximstats V1.47
996 Added 'Message too big' to the list of mail rejection
997 reasons (thanks to Marco Gaiarin).
999 SC/06 Eximstats V1.48
1000 Mainlog lines which have GMT offsets and are too short to
1001 have a flag are now skipped.
1003 SC/07 Eximstats V1.49 (Alain Williams)
1004 Added the -emptyok flag.
1006 SC/08 Eximstats V1.50
1007 Fixes for obtaining the IP address from reject messages.
1009 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1010 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1011 whitesspace changes from 4.64-PH/27
1013 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1016 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1017 are found, allow negative numbers in numeric criteria)
1019 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1021 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1022 to match changes made in 4.64-PH/43
1024 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1026 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1027 "domain" is actually an IP address, and force a failure. This locks out
1028 those revolvers/nameservers that support "A-for-A" lookups, in
1029 contravention of the specifications.
1031 PH/48 When a host name was looked up from an IP address, and the subsequent
1032 forward lookup of the name timed out, the host name was left in
1033 $sender_host_name, contrary to the specification.
1035 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1036 restricted to single-key lookups, Exim was not diagnosing an error if
1037 * or *@ was used with a query-style lookup.
1039 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1041 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1042 long ago, but noone interested enough thought of it. Let's just say that
1043 the "1.1" means that there are some new functions that weren't there at
1044 some point in the past.
1046 PH/51 Error processing for expansion failure of helo_data from an smtp
1047 transport during callout processing was broken.
1049 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1050 tested/used via the -bh/-bhc/-bs options.
1052 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1053 bug, fixed in subsequent PCRE releases).
1055 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1056 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1058 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1064 SC/01 Use a glob alias rather than an array ref in eximstats generated
1065 parser. This improves both readability and performance.
1067 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1068 Don't display local sender or destination tables in eximstats unless
1069 there is data to show.
1070 Added average volumes into the eximstats top table text output.
1072 SC/03 Collect data on the number of addresses (recipients) as well
1073 as the number of messages in eximstats.
1075 TF/01 Correct an error in the documentation for the redirect router. Exim
1076 does (usually) call initgroups() when daemonizing.
1078 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1079 with consistent privilege compared to when running as a daemon.
1081 TF/03 Note in the spec that $authenticated_id is not set for local
1082 submissions from trusted users.
1084 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1085 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1087 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1088 by adding some example configuration directives to the default
1089 configuration file. A little bit of work is required to uncomment the
1090 directives and define how usernames and passwords are checked, but
1091 there is now a framework to start from.
1093 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1094 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1095 without this. I don't know how relevant this is to other LDAP libraries.
1097 PH/02 Add the verb name to the "unknown ACL verb" error.
1099 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1101 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1103 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1104 directory not expanded when it should be if an expanded home directory
1105 was set for the address (which is overridden by the transport).
1107 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1110 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1111 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1112 because it is too late at that time, and has no effect.
1114 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1115 security issue with \' (bugzilla #107). I could not use the
1116 PQescapeStringConn() function, because it needs a PGconn value as one of
1119 PH/08 When testing addresses using -bt, indicate those final addresses that
1120 are duplicates that would not cause an additional delivery. At least one
1121 person was confused, thinking that -bt output corresponded to deliveries.
1122 (Suppressing duplicates isn't a good idea as you lose the information
1123 about possibly different redirections that led to the duplicates.)
1125 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1126 systems where poll() doesn't work, in particular OS X.
1128 PH/10 Added more information to debugging output for retry time not reached.
1130 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1131 operations in malware.c.
1133 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1136 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1137 syslog_duplication set false, log lines that would normally be written
1138 both the the main log and to the reject log were not written to syslog at
1141 PH/14 In the default configuration, change the use of "message" in ACL warn
1142 statements to "add_header".
1144 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1145 not followed by a command (e.g. "seen endif").
1147 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1148 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1151 PH/17 Added extra conditions to the default value of delay_warning_condition
1155 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1156 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1157 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1160 The Auto-Submitted: and various List- headers are standardised, whereas I
1161 don't think Precedence: ever was.
1163 PH/18 Refactored debugging code in route_finduser() to show more information,
1164 in particular, the error code if getpwnam() issues one.
1166 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1167 This is apparently needed in addition to the PH/07 change above to avoid
1168 any possible encoding problems.
1170 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1171 but not after initializing Perl.
1173 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1174 output them only if debugging. By default they are written stderr,
1175 apparently, which is not desirable.
1177 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1180 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1183 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1185 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1186 authenticated or an ident call has been made. Suppress the default
1187 values for $authenticated_id and $authenticated_sender (but permit -oMai
1188 and -oMas) when testing with -bh.
1190 PH/24 Re-jigged the order of the tests in the default configuration so that the
1191 tests for valid domains and recipients precede the DNS black list and CSA
1192 tests, on the grounds that those ones are more expensive.
1194 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1195 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1196 This bug exists in every version of Exim that I still have, right back to
1199 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1200 However, an attempt to turn on case-sensitivity in a regex key by
1201 including (?-i) didn't work because the subject string was already
1202 lowercased, and the effects were non-intuitive. It turns out that a
1203 one-line patch can be used to allow (?-i) to work as expected.
1209 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1210 other effects) broke the use of negated acl sub-conditions.
1212 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1215 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1216 "Deny" causes Exim to reject the incoming connection with a 554 error.
1217 Unfortunately, if there is a major crisis, such as a disk failure,
1218 tcp-wrappers gives "deny", whereas what one would like would be some
1219 kind of temporary error. A kludge has been added to help with this.
1220 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1221 554 error is used if errno is still zero or contains ENOENT (which occurs
1222 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1225 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1227 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1228 errors. Otherwise a message that provokes a temporary error (when other
1229 messages do not) can cause a whole host to time out.
1231 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1232 addresses were routed directly to files or pipes from a redirect router.
1233 File deliveries just didn't batch; pipe deliveries might have suffered
1236 PH/06 A failure to get a lock for a hints database would erroneously always say
1237 "Failed to get write lock", even when it was really a read lock.
1239 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1240 of 0600. This has been changed to use the value of the lockfile_mode
1241 option (which defaults to 0600).
1243 PH/08 Applied small patch from the Sieve maintainer.
1245 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1246 folder from quota calculations, a direct delivery into this folder messed
1247 up the contents of the maildirsize file. This was because the regex was
1248 used only to exclude .Trash (or whatever) when the size of the mailbox
1249 was calculated. There was no check that a delivery was happening into an
1250 excluded directory. This bug has been fixed by ignoring all quota
1251 processing for deliveries into excluded directories.
1253 PH/10 Added the maildirfolder_create_regex option to appendfile.
1259 PH/01 The code for finding all the local interface addresses on a FreeBSD
1260 system running IPv6 was broken. This may well have applied to all BSD
1261 systems, as well as to others that have similar system calls. The broken
1262 code found IPv4 interfaces correctly, but gave incorrect values for the
1263 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1264 that it would not match correctly against @[] and not recognize the IPv6
1267 PH/02 The ipliteral router was not recognizing addresses of the form user@
1268 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1270 PH/03 Added disable_ipv6.
1272 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1273 decoded form, because it is most often used to construct To: headers
1274 lines in autoreplies, and the decoded form may well be syntactically
1275 invalid. However, $reply_address has leading white space removed, and all
1276 newlines turned into spaces so that the autoreply transport does not
1279 PH/05 If group was specified without a user on a router, and no group or user
1280 was specified on a transport, the group from the router was ignored.
1282 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1283 for visible compile-time settings that can be used to change these
1284 numbers, for those that want even more. Backwards compatibility with old
1285 spool files has been maintained. However, going back to a previous Exim
1286 release will lost any variables that are in spool files.
1288 PH/07 Two small changes when running in the test harness: increase delay when
1289 passing a TCP/IP connection to a new process, in case the original
1290 process has to generate a bounce, and remove special handling of
1291 127.0.0.2 (sic), which is no longer necessary.
1293 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1294 be the same on different OS.
1296 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1299 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1300 whether --show-vars was specified or not
1302 JJ/02 exipick: Added support for new ACL variable spool format introduced
1305 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1306 syntactically invalid From: or Reply-to: line, and a filter used this to
1307 generate an autoreply, and therefore failed to obtain an address for the
1308 autoreply, Exim could try to deliver to a non-existent relative file
1309 name, causing unrelated and misleading errors. What now happens is that
1310 it logs this as a hard delivery error, but does not attempt to create a
1313 PH/11 The exinext utility has a -C option for testing purposes, but although
1314 the given file was scanned by exinext itself; it wasn't being passed on
1315 when Exim was called.
1317 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1318 an end-of-file indication when reading a command response.
1320 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1321 compiled. In many other places in Exim, IPv6 addresses are always
1322 recognized, so I have changed this. It also means that IPv4 domain
1323 literals of the form [IPV4:n.n.n.n] are now always recognized.
1325 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1326 used if the router is not running as root, for example, when verifying at
1327 ACL time, or when using -bh. The debugging output from this situation was
1328 non-existent - all you got was a failure to exec. I have made two
1331 (a) Failures to set uid/gid, the current directory, or a process leader
1332 in a subprocess such as that created by queryprogram now generate
1333 suitable debugging ouput when -d is set.
1335 (b) The queryprogram router detects when it is not running as root,
1336 outputs suitable debugging information if -d is set, and then runs
1337 the subprocess without attempting to change uid/gid.
1339 PH/15 Minor change to Makefile for building test_host (undocumented testing
1342 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1343 additional section of a DNS packet that returns MX or SRV records.
1344 Instead, it always explicitly searches for A/AAAA records. This avoids
1345 major problems that occur when a DNS server includes only records of one
1346 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1347 fixed another bug: if SRV records were looked up and the corresponding
1348 address records were *not* found in the additional section, the port
1349 values from the SRV records were lost.
1351 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1352 using the correct key (the original address) when searching the retry
1353 rules in order to find which one to use for generating the retry hint.
1355 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1356 adding the default one. Similarly, if it contains a Reply-To: header, the
1357 errors_reply_to option, if set, is not used.
1359 PH/19 When calculating a retry time, Exim used to measure the "time since
1360 failure" by looking at the "first failed" field in the retry record. Now
1361 it does not use this if it is later than than the arrival time of the
1362 message. Instead it uses the arrival time. This makes for better
1363 behaviour in cases where some deliveries succeed, thus re-setting the
1364 "first failed" field. An example is a quota failure for a huge message
1365 when small messages continue to be delivered. Without this change, the
1366 "time since failure" will always be short, possible causing more frequent
1367 delivery attempts for the huge message than are intended.
1368 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1370 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1371 $1, $2, $3) because the numerical variables can be reset during some
1372 expansion items (e.g. "match"), thereby losing the authentication data.
1374 PH/21 Make -bV show the size of off_t variables so that the test suite can
1375 decide whether to run tests for quotas > 2G.
1377 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1378 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1379 filecount value is greater than 2G or if a quota value is greater than 2G
1380 on a system where the size of off_t is not greater than 4, a panic error
1383 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1384 never match. The debug and -bh output now contains an explicit error
1385 message indicating a malformed IPv4 address or mask.
1387 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1388 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1389 PH/23 above applies.
1391 PH/25 Do not write to syslog when running in the test harness. The only
1392 occasion when this arises is a failure to open the main or panic logs
1393 (for which there is an explicit test).
1395 PH/26 Added the /no_tell option to "control=freeze".
1397 PH/27 If a host name lookup failed very early in a connection, for example, if
1398 the IP address matched host_lookup and the reverse lookup yielded a name
1399 that did not have a forward lookup, an error message of the form "no IP
1400 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1401 could be logged. Now it outputs the IP address instead of "NULL".
1403 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1404 allows the sender and the authenticated sender to be set when
1405 submitting a message from within Exim. Since child_open_exim() is
1406 documented for local_scan(), the new function should be too.
1408 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1409 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1410 results in an empty string is now treated as unset.
1412 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1414 PH/31 Added sender_verify_fail logging option.
1416 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1417 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1418 tidied the source and removed it altogether.
1420 PH/33 When a queue run was abandoned because the load average was too high, a
1421 log line was always written; now it is written only if the queue_run log
1422 selector is set. In addition, the log line for abandonment now contains
1423 information about the queue run such as the pid. This is always present
1424 in "start" and "stop" lines but was omitted from the "abandon" line.
1426 PH/34 Omit spaces between a header name and the colon in the error message that
1427 is given when verify = headers_syntax fails (if there are lots of them,
1428 the message gets confusing).
1430 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1431 names, as there are now some PTR records that contain slashes. This check
1432 is only to protect against broken name servers that fall over on strange
1433 characters, so the fact that it applies to all lookups doesn't matter.
1435 PH/36 Now that the new test suite is complete, we can remove some of the
1436 special code in Exim that was needed for the old test suite. For example,
1437 sorting DNS records because real resolvers return them in an arbitrary
1438 order. The new test suite's fake resolver always returns records in the
1441 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1442 bounces) except when queue_only is set, to avoid logging races between
1443 the different processes.
1445 PH/38 Panic-die if .include specifies a non-absolute path.
1447 PH/39 A tweak to the "H" retry rule from its user.
1449 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1450 a label. They prevented compilation on older perls.
1452 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1453 a warning to be raised on newish perls.
1455 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1456 on queue. Changes to match documented behaviour of showing count of
1457 messages matching specified criteria.
1459 PH/40 Changed the default ident timeout from 30s to 5s.
1461 PH/41 Added support for the use of login_cap features, on those BSD systems
1462 that have them, for controlling the resources used by pipe deliveries.
1464 PH/42 The content-scanning code uses fopen() to create files in which to put
1465 message data. Previously it was not paying any attention to the mode of
1466 the files. Exim runs with umask(0) because the rest of the code creates
1467 files with open(), and sets the required mode explicitly. Thus, these
1468 files were ending up world-writeable. This was not a big issue, because,
1469 being within the spool directory, they were not world-accessible. I have
1470 created a function called modefopen, which takes an additional mode
1471 argument. It sets umask(777), creates the file, chmods it to the required
1472 mode, then resets the umask. All the relevant calls to fopen() in the
1473 content scanning code have been changed to use this function.
1475 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1476 to 24 hours. This avoids potential overflow problems when processing G
1477 and H retry rules. I suspect nobody ever tinkers with this value.
1479 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1481 PH/45 When the plaintext authenticator is running as a client, the server's
1482 challenges are checked to ensure they are valid base64 strings. By
1483 default, the authentication attempt is cancelled if an invalid string is
1484 received. Setting client_ignore_invalid_base64 true ignores these errors.
1485 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1486 they are received. Thus, the responses can be made to depend on the
1487 challenges. If an invalid string is ignored, an empty string is placed in
1490 PH/46 Messages that are created by the autoreply transport now contains a
1491 References: header, in accordance with RFCs 2822 and 3834.
1493 PH/47 Added authenticated_sender_force to the smtp transport.
1495 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1497 PH/49 Installed latest patch from the Sieve maintainer.
1499 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1500 was also set, the appendfile transport was unnecessarily scanning a
1501 directory of message files (e.g. for maildir delivery) to find the count
1502 of files (along with the size), even though it did not need this
1503 information. It now does the scan only if it needs to find either the
1504 size of the count of files.
1506 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1508 PH/52 Two bugs concerned with error handling when the smtp transport is
1511 (i) Exim was not creating retry information for temporary errors given
1512 for individual recipients after the DATA command when the smtp transport
1513 was used in LMTP mode. This meant that they could be retried too
1514 frequently, and not timed out correctly.
1516 (ii) Exim was setting the flag that allows error details to be returned
1517 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1518 recipients that were returned after the DATA command.
1520 PH/53 This is related to PH/52, but is more general: for any failing address,
1521 when detailed error information was permitted to be returned to the
1522 sender, but the error was temporary, then after the final timeout, only
1523 "retry timeout exceeded" was returned. Now it returns the full error as
1524 well as "retry timeout exceeded".
1526 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1527 do this, and (what is worse) MTAs that accept it.
1529 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1530 will now be deprecated.
1532 PH/56 New os.c-cygwin from the Cygwin maintainer.
1534 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1535 formats (previously only available in exim formats via -bpr, -bpru,
1536 and -bpra. Now also available in native and exiqgrep formats)
1538 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1539 with very large, slow to parse queues
1541 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1543 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1545 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1546 responses to authentication challenges, though it was showing the
1547 challenges; (ii) I've removed the CR characters from the debug output for
1550 PH/58 Allow for the insertion of a newline as well as a space when a string
1551 is turned into more than one encoded-word during RFC 2047 encoding. The
1552 Sieve code now uses this.
1554 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1555 data_4xx, lost_connection, tls_required.
1557 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1558 message was being sent as an SMTP response.
1560 PH/61 Add -l and -k options to exicyclog.
1562 PH/62 When verifying, if an address was redirected to one new address, so that
1563 verification continued, and the new address failed or deferred after
1564 having set something in $address_data, the value of $address_data was not
1565 passed back to the ACL. This was different to the case when no
1566 redirection occurred. The value is now passed back in both cases.
1568 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1569 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1570 use login_cap.h, so on its own it isn't the distinguishing feature. The
1571 new name refers directly to the setclassresources() function.
1573 PH/65 Added configuration files for NetBSD3.
1575 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1577 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1578 is preferred over IPv4.
1580 PH/68 The bounce_return_message and bounce_return_body options were not being
1581 honoured for bounces generated during the reception of non-SMTP messages.
1582 In particular, this applied to messages rejected by the ACL. This bug has
1583 been fixed. However, if bounce_return_message is true and bounce_return_
1584 body is false, the headers that are returned for a non-SMTP message
1585 include only those that have been read before the error was detected.
1586 (In the case of an ACL rejection, they have all been read.)
1588 PH/69 The HTML version of the specification is now built in a directory called
1589 spec_html instead of spec.html, because the latter looks like a path with
1590 a MIME-type, and this confuses some software.
1592 PH/70 Catch two compiler warnings in sieve.c.
1594 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1595 function verify_get_ident() calls ip_connect() to connect a socket, but
1596 if the "connect()" function timed out, ip_connect() used to close the
1597 socket. However, verify_get_ident() also closes the socket later, and in
1598 between Exim writes to the log, which may get opened at this point. When
1599 the socket was closed in ip_connect(), the log could get the same file
1600 descriptor number as the socket. This naturally causes chaos. The fix is
1601 not to close the socket in ip_connect(); the socket should be closed by
1602 the function that creates it. There was only one place in the code where
1603 this was missing, in the iplookup router, which I don't think anybody now
1604 uses, but I've fixed it anyway.
1606 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1607 well as to direct DNS lookups. Otherwise the handling of names in host
1608 lists is inconsistent and therefore confusing.
1614 PH/01 Two changes to the default runtime configuration:
1616 (1) Move the checks for relay_from_hosts and authenticated clients from
1617 after to before the (commented out) DNS black list checks.
1619 (2) Add control=submission to the relay_from_hosts and authenticated
1620 clients checks, on the grounds that messages accepted by these
1621 statements are most likely to be submissions.
1623 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1625 (1) Generate an error if the third argument for the ${prvs expansion is
1628 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1631 (3) Reset the variables that are obtained from the first argument of
1632 ${prvscheck and used in the second argument before leaving the code,
1633 because their memory is reclaimed, so using them afterwards may do
1636 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1637 one (it's much easier than Tom thought :-).
1639 (5) Because of (4), we can now allow for the use of $prvscheck_result
1640 inside the third argument.
1642 PH/03 For some reason, the default setting of PATH when running a command from
1643 a pipe transport was just "/usr/bin". I have changed it to
1646 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1647 anything to be listed in the output from -bV.
1649 PH/05 When a filter generated an autoreply, the entire To: header line was
1650 quoted in the delivery log line, like this:
1652 => >A.N.Other <ano@some.domain> <original@ddress> ...
1654 This has been changed so that it extracts the operative address. There
1655 may be more than one such address. If so, they are comma-separated, like
1658 => >ano@some.domain,ona@other.domain <original@ddress> ...
1660 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1661 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1662 not being looked up in the rDNS to get a host name, Exim was showing the
1663 IP address twice in Received: lines, even though the IP addresses were
1664 identical. For example:
1666 Received: from [1.2.3.4] (helo=[1.2.3.4])
1668 However, if the real host name was known, it was omitting the HELO data
1669 if it matched the actual IP address. This has been tidied up so that it
1670 doesn't show the same IP address twice.
1672 PH/07 When both +timestamp and +memory debugging was on, the value given by
1673 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1674 called by the debug printing, thereby overwriting the timestamp buffer.
1675 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1677 PH/08 When the original message was included in an autoreply transport, it
1678 always said "this is a copy of the message, including all the headers",
1679 even if body_only or headers_only was set. It now gives an appropriate
1682 PH/09 Applied a patch from the Sieve maintainer which:
1684 o fixes some comments
1685 o adds the (disabled) notify extension core
1686 o adds some debug output for the result of if/elsif tests
1687 o points to the current vacation draft in the documentation
1688 and documents the missing references header update
1692 o fixes a bug in processing the envelope test (when testing
1693 multiple envelope elements, the last element determinted the
1696 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1697 Electronic Mail") by including:
1699 Auto-submitted: auto-generated
1701 in the messages that it generates (bounce messages and others, such as
1702 warnings). In the case of bounce messages for non-SMTP mesages, there was
1703 also a typo: it was using "Auto_submitted" (underscore instead of
1704 hyphen). Since every message generated by Exim is necessarily in response
1705 to another message, thes have all been changed to:
1707 Auto-Submitted: auto-replied
1709 in accordance with these statements in the RFC:
1711 The auto-replied keyword:
1713 - SHOULD be used on messages sent in direct response to another
1714 message by an automatic process,
1716 - MUST NOT be used on manually-generated messages,
1718 - MAY be used on Delivery Status Notifications (DSNs) and Message
1719 Disposition Notifications (MDNs),
1721 - MUST NOT be used on messages generated by automatic or periodic
1722 processes, except for messages which are automatic responses to
1725 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1726 to the default Received: header definition.
1728 PH/12 Added log selector acl_warn_skipped (default on).
1730 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1731 variables because (a) they are in the wrong storage pool and (b) even if
1732 they were copied, it wouldn't work properly because of the caching.
1734 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1735 checking when decoding. Apparently there are clients that generate
1736 overlong encoded strings. Why am I not surprised?
1738 PH/15 If the first argument of "${if match_address" was not empty, but did not
1739 contain an "@" character, Exim crashed. Now it writes a panic log message
1740 and treats the condition as false.
1742 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1744 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1745 extension "envelope-auth". The code is finished and in agreement with
1746 other implementations, but there is no documentation so far and in fact,
1747 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1748 not changing the active code.
1750 Print executed "if" and "elsif" statements when debugging is used. This
1751 helps a great deal to understand what a filter does.
1753 Document more things not specified clearly in RFC3028. I had all this
1754 sorted out, when out of a sudden new issues came to my mind. Oops."
1756 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1759 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1760 canonical form (as documented). However, after a host name lookup from
1761 the IP address, check_host() was doing a simple string comparison with
1762 addresses acquired from the DNS when checking that the found name did
1763 have the original IP as one of its addresses. Since any found IPv6
1764 addresses are likely to be in abbreviated form, the comparison could
1765 fail. Luckily, there already exists a function for doing the comparison
1766 by converting both addresses to binary, so now that is used instead of
1767 the text comparison.
1769 PH/20 There was another similar case to PH/19, when a complete host name was
1770 given in a host list; looking up its IP address could give an abbreviated
1771 form, whereas the current host's name might or might not be abbreviated.
1772 The same fix has been applied.
1778 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1779 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1782 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1783 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1785 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1787 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1788 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1789 Also, for "Auto-Submitted", treat anything other than "no" as
1790 non-personal, in accordance with RFC 3834. (Previously it treated
1791 anything starting "auto-" as non-personal.)
1793 TF/01 The control=submission/name=... option had a problem with syntax
1794 errors if the name included a slash character. The /name= option
1795 now slurps the rest of the string, so it can include any characters
1796 but it must come last in the list of options (after /sender_retain
1799 PH/05 Some modifications to the interface to the fake nameserver for the new
1807 TK/01 Added the "success_on_redirect" address verification option. See
1808 NewStuff for rationale and an example.
1810 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1812 PH/02 Patch to exigrep to allow it to work on syslog lines.
1814 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1815 fread() to skip over the body file's header line, because in Cygwin the
1816 header line is locked and is inaccessible.
1818 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1819 co-exist for some time) to make it clear that it is the Exim ID that is
1820 referenced, not the Message-ID: header line.
1822 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1823 string_format() function, because snprintf() does not exist on all
1826 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1827 ${stat: expansion item.
1829 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1830 protocol synchronization error", to keep the pedants happy.
1832 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1833 well as for IRIX systems, when gcc is being used. See the host.c source
1836 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1838 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1841 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1842 transport and to the smtp transport in LMTP mode.
1844 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1846 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1847 run a filter in a subprocess. This could lead to confusion in subsequent
1848 lookups in the parent process. There should also be a search_tidyup() at
1849 the end of the subprocess.
1851 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1852 only if the host matched helo_try_verify_hosts, which caused the
1853 verification to occur when the EHLO/HELO command was issued. The ACL just
1854 tested the remembered result. Now, if a previous verification attempt has
1855 not happened, "verify = helo" does it there and then.
1857 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1859 TK/03 Fix log output including CR from clamd.
1861 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1862 exist provoked a memory error which could cause a segfault.
1864 PH/15 Installed PCRE 6.2
1866 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1868 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1869 of the problem. Specifically, suggested +O2 rather than +O1 for the
1872 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1874 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1875 record, the port was not correctly logged when the outgoing_port log
1876 selector was set (it logged the transort's default port).
1878 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1879 fallback_hosts, and "hosts" in the smtp transport.
1881 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1882 host errors such as "Connection refused".
1884 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1885 authentication with radiusclient 0.4.9:
1887 - Error returned from rc_read_config was caught wrongly
1888 - Username/password not passed on to radius server due to wrong length.
1890 The presumption is that some radiusclient API changes for 4.51/PH/17
1891 were not taken care of correctly. The code is still untested by me (my
1892 Linux distribution still has 0.3.2 of radiusclient), but it was
1893 contributed by a Radius user.
1895 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1896 expanding the "port" option of the smtp transport.
1898 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1899 while reading a MIME header. Thanks to Tom Hughes for a patch.
1901 PH/24 Include config.h inside local_scan.h so that configuration settings are
1904 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1905 that in an ACL for RCPT (for example), you can examine exactly what was
1908 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1909 commands, but it was not correctly comparing the address with the actual
1910 client host address. Thus, it would show the EHLO address in Received:
1911 header lines when this was not necessary.
1913 PH/27 Added the % operator to ${eval:}.
1915 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1916 it should be ignoring failures (because with -C, for example, it has lost
1917 privilege). It wasn't ignoring creation failures other than "already
1920 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1923 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1924 because an input line was too long, either on its own, or by virtue of
1925 too many continuations, the temporary file was not being removed, and the
1926 return code was incorrect.
1928 PH/31 Missing "BOOL" in function definition in filtertest.c.
1930 PH/32 Applied Sieve patches from the maintainer.
1932 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1934 PH/33 Added "verify = not_blind".
1936 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1937 Local/Makefile (with some defaults set). These are used in built scripts
1938 such as exicyclog, but they have never been used in the exim_install
1939 script (though there are many overriding facilities there). I have
1940 arranged that the exim_install script now takes note of these two
1943 PH/35 Installed configuration files for Dragonfly.
1945 PH/36 When a locally submitted message by a trusted user did not contain a
1946 From: header, and the sender address was obtained from -f or from an SMTP
1947 MAIL command, and the trusted user did not use -F to supply a sender
1948 name, $originator_name was incorrectly used when constructing a From:
1949 header. Furthermore, $originator_name was used for submission mode
1950 messages from external hosts without From: headers in a similar way,
1951 which is clearly wrong.
1953 PH/37 Added control=suppress_local_fixups.
1955 PH/38 When log_selector = +received_sender was set, and the addition of the
1956 sender made the log line's construction buffer exactly full, or one byte
1957 less than full, an overflow happened when the terminating "\n" was
1960 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1961 when the result of a list match is failure because a DNS lookup failed.
1963 PH/40 RM_COMMAND is now used in the building process.
1965 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1966 the "build-* directories that it finds.
1968 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1969 address in a domain literal was a prefix of an interface address.
1971 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1972 when verifying a sender address, unless rewrite_headers is false.
1974 PH/44 Wrote a long comment about why errors_to addresses are verified as
1975 recipients, not senders.
1977 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1978 the ratelimit ACL was added.
1980 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1982 PH/46 Added extra information about PostgreSQL errors to the error string.
1984 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1985 avoiding the need to install special zones in a real server. This is
1986 backwards compatible; if it can't find the fake resolver, it drops back.
1987 Thus, both old and new test suites can be run.
1989 TF/02 Added util/ratelimit.pl
1991 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1992 clock is set back in time.
1994 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1995 Candler <B.Candler@pobox.com>.
1997 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1998 OK for addresses that are the result of redirections.
2000 PH/48 A number of further additions for the benefit of the new test suite,
2001 including a fake gethostbyname() that interfaces to the fake DNS resolver
2004 TF/06 The fix for widen_domains has also been applied to qualify_single and
2005 search_parents which are the other dnslookup options that can cause
2008 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2011 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2013 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2014 service" if it encountered an MX record with an empty target hostname.
2015 The message is now "an MX or SRV record indicated no SMTP service".
2017 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2018 if the DNS of the sending site is misconfigured. This is quite a
2019 common situation. This change restores the behaviour of treating a
2020 helo verification defer as a failure.
2022 PH/51 If self=fail was set on a router, the bounce message did not include the
2023 actual error message.
2029 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2031 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2032 command itself ended in error, the underlying message about the transport
2033 filter timeout was being overwritten with the pipe command error. Now the
2034 underlying error message should be appended to the second error message.
2036 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2038 PH/02 Reduce the amount of output that "make" produces by default. Full output
2039 can still be requested.
2041 PH/03 The warning log line about a condition test deferring for a "warn" verb
2042 was being output only once per connection, rather than after each
2043 occurrence (because it was using the same function as for successful
2044 "warn" verbs). This seems wrong, so I have changed it.
2046 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2047 it should not have, which might have caused a crash in the right
2048 circumstances, but probably never did.
2050 PH/04 Installed a modified version of Tony Finch's patch to make submission
2051 mode fix the return path as well as the Sender: header line, and to
2052 add a /name= option so that you can make the user's friendly name appear
2055 TF/03 Added the control = fakedefer ACL modifier.
2057 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2058 Mark Lowes for thorough testing.
2060 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2062 TK/03 Merged latest SRS patch from Miles Wilton.
2064 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2065 with the definition in sysexits.h (which is #included earlier).
2066 Fortunately, Exim does not actually use EX_OK. The code used to try to
2067 preserve the sysexits.h value, by assumimg that macro definitions were
2068 scanned for macro replacements. I have been disabused of this notion,
2069 so now the code just undefines EX_OK before #including unistd.h.
2071 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2072 in the smtp transport. When a block could not be written in a single
2073 write() function, the timeout was being re-applied to each part-write.
2074 This seems wrong - if the receiver was accepting one byte at a time it
2075 would take for ever. The timeout is now adjusted when this happens. It
2076 doesn't have to be particularly precise.
2078 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2079 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2081 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2082 without a domain if the check to <postmaster@domain> fails.
2084 SC/01 Eximstats: added -xls and the ability to specify output files
2085 (patch written by Frank Heydlauf).
2087 SC/02 Eximstats: use FileHandles for outputing results.
2089 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2091 SC/04 Eximstats: fixed display of large numbers with -nvr option
2093 SC/05 Eximstats: fixed merging of reports with empty tables.
2095 SC/06 Eximstats: added the -include_original_destination flag
2097 SC/07 Eximstats: removed tabs and trailing whitespace.
2099 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2101 TK/06 MBOX spool code: Add real "From " MBOX separator line
2102 so the .eml file is really in mbox format (even though
2103 most programs do not really care). Patch from Alex Miller.
2105 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2106 The latter is generated from $received_to and is only set if the
2107 message has one envelope recipient. SA can use these headers,
2108 obviously out-of-the-box. Patch from Alex Miller.
2110 PH/08 The ${def test on a variable was returning false if the variable's
2111 value was "0", contrary to what the specification has always said!
2112 The result should be true unless the variable is empty.
2114 PH/09 The syntax error of a character other than { following "${if
2115 def:variable_name" (after optional whitespace) was not being diagnosed.
2116 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2117 accidental colon was present, for example, could give incorrect results.
2119 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2120 result is used (not including appendfile, where other changes are about
2123 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2124 This involved changing a lot of size variables from int to off_t. It
2125 should work with maildirs and everything.
2127 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2128 spamd dying while we are connected to it.
2130 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2133 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2134 function for BDB 4.3.
2136 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2138 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2139 the sources for PCRE have been re-arranged and more files are now
2142 PH/15 The code I had for printing potentially long long variables in PH/11
2143 above was not the best (it lost precision). The length of off_t variables
2144 is now inspected at build time, and an appropriate printing format (%ld
2145 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2146 to be "long long int" or "long int". This is needed for the internal
2147 formatting function string_vformat().
2149 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2150 the configuration file to be ":syslog", then the script "guesses" where
2151 the logs files are, rather than using the compiled in default. In our
2152 case the guess is not the same as the compiled default, so the script
2153 suddenly stopped working when I started to use syslog. The patch checks
2154 to see if log_file_path is "". If so, it attempts to read it from exim
2155 with no configuration file to get the compiled in version, before it
2156 falls back to the previous guessing code."
2158 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2159 implementing BATV in an Exim configuration. See NewStuff for the gory
2162 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2163 Makefile that are specific to HP-UX.
2165 PH/18 If the "use_postmaster" option was set for a recipient callout together
2166 with the "random" option, the postmaster address was used as the MAIL
2167 FROM address for the random test, but not for the subsequent recipient
2168 test. It is now used for both.
2170 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2171 patch removes a few documentation additions to RFC 3028, because the
2172 latest draft now contains them. It adds the new en;ascii-case comparator
2173 and a new error check for 8bit text in MIME parts. Comparator and
2174 require names are now matched exactly. I enabled the subaddress
2175 extension, but it is not well tested yet (read: it works for me)."
2177 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2178 rework some of the code of TK/09 above to avoid the hardwired use of
2179 "%lld" and "long long". Replaced the call to snprintf() with a call to
2182 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2183 records point to non-existent hosts", "retry timeout exceeded", and
2184 "retry time not reached for any host after a long failure period".
2186 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2187 experimental DomainKeys support:
2189 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2190 (2) On an error such as an illegally used "control", the wrong name for
2191 the control was given.
2193 These problems did NOT occur unless DomainKeys support was compiled.
2195 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2197 PH/24 Added ${if match_ip condition.
2199 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2200 databases so that it will be absolutely obvious if a crash occurs in the
2201 DB library. This is a regular occurrence (often caused by mis-matched
2204 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2205 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2206 picked up on a user's system that detects such things. There doesn't seem
2207 to be a gcc warning option for this - only an attribute that has to be
2208 put on the function's prototype. It seems that in Fedora Core 4 they have
2209 set this on a number of new functions. No doubt there will be more in due
2212 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2213 specify a transport. However, if an address that was verified by such a
2214 router was the subject of a callout, Exim crashed because it tried to
2215 read the rcpt_include_affixes from the non-existent transport. Now it
2216 just assumes that the setting of that option is false. This bug was
2217 introduced by 4.51/PH/31.
2219 PH/28 Changed -d+all to exclude +memory, because that information is very
2220 rarely of interest, but it makes the output a lot bigger. People tend to
2221 do -d+all out of habit.
2223 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2224 code in os-type was giving problems when libc.so lives in lib64, like on
2227 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2228 aren't the modern standard, and it seems that some systems' include files
2229 don't always have them. Exim was already checking for some of the newer
2230 ones like T_AAAA, and defining it itself. I've added checks for all the
2231 record types that Exim uses.
2233 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2234 not automatically generating a new one, as it is supposed to. This
2235 prevented TLS from working. If the file did exist, but contained invalid
2236 data, a new version was generated, as expected. It was only the case of a
2237 non-existent file that was broken.
2239 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2240 with a change in libdomainkeys > 0.64.
2242 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2243 from DNS. If the selector record carries the flag, it now has
2244 precedence over the domain-wide flag.
2246 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2248 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2249 the use of a port name that isn't defined in /etc/services) occurred, the
2250 message was deferred as in a normal delivery, and thus remained on the
2251 spool, instead of being failed because of the mua_wrapper setting. This
2252 is now fixed, and I tidied up some of the mua_wrapper messages at the
2255 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2256 the messages in a hash of arrays rather than using individual hashes.
2257 This is a bit cleaner and results in dramatic memory savings, albeit
2258 at a slight CPU cost.
2260 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2261 as requested by Marc Sherman.
2263 SC/10 Eximstats: added histograms for user specified patterns as requested
2266 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2268 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2269 fopen() in the content-scanning modules that did not already have it.
2275 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2276 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2278 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2280 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2282 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2283 to test Sieve filters that use "vacation".
2285 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2286 that changes the way the GnuTLS parameters are stored in the cache file.
2287 The new format can be generated externally. For backward compatibility,
2288 if the data in the cache doesn't make sense, Exim assumes it has read an
2289 old-format file, and it generates new data and writes a new file. This
2290 means that you can't go back to an older release without removing the
2293 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2294 work if there are any delivery delays because "one_time" forces the
2295 parent to be marked "delivered", so its unseen clone is never tried
2296 again. For this reason, Exim now forbids the simultaneous setting of
2299 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2300 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2301 entry if you want to know the details. The fix, however, neglected to
2302 consider the case when local delivery batching is involved. The test for
2303 "previously delivered" was not happening when checking to see if an
2304 address could be batched with a previous (undelivered) one; under
2305 certain circumstances this could lead to multiple deliveries to the same
2308 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2309 in its include files, and this causes problems building Exim.
2311 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2312 header_syntax) but Exim was just ignoring anything given after a slash.
2313 In particular, this caused confusion with an attempt to use "verify =
2314 reverse_host_lookup/defer_ok". An error is now given when options are
2315 supplied for verify items that do not have them. (Maybe reverse_host_
2316 lookup should have a defer_ok option, but that's a different point.)
2318 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2319 defined by RFC 821) to 2048, because there were problems with some AUTH
2320 commands, and RFC 1869 says the size should be increased for extended
2321 SMTP commands that take arguments.
2323 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2326 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2327 "unknown" error; now it says that the functionality isn't in the binary.
2329 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2330 an address' error message when a string expansion fails (syntax or
2331 whatever). Otherwise the password may appear in the log. Following change
2332 PH/42 below, there is no longer a chance of it appearing in a bounce
2335 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2337 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2338 of its addresses was ever tried. (Bugzilla bug #2.)
2340 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2341 the result incorrectly in the debug output. (It correctly added a newline
2342 to what was transported.)
2344 TF/01 Added $received_time.
2346 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2347 commented out examples of how to interface to a virus scanner and to
2348 SpamAssassin. Also added commented examples of av_scanner and
2349 spamd_address settings.
2351 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2352 and controls are allowed in which ACLs. There were a couple of minor
2353 errors. Some of the entries in the conditions table (which is a table of
2354 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2355 negation of where the condition IS allowed.
2357 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2359 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2360 header file does not have a version number, so I've had to invent a new
2361 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2362 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2363 radiusclient), but it was contributed by a Radius user.
2365 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2366 files or directories, for OpenSSL.
2368 PH/19 When an Exim process that is running as root has to create an Exim log
2369 file, it does so in a subprocess that runs as exim:exim so as to get the
2370 ownership right at creation (otherwise, other Exim processes might see
2371 the file with the wrong ownership). There was no test for failure of this
2372 fork() call, which would lead to the process getting stuck as it waited
2373 for a non-existent subprocess. Forks do occasionally fail when resources
2374 run out. I reviewed all the other calls to fork(); they all seem to check
2377 PH/20 When checking for unexpected SMTP input at connect time (before writing
2378 the banner), Exim was not dealing correctly with a non-positive return
2379 from the read() function. If the client had disconnected by this time,
2380 the result was a log entry for a synchronization error with an empty
2381 string after "input=" when read() returned zero. If read() returned -1
2382 (an event I could not check), uninitialized data bytes were printed.
2383 There were reports of junk text (parts of files, etc) appearing after
2386 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2388 PH/22 Added support for macro redefinition, and (re)definition in between
2389 driver and ACL definitions.
2391 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2392 forgetting to use the resulting value; it was using the unexpanded value.
2394 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2395 hadn't been configured. The fix is from Juergen Kreileder, who
2396 understands it better than I do:
2398 "Here's what I see happening with three configured cyrus_sasl
2399 authenticators configured (plain, login, cram-md5):
2401 On startup auth_cyrus_sasl_init() gets called for each of these.
2402 This means three calls to sasl_listmech() without a specified mech_list.
2403 => SASL tests which mechs of all available mechs actually work
2404 => three warnings about OTP not working
2405 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2407 With the patch, sasl_listmech() also gets called three times. But now
2408 SASL's mech_list option is set to the server_mech specified in the the
2409 authenticator. Or in other words, the answer from sasl_listmech()
2410 gets limited to just the mech you're testing for (which is different
2412 => the return list contains just 'plain' or 'login', 'cram-md5' or
2413 nothing depending on the value of ob->server_mech.
2415 I've just tested the patch: Authentication still works fine,
2416 unavailable mechs specified in the exim configuration are still
2417 caught, and the auth.log warnings about OTP are gone."
2419 PH/25 When debugging is enabled, the contents of the command line are added
2420 to the debugging output, even when log_selector=+arguments is not
2423 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2424 answer is "GNU", and only if the return is "GNU/something" is the answer
2427 PH/27 $acl_verify_message is now set immediately after the failure of a
2428 verification in an ACL, and so is available in subsequent modifiers. In
2429 particular, the message can be preserved by coding like this:
2431 warn !verify = sender
2432 set acl_m0 = $acl_verify_message
2434 Previously, $acl_verify_message was set only while expanding "message"
2435 and "log_message" when a very denied access.
2437 PH/28 Modified OS/os.c-Linux with
2439 -#ifndef OS_LOAD_AVERAGE
2440 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2442 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2443 nomenclature these days.)
2445 PH/29 Installed patch from the Sieve maintainer that adds the options
2446 sieve_useraddress and sieve_subaddress to the redirect router.
2448 PH/30 In these circumstances:
2449 . Two addresses routed to the same list of hosts;
2450 . First host does not offer TLS;
2451 . First host accepts first address;
2452 . First host gives temporary error to second address;
2453 . Second host offers TLS and a TLS session is established;
2454 . Second host accepts second address.
2455 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2456 and peerdn, if requested) that were in fact used only for the second
2459 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2460 attention to any local part prefix or suffix that was matched by the
2461 router that accepted the address. It now behaves in the same way as it
2462 does for delivery: the affixes are removed from the local part unless
2463 rcpt_include_affixes is set on the transport.
2465 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2466 timeout during the DATA phase of an incoming message.
2468 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2469 applied a patch sanctioned by the Sieve maintainer.
2471 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2472 the uid or gid is negative. A case of a negative gid caused this to be
2473 noticed. The fix allows for either to be negative.
2475 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2476 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2479 PH/36 The change PH/12 above was broken. Fixed it.
2481 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2482 the grounds that routing the same address twice would always produce the
2483 same answer. This might have been true once, but it is certainly no
2484 longer true now. Routing a child address may depend on the previous
2485 routing that produced that child. Some complicated redirection strategies
2486 went wrong when messages had multiple recipients, and made Exim's
2487 behaviour dependent on the order in which the addresses were given.
2489 I have moved the duplicate checking until after the routing is complete.
2490 Exim scans the addresses that are assigned to local and remote
2491 transports, and removes any duplicates. This means that more work will be
2492 done, as duplicates will always all be routed, but duplicates are
2493 presumably rare, so I don't expect this is of any significance.
2495 For deliveries to pipes, files, and autoreplies, the duplicate checking
2496 still happens during the routing process, since they are not going to be
2499 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2500 It corrects a timeout issue with spamd. This is Ian's comment: "The
2501 background is that sometimes spamd either never reads data from a
2502 connection it has accepted, or it never writes response data. The exiscan
2503 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2504 blindly assumes that writes won't block so it may never time out."
2506 PH/39 Allow G after quota size as well as K and M.
2508 PH/40 The value set for $authenticated_id in an authenticator may not contain
2509 binary zeroes or newlines because the value is written to log lines and
2510 to spool files. There was no check on this. Now the value is run through
2511 the string_printing() function so that such characters are converted to
2512 printable escape sequences.
2514 PH/41 $message_linecount is a new variable that contains the total number of
2515 lines in the message. Compare $body_linecount, which is the count for the
2518 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2519 bounce and delay warning messages, except in certain special cases, which
2522 (a) An SMTP error message from a remote host;
2523 (b) A message specified in a :fail: redirection;
2524 (c) A message specified in a "fail" command in a system filter;
2525 (d) A message specified in a FAIL return from the queryprogram router;
2526 (e) A message specified by the cannot_route_message router option.
2528 In these cases only, Exim does include the error details in bounce and
2529 warning messages. There are also a few cases where bland messages such
2530 as "unrouteable address" or "local delivery error" are given.
2532 PH/43 $value is now also set for the "else" part of a ${run expansion.
2534 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2535 being worked on, but at least Exim now implements the latest version to
2538 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2539 process to complete was treated as a delivery failure, a timeout while
2540 writing the message to the pipe was logged, but erroneously treated as a
2541 successful delivery. Such timeouts include transport filter timeouts. For
2542 consistency with the overall process timeout, these timeouts are now
2543 treated as errors, giving rise to delivery failures by default. However,
2544 there is now a new Boolean option for the pipe transport called
2545 timeout_defer, which, if set TRUE, converts the failures into defers for
2546 both kinds of timeout. A transport filter timeout is now identified in
2549 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2550 systems where "make" and "gmake" are different, calling "gmake" at top
2551 level broke things. I've arranged for the value of $(MAKE) to be passed
2552 from the Makefile to this script so that it can call the same version of
2556 A note about Exim versions 4.44 and 4.50
2557 ----------------------------------------
2559 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2560 changes of various kinds. As a consequence, a big documentation update was
2561 needed. This delayed the release for rather longer than seemed good, especially
2562 in the light of a couple of (minor) security issues. Therefore, the changes
2563 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2564 release. So 4.44 and 4.50 are in effect two different branches that both start
2567 I have left the 4.50 change log unchanged; it contains all the changes since
2568 4.43. The change log for 4.44 is below; many of its items are identical to
2569 those for 4.50. This seems to be the most sensible way to preserve the
2570 historical information.
2576 1. Minor wording change to the doc/README.SIEVE file.
2578 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2579 computation of the current number of files was incorrect.
2581 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2582 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2585 4. Give more explanation in the error message when the command for a transport
2586 filter fails to execute.
2588 5. There are several places where Exim runs a non-Exim command in a
2589 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2590 was being done only for the command run by the queryprogram router. It is
2591 now done for all such subprocesses. The other cases are: ${run, transport
2592 filters, and the commands run by the lmtp and pipe transports.
2594 6. Added CONFIGURE_GROUP build-time option.
2596 7. Some older OS have a limit of 256 on the maximum number of file
2597 descriptors. Exim was using setrlimit() to set 1000 as a large value
2598 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2599 systems. I've change it so that if it can't get 1000, it tries for 256.
2601 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2602 was an oversight, and furthermore, ever since the addition of extra
2603 controls (e.g. 4.43/32), the checks on when to allow different forms of
2604 "control" were broken. There should now be diagnostics for all cases when a
2605 control that does not make sense is encountered.
2607 9. Added the /retain_sender option to "control=submission".
2609 10. $recipients is now available in the predata ACL (oversight).
2611 11. Tidy the search cache before the fork to do a delivery from a message
2612 received from the command line. Otherwise the child will trigger a lookup
2613 failure and thereby defer the delivery if it tries to use (for example) a
2614 cached ldap connection that the parent has called unbind on.
2616 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2617 of $address_data from the recipient verification was clobbered by the
2618 sender verification.
2620 13. The value of address_data from a sender verification is now available in
2621 $sender_address_data in subsequent conditions in the ACL statement.
2623 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2625 15. Added a new option "connect=<time>" to callout options, to set a different
2628 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2629 was its contents. (It was OK if the option was not defined at all.)
2631 17. A "Completed" log line is now written for messages that are removed from
2632 the spool by the -Mrm option.
2634 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2635 information about exactly what failed.
2637 19. Added -dd to debug only the daemon process.
2639 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2640 handles timeouts, both on the server side and network timeouts. Renamed the
2641 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2643 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2644 It is now set to "smtps".
2646 22. $host_address is now set to the target address during the checking of
2647 ignore_target_hosts.
2649 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2650 being passed; this would have caused $sender_host_name to have been used if
2651 matching the list had actually called for a host name (not very likely,
2652 since this list is usually IP addresses). A host name is now passed as
2655 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2656 code that specifies a non-restarting handler (typically sigaction() in
2657 modern systems) in an attempt to fix a rare and obscure crash bug.
2659 25. Narrowed the window for a race in the daemon that could cause it to ignore
2660 SIGCHLD signals. This is not a major problem, because they are used only to
2661 wake it up if nothing else does.
2663 26. A malformed maildirsize file could cause Exim to calculate negative values
2664 for the mailbox size or file count. Odd effects could occur as a result.
2665 The maildirsize information is now recalculated if the size or filecount
2668 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2669 support for a long time. Removed HAVE_SYS_VFS_H.
2671 28. Installed the latest version of exipick from John Jetmore.
2673 29. In an address list, if the pattern was not a regular expression, an empty
2674 subject address (from a bounce message) matched only if the pattern was an
2675 empty string. Non-empty patterns were not even tested. This was the wrong
2676 because it is perfectly reasonable to use an empty address as part of a
2677 database query. An empty address is now tested by patterns that are
2678 lookups. However, all the other forms of pattern expect the subject to
2679 contain a local part and a domain, and therefore, for them, an empty
2680 address still always fails if the pattern is not itself empty.
2682 30. Exim went into a mad DNS loop when attempting to do a callout where the
2683 host was specified on an smtp transport, and looking it up yielded more
2684 than one IP address.
2686 31. Re-factored the code for checking spool and log partition space into a
2687 function that finds that data and another that does the check. The former
2688 is then used to implement four new variables: $spool_space, $log_space,
2689 $spool_inodes, and $log_inodes.
2691 32. The RFC2047 encoding function was originally intended for short strings
2692 such as real names; it was not keeping to the 75-character limit for
2693 encoded words that the RFC imposes. It now respects the limit, and
2694 generates multiple encoded words if necessary. To be on the safe side, I
2695 have increased the buffer size for the ${rfc2047: expansion operator from
2698 33. It is now permitted to omit both strings after an "if" condition; if the
2699 condition is true, the result is "true". As before, when the second string
2700 is omitted, a false condition yields an empty string. This makes it less
2701 cumbersome to write custom ACL and router conditions.
2703 34. Failure to deliver a bounce message always caused it to be frozen, even if
2704 there was an errors_to setting on the router. The errors_to setting is now
2707 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2708 canonical form (fully expanded) before being placed in
2709 $sender_host_address.
2711 36. The table in the code that translates DNS record types into text (T_A to
2712 "A" for instance) was missing entries for NS and CNAME. It is just possible
2713 that this could have caused confusion if both these types were looked up
2714 for the same domain, because the text type is used as part of Exim's
2715 per-process caching. But the chance of anyone hitting this buglet seems
2718 37. The dnsdb lookup has been extended in a number of ways.
2720 (1) There is a new type, "zns", which walks up the domain tree until it
2721 finds some nameserver records. It should be used with care.
2723 (2) There is a new type, "mxh", which is like "mx" except that it returns
2724 just the host names, not the priorities.
2726 (3) It is now possible to give a list of domains (or IP addresses) to be
2727 looked up. The behaviour when one of the lookups defers can be
2728 controlled by a keyword.
2730 (4) It is now possible to specify the separator character for use when
2731 multiple records are returned.
2733 38. The dnslists ACL condition has been extended: it is now possible to supply
2734 a list of IP addresses and/or domains to be looked up in a particular DNS
2737 39. Added log_selector=+queue_time_overall.
2739 40. When running the queue in the test harness, wait just a tad after forking a
2740 delivery process, to get repeatability of debugging output.
2742 41. Include certificate and key file names in error message when GnuTLS fails
2743 to set them up, because the GnuTLS error message doesn't include the name
2744 of the failing file when there is a problem reading it.
2746 42. Allow both -bf and -bF in the same test run.
2748 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2750 44. The "Exiscan patch" is now merged into the mainline Exim source.
2752 45. Sometimes the final signoff response after QUIT could fail to get
2753 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2754 < 0 before doing a fflush(). This bug looks as though it goes back to the
2755 introduction of TLS in release 3.20, but "sometimes" must have been rare
2756 because the tests only now provoked it.
2758 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2759 (this can affect the format of dates).
2761 47. exim_tidydb, when checking for the continued existence of a message for
2762 which it has found a message-specific retry record, was not finding
2763 messages that were in split spool directories. Consequently, it was
2764 deleting retry records that should have stayed in existence.
2766 48. Steve fixed some bugs in eximstats.
2768 49. The SPA authentication driver was not abandoning authentication and moving
2769 on to the next authenticator when an expansion was forced to fail,
2770 contradicting the general specification for all authenticators. Instead it
2771 was generating a temporary error. It now behaves as specified.
2773 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2774 (the order specifies the preference for clients). The order is now AES256,
2775 AES128, 3DES, ARCFOUR128.
2777 51. Small patch to Sieve code - explicitly set From: when generating an
2780 52. Exim crashed if a remote delivery caused a very long error message to be
2781 recorded - for instance if somebody sent an entire SpamAssassin report back
2782 as a large number of 550 error lines. This bug was coincidentally fixed by
2783 increasing the size of one of Exim's internal buffers (big_buffer) that
2784 happened as part of the Exiscan merge. However, to be on the safe side, I
2785 have made the code more robust (and fixed the comments that describe what
2788 53. Now that there can be additional text after "Completed" in log lines (if
2789 the queue_time_overall log selector is set), a one-byte patch to exigrep
2790 was needed to allow it to recognize "Completed" as not the last thing in
2793 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2794 patch that reportedly fixes this has been added. I am not expert enough to
2795 create a test for it. This is what the patch creator wrote:
2797 "I found a little strange behaviour of ldap code when working with
2798 Windows 2003 AD Domain, where users was placed in more than one
2799 Organization Units. When I tried to give exim partial DN, the exit code
2800 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2801 But simultaneously result of request was absolutely normal ldap result,
2802 so I produce this patch..."
2804 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2805 REFERENCE, so I have modified the code to exclude the patch when that macro
2808 55. Some experimental protocols are using DNS PTR records for new purposes. The
2809 keys for these records are domain names, not reversed IP addresses. The
2810 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2811 leaves it alone. Component reversal etc. now happens only for IP addresses.
2814 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2816 57. Double the size of the debug message buffer (to 2048) so that more of very
2817 long debug lines gets shown.
2819 58. The exicyclog utility now does better if the number of log files to keep
2820 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2822 59. Two changes related to the smtp_active_hostname option:
2824 (1) $smtp_active_hostname is now available as a variable.
2825 (2) The default for smtp_banner uses $smtp_active_hostname instead
2826 of $primary_hostname.
2828 60. The host_aton() function is supposed to be passed a string that is known
2829 to be a valid IP address. However, in the case of IPv6 addresses, it was
2830 not checking this. This is a hostage to fortune. Exim now panics and dies
2831 if the condition is not met. A case was found where this could be provoked
2832 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2833 components; fortuitously, this particular loophole had already been fixed
2834 by change 4.50/55 above.
2836 If there are any other similar loopholes, the new check in host_aton()
2837 itself should stop them being exploited. The report I received stated that
2838 data on the command line could provoke the exploit when Exim was running as
2839 exim, but did not say which command line option was involved. All I could
2840 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2841 running as the user.
2844 61. There was a buffer overflow vulnerability in the SPA authentication code
2845 (which came originally from the Samba project). I have added a test to the
2846 spa_base64_to_bits() function which I hope fixes it.
2849 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2850 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2852 63. The daemon start-up calls getloadavg() while still root for those OS that
2853 need the first call to be done as root, but it missed one case: when
2854 deliver_queue_load_max is set with deliver_drop_privilege. This is
2855 necessary for the benefit of the queue runner, because there is no re-exec
2856 when deliver_drop_privilege is set.
2858 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2859 This has been fixed.
2861 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2862 was in use, was not putting the data itself into the right store pool;
2863 consequently, it could be overwritten for a subsequent message in the same
2864 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2867 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2869 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2870 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2871 the function were treating the return as a boolean value, which happened to
2872 work because 0=false and not-0=true, but is not correct code.
2874 68. The host_aton() function was not handling scoped IPv6 addresses (those
2875 with, for example, "%eth0" on the end) correctly.
2877 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2878 negated items (that is, ~something) in unsigned ints. Some compilers
2879 apparently mutter when there is no cast.
2881 70. If an address verification called from an ACL failed, and did not produce a
2882 user-specific message (i.e. there was only a "system" message), nothing was
2883 put in $acl_verify_message. In this situation, it now puts the system
2886 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2887 session; change 4.31/43 added the unwanted input to the log line - except
2888 that it did not do this in the start of session case. It now does.
2890 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2891 This is wrong and can cause the other end to generate a synchronization
2892 error if it is another Exim or anything else that does the synchronization
2893 check. A QUIT command is no longer sent after a timeout.
2895 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2896 during host lookups.
2898 74. The defer_ok option of callout verification was not working if it was used
2899 when verifying addresses in header lines, that is, for this case:
2901 verify = header_sender/callout=defer_ok
2903 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2904 those file descriptors could be used for SMTP connections. If anything
2905 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2906 could be sent to the SMTP client, causing chaos. The daemon now opens
2907 stdin, stdout, and stderr to /dev/null when it puts itself into the
2910 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2911 log by default. The user can override this with suitable Perl magic.
2913 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2914 the log message when discard triggers, was not working for the DATA ACL or
2915 for the non-SMTP ACL.
2917 78. Error message wording change in sieve.c.
2919 79. If smtp_accept_max_per_host was set, the number of connections could be
2920 restricted to fewer than expected, because the daemon was trying to set up
2921 a new connection before checking whether the processes handling previous
2922 connections had finished. The check for completed processes is now done
2923 earlier. On busy systems, this bug wouldn't be noticed because something
2924 else would have woken the daemon, and it would have reaped the completed
2927 80. If a message was submitted locally by a user whose login name contained one
2928 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2929 It caused a spool format error. I have fixed the spool reading code. A
2930 related problem was that the "from" clause in the Received: line became
2931 illegal because of the space(s). It is now covered by ${quote_local_part.
2933 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2936 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2938 83. Patch from Sieve maintainer to fix unterminated string problem in
2939 "vacation" handling.
2941 84. Some minor changes to the Linux configuration files to help with other
2942 OS variants using glibc.
2944 85. One more patch for Sieve to update vacation handling to latest spec.
2947 ----------------------------------------------------
2948 See the note above about the 4.44 and 4.50 releases.
2949 ----------------------------------------------------
2955 1. Change 4.43/35 introduced a bug that caused file counts to be
2956 incorrectly computed when quota_filecount was set in an appendfile
2959 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2960 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2963 3. Give more explanation in the error message when the command for a transport
2964 filter fails to execute.
2966 4. There are several places where Exim runs a non-Exim command in a
2967 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2968 was being done only for the command run by the queryprogram router. It is
2969 now done for all such subprocesses. The other cases are: ${run, transport
2970 filters, and the commands run by the lmtp and pipe transports.
2972 5. Some older OS have a limit of 256 on the maximum number of file
2973 descriptors. Exim was using setrlimit() to set 1000 as a large value
2974 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2975 systems. I've change it so that if it can't get 1000, it tries for 256.
2977 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2978 was an oversight, and furthermore, ever since the addition of extra
2979 controls (e.g. 4.43/32), the checks on when to allow different forms of
2980 "control" were broken. There should now be diagnostics for all cases when a
2981 control that does not make sense is encountered.
2983 7. $recipients is now available in the predata ACL (oversight).
2985 8. Tidy the search cache before the fork to do a delivery from a message
2986 received from the command line. Otherwise the child will trigger a lookup
2987 failure and thereby defer the delivery if it tries to use (for example) a
2988 cached ldap connection that the parent has called unbind on.
2990 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2991 of $address_data from the recipient verification was clobbered by the
2992 sender verification.
2994 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2995 was its contents. (It was OK if the option was not defined at all.)
2997 11. A "Completed" log line is now written for messages that are removed from
2998 the spool by the -Mrm option.
3000 12. $host_address is now set to the target address during the checking of
3001 ignore_target_hosts.
3003 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3004 being passed; this would have caused $sender_host_name to have been used if
3005 matching the list had actually called for a host name (not very likely,
3006 since this list is usually IP addresses). A host name is now passed as
3009 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3010 code that specifies a non-restarting handler (typically sigaction() in
3011 modern systems) in an attempt to fix a rare and obscure crash bug.
3013 15. Narrowed the window for a race in the daemon that could cause it to ignore
3014 SIGCHLD signals. This is not a major problem, because they are used only to
3015 wake it up if nothing else does.
3017 16. A malformed maildirsize file could cause Exim to calculate negative values
3018 for the mailbox size or file count. Odd effects could occur as a result.
3019 The maildirsize information is now recalculated if the size or filecount
3022 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3023 support for a long time. Removed HAVE_SYS_VFS_H.
3025 18. Updated exipick to current release from John Jetmore.
3027 19. Allow an empty sender to be matched against a lookup in an address list.
3028 Previously the only cases considered were a regular expression, or an
3031 20. Exim went into a mad DNS lookup loop when doing a callout where the
3032 host was specified on the transport, if the DNS lookup yielded more than
3035 21. The RFC2047 encoding function was originally intended for short strings
3036 such as real names; it was not keeping to the 75-character limit for
3037 encoded words that the RFC imposes. It now respects the limit, and
3038 generates multiple encoded words if necessary. To be on the safe side, I
3039 have increased the buffer size for the ${rfc2047: expansion operator from
3042 22. Failure to deliver a bounce message always caused it to be frozen, even if
3043 there was an errors_to setting on the router. The errors_to setting is now
3046 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3047 canonical form (fully expanded) before being placed in
3048 $sender_host_address.
3050 24. Updated eximstats to version 1.33
3052 25. Include certificate and key file names in error message when GnuTLS fails
3053 to set them up, because the GnuTLS error message doesn't include the name
3054 of the failing file when there is a problem reading it.
3056 26. Expand error message when OpenSSL has problems setting up cert/key files.
3059 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3060 (this can affect the format of dates).
3062 28. exim_tidydb, when checking for the continued existence of a message for
3063 which it has found a message-specific retry record, was not finding
3064 messages that were in split spool directories. Consequently, it was
3065 deleting retry records that should have stayed in existence.
3067 29. eximstats updated to version 1.35
3068 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3069 1.35 - bugfix such that pie charts by volume are generated correctly
3071 30. The SPA authentication driver was not abandoning authentication and moving
3072 on to the next authenticator when an expansion was forced to fail,
3073 contradicting the general specification for all authenticators. Instead it
3074 was generating a temporary error. It now behaves as specified.
3076 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3077 (the order specifies the preference for clients). The order is now AES256,
3078 AES128, 3DES, ARCFOUR128.
3080 31. Small patch to Sieve code - explicitly set From: when generating an
3083 32. Exim crashed if a remote delivery caused a very long error message to be
3084 recorded - for instance if somebody sent an entire SpamAssassin report back
3085 as a large number of 550 error lines. This bug was coincidentally fixed by
3086 increasing the size of one of Exim's internal buffers (big_buffer) that
3087 happened as part of the Exiscan merge. However, to be on the safe side, I
3088 have made the code more robust (and fixed the comments that describe what
3091 33. Some experimental protocols are using DNS PTR records for new purposes. The
3092 keys for these records are domain names, not reversed IP addresses. The
3093 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3094 leaves it alone. Component reversal etc. now happens only for IP addresses.
3097 34. The host_aton() function is supposed to be passed a string that is known
3098 to be a valid IP address. However, in the case of IPv6 addresses, it was
3099 not checking this. This is a hostage to fortune. Exim now panics and dies
3100 if the condition is not met. A case was found where this could be provoked
3101 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3102 components; fortuitously, this particular loophole had already been fixed
3103 by change 4.50/55 or 4.44/33 above.
3105 If there are any other similar loopholes, the new check in host_aton()
3106 itself should stop them being exploited. The report I received stated that
3107 data on the command line could provoke the exploit when Exim was running as
3108 exim, but did not say which command line option was involved. All I could
3109 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3110 running as the user.
3113 35. There was a buffer overflow vulnerability in the SPA authentication code
3114 (which came originally from the Samba project). I have added a test to the
3115 spa_base64_to_bits() function which I hope fixes it.
3118 36. The daemon start-up calls getloadavg() while still root for those OS that
3119 need the first call to be done as root, but it missed one case: when
3120 deliver_queue_load_max is set with deliver_drop_privilege. This is
3121 necessary for the benefit of the queue runner, because there is no re-exec
3122 when deliver_drop_privilege is set.
3124 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3125 was in use, was not putting the data itself into the right store pool;
3126 consequently, it could be overwritten for a subsequent message in the same
3127 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3130 38. Sometimes the final signoff response after QUIT could fail to get
3131 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3132 < 0 before doing a fflush(). This bug looks as though it goes back to the
3133 introduction of TLS in release 3.20, but "sometimes" must have been rare
3134 because the tests only now provoked it.
3140 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3141 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3142 Now it returns an integer. A number of calls were still expecting a T/F
3143 return. Fortuitously, in all cases, the tests worked in OK situations,
3144 which is the norm. However, things would have gone wrong on any write
3145 failures on the smtp file descriptor. This function is used when sending
3146 messages over SMTP and also when doing verify callouts.
3148 2. When Exim is called to do synchronous delivery of a locally submitted
3149 message (the -odf or -odi options), it no longer closes stderr before doing
3152 3. Implemented the mua_wrapper option.
3154 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3156 5. Implemented the functions header_remove(), header_testname(),
3157 header_add_at_position(), and receive_remove_recipient(), and exported them
3160 6. If an ACL "warn" statement specified the addition of headers, Exim already
3161 inserted X-ACL-Warn: at the start if there was no header name. However, it
3162 was not making this test for the second and subsequent header lines if
3163 there were newlines in the string. This meant that an invalid header could
3164 be inserted if Exim was badly configured.
3166 7. Allow an ACL "warn" statement to add header lines at the start or after all
3167 the Received: headers, as well as at the end.
3169 8. Added the rcpt_4xx retry error code.
3171 9. Added postmaster_mailfrom=xxx to callout verification option.
3173 10. Added mailfrom=xxxx to the callout verification option, for verify=
3176 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3177 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3179 12. Inserted some casts to stop certain compilers warning when using pointer
3180 differences as field lengths or precisions in printf-type calls (mostly
3181 affecting debugging statements).
3183 13. Added optional readline() support for -be (dynamically loaded).
3185 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3186 same clock tick as a message's arrival, so that its received time was the
3187 same as the "first fail" time on the retry record, and that message
3188 remained on the queue past the ultimate address timeout, every queue runner
3189 would try a delivery (because it was past the ultimate address timeout) but
3190 after another failure, the ultimate address timeout, which should have then
3191 bounced the address, did not kick in. This was a "< instead of <=" error;
3192 in most cases the first failure would have been in the next clock tick
3193 after the received time, and all would be well.
3195 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3196 being recognized when the domain list was tested by the match_domain
3197 condition in an expansion string.
3199 16. Added the ${str2b64: operator.
3201 17. Exim was always calling setrlimit() to set a large limit for the number of
3202 processes, without checking whether the existing limit was already
3203 adequate. (It did check for the limit on file descriptors.) Furthermore,
3204 errors from getrlimit() and setrlimit() were being ignored. Now they are
3205 logged to the main and panic logs, but Exim does carry on, to try to do its
3206 job under whatever limits there are.
3208 18. Imported PCRE 5.0.
3210 19. Trivial typo in log message " temporarily refused connection" (the leading
3213 20. If the log selector return_path_on_delivery was set and an address was
3214 redirected to /dev/null, the delivery process crashed because it assumed
3215 that a return path would always be set for a "successful" delivery. In this
3216 case, the whole delivery is bypassed as an optimization, and therefore no
3219 21. Internal re-arrangement: the function for sending a challenge and reading
3220 a response while authentication was assuming a zero-terminated challenge
3221 string. It's now changed to take a pointer and a length, to allow for
3222 binary data in such strings.
3224 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3226 23. Exim was not respecting finduser_retries when seeking the login of the
3227 uid under which it was called; it was always trying 10 times. (The default
3228 setting of finduser_retries is zero.) Also, it was sleeping after the final
3229 failure, which is pointless.
3231 24. Implemented tls_on_connect_ports.
3233 25. Implemented acl_smtp_predata.
3235 26. If the domain in control=submission is set empty, Exim assumes that the
3236 authenticated id is a complete email address when it generates From: or
3237 Sender: header lines.
3239 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3240 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3241 chown and chgrp in /bin and hostname in /usr/bin.
3243 28. Exim was keeping the "process log" file open after each use, just as it
3244 does for the main log. This opens the possibility of it remaining open for
3245 long periods when the USR1 signal hits a daemon. Occasional processlog
3246 errors were reported, that could have been caused by this. Anyway, it seems
3247 much more sensible not to leave this file open at all, so that is what now
3250 29. The long-running daemon process does not normally write to the log once it
3251 has entered its main loop, and it closes the log before doing so. This is
3252 so that log files can straightforwardly be renamed and moved. However,
3253 there are a couple of unusual error situations where the daemon does write
3254 log entries, and I had neglected to close the log afterwards.
3256 30. The text of an SMTP error response that was received during a remote
3257 delivery was being truncated at 512 bytes. This is too short for some of
3258 the long messages that one sometimes sees. I've increased the limit to
3261 31. It is now possible to make retry rules that apply only when a message has a
3262 specific sender, in particular, an empty sender.
3264 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3265 it possible to be selective about when SMTP synchronization is enforced.
3267 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3269 32. Implemented hosts_connection_nolog.
3271 33. Added an ACL for QUIT.
3273 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3276 35. Added mailbox_size and mailbox_filecount to appendfile.
3278 36. Added control = no_multiline_responses to ACLs.
3280 37. There was a bug in the logic of the code that waits for the clock to tick
3281 in the case where the clock went backwards by a substantial amount such
3282 that the microsecond fraction of "now" was more than the microsecond
3283 fraction of "then" (but the whole seconds number was less).
3285 38. Added support for the libradius Radius client library this is found on
3286 FreeBSD (previously only the radiusclient library was supported).
3292 1. When certain lookups returned multiple values in the form name=value, the
3293 quoting of the values was not always being done properly. Specifically:
3294 (a) If the value started with a double quote, but contained no whitespace,
3296 (b) If the value contained whitespace other than a space character (i.e.
3297 tabs or newlines or carriage returns) it was not quoted.
3298 This fix has been applied to the mysql and pgsql lookups by writing a
3299 separate quoting function and calling it from the lookup code. The fix
3300 should probably also be applied to nisplus, ibase and oracle lookups, but
3301 since I cannot test any of those, I have not disturbed their existing code.
3303 2. A hit in the callout cache for a specific address caused a log line with no
3304 reason for rejecting RCPT. Now it says "Previous (cached) callout
3305 verification failure".
3307 3. There was an off-by-one bug in the queryprogram router. An over-long
3308 return line was truncated at 256 instead of 255 characters, thereby
3309 overflowing its buffer with the terminating zero. As well as fixing this, I
3310 have increased the buffer size to 1024 (and made a note to document this).
3312 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3313 when Exim is waiting for an SMTP response from a remote server, Exim
3314 restarts its select() call on the socket, thereby resetting its timeout.
3315 This is not a problem when such interrupts are rare. Somebody set up a cron
3316 job to run exiwhat every 2 minutes, which is less than the normal select()
3317 timeout (5 or 10 minutes). This meant that the select() timeout never
3318 kicked in because it was always reset. I have fixed this by comparing the
3319 time when an interrupt arrives with the time at the start of the first call
3320 to select(). If more time than the timeout has elapsed, the interrupt is
3321 treated as a timeout.
3323 5. Some internal re-factoring in preparation for the addition of Sieve
3324 extensions (by MH). In particular, the "personal" test is moved to a
3325 separate function, and given an option for scanning Cc: and Bcc: (which is
3326 not set for Exim filters).
3328 6. When Exim created an email address using the login of the caller as the
3329 local part (e.g. when creating a From: or Sender: header line), it was not
3330 quoting the local part when it contained special characters such as @.
3332 7. Installed new OpenBSD configuration files.
3334 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3335 try to make them clearer.
3337 9. Callout options, other than the timeout value, were being ignored when
3338 verifying sender addresses in header lines. For example, when using
3340 verify = header_sender/callout=no_cache
3342 the cache was (incorrectly) being used.
3344 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3345 only the Cygwin environment.
3347 11. When return_path_on_delivery was set as a log selector, if different remote
3348 addresses in the same message used different return paths and parallel
3349 remote delivery occurred, the wrong values would sometimes be logged.
3350 (Whenever a remote delivery process finished, the return path value from
3351 the most recently started remote delivery process was logged.)
3353 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3354 lines when AUTH and/or TLS are in use. This is the "received protocol"
3355 field. Exim used to use "asmtp" for authenticated SMTP, without any
3356 indication (in the protocol name) for TLS use. Now it follows the RFC and
3357 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3358 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3359 names appear in log lines as well as in Received: header lines.
3361 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3362 extensions, and comparison tests, and to fix some bugs.
3364 14. Changes to the "personal" filter test:
3366 (1) The test was buggy in that it was just doing the equivalent of
3367 "contains" tests on header lines. For example, if a user's address was
3368 anne@some.where, the "personal" test would incorrectly be true for
3370 To: susanne@some.where
3372 This test is now done by extracting each address from the header in turn,
3373 and checking the entire address. Other tests that are part of "personal"
3374 are now done using regular expressions (for example, to check local parts
3375 of addresses in From: header lines).
3377 (2) The list of non-personal local parts in From: addresses has been
3378 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3379 taken from the Sieve specification recommendations.
3381 (3) If the message contains any header line starting with "List-" it is
3382 treated as non-personal.
3384 (4) The test for "circular" in the Subject: header line has been removed
3385 because it now seems ill-conceived.
3387 15. Minor typos in src/EDITME comments corrected.
3389 16. Installed latest exipick from John Jetmore.
3391 17. If headers_add on a router specified a text string that was too long for
3392 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3393 of string_sprintf() is now avoided.
3395 18. $message_body_size was not set (it was always zero) when running the DATA
3396 ACL and the local_scan() function.
3398 19. For the "mail" command in an Exim filter, no default was being set for
3399 the once_repeat time, causing a random time value to be used if "once" was
3400 specified. (If the value happened to be <= 0, no repeat happened.) The
3401 default is now 0s, meaning "never repeat". The "vacation" command was OK
3402 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3403 (I found it when inspecting the code).
3405 20. There is now an overall timeout for performing a callout verification. It
3406 defaults to 4 times the callout timeout, which applies to individual SMTP
3407 commands during the callout. The overall timeout applies when there is more
3408 than one host that can be tried. The timeout is checked before trying the
3409 next host. This prevents very long delays if there are a large number of
3410 hosts and all are timing out (e.g. when the network connections are timing
3411 out). The value of the overall timeout can be changed by specifying an
3412 additional sub-option for "callout", called "maxwait". For example:
3414 verify = sender/callout=5s,maxwait=20s
3416 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3417 to the end before writing, but this should make it even safer).
3419 22. Exim was forgetting that it had advertised PIPELINING for the second and
3420 subsequent messages on an SMTP connection. It was also not resetting its
3421 memory on STARTTLS and an internal HELO.
3423 23. When Exim logs an SMTP synchronization error within a session, it now
3424 records whether PIPELINING has been advertised or not.
3426 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3427 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3428 rather than long int.
3430 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3432 26. Added the never_mail option to autoreply.
3438 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3439 crash if the getsockname() call failed; this can happen if a connection is
3440 closed very soon after it is established. The problem was simply in the
3441 order in which certain operations were done, causing Exim to try to write
3442 to the SMTP stream before it had set up the file descriptor. The bug has
3443 been fixed by making things happen in the correct order.
3449 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3450 before the connection was closed, thus losing the rejection response.
3452 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3453 some early Solaris releases, but causes trouble in current releases where
3454 socklen_t is defined.
3456 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3459 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3462 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3463 that releasing the top of it at the end releases what was used for sub-
3464 expansions (unless the block got too big). However, discard this block if
3465 the first thing is a variable or header, so that we can use its block when
3466 it is dynamic (useful for very large $message_headers, for example).
3468 6. Lookups now cache *every* query, not just the most recent. A new, separate
3469 store pool is used for this. It can be recovered when all lookup caches are
3470 flushed. Lookups now release memory at the end of their result strings.
3471 This has involved some general refactoring of the lookup sources.
3473 7. Some code has been added to the store_xxx() functions to reduce the amount
3474 of flapping under certain conditions.
3476 8. log_incoming_interface used to affect only the <= reception log lines. Now
3477 it causes the local interface and port to be added to several more SMTP log
3478 lines, for example "SMTP connection from", and rejection lines.
3480 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3482 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3484 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3485 could be overwritten at the end of the current message (or the start of a
3486 new message if it was set in a HELO ACL). The value is now preserved for
3487 the duration of the SMTP connection.
3489 12. If a transport had a headers_rewrite setting, and a matching header line
3490 contained an unqualified address, that address was qualified, even if it
3491 did not match any rewriting rules. The underlying bug was that the values
3492 of the flags that permit the existence of unqualified sender and recipient
3493 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3494 non-local messages, and by -bnq for local messages) were not being
3495 preserved with the message after it was received.
3497 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3498 "next input=" as part of the text comprising the host identity instead of
3499 the correct text. The code was using the same buffer for two different
3500 strings. However, depending on which order the printing function evaluated
3501 its arguments, the bug did not always show up. Under Linux, for example, my
3502 test suite worked just fine.
3504 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3505 doesn't work with some older versions of Perl. It has been changed to "my",
3506 which in any case is probably the better facility to use.
3508 15. A really picky compiler found some instances of statements for creating
3509 error messages that either had too many or two few arguments for the format
3512 16. The size of the buffer for calls to the DNS resolver has been increased
3513 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3514 for addresses that have a lot of PTR records. This alleviates a problem; it
3515 does not fully solve it.
3517 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3518 buffer now truncates the list and logs the incident, which is the same
3519 action as happens when Exim is looking up a host name and its aliases.
3520 Previously in this situation something unpredictable would happen;
3521 sometimes it was "internal error: store_reset failed".
3523 18. If a server dropped the connection unexpectedly when an Exim client was
3524 using GnuTLS and trying to read a response, the client delivery process
3525 crashed while trying to generate an error log message.
3527 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3528 string, for example:
3530 warn message = H1: something\nH2: something
3532 the text was added as a single header line from Exim's point of view
3533 though it ended up OK in the delivered message. However, searching for the
3534 second and subsequent header lines using $h_h2: did not work. This has been
3535 fixed. Similarly, if a system filter added multiple headers in this way,
3536 the routers could not see them.
3538 20. Expanded the error message when iplsearch is called with an invalid key to
3539 suggest using net-iplsearch in a host list.
3541 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3542 ACLs are no longer actually imposed (and a message to that effect is
3545 22. If a "gecos" field in a passwd entry contained escaped characters, in
3546 particular, if it contained a \" sequence, Exim got it wrong when building
3547 a From: or a Sender: header from that name. A second bug also caused
3548 incorrect handling when an unquoted " was present following a character
3549 that needed quoting.
3551 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3552 was not being matched caselessly.
3554 24. Arranged for all hyphens in the exim.8 source to be escaped with
3557 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3558 properly. Recipient callout cache records were still being keyed to include
3559 the sender, even when use_sender was set false. This led to far more
3560 callouts that were necessary. The sender is no longer included in the key
3561 when use_sender is false.
3563 26. Added "control = submission" modifier to ACLs.
3565 27. Added the ${base62d: operator to decode base 62 numbers.
3567 28. dnsdb lookups can now access SRV records.
3569 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3570 the configuration file.
3572 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3573 (-v) mode. This makes the output for a verbose queue run more intelligible.
3575 31. Added a use_postmaster feature to recipient callouts.
3577 32. Added the $body_zerocount variable, containing the number of binary zero
3578 bytes in the message body.
3580 33. The time of last modification of the "new" subdirectory is now used as the
3581 "mailbox time last read" when there is a quota error for a maildir
3584 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3586 35. Added +ignore_unknown as a special item in host lists.
3588 36. Code for decoding IPv6 addresses in host lists is now included, even if
3589 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3590 address was recognized as an IP address, but was then not correctly decoded
3591 into binary, causing unexpected and incorrect effects when compared with
3598 1. Very minor rewording of debugging text in manualroute to say "list of
3599 hosts" instead of "hostlist".
3601 2. If verify=header_syntax was set, and a header line with an unqualified
3602 address (no domain) and a large number of spaces between the end of the
3603 name and the colon was received, the reception process suffered a buffer
3604 overflow, and (when I tested it) crashed. This was caused by some obsolete
3605 code that should have been removed. The fix is to remove it!
3607 3. When running in the test harness, delay a bit after writing a bounce
3608 message to get a bit more predictability in the log output.
3610 4. Added a call to search_tidyup() just before forking a reception process. In
3611 theory, someone could use a lookup in the expansion of smtp_accept_max_
3612 per_host which, without the tidyup, could leave open a database connection.
3614 5. Added the variables $recipient_data and $sender_data which get set from a
3615 lookup success in an ACL "recipients" or "senders" condition, or a router
3616 "senders" option, similar to $domain_data and $local_part_data.
3618 6. Moved the writing of debug_print from before to after the "senders" test
3621 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3622 problems for message scanning, either using a data ACL, or using
3623 local_scan() because the Received: header was not generated till after they
3624 were called (in order to set the time as the time of reception completion).
3625 I have revised the way this works. The header is now generated after the
3626 body is received, but before the ACL or local_scan() are called. After they
3627 are run, the timestamp in the header is updated.
3633 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3634 before starting a queue runner without re-exec. This happened only when
3635 deliver_drop_privilege was set or when the Exim user was set to root. The
3636 effect of the bug was that timeouts during subsequent deliveries caused
3637 crashes instead of being properly handled. The handler is now left at its
3638 default (and expected) setting.
3640 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3641 message, again when deliver_drop_privilege is set or Exim is run as root.
3642 The bug described in (1) was not present in this case, but the tidying up
3643 of the other signals was missing. I have made the two cases consistent.
3645 3. The ignore_target_hosts setting on a manualroute router was being ignored
3646 for hosts that were looked up using the /MX notation.
3648 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3651 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3652 operated on the sender address. After changing the $sender_address to <>
3653 for the sender address verify, Exim was re-instated it as the original
3654 (before rewriting) address, but remembering that it had rewritten it, so it
3655 wasn't rewriting it again. This bug also had the effect of breaking the
3656 sender address verification caching when the sender address was rewritten.
3658 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3659 This has been changed so that if the ip literal address matches
3660 ignore_target_hosts, the router declines.
3662 7. Added expansion conditions match_domain, match_address, and match_local_
3663 part (NOT match_host).
3665 8. The placeholder for the Received: header didn't have a length field set.
3667 9. Added code to Exim itself and to exim_lock to test for a specific race
3668 condition that could lead to file corruption when using MBX delivery. The
3669 issue is with the lockfile that is created in /tmp. If this file is removed
3670 after a process has opened it but before that process has acquired a lock,
3671 there is the potential for a second process to recreate the file and also
3672 acquire a lock. This could lead to two Exim processes writing to the file
3673 at the same time. The added code performs the same test as UW imapd; it
3674 checks after acquiring the lock that its file descriptor still refers to
3675 the same named file.
3677 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3678 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3681 11. Added the smtp_active_hostname option. If used, this will typically be made
3682 to depend on the incoming interface address. Because $interface_address is
3683 not set up until the daemon has forked a reception process, error responses
3684 that can happen earlier (such as "too many connections") no longer contain
3687 12. If an expansion in a condition on a "warn" statement fails because a lookup
3688 defers, the "warn" statement is abandoned, and the next ACL statement is
3689 processed. Previously this caused the whole ACL to be aborted.
3691 13. Added the iplsearch lookup type.
3693 14. Added ident_timeout as a log selector.
3695 15. Added tls_certificate_verified as a log selector.
3697 16. Added a global option tls_require_ciphers (compare the smtp transport
3698 option of the same name). This controls incoming TLS connections.
3700 17. I finally figured out how to make tls_require_ciphers do a similar thing
3701 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3702 before starting the TLS session.
3704 18. Tabs are now shown as \t in -bP output.
3706 19. If the log selector return_path_on_delivery was set, Exim crashed when
3707 bouncing a message because it had too many Received: header lines.
3709 20. If two routers both had headers_remove settings, and the first one included
3710 a superfluous trailing colon, the final name in the first list and the
3711 first name in the second list were incorrectly joined into one item (with a
3712 colon in the middle).
3718 1. Added -C and -D options to the exinext utility, mainly to make it easier
3719 to include in the automated testing, but these could be helpful when
3720 multiple configurations are in use.
3722 2. The exinext utility was not formatting the output nicely when there was
3723 an alternate port involved in the retry record key, nor when there was a
3724 message id as well (for retries that were specific to a specific message
3725 and a specific host). It was also confused by IPv6 addresses, because of
3726 the additional colons they contain. I have fixed the IPv4 problem, and
3727 patched it up to do a reasonable job for IPv6.
3729 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3730 delivery, the log line now contains "pipelined" if PIPELINING was used.
3732 4. An SMTP transport process used to panic and die if the bind() call to set
3733 an explicit outgoing interface failed. This has been changed; it is now
3734 treated in the same way as a connect() failure.
3736 5. A reference to $sender_host_name in the part of a conditional expansion
3737 that was being skipped was still causing a DNS lookup. This no longer
3740 6. The def: expansion condition was not recognizing references to header lines
3741 that used bh_ and bheader_.
3743 7. Added the _cache feature to named lists.
3745 8. The code for checking quota_filecount in the appendfile transport was
3746 allowing one more file than it should have been.
3748 9. For compatibility with Sendmail, the command line option
3756 and sets the incoming protocol and host name (for trusted callers). The
3757 host name and its colon can be omitted when only the protocol is to be set.
3758 Note the Exim already has two private options, -pd and -ps, that refer to
3759 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3760 "s", but I don't think that's a major issue.
3762 10. A number of refactoring changes to the code, none of which should affect
3765 (a) The number of logging options was getting close to filling up the
3766 32-bit word that was used as a bit map. I have split them into two classes:
3767 those that are passed in the argument to log_write(), and those that are
3768 only ever tested independently outside of that function. These are now in
3769 separate 32-bit words, so there is plenty of room for expansion again.
3770 There is no change in the user interface or the logging behaviour.
3772 (b) When building, for example, log lines, the code previously used a
3773 macro that called string_cat() twice, in order to add two strings. This is
3774 not really sufficiently general. Furthermore, there was one instance where
3775 it was actually wrong because one of the argument was used twice, and in
3776 one call a function was used. (As it happened, calling the function twice
3777 did not affect the overall behaviour.) The macro has been replaced by a
3778 function that can join an arbitrary number of extra strings onto a growing
3781 (c) The code for expansion conditions now uses a table and a binary chop
3782 instead of a serial search (which was left over from when there were very
3783 few conditions). Also, it now recognizes conditions like "pam" even when
3784 the relevant support is not compiled in: a suitably worded error message is
3785 given if an attempt is made to use such a condition.
3787 11. Added ${time_interval:xxxxx}.
3789 12. A bug was causing one of the ddress fields not to be passed back correctly
3790 from remote delivery subprocesses. The field in question was not being
3791 subsequently used, so this caused to problems in practice.
3793 13. Added new log selectors queue_time and deliver_time.
3795 14. Might have fixed a bug in maildirsizefile handling that threw up
3796 "unexpected character" debug warnings, and recalculated the data
3797 unnecessarily. In any case, I expanded the warning message to give more
3800 15. Added the message "Restricted characters in address" to the statements in
3801 the default ACL that block characters like @ and % in local parts.
3803 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3804 Three changes have been made:
3806 (a) There was a serious bug; a negative response to MAIL caused the whole
3807 recipient domain to be cached as invalid, thereby blocking all messages
3808 to all local parts at the same domain, from all senders. This bug has
3809 been fixed. The domain is no longer cached after a negative response to
3810 MAIL if the sender used is not empty.
3812 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3815 (c) A new callout option, "use_sender" has been added for people who want
3816 the modified behaviour.
3822 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3825 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3826 indeed breaks things for older releases.
3828 3. Added additional logging to the case where there is a problem reading data
3829 from a filter that is running in a subprocess using a pipe, in order to
3830 try to track down a specific problem.
3832 4. Testing facility fudge: when running in the test harness and attempting
3833 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3834 getting "No route to host". Convert this to a timeout.
3836 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3839 6. Some OS don't have socklen_t but use size_t instead. This affects the
3840 fifth argument of getsockopt() amongst other things. This is now
3841 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3842 can be set for individual OS. I have set it for SunOS5, OSF1, and
3843 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3844 some earlier ones do not.
3846 7. Change 4.30/15 was not doing the test caselessly.
3848 8. The standard form for an IPv6 address literal was being rejected by address
3849 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3850 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3851 this, as well as the form without the "IPv6" on the front (but only when
3852 address literals are enabled, of course).
3854 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3856 10. Exim crashed if a message with an empty sender address specified by -f
3857 encountered a router with an errors_to setting. This could be provoked only
3858 by a command such as
3862 where an empty string was supplied; "<>" did not hit this bug.
3864 11. Installed PCRE release 4.5.
3866 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3867 remained set. It is now erased.
3869 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3870 times from message ids (which are base 36 rather than the normal 62).
3872 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3873 were being counted as actual protocol errors, and logged if the log
3874 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3875 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3876 and DATA following a set of rejected RCPTs do not count as protocol errors.
3877 In other words, Exim assumes they were pipelined, though this may not
3878 actually be the case. Of course, in all cases the client gets an
3879 appropriate error code.
3881 15. If a lookup fails in an ACL condition, a message about the failure may
3882 be available; it is used if testing the ACL cannot continue, because most
3883 such messages specify what the cause of the deferral is. However, some
3884 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3885 that caused an old message to be retained and used if a later statement
3886 caused a defer, replacing the real cause of the deferral.
3888 16. If an IP address had so many PTR records that the DNS lookup buffer
3889 was not large enough to hold them, Exim could crash while trying to process
3890 the truncated data. It now detects and logs this case.
3892 17. Further to 4.21/58, another change has been made: if (and only if) the
3893 first line of a message (the first header line) ends with CRLF, a bare LF
3894 in a subsequent header line has a space inserted after it, so as not to
3895 terminate the header.
3897 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3898 unnecessarily, used atoi() instead of strtol(), and didn't check the
3899 termination when getting file sizes from file names by regex.
3901 19. Completely re-implemented the support for maildirsize files, in the light
3902 of a number of problems with the previous contributed implementation
3903 (4.30/29). In particular:
3905 . If the quota is zero, the maildirsize file is maintained, but no quota is
3908 . If the maildir directory does not exist, it is created before any attempt
3909 to write a maildirsize file.
3911 . The quota value in the file is just a cache; if the quota is changed in
3912 the transport, the new value overrides.
3914 . A regular expression is available for excluding directories from the
3917 20. The autoreply transport checks the characters in options that define the
3918 message's headers; it allows continued headers, but it was checking with
3919 isspace() after an embedded newline instead of explicitly looking for a
3922 21. If all the "regular" hosts to which an address was routed had passed their
3923 expiry times, and had not reached their retry times, the address was
3924 bounced, even if fallback hosts were defined. Now Exim should go on to try
3927 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3928 equivalent code in the SMTP transport. Some hosts send humungous responses
3929 to HELO/EHLO, more than 1024 it seems.
3931 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3932 gives compiler warnings in some environments. I've now done it "properly",
3935 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3936 (because of problems with the built-in one) was declared to return uschar *
3937 instead of char *, causing compiler failure.
3939 25. Fixed a file descriptor leak when processing alias/forward files.
3941 26. Fixed a minor format string issue in dbfn.c.
3943 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3945 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3946 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3949 29. When a sender address is verified, it is cached, to save repeating the test
3950 when there is more than one recipient in a message. However, when the
3951 verification involves a callout, it is possible for different callout
3952 options to be set for different recipients. It is too complicated to keep
3953 track of this in the cache, so now Exim always runs a verification when a
3954 callout is required, relying on the callout cache for the optimization.
3955 The overhead is duplication of the address routing, but this should not be
3958 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3959 to be verified with callout=postmaster, and the main callout worked but the
3960 postmaster check failed, the verification correctly failed. However, if a
3961 subsequent RCPT command asked for sender verification *without* the
3962 postmaster check, incorrect caching caused this verification also to fail,
3965 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3966 it was not caching the DNS options (qualify_single, search_parents) that
3967 were used when the lookup failed. A subsequent lookup with different
3968 options therefore always gave the same answer, though there were cases
3969 where it should not have. (Example: a "domains = !$mx_any" option on a
3970 dnslookup router: the "domains" option is always processed without any
3971 widening, but the router might have qualify_single set.) Now Exim uses the
3972 cached value only when the same options are set.
3974 32. Added John Jetmore's "exipick" utility to the distribution.
3976 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3977 than a timeout (e.g. a certificate is required, and is not provided), an
3978 Exim server now closes the connection immediately. Previously it waited for
3979 the client to close - but if the client is SSL, it seems that they each
3980 wait for each other, leading to a delay before one of them times out.
3982 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3983 maintained 0.8.x compatibility because I don't think many are using it, and
3984 it is clearly obsolete.
3986 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3989 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3990 client certificate was expired. A simple patch fixes this, though I don't
3991 understand the full logic of why the verify callback is called multiple
3994 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3995 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3996 which causes problems with some clients (such as the Certicom SSL Plus
3997 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3998 disables the coutermeasure allowing Eudora to connect."
4000 38. Exim was not checking that a write() to a log file succeeded. This could
4001 lead to Bad Things if a log got too big, in particular if it hit a file
4002 size limit. Exim now panics and dies if it cannot write to a log file, just
4003 as it does if it cannot open a log file.
4005 39. Modified OS/Makefile-Linux so that it now contains
4007 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4009 The two -D definitions ensure that Exim is compiled with large file
4010 support, which makes it possible to handle log files that are bigger than
4013 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4014 instance) a domain was checked against a named list that involved a lookup,
4015 causing $domain_data to be set, then another domain was checked against the
4016 same list, then the first domain was re-checked, the value of $domain_data
4017 after the final check could be wrong. In particular, if the second check
4018 failed, it could be set empty. This bug probably also applied to
4021 41. The strip_trailing_dot option was not being applied to the address given
4022 with the -f command-line option.
4024 42. The code for reading a message's header from the spool was incrementing
4025 $received_count, but never initializing it. This meant that the value was
4026 incorrect (doubled) while delivering a message in the same process in which
4027 it was received. In the most common configuration of Exim, this never
4028 happens - a fresh exec is done - but it can happen when
4029 deliver_drop_privilege is set.
4031 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4032 it now includes up to 150 characters of the unexpected data in the log
4035 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4036 and building data strings. The size of both of these buffers was 10 000
4037 bytes - far larger than anybody would *ever* want, thought I. Needless to
4038 say, somebody hit the limit. I have increased the maximum line length to
4039 20 000 and the maximum data length of concatenated lines to 100 000. I have
4040 also fixed two bugs, because there was no checking on these buffers. Tsk,
4041 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4042 buffer is too small.
4044 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4045 lsearch lookups. Now it does.
4047 46. When parsing a route_list item in a manualroute router, a fixed-length
4048 buffer was used for the list of hosts. I made this 1024 bytes long,
4049 thinking that nobody would ever have a list of hosts that long. Wrong.
4050 Somebody had a whole pile of complicated expansion conditions, and the
4051 string was silently truncated, leading to an expansion error. It turns out
4052 that it is easier to change to an unlimited length (owing to other changes
4053 that have happened since this code was originally written) than to build
4054 structure for giving a limitation error. The length of the item that
4055 expands into the list of hosts is now unlimited.
4057 47. The lsearch lookup could not handle data where the length of text line was
4058 more than 4095 characters. Such lines were truncated, leading to shortened
4059 data being returned. It should now handle lines of any length.
4061 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4062 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4065 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4066 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4067 no longer happens in comment lines. A list of replacements is now placed
4068 at the head of all of the source files, except those whose only change is
4069 to replace PERL_COMMAND in the very first #! line.
4071 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4072 messages on the queue, with a bottom-up merge sort, using code contributed
4073 by Michael Haardt. This should make operations like -bp somewhat faster on
4074 large queues. It won't affect queue runners, except when queue_run_in_order
4077 51. Installed eximstats 1.31 in the distribution.
4079 52. Added support for SRV lookups to the dnslookup router.
4081 53. If an ACL referred to $message_body or $message_body_end, the value was not
4082 reset for any messages that followed in the same SMTP session.
4084 54. The store-handling optimization for building very long strings was not
4085 differentiating between the different store pools. I don't think this
4086 actually made any difference in practice, but I've tidied it.
4088 55. While running the routers to verify a sender address, $sender_address
4089 was still set to the sender address. This is wrong, because when routing to
4090 send a bounce to the sender, it would be empty. Therefore, I have changed
4091 it so that, while verifying a sender address, $sender_address is set to <>.
4092 (There is no change to what happens when verifying a recipient address.)
4094 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4095 target A or AAAA records (if not already returned) without resetting the
4096 qualify_single or search_parents options of the DNS resolver. These are
4097 inappropriate in this case because the targets of MX and SRV records must
4098 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4099 target that, when qualified, matched something in the local domain. These
4100 two options are now turned off when doing these lookups.
4102 57. It seems that at least some releases of Reiserfs (which does not have the
4103 concept of a fixed number of inodes) returns zero and not -1 for the
4104 number of available inodes. This interacted badly with check_spool_inodes,
4105 which assumed that -1 was the "no such thing" setting. What I have done is
4106 to check that the total number of inodes is greater than zero before doing
4107 the test of how many are available.
4109 58. When a "warn" ACL statement has a log_message modifier, the message is
4110 remembered, and not repeated. This is to avoid a lot of repetition when a
4111 message has many recipients that cause the same warning to be written.
4112 Howewer, Exim was preserving the list of already written lines for an
4113 entire SMTP session, which doesn't seem right. The memory is now reset if a
4114 new message is started.
4116 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4117 debugging output unless log_rewrite was also set.
4119 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4120 of (int)(handle) when we know that handle contains (void *)(-1).
4122 61. The Exim daemon panic-logs an error return when it closes the incoming
4123 connection. However "connection reset by peer" seems to be common, and
4124 isn't really an error worthy of noting specially, so that particular error
4127 62. When Exim is trying to find all the local interfaces, it used to panic and
4128 die if the ioctl to get the interface flags failed. However, it seems that
4129 on at least one OS (Solaris 9) it is possible to have an interface that is
4130 included in the list of interfaces, but for which you get a failure error
4131 for this call. This happens when the interface is not "plumbed" into a
4132 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4133 failure of the "get flags" call assumes that the interface is down.
4135 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4136 makes life easier for people who are doing arithmetic on fields extracted
4137 from dates, where you often get leading zeros that should not be
4138 interpreted as octal.
4140 64. Added qualify_domain to the redirect router, to override the global
4143 65. If a pathologically long header line contained very many addresses (the
4144 report of this problem mentioned 10 000) and each of them was rewritten,
4145 Exim could use up a very large amount of memory. (It kept on making new
4146 copies of the header line as it rewrote, and never released the old ones.)
4147 At the expense of a bit more processing, the header rewriting function has
4148 been changed so that it no longer eats memory in this way.
4150 66. The generation of the Received: header has been moved from the time that a
4151 message starts to be received, to the time that it finishes. The timestamp
4152 in the Received: header should now be very close to that of the <= log
4153 line. There are two side-effects of this change:
4155 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4156 logged header lines no longer include the local Received: line, because
4157 it has not yet been created. The same applies to a copy of the message
4158 that is returned to a non-SMTP sender when a message is rejected.
4160 (b) When a filter file is tested using -bf, no additional Received: header
4161 is added to the test message. After some thought, I decided that this
4164 This change does not affect the value of $received_for. It is still set
4165 after address rewriting, but before local_scan() is called.
4167 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4169 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4170 gave an unhelpful panic error message, and a defer error. I have managed to
4171 change this behaviour so that it now rejects any supplied certificate,
4172 which seems right, as the list of acceptable certificates is empty.
4174 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4175 gave an unhelpful defer error. I have not managed to make this reject any
4176 supplied certificates, but the error message it gives is "no certificate
4177 supplied", which is not helpful.
4179 70. exigrep's output now also includes lines that are not associated with any
4180 message, but which match the given pattern. Implemented by a patch from
4181 Martin Sluka, which also tidied up the Perl a bit.
4183 71. Recipient callout verification, like sender verification, was using <> in
4184 the MAIL FROM command. This isn't really the right thing, since the actual
4185 sender may affect whether the remote host accepts the recipient or not. I
4186 have changed it to use the actual sender in the callout; this means that
4187 the cache record is now keyed on a recipient/sender pair, not just the
4188 recipient address. There doesn't seem to be a real danger of callout loops,
4189 since a callout by the remote host to check the sender would use <>.
4190 [SEE ABOVE: changed after hitting problems.]
4192 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4193 temporary errors. However, in the case of such a code being given after
4194 the end of a data transmission (i.e. after ".") Exim was failing to write
4195 a retry record for the message. (Yes, there was some broken host that was
4196 actually sending 8xx at this point.)
4198 73. An unknown lookup type in a host list could cause Exim to panic-die when
4199 the list was checked. (An example that provoked this was putting <; in the
4200 middle of a list instead of at the start.) If this happened during a DATA
4201 ACL check, a -D file could be left lying around. This kind of configuration
4202 error no longer causes Exim to die; instead it causes a defer errror. The
4203 incident is still logged to the main and panic logs.
4205 74. Buglet left over from Exim 3 conversion. The message "too many messages
4206 in one connection" was written to the rejectlog but not the mainlog, except
4207 when address rewriting (yes!) was being logged.
4209 75. Added write_rejectlog option.
4211 76. When a system filter was run not as root (that is, when system_filter_user
4212 was set), the values of the $n variables were not being returned to the
4213 main process; thus, they were not subsequently available in the $sn
4216 77. Added +return_path_on_delivery log selector.
4218 78. A connection timeout was being treated differently from recipients deferred
4219 when testing hosts_max_try with a message that was older than the host's
4220 retry timeout. (The host should not be counted, thus allowing all hosts to
4221 be tried at least once before bouncing.) This may have been the cause of an
4222 occasionally reported bug whereby a message would remain on the queue
4223 longer than the retry timeout, but would be bounced if a delivery was
4224 forced. I say "may" because I never totally pinned down the problem;
4225 setting up timeout/retry tests is difficult. See also the next item.
4227 79. The ultimate address timeout was not being applied to errors that involved
4228 a combination of host plus message (for example, a timeout on a MAIL
4229 command). When an address resolved to a number of possible hosts, and they
4230 were not all tried for each delivery (e.g. because of hosts_max_try), a
4231 message could remain on the queue longer than the retry timeout.
4233 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4236 81. Fixed an obscure SMTP outgoing bug which required at least the following
4237 conditions: (a) there was another message waiting for the same server;
4238 (b) the server returned 5xx to all RCPT commands in the first message so
4239 that the message was not completed; (c) the server dropped the connection
4240 or gave a negative response to the RSET that Exim sends to abort the
4241 transaction. The observed case was a dropped connection after DATA that had
4242 been sent in pipelining mode. That is, the server had advertised PIPELINING
4243 but was not implementing it correctly. The effect of the bug was incorrect
4244 behaviour, such as trying another host, and this could lead to a crash.
4250 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4251 and daemon.c were passed as pointers to ints; they should have been
4252 pointers to socklen_t variables (which are typically unsigned ints).
4254 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4257 3. Fixed a really odd bug that affected only the testing scheme; patching a
4258 certain fixed string in the binary changed the value of another string that
4259 happened to be identical to the end of the original first string.
4261 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4262 name", it returns that address as the IP address. On some operating
4263 systems (e.g. Solaris), it also passes back the IP address string as the
4264 "host name". However, on others (e.g. Linux), it passes back an empty
4265 string. Exim wasn't checking for this, and was changing the host name to an
4266 empty string, assuming it had been canonicized.
4268 5. Although rare, it is permitted to have more than one PTR record for a given
4269 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4270 all the names associated with an address, because they do in Solaris.
4271 However, it seems that they do not in Linux for data that comes from the
4272 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4273 I found this out when I moved to a new Linux workstation and tried to run
4274 the Exim test suite.
4276 To get round this problem I have changed the code so that it now does its
4277 own call to the DNS to look up PTR records when searching for a host name.
4278 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4279 addresses that are only in /etc/hosts are still found.
4281 This behaviour is, however, controlled by an option called host_lookup_
4282 order, which defaults to "bydns:byaddr". If people want to use the other
4283 order, or indeed, just use one or the other means of lookup, they can
4284 specify it in this variable.
4286 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4287 some operating systems, this comes back from gethostbyaddr() as an empty
4288 string, and this is what Exim used to test for. However, it seems that in
4289 other systems, "." is yielded. Exim now tests for this case too.
4291 7. The values of check_spool_space and check_log_space are now held internally
4292 as a number of kilobytes instead of an absolute number of bytes. If a
4293 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4294 kilobyte. This means that much larger values can be stored.
4296 8. Exim monitor: an attempt to get the action menu when not actually pointing
4297 at a message produces an empty menu entitled "No message selected". This
4298 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4299 no entries in it ("Shell widget menu has zero width and/or height"). So I
4300 have added a single, blank menu entry in this case.
4302 9. Added ${quote_local_part.
4304 10. MIME decoding is now applied to the contents of Subject: header lines when
4307 11. Now that a reference to $sender_host_address automatically causes a reverse
4308 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4309 host lookup before query-style lookups in lists that might use this
4310 variable. This has therefore been abolished, and the "net-" prefix is no
4311 longer necessary for query-style lookups.
4313 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4314 have been a typo for LFLAGS, so it has been changed.
4316 13. The install script calls Exim with "-C /dev/null" in order to find the
4317 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4318 to be output. Howeve, since Exim outputs its version number before the
4319 error, it didn't break the script. It just looked ugly. I fixed this by
4320 always allowing "-C /dev/null" if the caller is root.
4322 14. Ignore overlarge ACL variable number when reading spool file - insurance
4323 against a later release with more variables having written the file.
4325 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4326 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4327 this, as well as the form without the "IPv6" on the front.
4329 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4330 OS/Makefile-Darwin file.
4332 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4333 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4335 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4336 variable. However, some OS use ints for inodes. I've added cast to long int
4337 to get rid of the compiler warning.
4339 19. I had forgotten to lock out "/../" in configuration file names when
4340 ALT_CONFIG_PREFIX was set.
4342 20. Routers used for verification do not need to specify transports. However,
4343 if such a router generated a host list, and callout was configured, Exim
4344 crashed, because it could not find a port number from the (non-existent)
4345 transport. It now assumes port 25 in this circumstance.
4347 21. Added the -t option to exigrep.
4349 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4350 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4351 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4352 them is set, LOOKUP_LSEARCH is forced.
4354 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4356 24. Added sender and host information to the "rejected by local_scan()" log
4357 line; previously there was no indication of these.
4359 25. Added .include_if_exists.
4361 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4362 Linux. It turns out that not all file systems support this. Apparently some
4363 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4364 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4365 not supported on the file descriptor, is now ignored when Exim is trying to
4366 sync a directory. This applies only to Linux.
4368 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4370 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4371 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4374 29. Added support for maildirsize files from supplied patch (modified a bit).
4376 30. The use of :fail: followed by an empty string could lead Exim to respond to
4377 sender verification failures with (e.g.):
4379 550 Verification failed for <xxx>
4380 550 Sender verify failed
4382 where the first response line was missing the '-' that indicates it is not
4383 the final line of the response.
4385 31. The loop for finding the name of the user that called Exim had a hardwired
4386 limit of 10; it now uses the value of finduser_retries, which is used for
4387 all other user lookups.
4389 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4392 33. Exim was neglecting to zero errno before one call of strtol() when
4393 expanding a string and expecting an integer value. On some systems this
4394 resulted in spurious "integer overflow" errors. Also, it was casting the
4395 result into an int without checking.
4397 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4398 did not work. The code looks as if it has *never* worked, though it appears
4399 to have been documented since at least releast 1.62. I have made it work.
4401 35. The "timeout_DNS" error in retry rules, also documented since at least
4402 1.62, also never worked. As it isn't clear exactly what this means, and
4403 clearly it isn't a major issue, I have abolished the feature by treating it
4404 as "timeout", and writing a warning to the main and panic logs.
4406 36. The display of retry rules for -brt wasn't always showing the error code
4409 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4410 timeout_connect_A, timeout_connect_MX.
4412 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4413 to the empty sender.
4415 39. The daemon was not analysing the content of -oX till after it had closed
4416 stderr and disconnected from the controlling terminal. This meant that any
4417 syntax errors were only noted on the panic log, and the return code from
4418 the command was 0. By re-arranging the code a little, I've made the
4419 decoding happen first, so such errors now appear on stderr, and the return
4420 code is 1. However, the actual setting up of the sockets still happens in
4421 the disconnected process, so errors there are still only recorded on the
4424 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4425 connections (as happens on some IP stacks) was logged at start up time as
4426 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4427 it from "IPv6 and IPv4", which means that two separate sockets are being
4430 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4431 says whether AF_INET or AF_INET6 was passed as an argument.
4433 42. Exiwhat output was messed up when time zones were included in log
4436 43. Exiwhat now gives more information about the daemon's listening ports,
4437 and whether -tls-on-connect was used.
4439 44. The "port" option of the smtp transport is now expanded.
4441 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4442 silently ignored. Now an error message is written to the main and panic
4445 46. There's a new ACL modifier called "logwrite" which writes to a log file
4446 as soon as it is encountered.
4448 47. Added $local_user_uid and $local_user_gid at routing time.
4450 48. Exim crashed when trying to verify a sender address that was being
4453 49. Exim was recognizing only a space character after ".include". It now also
4454 recognizes a tab character.
4456 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4457 extracting the relevant information from the specification. The man page no
4458 longer contains scrambled data for the -d option, and I've added a section
4459 at the front about calling Exim under different names.
4461 51. Added "extra_headers" argument to the "mail" command in filter files.
4463 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4466 53. Installed eximstats 1.29.
4468 54. Added transport_filter_timeout as a generic transport option.
4470 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4471 Cc: header lines. This was required by RFC 822, but it not required by RFC
4474 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4475 incoming messages that did not have them. Now it does so only if the
4476 message originates locally, that is, if there is no associated remote host
4477 address. When Resent- header lines are present, this applies to the Resent-
4478 lines rather than the non-Resent- lines.
4480 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4481 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4483 58. Messages for configuration errors now include the name of the main
4484 configuration file - useful now that there may be more than one file in a
4485 list (.included file names were always shown).
4487 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4488 for those rare installations that do not start the daemon as root or run it
4489 setuid root. I've cut out the call to initgroups() if the daemon is not
4492 60. The Exim user and group can now be bound into the binary as text strings
4493 that are looked up at the start of Exim's processing.
4495 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4497 62. Added $mailstore_basename variable.
4499 63. Installed patch to sieve.c from Michael Haardt.
4501 64. When Exim failed to open the panic log after failing to open the main log,
4502 the original message it was trying to log was written to stderr and debug
4503 output, but if they were not available (the usual case in production), it
4504 was lost. Now it is written to syslog before the two lines that record the
4505 failures to open the logs.
4507 65. Users' Exim filters run in subprocesses under the user's uid. It is
4508 possible for a "deliver" command or an alias in a "personal" command to
4509 provoke an address rewrite. If logging of address rewriting is configured,
4510 this fails because the process is not running as root or exim. There may be
4511 a better way of dealing with this, but for the moment (because 4.30 needs
4512 to be released), I have disabled address rewrite logging when running a
4513 filter in a non-root, non-exim process.
4519 1. The buildconfig auxiliary program wasn't quoting the value set for
4520 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4521 not defined. This bug was masked in 4.22 by the effect that was fixed in
4524 2. Some messages that were rejected after a message id was allocated were
4525 shown as "incomplete" by exigrep. It no longer does this for messages that
4526 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4528 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4529 have allow_domain_literals set, the ID did not get logged in the <= line.
4530 Domain literals are now always recognized in Message-ID: header lines.
4532 4. The first argument for a ${extract expansion item is the key name or field
4533 number. Leading and trailing spaces in this item were not being ignored,
4534 causing some misleading effects.
4536 5. When deliver_drop_privilege was set, single queue runner processes started
4537 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4538 same command in the process it spins off) were not dropping privilege.
4540 6. When the daemon running as "exim" started a queue runner, it always
4541 re-executed Exim in the spun-off process. This is a waste of effort when
4542 deliver_drop_privilege is set. The new process now just calls the
4543 queue-runner function directly.
4549 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4552 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4553 ignored. Though the use of -oP was forcing the writing of a pid file, it
4554 was always written to the default place.
4556 3. If the message "no IP address found for host xxxx" is generated during
4557 incoming verification, it is now followed by identification of the incoming
4558 connection (so you can more easily find what provoked it).
4560 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4562 5. Added some features to "harden" Exim a bit more against certain attacks:
4564 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4565 be put in Local/Makefile. This is like the never_users runtime option,
4566 but it cannot be overridden. The default setting is "root".
4568 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4569 prefix string with which any file named in a -C command line option
4572 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4573 is retained for -C and -D only if the caller of Exim is root. Without
4574 it, the exim user may also use -C and -D and retain privilege.
4576 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4577 command line option is disabled.
4579 6. Macro names set by the -D option must start with an upper case letter, just
4580 like macro names defined in the configuration file.
4582 7. Added "dereference=" facility to LDAP.
4584 8. Two instances of the typo "uknown" in the source files are fixed.
4586 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4587 the Configure-Makefile script screwed up while processing it.
4589 10. Incorporated PCRE 4.4.
4591 11. The SMTP synchronization check was not operating right at the start of an
4592 SMTP session. For example, it could not catch a HELO sent before the client
4593 waited for the greeting. There is now a check for outstanding input at the
4594 point when the greeting is written. Because of the duplex, asynchronous
4595 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4596 way, but not yet received, when the check is performed.
4598 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4599 on TCP/IP sockets, because this apparently causes some broken clients to
4602 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4603 unchanged) from the Cygwin maintainer.
4605 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4606 was supported instead of testing for mbx. Effectively a typo.
4608 15. The spa authenticator server code was not checking that the input it
4609 received was valid base64.
4611 16. The debug output line for the "set" modifier in ACLs was not showing the
4612 name of the variable that was being set.
4614 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4616 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4617 lookup on its own. Something else was needed to trigger the lookup. For
4618 example, a match in host_lookup or the need for a host name in a host list.
4619 Now, if $sender_host_name is referenced and the host name has not yet been
4620 looked up, a lookup is performed. If the lookup fails, the variable remains
4621 empty, and $host_lookup_failed is set to "1".
4623 19. Added "eqi" as a case-independent comparison operator.
4625 20. The saslauthd authentication condition could segfault if neither service
4626 nor realm was specified.
4628 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4629 error message that was logged was misleading, and incoming SMTP
4630 connections were dropped. The message is now more accurate, and temporary
4631 errors are given to SMTP connections.
4633 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4634 (except RSET) with a 421 error, until QUIT is received. However, it was
4635 failing to send a response to QUIT.
4637 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4638 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4639 ACL is now run after the helo_try_verify_hosts code.
4641 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4642 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4643 case-independent, so other case variants are also recognized). Apparently
4644 some systems use these upper case variants.
4646 25. If more than two messages were waiting for the same host, and a transport
4647 filter was specified for the transport, Exim sent two messages over the
4648 same TCP/IP connection, and then failed with "socket operation on non-
4649 socket" when it tried to send the third.
4651 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4653 27. The extern definition of crypt16() in expand.c was not being excluded when
4654 the OS had its own crypt16() function.
4656 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4657 as a preferred synonym for return_size_limit, both as an option and as an
4660 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4662 30. Changed the default configuration ACL to relax the local part checking rule
4663 for addresses that are not in any local domains. For these addresses,
4664 slashes and pipe symbols are allowed within local parts, but the sequence
4665 /../ is explicitly forbidden.
4667 31. SPA server authentication was not clearing the challenge buffer before
4670 32. log_message in a "warn" ACL statement was writing to the reject log as
4671 well as to the main log, which contradicts the documentation and doesn't
4672 seem right (because no rejection is happening). So I have stopped it.
4674 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4675 However, I am unable to do any testing of this.
4677 34. Fixed an infelicity in the appendfile transport. When checking directories
4678 for a mailbox, to see if any needed to be created, it was accidentally
4679 using path names with one or more superfluous leading slashes; tracing
4680 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4682 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4683 contents are added to the log line that is written for every discarded
4684 recipient. (Previously a log_message setting was ignored.)
4686 36. The ${quote: operator now quotes the string if it is empty.
4688 37. The install script runs exim in order to find its version number. If for
4689 some reason other than non-existence or emptiness, which it checks, it
4690 could not run './exim', it was installing it with an empty version number,
4691 i.e. as "exim-". This error state is now caught, and the installation is
4694 38. An argument was missing from the function that creates an error message
4695 when Exim fails to connect to the socket for saslauthd authentication.
4696 This could cause Exim to crash, or give a corrupted message.
4698 39. Added isip, isip4, and isip6 to ${if conditions.
4700 40. The ACL variables $acl_xx are now saved with the message, and can be
4701 accessed later in routers, transports, and filters.
4703 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4704 strings in the file are not string-expanded.
4706 42. If a MAIL command specified a SIZE value that was too large to fit into an
4707 int variable, the check against message_size_limit failed. Such values are
4708 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4709 day this will have to be increased, but I don't think I want to be around
4710 when emails are that large.
4717 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4718 iconv() is not standard in FreeBSD.
4720 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4721 IPv6 enabled. The observed symptom was a segmentation fault on return from
4722 the function os_common_find_running_interfaces() in src/os.c.
4724 3. In the check_special_case() function in daemon.c I had used "errno" as an
4725 argument name, which causes warnings on some systems. This was basically a
4726 typo, since it was named "eno" in the comments!
4728 4. The code that waits for the clock to tick (at a resolution of some fraction
4729 of a second) so as to ensure message-id uniqueness was always waiting for
4730 at least one whole tick, when it could have waited for less. [This is
4731 almost certainly not relevant at current processor speeds, where it is
4732 unlikely to ever wait at all. But we try to future-proof.]
4734 5. The function that sleeps for a time interval that includes fractions of a
4735 second contained a race. It did not block SIGALRM between setting the
4736 timer, and suspending (a couple of lines later). If the interval was short
4737 and the sigsuspend() was delayed until after it had expired, the suspension
4738 never ended. On busy systems this could lead to processes getting stuck for
4741 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4742 process, before it forks any delivery processes. The open lookup caching
4743 mechanism meant that the open file or database connection was passed into
4744 the delivery process. The problem was that delivery processes always tidy
4745 up cached lookup data. This could cause a problem for the next delivery
4746 process started by the queue runner, because the external queue runner
4747 process does not know about the closure. So the next delivery process
4748 still has data in the lookup cache. In the case of a file lookup, there was
4749 no problem because closing a file descriptor in a subprocess doesn't affect
4750 the parent. However, if the lookup was caching a connection to a database,
4751 the connection was closed, and the second delivery process was likely to
4752 see errors such as "PGSQL: query failed: server closed the connection
4753 unexpectedly". The problem has been fixed by closing all cached lookups
4754 in a queue runner before running a delivery process.
4756 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4757 seem to have the "const" qualifier which it has on other OS. I've
4760 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4761 *of the same type* (client or server) with the same public name that an
4762 error should be diagnosed.
4764 9. When Exim looked up a host name for an IP address, but failed to find the
4765 original IP address when looking up the host name (a safety check), it
4766 output the message "<ip address> does not match any IP for NULL", which was
4767 confusing, to say the least. The bug was that the host name should have
4768 appeared instead of "NULL".
4770 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4771 user that is built into the binary, and the -C or -D options is used, root
4772 privilege is dropped before the configuration file is read. In addition,
4773 logging is switched to stderr instead of the normal log files. If the
4774 configuration then re-defines the Exim user, the unprivileged environment
4775 is probably not what is expected, so Exim logs a panic warning message (but
4778 However, if deliver_drop_privilege is set, the unprivileged state may well
4779 be exactly what is intended, so the warning has been cut out in that case,
4780 and Exim is allowed to try to write to its normal log files.
4786 1. smtp_return_error_details was not giving details for temporary sender
4787 or receiver verification errors.
4789 2. Diagnose a configuration error if two authenticators have the same public
4792 3. Exim used not to create the message log file for a message until the first
4793 delivery attempt. This could be confusing when incoming messages were held
4794 for policy or load reasons. The message log file is now created at the time
4795 the message is received, and an initial "Received" line is written to it.
4797 4. The automatically generated man page for command line options had a minor
4798 bug that caused no ill effects; however, a more serious problem was that
4799 the procedure for building the man page automatically didn't always
4800 operate. Consequently, release 4.20 contains an out-of-date version. This
4801 shouldn't happen again.
4803 5. When building Exim with embedded Perl support, the script that builds the
4804 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4805 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4807 6. The freeze_tell option was not being used for messages that were frozen on
4808 arrival, either by an ACL or by local_scan().
4810 7. Added the smtp_incomplete_transaction log selector.
4812 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4813 was accepting AUTH without a new EHLO.
4815 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4816 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4819 10. Logging of TCP/IP connections (when configured) now happens in the main
4820 daemon process instead of the child process, so that the TCP/IP connection
4821 count is more accurate (but it can never be perfect).
4823 11. The use of "drop" in a nested ACL was not being handled correctly in the
4824 outer ACL. Now, if condition failure induced by the nested "drop" causes
4825 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4826 or "require"), the connection is dropped.
4828 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4829 that yield "discard" can now be used with an "accept" or a "discard" verb,
4830 but an error is generated for any others (because I can't see a useful way
4831 to define what should happen).
4833 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4834 are now processed in the same way as lines in the Exim configuration file.
4835 In particular, continuation lines are supported.
4837 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4839 15. Added -ti meaning -t -i.
4841 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4842 domains, and warn by logging if others are found.
4844 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4845 returned from GIFCONF when Exim is trying to find the list of interfaces on
4846 a host. The code in os.c has been modified to copy each ifreq to an aligned
4847 structure in all cases.
4849 Also, in some cases, the returned ifreq's were being copied to a 'struct
4850 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4851 means the last couple of bytes of an IPv6 address could be chopped if the
4852 ifreq contained only a normal sockaddr (14 bytes storage).
4854 18. Named domain lists were not supported in the hosts_treat_as_local option.
4855 An entry such as +xxxx was not recognized, and was treated as a literal
4858 19. Ensure that header lines added by a DATA ACL are included in the reject log
4859 if the ACL subsequently rejects the message.
4861 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4862 MD5 (which is deprecated).
4864 21. When testing a filter file using -bf, Exim was writing a message when it
4865 took the sender from a "From " line in the message, but it was not doing so
4866 when it took $return_path from a Return-Path: header line. It now does.
4868 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4869 with a valid header line field name (a series of printing characters
4870 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4872 23. Changed "disc" in the source to "disk" to conform to the documentation and
4873 the book and for uniformity.
4875 24. Ignore Sendmail's -Ooption=value command line item.
4877 25. When execve() failed while trying to run a command in a pipe transport,
4878 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4879 could be confused with a return value of 69 from the command itself. This
4880 has been changed to 127, the value the shell returns if it is asked to run
4881 a non-existent command. The wording for the related log line suggests a
4882 non-existent command as the problem.
4884 26. If received_header_text expands to an empty string, do not add a Received:
4885 header line to the message. (Well, it adds a token one on the spool, but
4886 marks it "old" so that it doesn't get used or transmitted.)
4888 27. Installed eximstats 1.28 (addition of -nt option).
4890 28. There was no check for failure on the call to getsockname() in the daemon
4891 code. This can fail if there is a shortage of resources on the system, with
4892 ENOMEM, for example. A temporary error is now given on failure.
4894 29. Contrary to the C standard, it seems that in some environments, the
4895 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4896 program. Exim now does this explicitly; it affects the formatting of
4897 timestamps using strftime().
4899 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4900 complaints. I've now initialized all the variables, to avoid this.
4902 32. Header lines added by a system filter were not being "seen" during
4903 transport-time rewrites.
4905 33. The info_callback() function passed to OpenSSL is set up with type void
4906 (*)(SSL *, int, int), as described somewhere. However, when calling the
4907 function (actually a macro) that sets it up, the type void(*)() is
4908 expected. I've put in a cast to prevent warnings from picky compilers.
4910 34. If a DNS black list lookup found a CNAME record, but there were no A
4911 records associated with the domain it pointed at, Exim crashed.
4913 35. If a DNS black list lookup returned more than one A record, Exim ignored
4914 all but the first. It now scans all returned addresses if a particular IP
4915 value is being sought. In this situation, the contents of the
4916 $dnslist_value variable are a list of all the addresses, separated by a
4919 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4920 to accept a host name and all its aliases if the forward lookup for any of
4921 them yielded the IP address of the incoming connection. Now it accepts only
4922 those names whose forward lookup yields the correct IP address. Any other
4923 names are discarded. This closes a loophole whereby a rogue DNS
4924 administrator could create reverse DNS records to break through a
4925 wildcarded host restriction in an ACL.
4927 37. If a user filter or a system filter that ran in a subprocess used any of
4928 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4929 the wrong values were passed to the pipe command ($thisaddress had the
4930 value of $0, $0 had the value of $1, etc). This bug was introduced by
4931 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4933 38. Improved the line breaking for long SMTP error messages from ACLs.
4934 Previously, if there was no break point between 40 and 75 characters, Exim
4935 left the rest of the message alone. Two changes have been made: (a) I've
4936 reduced the minimum length to 35 characters; (b) if it can't find a break
4937 point between 35 and 75 characters, it looks ahead and uses the first one
4938 that it finds. This may give the occasional overlong line, but at least the
4939 remaining text gets split now.
4941 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4942 file descriptors might be low, and that setting 1000 would always raise it.
4943 It turns out that in some environments, the limit is already over 1000 and
4944 that lowering it causes trouble. So now Exim takes care not to decrease it.
4946 40. When delivering a message, the value of $return_path is set to $sender_
4947 address at the start of routing (routers may change the value). By an
4948 oversight, this default was not being set up when an address was tested by
4949 -bt or -bv, which affected the outcome if any router or filter referred to
4952 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4953 log only when "message" or "log_message" are set. However, if one of the
4954 conditions was an address verification, or a call to a nested ACL, the
4955 messages generated by the underlying test were being passed through. This
4956 no longer happens. The underlying message is available in $acl_verify_
4957 message for both "message" and "log_message" expansions, so it can be
4958 passed through if needed.
4960 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4961 new expansion $bh_ to give the encoded byte string without charset
4962 translation. Translation happens only if iconv() is available; HAVE_ICONV
4963 indicates this at build time. HEADERS_CHARSET gives the charset to
4964 translate to; headers_charset can change it in the configuration, and
4965 "headers charset" can change it in an individual filter file.
4967 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4968 that creates RFC 2047 encoded "words" labels them as that charset instead
4969 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4970 expansion operator; (ii) when Exim creates a From: line for a local
4971 message; (iii) when a header line is rewritten to include a "phrase" part.
4973 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4974 buggy, causing it to skip the first lines of messages whose message ID
4975 ended in 'D'. This would not have bitten before Exim release 4.14, because
4976 message IDs were unlikely to end in 'D' before then. The effect was to have
4977 incorrect size information for certain domains.
4979 45. #include "config.h" was missing at the start of the crypt16.c module. This
4980 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4983 46. If there was a timeout during a "random" callout check, Exim treated it as
4984 a failure of the random address, and carried on sending RSET and the real
4985 address. If the delay was just some slowness somewhere, the response to the
4986 original RCPT would be taken as a response to RSET and so on, causing
4987 mayhem of various kinds.
4989 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4990 when I implemented it. It didn't allow for the fact that some option values
4991 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4992 the right test for positive values.
4994 48. Domain names in DNS records are case-independent. Exim always looks them up
4995 in lower case. Some resolvers return domain names in exactly the case they
4996 appear in the zone file, that is, they may contain uppercase letters. Not
4997 all resolvers do this - some return always lower case. Exim was treating a
4998 change of case by a resolver as a change of domain, similar to a widening
4999 of a domain abbreviation. This triggered its re-routing code and so it was
5000 trying to route what was effectively the same domain again. This normally
5001 caused routing to fail (because the router wouldn't handle the domain
5002 twice). Now Exim checks for this case specially, and just changes the
5003 casing of the domain that it ultimately uses when it transmits the message
5006 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5009 50. If a filter generated a file delivery with a non-absolute name (possible if
5010 no home directory exists for the router), the forbid_file option was not
5013 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5014 the existing equality matching.
5016 52. Exim was using ints instead of ino_t variables in some places where it was
5017 dealing with inode numbers.
5019 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5020 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5021 called TMPDIR, and if it finds it is different, it changes its value.
5023 54. The smtp_printf() function is now made available to local_scan() so
5024 additional output lines can be written before returning. There is also an
5025 smtp_fflush() function to enable the detection of a dropped connection.
5026 The variables smtp_input and smtp_batched_input are exported to
5029 55. Changed the default runtime configuration: the message "Unknown user"
5030 has been removed from the ACL, and instead placed on the localuser router,
5031 using the cannot_route_message feature. This means that any verification
5032 failures that generate their own messages won't get overridden. Similarly,
5033 the "Unrouteable address" message that was in the ACL for unverifiable
5034 relay addresses has also been removed.
5036 56. Added hosts_avoid_esmtp to the smtp transport.
5038 57. The exicyclog script was not checking for the esoteric option
5039 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5040 will work only if exicyclog is run under the appropriate euid.
5042 58. Following a discussion on the list, the rules by which Exim recognises line
5043 endings on incoming messages have been changed. The -dropcr and drop_cr
5044 options are now no-ops, retained only for backwards compatibility. The
5045 following line terminators are recognized: LF CRLF CR. However, special
5046 processing applies to CR:
5048 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5049 nor a local message in the state where . is a terminator.
5051 (ii) If a bare CR is encountered in a header line, an extra space is added
5052 after the line terminator so as not to end the header. The reasoning
5053 behind this is that bare CRs in header lines are most likely either
5054 to be mistakes, or people trying to play silly games.
5056 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5057 was being incorrectly given as 18 bytes larger than it should have been.
5058 This is a VOB (very old bug).
5060 60. This may never have affected anything current, but just in case it has:
5061 When the local host is found other than at the start of a list of hosts,
5062 the local host, those with the same MX, and any that follow, are discarded.
5063 When the list in question was part of a longer list of hosts, the following
5064 hosts (not currently being processed) were also being discarded. This no
5065 longer happens. I'm not sure if this situation could ever has previously
5068 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5071 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5072 header. This is something that is recommended in a new Internet Draft, and
5073 is something that is documented as being done by Sendmail. There are two
5074 possible values. For messages generated by the autoreply transport, Exim
5077 Auto-Submitted: auto-replied
5079 whereas for all other generated messages (e.g. bounces) it adds
5081 Auto-Submitted: auto-generated
5083 63. The "personal" condition in filters now includes a test for the
5084 Auto-Submitted: header. If it contains the string "auto-" the message it
5085 not considered personal.
5087 64. Added rcpt_include_affixes as a generic transport option.
5089 65. Added queue_only_override (default true).
5091 66. Added the syslog_duplication option.
5093 67. If what should have been the first header line of a message consisted of
5094 a space followed by a colon, Exim was mis-interpreting it as a header line.
5095 It isn't of course - it is syntactically invalid and should therefore be
5096 treated as the start of the message body. The misbehaviour could have
5097 caused a number of strange effects, including loss of data in subsequent
5098 header lines, and spool format errors.
5100 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5101 client host had authenticated. This control can now be exercised by an ACL
5102 for more flexibility.
5104 69. By default, callouts do not happen when testing with -bh. There is now a
5105 variant, -bhc, which does actually run the callout code, including
5106 consulting and updating the callout cache.
5108 70. Added support for saslauthd authentication, courtesy of Alexander
5111 71. If statvfs() failed on the spool or log directories while checking their
5112 size for availability, Exim confusingly gave the error "space shortage".
5113 Furthermore, in debugging mode it crashed with a floating point exception.
5114 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5115 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5116 serious problem, Exim now writes to the main and panic logs when this
5117 happens, with details of the failure. It then refuses to accept the
5118 incoming message, giving the message "spool directory problem" or "log
5119 directory problem" with a 421 code for SMTP messages.
5121 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5122 0, 1, and 2 exist, because some OS complain for execs without them (see
5123 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5124 descriptors. However, the code omitted to check that the open succeeded,
5125 causing mysterious errors if for some reason the permissions on /dev/null
5126 got screwed. Now Exim writes a message to the main and panic logs, and
5127 bombs out if it can't open /dev/null.
5129 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5130 interact so that it is all more flexible. It is supposed to remain
5131 backwards compatible. Also added extra_local_interfaces.
5133 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5134 to bomb out with an assertion failure - to the client this appears as a
5135 connection drop. This problem occurs in the part of the code that was taken
5136 from the Samba project. Fortunately, the assertion is in a very simple
5137 function, so I have fixed this by reproducing the function inline in the
5138 one place where it is called, and arranging for authentication to fail
5139 instead of killing the process with assert().
5141 75. The SPA client code was not working when the server requested OEM rather
5142 than Unicode encoding.
5144 76. Added code to make require_files with a specific uid setting more usable in
5145 the case where statting the file as root fails - usually a non-root-mounted
5146 NFS file system. When this happens and the failure is EACCES, Exim now
5147 forks a subprocess and does the per-uid checking as the relevant uid.
5149 77. Added process_log_path.
5151 78. If log_file_path was not explicitly set, a setting of check_log_space or
5152 check_log_inodes was ignored.
5154 79. If a space check for the spool or log partitions fails, the incident is now
5155 logged. Of course, in the latter case the data may get lost...
5157 80. Added the %p formatting code to string_format() so that it can be used to
5158 print addresses in debug_print(). Adjusted all the address printing in the
5159 debugging in store.c to use %p rather than %d.
5161 81. There was a concern that a line of code in smtp_in.c could overflow a
5162 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5163 initially expressed, the concern was not well-founded, because trailing
5164 spaces are removed early. However, if the trailing spaces were followed by
5165 a NULL, they did not get removed, so the overflow was possible. Two fixes
5168 (a) I re-wrote the offending code in a cleaner fashion.
5169 (b) If an incoming SMTP command contains a NULL character, it is rejected
5172 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5173 runs initgroups(), so that if the Exim user is in any additional groups,
5174 they will be used during message reception.
5180 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob