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 The new ldap_require_cert option would segfault if used. Fixed.
11 PP/02 Harmonised TLS library version reporting; only show if debugging.
12 Layout now matches that introduced for other libraries in 4.74 PP/03.
14 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
20 NM/01 Workround for PCRE version dependency in version reporting
23 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
24 This fixes portability to compilers other than gcc, notably
25 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
27 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
28 makefiles for portability to HP-UX and POSIX correctness.
30 PP/01 Permit LOOKUP_foo enabling on the make command-line.
31 Also via indented variable definition in the Makefile.
32 (Debugging by Oliver Heesakkers).
34 PP/02 Restore caching of spamd results with expanded spamd_address.
35 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
37 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
38 Improves build reliability. Fix from: Frank Elsner
40 NM/02 Fix wide character breakage in the rfc2047 coding
41 Fixes bug 1064. Patch from Andrey N. Oktyabrski
43 NM/03 Allow underscore in dnslist lookups
44 Fixes bug 1026. Patch from Graeme Fowler
46 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
47 Code patches from Adam Ciarcinski of NetBSD.
49 NM/04 Fixed exiqgrep to cope with mailq missing size issue
52 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
53 is logged, to avoid truncation. Patch from John Horne.
55 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
56 Patch from Jakob Hirsch.
58 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
59 SQL string expansion failure details.
60 Patch from Andrey Oktyabrski.
62 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
63 Patch from Simon Arlott.
65 PP/09 New lookups functionality failed to compile on old gcc which rejects
66 extern declarations in function scope.
67 Patch from Oliver Fleischmann
69 PP/10 Use sig_atomic_t for flags set from signal handlers.
70 Check getgroups() return and improve debugging.
71 Fixed developed for diagnosis in bug 927 (which turned out to be
74 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
75 Patch from Mark Zealey.
77 PP/12 Bugzilla 1056: Improved spamd server selection.
78 Patch from Mark Zealey.
80 PP/13 Bugzilla 1086: Deal with maildir quota file races.
81 Based on patch from Heiko Schlittermann.
83 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
84 Patch from Uwe Doering, sign-off by Michael Haardt.
86 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
87 variable declaration deep within a block. Bug and patch from
90 PP/15 lookups-Makefile IRIX compatibilty coercion.
92 PP/16 Make DISABLE_DKIM build knob functional.
94 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
97 TF/03 Fix valgrind.h portability to C89 compilers that do not support
98 variable argument macros. Our copy now differs from upstream.
104 TF/01 Failure to get a lock on a hints database can have serious
105 consequences so log it to the panic log.
107 TF/02 Log LMTP confirmation messages in the same way as SMTP,
108 controlled using the smtp_confirmation log selector.
110 TF/03 Include the error message when we fail to unlink a spool file.
112 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
113 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
114 for maintaining out-of-tree patches for some time.
116 PP/01 Bugzilla 139: Documentation and portability issues.
117 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
118 Handle per-OS dynamic-module compilation flags.
120 PP/02 Let /dev/null have normal permissions.
121 The 4.73 fixes were a little too stringent and complained about the
122 permissions on /dev/null. Exempt it from some checks.
123 Reported by Andreas M. Kirchwitz.
125 PP/03 Report version information for many libraries, including
126 Exim version information for dynamically loaded libraries. Created
127 version.h, now support a version extension string for distributors
128 who patch heavily. Dynamic module ABI change.
130 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
131 privilege escalation vulnerability whereby the Exim run-time user
132 can cause root to append content of the attacker's choosing to
135 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
138 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
139 If dropping privileges for untrusted macros, we disabled normal logging
140 on the basis that it would fail; for the Exim run-time user, this is not
141 the case, and it resulted in successful deliveries going unlogged.
142 Fixed. Reported by Andreas Metzler.
148 PP/01 Date: & Message-Id: revert to normally being appended to a message,
149 only prepend for the Resent-* case. Fixes regression introduced in
150 Exim 4.70 by NM/22 for Bugzilla 607.
152 PP/02 Include check_rfc2047_length in configure.default because we're seeing
153 increasing numbers of administrators be bitten by this.
155 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
157 PP/03 Bugzilla 994: added openssl_options main configuration option.
159 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
161 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
163 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
165 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
166 without a peer certificate, leading to a segfault because of an
167 assumption that peers always have certificates. Be a little more
168 paranoid. Problem reported by Martin Tscholak.
170 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
171 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
172 NB: ClamAV planning to remove STREAM in "middle of 2010".
173 CL also introduces -bmalware, various -d+acl logging additions and
174 more caution in buffer sizes.
176 PP/09 Implemented reverse_ip expansion operator.
178 PP/10 Bugzilla 937: provide a "debug" ACL control.
180 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
182 PP/12 Bugzilla 973: Implement --version.
184 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
186 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
188 PP/15 Bugzilla 816: support multiple condition rules on Routers.
190 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
191 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
192 ignore trailing whitespace.
194 JJ/02 prevent non-panic DKIM error from being sent to paniclog
196 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
199 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
200 Notification from Dr Andrew Aitchison.
202 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
203 ExtendedDetectionInfo response format.
204 Notification from John Horne.
206 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
209 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
210 XSL and documented dependency on system catalogs, with examples of how
213 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
216 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
217 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
218 configuration file which is writeable by the Exim user or group.
220 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
221 of configuration files to cover files specified with the -C option if
222 they are going to be used with root privileges, not just the default
225 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
226 option (effectively making it always true).
228 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
229 files to be used while preserving root privileges.
231 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
232 that rogue child processes cannot use them.
234 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
235 run-time user, instead of root.
237 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
238 Exim run-time user without dropping privileges.
240 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
241 result string, instead of calling string_vformat() twice with the same
244 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
245 for other users. Others should always drop root privileges if they use
246 -C on the command line, even for a whitelisted configure file.
248 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
250 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
256 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
257 $data_path, and $header_path variables; fixed documentation bugs and
260 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
261 exipick to access non-standard spools, including the "frozen" queue
264 NM/01 Bugzilla 965: Support mysql stored procedures.
265 Patch from Alain Williams
267 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
269 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
270 Patch from Andreas Metzler
272 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
273 Patch from Kirill Miazine
275 NM/05 Bugzilla 671: Added umask to procmail example.
277 JJ/03 installed exipick 20100323.0, fixing doc bug
279 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
280 directory. Notification and patch from Dan Rosenberg.
282 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
284 TK/02 Improve log output when DKIM signing operation fails.
286 MH/01 Treat the transport option dkim_domain as a colon separated
287 list, not as a single string, and sign the message with each element,
288 omitting multiple occurences of the same signer.
290 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
291 Bugzilla 985, 986. Patch by Simon Arlott
293 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
294 Patch by Simon Arlott
296 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
297 MBX locking. Notification from Dan Rosenberg.
303 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
305 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
307 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
309 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
311 NM/04 Bugzilla 915: Flag broken perl installation during build.
317 TK/01 Added patch by Johannes Berg that expands the main option
318 "spamd_address" if it starts with a dollar sign.
320 TK/02 Write list of recipients to X-Envelope-Sender header when building
321 the mbox-format spool file for content scanning (suggested by Jakob
324 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
325 (http://www.dcc-servers.net/) support via dccifd. Activated by
326 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
328 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
329 by Mark Daniel Reidel <mr@df.eu>.
331 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
332 When building exim an external PCRE library is now needed -
333 PCRE is a system library on the majority of modern systems.
334 See entry on PCRE_LIBS in EDITME file.
336 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
337 conversation. Added nologin parameter to request.
338 Patch contributed by Kirill Miazine.
340 TF/01 Do not log submission mode rewrites if they do not change the address.
342 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
344 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
345 log files in place. Contributed by Roberto Lima.
347 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
349 TF/03 Bugzilla 615: When checking the local_parts router precondition
350 after a local_part_suffix or local_part_prefix option, Exim now
351 does not use the address's named list lookup cache, since this
352 contains cached lookups for the whole local part.
354 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
355 Robert Millan. Documentation is in experimental-spec.txt.
357 TF/04 Bugzilla 668: Fix parallel build (make -j).
359 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
361 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
362 Patch provided by Jan Srzednicki.
364 TF/05 Leading white space used to be stripped from $spam_report which
365 wrecked the formatting. Now it is preserved.
367 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
368 that they are available at delivery time.
370 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
372 TF/08 TLS error reporting now respects the incoming_interface and
373 incoming_port log selectors.
375 TF/09 Produce a more useful error message if an SMTP transport's hosts
376 setting expands to an empty string.
378 NM/06 Bugzilla 744: EXPN did not work under TLS.
379 Patch provided by Phil Pennock.
381 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
382 Patch provided by Richard Godbee.
384 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
385 acl_smtp_notquit, added index entry.
387 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
388 Patch provided by Eugene Bujak.
390 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
391 accept(). Patch provided by Maxim Dounin.
393 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
394 Patch provided by Phil Pennock.
396 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
398 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
399 Patch provided by Brad "anomie" Jorsch.
401 NM/14 Bugzilla 622: Added timeout setting to transport filter.
402 Patch provided by Dean Brooks.
404 TK/05 Add native DKIM support (does not depend on external libraries).
406 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
407 Patch provided by Graeme Fowler.
409 NM/16 Bugzilla 851: Documentation example syntax fix.
411 NM/17 Changed NOTICE file to remove references to embedded PCRE.
413 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
416 NM/19 Bugzilla 745: TLS version reporting.
417 Patch provided by Phil Pennock.
419 NM/20 Bugzilla 167: bool: condition support.
420 Patch provided by Phil Pennock.
422 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
423 clients. Patch provided by Phil Pennock.
425 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
426 Patch provided by Brad "anomie" Jorsch.
428 NM/23 Bugzilla 687: Fix misparses in eximstats.
429 Patch provided by Heiko Schlittermann.
431 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
432 Patch provided by Heiko Schlittermann.
434 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
435 plus update to original patch.
437 NM/26 Bugzilla 799: Documentation correction for ratelimit.
439 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
440 Patch provided by David Brownlee.
442 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
444 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
446 NM/30 Bugzilla 888: TLS documentation bugfixes.
448 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
450 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
451 Unlike the original bugzilla I have changed all shell scripts in src tree.
453 NM/33 Bugzilla 898: Transport filter timeout fix.
454 Patch by Todd Rinaldo.
456 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
457 Patch by Serge Demonchaux.
459 NM/35 Bugzilla 39: Base64 decode bug fixes.
460 Patch by Jakob Hirsch.
462 NM/36 Bugzilla 909: Correct connect() call in dcc code.
464 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
466 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
468 NM/39 Bugzilla 911: Fixed MakeLinks build script.
474 TK/01 Add preliminary DKIM support. Currently requires a forked version of
475 ALT-N's libdkim that I have put here:
476 http://duncanthrax.net/exim-experimental/
478 Note to Michael Haardt: I had to rename some vars in sieve.c. They
479 were called 'true' and it seems that C99 defines that as a reserved
480 keyword to be used with 'bool' variable types. That means you could
481 not include C99-style headers which use bools without triggering
482 build errors in sieve.c.
484 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
485 as mailq or other aliases. Changed the --help handling significantly
486 to do whats expected. exim_usage() emits usage/help information.
488 SC/01 Added the -bylocaldomain option to eximstats.
490 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
492 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
494 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
500 PH/01 Another patch from the Sieve maintainer.
502 PH/02 When an IPv6 address is converted to a string for single-key lookup
503 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
504 dots are used instead of colons so that keys in lsearch files need not
505 contain colons. This was done some time before quoting was made available
506 in lsearch files. However, iplsearch files do require colons in IPv6 keys
507 (notated using the quote facility) so as to distinguish them from IPv4
508 keys. This meant that lookups for IP addresses in host lists did not work
509 for iplsearch lookups.
511 This has been fixed by arranging for IPv6 addresses to be expressed with
512 colons if the lookup type is iplsearch. This is not incompatible, because
513 previously such lookups could never work.
515 The situation is now rather anomolous, since one *can* have colons in
516 ordinary lsearch keys. However, making the change in all cases is
517 incompatible and would probably break a number of configurations.
519 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
522 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
523 conversion specification without a maximum field width, thereby enabling
524 a rogue spamd server to cause a buffer overflow. While nobody in their
525 right mind would setup Exim to query an untrusted spamd server, an
526 attacker that gains access to a server running spamd could potentially
527 exploit this vulnerability to run arbitrary code as the Exim user.
529 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
530 $primary_hostname instead of what libspf2 thinks the hosts name is.
532 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
533 a directory entry by the name of the lookup key. Previously, if a
534 symlink pointed to a non-existing file or a file in a directory that
535 Exim lacked permissions to read, a lookup for a key matching that
536 symlink would fail. Now it is enough that a matching directory entry
537 exists, symlink or not. (Bugzilla 503.)
539 PH/03 The body_linecount and body_zerocount variables are now exported in the
542 PH/04 Added the $dnslist_matched variable.
544 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
545 This means they are set thereafter only if the connection becomes
548 PH/06 Added the client_condition to authenticators so that some can be skipped
549 by clients under certain conditions.
551 PH/07 The error message for a badly-placed control=no_multiline_responses left
552 "_responses" off the end of the name.
554 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
556 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
557 (without spaces) instead of just copying the configuration text.
559 PH/10 Added the /noupdate option to the ratelimit ACL condition.
561 PH/11 Added $max_received_linelength.
563 PH/12 Added +ignore_defer and +include_defer to host lists.
565 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
566 way in which PCRE > 7.0 is built.
568 PH/14 Implemented queue_only_load_latch.
570 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
571 MAIL command. The effect was to mangle the value on 64-bit systems.
573 PH/16 Another patch from the Sieve maintainer.
575 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
577 PH/18 If a system quota error occurred while trying to create the file for
578 a maildir delivery, the message "Mailbox is full" was not appended to the
579 bounce if the delivery eventually timed out. Change 4.67/27 below applied
580 only to a quota excession during the actual writing of the file.
582 PH/19 It seems that peer DN values may contain newlines (and other non-printing
583 characters?) which causes problems in log lines. The DN values are now
584 passed through string_printing() before being added to log lines.
586 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
587 and InterBase are left for another time.)
589 PH/21 Added message_body_newlines option.
591 PH/22 Guard against possible overflow in moan_check_errorcopy().
593 PH/23 POSIX allows open() to be a macro; guard against that.
595 PH/24 If the recipient of an error message contained an @ in the local part
596 (suitably quoted, of course), incorrect values were put in $domain and
597 $local_part during the evaluation of errors_copy.
603 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
604 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
607 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
608 issue a MAIL command.
610 PH/02 In an ACL statement such as
612 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
614 if a client was not listed at all, or was listed with a value other than
615 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
616 the condition was not true (as it should be), so access was not denied.
617 The bug was that the ! inversion was incorrectly passed on to the second
618 item. This has been fixed.
620 PH/03 Added additional dnslists conditions == and =& which are different from
621 = and & when the dns lookup returns more than one IP address.
623 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
624 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
626 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
627 FSYNC, which compiles an option called disable_fsync that allows for
628 bypassing fsync(). The documentation is heavily laced with warnings.
630 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
632 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
633 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
634 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
635 including adding "make clean"; (3) Added -fPIC when compiling the test
636 dynamically loaded module, to get rid of a warning.
638 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
639 message fails, move_frozen_messages = true and ignore_bounce_errors_after
640 = 0s. The bug is otherwise harmless.
642 PH/07 There was a bug in the dovecot authenticator such that the value of
643 $auth1 could be overwritten, and so not correctly preserved, after a
644 successful authentication. This usually meant that the value preserved by
645 the server_setid option was incorrect.
647 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
649 PH/09 Installed PCRE release 7.0.
651 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
652 run for batched SMTP input. It is now run at the start of every message
653 in the batch. While fixing this I discovered that the process information
654 (output by running exiwhat) was not always getting set for -bs and -bS
655 input. This is fixed, and it now also says "batched" for BSMTP.
657 PH/11 Added control=no_pipelining.
659 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
660 patch, slightly modified), and move the expansion of helo_data till after
661 the connection is made in the smtp transport (so it can use these
664 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
666 PH/14 Added log_selector = +pid.
668 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
670 PH/16 Add ${if forany and ${if forall.
672 PH/17 Added dsn_from option to vary the From: line in DSNs.
674 PH/18 Flush SMTP output before performing a callout, unless control =
675 no_callout_flush is set.
677 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
678 was true (the default) a successful delivery failed to delete the retry
679 item, thus causing premature timeout of the address. The bug is now
682 PH/20 Added hosts_avoid_pipelining to the smtp transport.
684 PH/21 Long custom messages for fakedefer and fakereject are now split up
685 into multiline reponses in the same way that messages for "deny" and
686 other ACL rejections are.
688 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
689 with slight modification.
691 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
692 draft, changing the syntax and factoring some duplicate code".
694 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
695 for deliveries of the second and subsequent messages over the same SMTP
698 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
699 ${reduce, with only minor "tidies".
701 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
703 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
704 expansion side effects.
706 PH/27 When a message times out after an over-quota error from an Exim-imposed
707 quota, the bounce message says "mailbox is full". This message was not
708 being given when it was a system quota that was exceeded. It now should
711 MH/03 Made $recipients available in local_scan(). local_scan() already has
712 better access to the recipient list through recipients_list[], but
713 $recipients can be useful in postmaster-provided expansion strings.
715 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
716 in the case of a MAIL command with additional options following the
717 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
718 were accidentally chopped off.
720 PH/29 SMTP synchronization checks are implemented when a command is read -
721 there is a check that no more input is waiting when there shouldn't be
722 any. However, for some commands, a delay in an ACL can mean that it is
723 some time before the response is written. In this time, more input might
724 arrive, invalidly. So now there are extra checks after an ACL has run for
725 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
726 pipelining has not been advertised.
728 PH/30 MH's patch to allow iscntrl() characters to be list separators.
730 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
731 returned in the SMTP response when smtp_return_error_details was false.
734 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
735 instead of the C I/O that was originally supplied, because problems were
738 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
739 Exim which did not show up earlier: it was assuming that a call to
740 SSL_CTX_set_info_callback() might give an error value. In fact, there is
741 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
742 was a macro that became an assignment, so it seemed to work. This has
743 changed to a proper function call with a void return, hence the compile
744 error. Exim's code has been fixed.
746 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
749 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
751 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
753 PH/37 If a message is not accepted after it has had an id assigned (e.g.
754 because it turns out to be too big or there is a timeout) there is no
755 "Completed" line in the log. When some messages of this type were
756 selected by exigrep, they were listed as "not completed". Others were
757 picked up by some special patterns. I have improved the selection
758 criteria to be more general.
760 PH/38 The host_find_failed option in the manualroute router can now be set
761 to "ignore", to completely ignore a host whose IP address cannot be
762 found. If all hosts are ignored, the behaviour is controlled by the new
763 host_all_ignored option.
765 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
766 homing or because of multiple MX records with /mx) generated more than
767 one IP address, and the following item turned out to be the local host,
768 all the secondary addresses of the first item were incorrectly removed
769 from the list, along with the local host and any following hosts (which
770 is what is supposed to happen).
772 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
773 whoever called Exim into the -H file. In the case of the daemon it was
774 behaving confusingly. When first started, it used values for whoever
775 started the daemon, but after a SIGHUP it used the Exim user (because it
776 calls itself on a restart). I have changed the code so that it now always
779 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
780 message are rejected with the same error (e.g. no authentication or bad
781 sender address), and a DATA command is nevertheless sent (as can happen
782 with PIPELINING or a stupid MUA), the error message that was given to the
783 RCPT commands is included in the rejection of the DATA command. This is
784 intended to be helpful for MUAs that show only the final error to their
787 PH/42 Another patch from the Sieve maintainer.
789 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
790 Eximstats - Fixed some broken HTML links and added missing column headers
792 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
793 columns for Rejects, Temp Rejects, Ham, and Spam rows.
795 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
797 PH/43 Yet another patch from the Sieve maintainer.
799 PH/44 I found a way to check for a TCP/IP connection going away before sending
800 the response to the final '.' that terminates a message, but only in the
801 case where the client has not sent further data following the '.'
802 (unfortunately, this is allowed). However, in many cases there won't be
803 any further data because there won't be any more messages to send. A call
804 to select() can be used: if it shows that the input is "ready", there is
805 either input waiting, or the socket has been closed. An attempt to read
806 the next input character can distinguish the two cases. Previously, Exim
807 would have sent an OK response which the client would never have see.
808 This could lead to message repetition. This fix should cure that, at
809 least in a lot of common cases.
811 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
812 advertised in response to EHLO.
818 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
819 fixed by 4.65/MH/01 (is this a record?) are fixed:
821 (i) An empty string was always treated as zero by the numeric comparison
822 operators. This behaviour has been restored.
824 (ii) It is documented that the numeric comparison operators always treat
825 their arguments as decimal numbers. This was broken in that numbers
826 starting with 0 were being interpreted as octal.
828 While fixing these problems I realized that there was another issue that
829 hadn't been noticed. Values of message_size_limit (both the global option
830 and the transport option) were treated as octal if they started with 0.
831 The documentation was vague. These values are now always treated as
832 decimal, and I will make that clear in the documentation.
838 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
839 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
842 MH/01 Don't check that the operands of numeric comparison operators are
843 integers when their expansion is in "skipping" mode (fixes bug
844 introduced by 4.64-PH/07).
846 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
847 child addresses, Exim now panics and dies. Previously, because the count
848 is held in a short int, deliveries were likely to be lost. As such a
849 large number of recipients for a single message is ridiculous
850 (performance will be very, very poor), I have chosen to impose a limit
851 rather than extend the field.
857 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
858 leftover -K file (the existence of which was triggered by #402).
859 While we were at it, introduced process PID as part of the -K
860 filename. This should rule out race conditions when creating
863 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
864 processing considerably. Previous code took too long for large mails,
865 triggering a timeout which in turn triggers #401.
867 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
868 in the DK code in transports.c. sendfile() is not really portable,
869 hence the _LINUX specificness.
871 TF/01 In the add_headers option to the mail command in an Exim filter,
872 there was a bug that Exim would claim a syntax error in any
873 header after the first one which had an odd number of characters
876 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
877 callout verification, Exim cached a "reject" for the entire domain. This
878 is correct for most verifications, but it is not correct for a recipient
879 verification with use_sender or use_postmaster set, because in that case
880 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
881 case of MAIL FROM:<> rejection from other early rejections (e.g.
882 rejection of HELO). When verifying a recipient using a non-null MAIL
883 address, the cache is ignored if it shows MAIL FROM:<> rejection.
884 Whatever the result of the callout, the value of the domain cache is
885 left unchanged (for any other kind of callout, getting as far as trying
886 RCPT means that the domain itself is ok).
888 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
891 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
892 manifest itself as EPIPE rather than ECONNECT. When tidying away a
893 session, the daemon ignores ECONNECT errors and logs others; it now
894 ignores EPIPE as well.
896 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
897 (quoted-printable decoding).
899 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
900 later the small subsequent patch to fix an introduced bug.
902 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
904 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
906 PH/08 An error is now given if message_size_limit is specified negative.
908 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
909 to be given (somewhat) arbitrary names.
911 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
914 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
915 miscellaneous code fixes
917 PH/10 Added the log_reject_target ACL modifier to specify where to log
920 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
921 hostname. This is wrong, because it relates to the incoming message (and
922 probably the interface on which it is arriving) and not to the outgoing
923 callout (which could be using a different interface). This has been
924 changed to use the value of the helo_data option from the smtp transport
925 instead - this is what is used when a message is actually being sent. If
926 there is no remote transport (possible with a router that sets up host
927 addresses), $smtp_active_hostname is used.
929 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
930 tweaks were necessary in order to get it to work (see also 21 below):
931 (a) The code assumed that strncpy() returns a negative number on buffer
932 overflow, which isn't the case. Replaced with Exim's string_format()
934 (b) There were several signed/unsigned issues. I just did the minimum
935 hacking in of casts. There is scope for a larger refactoring.
936 (c) The code used strcasecmp() which is not a standard C function.
937 Replaced with Exim's strcmpic() function.
938 (d) The code set only $1; it now sets $auth1 as well.
939 (e) A simple test gave the error "authentication client didn't specify
940 service in request". It would seem that Dovecot has changed its
941 interface. Fortunately there's a specification; I followed it and
942 changed what the client sends and it appears to be working now.
944 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
947 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
948 address A is aliased to B and C, where B exists and C does not. Without
949 -v the output is "A verified" because verification stops after a
950 successful redirection if more than one address is generated. However,
951 with -v the child addresses are also verified. Exim was outputting "A
952 failed to verify" and then showing the successful verification for C,
953 with its parentage. It now outputs "B failed to verify", showing B's
954 parentage before showing the successful verification of C.
956 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
957 look up a TXT record in a specific list after matching in a combined
960 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
961 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
962 they consult the DNS. I had assumed they would set it the way they
963 wanted; and indeed my experiments on Linux seem to show that in some
964 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
965 To be on the safe side, however, I have now made the interface to
966 host_find_byname() similar to host_find_bydns(), with an argument
967 containing the DNS resolver options. The host_find_byname() function now
968 sets these options at its start, just as host_find_bydns() does. The smtp
969 transport options dns_qualify_single and dns_search_parents are passed to
970 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
971 of host_find_byname() use the default settings of RES_DEFNAMES
972 (qualify_single) but not RES_DNSRCH (search_parents).
974 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
975 spool_read_header() do less string testing, by means of a preliminary
976 switch on the second character of optional "-foo" lines. (This is
977 overdue, caused by the large number of possibilities that now exist.
978 Originally there were few.) While I was there, I also converted the
979 str(n)cmp tests so they don't re-test the leading "-" and the first
980 character, in the hope this might squeeze out yet more improvement.
982 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
983 flag allowing group syntax was set by the header_syntax check but not
984 turned off, possible causing trouble later; (2) The flag was not being
985 set at all for the header_verify test, causing "group"-style headers to
986 be rejected. I have now set it in this case, and also caused header_
987 verify to ignore an empty address taken from a group. While doing this, I
988 came across some other cases where the code for allowing group syntax
989 while scanning a header line wasn't quite right (mostly, not resetting
990 the flag correctly in the right place). These bugs could have caused
991 trouble for malformed header lines. I hope it is now all correct.
993 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
994 with the "reply" argument non-NULL. The code, however (which originally
995 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
996 but it didn't always do it. This confused somebody who was copying the
997 code for some other use. I have removed all the tests.
999 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1000 feature that was used to support insecure browsers during the U.S. crypto
1001 embargo. It requires special client support, and Exim is probably the
1002 only MTA that supported it -- and would never use it because real RSA is
1003 always available. This code has been removed, because it had the bad
1004 effect of slowing Exim down by computing (never used) parameters for the
1005 RSA_EXPORT functionality.
1007 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1008 authenticator to fail if there's a tab character in the incoming data
1009 (there should never be unless someone is messing about, as it's supposed
1010 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1011 if the connection is using TLS or if the remote IP is the same as the
1012 local IP, and the "valid-client-cert option" if a client certificate has
1015 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1016 authenticators. This can be used for authorization after authentication
1017 succeeds. (In the case of plaintext, it servers for both authentication
1020 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1021 if any retry times were supplied.
1023 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1024 connection, where there is no client IP address to check. In this
1025 situation, the verify now always succeeds.
1027 PH/25 Applied John Jetmore's -Mset patch.
1029 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1031 PH/27 In a string expansion for a processed (not raw) header when multiple
1032 headers of the same name were present, leading whitespace was being
1033 removed from all of them, but trailing whitespace was being removed only
1034 from the last one. Now trailing whitespace is removed from each header
1035 before concatenation. Completely empty headers in a concatenation (as
1036 before) are ignored.
1038 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1039 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1041 PH/29 [Removed. This was a change that I later backed out, and forgot to
1042 correct the ChangeLog entry (that I had efficiently created) before
1043 committing the later change.]
1045 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1046 address errors (4xx response to RCPT) over the same connection as other
1047 messages routed to the same hosts. Such deliveries are always "forced",
1048 so retry times are not inspected. This resulted in far too many retries
1049 for the affected addresses. The effect occurred only when there were more
1050 hosts than the hosts_max_try setting in the smtp transport when it had
1051 the 4xx errors. Those hosts that it had tried were not added to the list
1052 of hosts for which the message was waiting, so if all were tried, there
1053 was no problem. Two fixes have been applied:
1055 (i) If there are any address or message errors in an SMTP delivery, none
1056 of the hosts (tried or untried) are now added to the list of hosts
1057 for which the message is waiting, so the message should not be a
1058 candidate for sending over the same connection that was used for a
1059 successful delivery of some other message. This seems entirely
1060 reasonable: after all the message is NOT "waiting for some host".
1061 This is so "obvious" that I'm not sure why it wasn't done
1062 previously. Hope I haven't missed anything, but it can't do any
1063 harm, as the worst effect is to miss an optimization.
1065 (ii) If, despite (i), such a delivery is accidentally attempted, the
1066 routing retry time is respected, so at least it doesn't keep
1067 hammering the server.
1069 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1070 in ${readsocket because some servers need this prod.
1072 PH/32 Added some extra debug output when updating a wait-xxx database.
1074 PH/33 The hint "could be header name not terminated by colon", which has been
1075 given for certain expansion errors for a long time, was not being given
1076 for the ${if def:h_colon_omitted{... case.
1078 PH/34 The spec says: "With one important exception, whenever a domain list is
1079 being scanned, $domain contains the subject domain." There was at least
1080 one case where this was not true.
1082 PH/35 The error "getsockname() failed: connection reset by peer" was being
1083 written to the panic log as well as the main log, but it isn't really
1084 panic-worthy as it just means the connection died rather early on. I have
1085 removed the panic log writing for the ECONNRESET error when getsockname()
1088 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1089 runs only) independently of the message's sender address. This meant
1090 that, if the 4xx error was in fact related to the sender, a different
1091 message to the same recipient with a different sender could confuse
1092 things. In particualar, this can happen when sending to a greylisting
1093 server, but other circumstances could also provoke similar problems.
1094 I have changed the default so that the retry time for these errors is now
1095 based a combination of the sender and recipient addresses. This change
1096 can be overridden by setting address_retry_include_sender=false in the
1099 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1100 remote server are returned as part of bounce messages. This was not
1101 happening for LMTP over a pipe (the lmtp transport), but now it is the
1102 same for both kinds of LMTP.
1104 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1105 in header lines that were in fact CNAMEs. This is no longer the case.
1107 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1108 and queue runs started by the daemon processed all messages. This has
1109 been fixed so that -R and -S can now usefully be given with -q<time>.
1111 PH/40 Import PCRE release 6.7 (fixes some bugs).
1113 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1115 PH/42 Give an error if -q is specified more than once.
1117 PH/43 Renamed the variables $interface_address and $interface_port as
1118 $received_ip_address and $received_port, to make it clear that these
1119 values apply to message reception, and not to the outgoing interface when
1120 a message is delivered. (The old names remain recognized, of course.)
1122 PH/44 There was no timeout on the connect() call when using a Unix domain
1123 socket in the ${readsocket expansion. There now is.
1125 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1126 be meaningful with "accept".
1128 SC/01 Eximstats V1.43
1129 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1131 SC/02 Eximstats V1.44
1132 Use a glob alias rather than an array ref in the generated
1133 parser. This improves both readability and performance.
1135 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1136 Collect SpamAssassin and rejection statistics.
1137 Don't display local sender or destination tables unless
1138 there is data to show.
1139 Added average volumes into the top table text output.
1141 SC/04 Eximstats V1.46
1142 Collect data on the number of addresses (recipients)
1143 as well as the number of messages.
1145 SC/05 Eximstats V1.47
1146 Added 'Message too big' to the list of mail rejection
1147 reasons (thanks to Marco Gaiarin).
1149 SC/06 Eximstats V1.48
1150 Mainlog lines which have GMT offsets and are too short to
1151 have a flag are now skipped.
1153 SC/07 Eximstats V1.49 (Alain Williams)
1154 Added the -emptyok flag.
1156 SC/08 Eximstats V1.50
1157 Fixes for obtaining the IP address from reject messages.
1159 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1160 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1161 whitesspace changes from 4.64-PH/27
1163 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1166 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1167 are found, allow negative numbers in numeric criteria)
1169 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1171 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1172 to match changes made in 4.64-PH/43
1174 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1176 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1177 "domain" is actually an IP address, and force a failure. This locks out
1178 those revolvers/nameservers that support "A-for-A" lookups, in
1179 contravention of the specifications.
1181 PH/48 When a host name was looked up from an IP address, and the subsequent
1182 forward lookup of the name timed out, the host name was left in
1183 $sender_host_name, contrary to the specification.
1185 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1186 restricted to single-key lookups, Exim was not diagnosing an error if
1187 * or *@ was used with a query-style lookup.
1189 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1191 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1192 long ago, but noone interested enough thought of it. Let's just say that
1193 the "1.1" means that there are some new functions that weren't there at
1194 some point in the past.
1196 PH/51 Error processing for expansion failure of helo_data from an smtp
1197 transport during callout processing was broken.
1199 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1200 tested/used via the -bh/-bhc/-bs options.
1202 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1203 bug, fixed in subsequent PCRE releases).
1205 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1206 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1208 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1214 SC/01 Use a glob alias rather than an array ref in eximstats generated
1215 parser. This improves both readability and performance.
1217 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1218 Don't display local sender or destination tables in eximstats unless
1219 there is data to show.
1220 Added average volumes into the eximstats top table text output.
1222 SC/03 Collect data on the number of addresses (recipients) as well
1223 as the number of messages in eximstats.
1225 TF/01 Correct an error in the documentation for the redirect router. Exim
1226 does (usually) call initgroups() when daemonizing.
1228 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1229 with consistent privilege compared to when running as a daemon.
1231 TF/03 Note in the spec that $authenticated_id is not set for local
1232 submissions from trusted users.
1234 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1235 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1237 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1238 by adding some example configuration directives to the default
1239 configuration file. A little bit of work is required to uncomment the
1240 directives and define how usernames and passwords are checked, but
1241 there is now a framework to start from.
1243 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1244 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1245 without this. I don't know how relevant this is to other LDAP libraries.
1247 PH/02 Add the verb name to the "unknown ACL verb" error.
1249 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1251 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1253 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1254 directory not expanded when it should be if an expanded home directory
1255 was set for the address (which is overridden by the transport).
1257 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1260 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1261 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1262 because it is too late at that time, and has no effect.
1264 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1265 security issue with \' (bugzilla #107). I could not use the
1266 PQescapeStringConn() function, because it needs a PGconn value as one of
1269 PH/08 When testing addresses using -bt, indicate those final addresses that
1270 are duplicates that would not cause an additional delivery. At least one
1271 person was confused, thinking that -bt output corresponded to deliveries.
1272 (Suppressing duplicates isn't a good idea as you lose the information
1273 about possibly different redirections that led to the duplicates.)
1275 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1276 systems where poll() doesn't work, in particular OS X.
1278 PH/10 Added more information to debugging output for retry time not reached.
1280 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1281 operations in malware.c.
1283 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1286 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1287 syslog_duplication set false, log lines that would normally be written
1288 both the the main log and to the reject log were not written to syslog at
1291 PH/14 In the default configuration, change the use of "message" in ACL warn
1292 statements to "add_header".
1294 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1295 not followed by a command (e.g. "seen endif").
1297 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1298 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1301 PH/17 Added extra conditions to the default value of delay_warning_condition
1305 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1306 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1307 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1310 The Auto-Submitted: and various List- headers are standardised, whereas I
1311 don't think Precedence: ever was.
1313 PH/18 Refactored debugging code in route_finduser() to show more information,
1314 in particular, the error code if getpwnam() issues one.
1316 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1317 This is apparently needed in addition to the PH/07 change above to avoid
1318 any possible encoding problems.
1320 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1321 but not after initializing Perl.
1323 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1324 output them only if debugging. By default they are written stderr,
1325 apparently, which is not desirable.
1327 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1330 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1333 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1335 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1336 authenticated or an ident call has been made. Suppress the default
1337 values for $authenticated_id and $authenticated_sender (but permit -oMai
1338 and -oMas) when testing with -bh.
1340 PH/24 Re-jigged the order of the tests in the default configuration so that the
1341 tests for valid domains and recipients precede the DNS black list and CSA
1342 tests, on the grounds that those ones are more expensive.
1344 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1345 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1346 This bug exists in every version of Exim that I still have, right back to
1349 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1350 However, an attempt to turn on case-sensitivity in a regex key by
1351 including (?-i) didn't work because the subject string was already
1352 lowercased, and the effects were non-intuitive. It turns out that a
1353 one-line patch can be used to allow (?-i) to work as expected.
1359 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1360 other effects) broke the use of negated acl sub-conditions.
1362 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1365 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1366 "Deny" causes Exim to reject the incoming connection with a 554 error.
1367 Unfortunately, if there is a major crisis, such as a disk failure,
1368 tcp-wrappers gives "deny", whereas what one would like would be some
1369 kind of temporary error. A kludge has been added to help with this.
1370 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1371 554 error is used if errno is still zero or contains ENOENT (which occurs
1372 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1375 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1377 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1378 errors. Otherwise a message that provokes a temporary error (when other
1379 messages do not) can cause a whole host to time out.
1381 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1382 addresses were routed directly to files or pipes from a redirect router.
1383 File deliveries just didn't batch; pipe deliveries might have suffered
1386 PH/06 A failure to get a lock for a hints database would erroneously always say
1387 "Failed to get write lock", even when it was really a read lock.
1389 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1390 of 0600. This has been changed to use the value of the lockfile_mode
1391 option (which defaults to 0600).
1393 PH/08 Applied small patch from the Sieve maintainer.
1395 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1396 folder from quota calculations, a direct delivery into this folder messed
1397 up the contents of the maildirsize file. This was because the regex was
1398 used only to exclude .Trash (or whatever) when the size of the mailbox
1399 was calculated. There was no check that a delivery was happening into an
1400 excluded directory. This bug has been fixed by ignoring all quota
1401 processing for deliveries into excluded directories.
1403 PH/10 Added the maildirfolder_create_regex option to appendfile.
1409 PH/01 The code for finding all the local interface addresses on a FreeBSD
1410 system running IPv6 was broken. This may well have applied to all BSD
1411 systems, as well as to others that have similar system calls. The broken
1412 code found IPv4 interfaces correctly, but gave incorrect values for the
1413 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1414 that it would not match correctly against @[] and not recognize the IPv6
1417 PH/02 The ipliteral router was not recognizing addresses of the form user@
1418 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1420 PH/03 Added disable_ipv6.
1422 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1423 decoded form, because it is most often used to construct To: headers
1424 lines in autoreplies, and the decoded form may well be syntactically
1425 invalid. However, $reply_address has leading white space removed, and all
1426 newlines turned into spaces so that the autoreply transport does not
1429 PH/05 If group was specified without a user on a router, and no group or user
1430 was specified on a transport, the group from the router was ignored.
1432 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1433 for visible compile-time settings that can be used to change these
1434 numbers, for those that want even more. Backwards compatibility with old
1435 spool files has been maintained. However, going back to a previous Exim
1436 release will lost any variables that are in spool files.
1438 PH/07 Two small changes when running in the test harness: increase delay when
1439 passing a TCP/IP connection to a new process, in case the original
1440 process has to generate a bounce, and remove special handling of
1441 127.0.0.2 (sic), which is no longer necessary.
1443 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1444 be the same on different OS.
1446 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1449 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1450 whether --show-vars was specified or not
1452 JJ/02 exipick: Added support for new ACL variable spool format introduced
1455 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1456 syntactically invalid From: or Reply-to: line, and a filter used this to
1457 generate an autoreply, and therefore failed to obtain an address for the
1458 autoreply, Exim could try to deliver to a non-existent relative file
1459 name, causing unrelated and misleading errors. What now happens is that
1460 it logs this as a hard delivery error, but does not attempt to create a
1463 PH/11 The exinext utility has a -C option for testing purposes, but although
1464 the given file was scanned by exinext itself; it wasn't being passed on
1465 when Exim was called.
1467 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1468 an end-of-file indication when reading a command response.
1470 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1471 compiled. In many other places in Exim, IPv6 addresses are always
1472 recognized, so I have changed this. It also means that IPv4 domain
1473 literals of the form [IPV4:n.n.n.n] are now always recognized.
1475 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1476 used if the router is not running as root, for example, when verifying at
1477 ACL time, or when using -bh. The debugging output from this situation was
1478 non-existent - all you got was a failure to exec. I have made two
1481 (a) Failures to set uid/gid, the current directory, or a process leader
1482 in a subprocess such as that created by queryprogram now generate
1483 suitable debugging ouput when -d is set.
1485 (b) The queryprogram router detects when it is not running as root,
1486 outputs suitable debugging information if -d is set, and then runs
1487 the subprocess without attempting to change uid/gid.
1489 PH/15 Minor change to Makefile for building test_host (undocumented testing
1492 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1493 additional section of a DNS packet that returns MX or SRV records.
1494 Instead, it always explicitly searches for A/AAAA records. This avoids
1495 major problems that occur when a DNS server includes only records of one
1496 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1497 fixed another bug: if SRV records were looked up and the corresponding
1498 address records were *not* found in the additional section, the port
1499 values from the SRV records were lost.
1501 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1502 using the correct key (the original address) when searching the retry
1503 rules in order to find which one to use for generating the retry hint.
1505 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1506 adding the default one. Similarly, if it contains a Reply-To: header, the
1507 errors_reply_to option, if set, is not used.
1509 PH/19 When calculating a retry time, Exim used to measure the "time since
1510 failure" by looking at the "first failed" field in the retry record. Now
1511 it does not use this if it is later than than the arrival time of the
1512 message. Instead it uses the arrival time. This makes for better
1513 behaviour in cases where some deliveries succeed, thus re-setting the
1514 "first failed" field. An example is a quota failure for a huge message
1515 when small messages continue to be delivered. Without this change, the
1516 "time since failure" will always be short, possible causing more frequent
1517 delivery attempts for the huge message than are intended.
1518 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1520 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1521 $1, $2, $3) because the numerical variables can be reset during some
1522 expansion items (e.g. "match"), thereby losing the authentication data.
1524 PH/21 Make -bV show the size of off_t variables so that the test suite can
1525 decide whether to run tests for quotas > 2G.
1527 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1528 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1529 filecount value is greater than 2G or if a quota value is greater than 2G
1530 on a system where the size of off_t is not greater than 4, a panic error
1533 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1534 never match. The debug and -bh output now contains an explicit error
1535 message indicating a malformed IPv4 address or mask.
1537 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1538 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1539 PH/23 above applies.
1541 PH/25 Do not write to syslog when running in the test harness. The only
1542 occasion when this arises is a failure to open the main or panic logs
1543 (for which there is an explicit test).
1545 PH/26 Added the /no_tell option to "control=freeze".
1547 PH/27 If a host name lookup failed very early in a connection, for example, if
1548 the IP address matched host_lookup and the reverse lookup yielded a name
1549 that did not have a forward lookup, an error message of the form "no IP
1550 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1551 could be logged. Now it outputs the IP address instead of "NULL".
1553 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1554 allows the sender and the authenticated sender to be set when
1555 submitting a message from within Exim. Since child_open_exim() is
1556 documented for local_scan(), the new function should be too.
1558 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1559 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1560 results in an empty string is now treated as unset.
1562 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1564 PH/31 Added sender_verify_fail logging option.
1566 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1567 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1568 tidied the source and removed it altogether.
1570 PH/33 When a queue run was abandoned because the load average was too high, a
1571 log line was always written; now it is written only if the queue_run log
1572 selector is set. In addition, the log line for abandonment now contains
1573 information about the queue run such as the pid. This is always present
1574 in "start" and "stop" lines but was omitted from the "abandon" line.
1576 PH/34 Omit spaces between a header name and the colon in the error message that
1577 is given when verify = headers_syntax fails (if there are lots of them,
1578 the message gets confusing).
1580 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1581 names, as there are now some PTR records that contain slashes. This check
1582 is only to protect against broken name servers that fall over on strange
1583 characters, so the fact that it applies to all lookups doesn't matter.
1585 PH/36 Now that the new test suite is complete, we can remove some of the
1586 special code in Exim that was needed for the old test suite. For example,
1587 sorting DNS records because real resolvers return them in an arbitrary
1588 order. The new test suite's fake resolver always returns records in the
1591 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1592 bounces) except when queue_only is set, to avoid logging races between
1593 the different processes.
1595 PH/38 Panic-die if .include specifies a non-absolute path.
1597 PH/39 A tweak to the "H" retry rule from its user.
1599 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1600 a label. They prevented compilation on older perls.
1602 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1603 a warning to be raised on newish perls.
1605 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1606 on queue. Changes to match documented behaviour of showing count of
1607 messages matching specified criteria.
1609 PH/40 Changed the default ident timeout from 30s to 5s.
1611 PH/41 Added support for the use of login_cap features, on those BSD systems
1612 that have them, for controlling the resources used by pipe deliveries.
1614 PH/42 The content-scanning code uses fopen() to create files in which to put
1615 message data. Previously it was not paying any attention to the mode of
1616 the files. Exim runs with umask(0) because the rest of the code creates
1617 files with open(), and sets the required mode explicitly. Thus, these
1618 files were ending up world-writeable. This was not a big issue, because,
1619 being within the spool directory, they were not world-accessible. I have
1620 created a function called modefopen, which takes an additional mode
1621 argument. It sets umask(777), creates the file, chmods it to the required
1622 mode, then resets the umask. All the relevant calls to fopen() in the
1623 content scanning code have been changed to use this function.
1625 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1626 to 24 hours. This avoids potential overflow problems when processing G
1627 and H retry rules. I suspect nobody ever tinkers with this value.
1629 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1631 PH/45 When the plaintext authenticator is running as a client, the server's
1632 challenges are checked to ensure they are valid base64 strings. By
1633 default, the authentication attempt is cancelled if an invalid string is
1634 received. Setting client_ignore_invalid_base64 true ignores these errors.
1635 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1636 they are received. Thus, the responses can be made to depend on the
1637 challenges. If an invalid string is ignored, an empty string is placed in
1640 PH/46 Messages that are created by the autoreply transport now contains a
1641 References: header, in accordance with RFCs 2822 and 3834.
1643 PH/47 Added authenticated_sender_force to the smtp transport.
1645 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1647 PH/49 Installed latest patch from the Sieve maintainer.
1649 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1650 was also set, the appendfile transport was unnecessarily scanning a
1651 directory of message files (e.g. for maildir delivery) to find the count
1652 of files (along with the size), even though it did not need this
1653 information. It now does the scan only if it needs to find either the
1654 size of the count of files.
1656 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1658 PH/52 Two bugs concerned with error handling when the smtp transport is
1661 (i) Exim was not creating retry information for temporary errors given
1662 for individual recipients after the DATA command when the smtp transport
1663 was used in LMTP mode. This meant that they could be retried too
1664 frequently, and not timed out correctly.
1666 (ii) Exim was setting the flag that allows error details to be returned
1667 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1668 recipients that were returned after the DATA command.
1670 PH/53 This is related to PH/52, but is more general: for any failing address,
1671 when detailed error information was permitted to be returned to the
1672 sender, but the error was temporary, then after the final timeout, only
1673 "retry timeout exceeded" was returned. Now it returns the full error as
1674 well as "retry timeout exceeded".
1676 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1677 do this, and (what is worse) MTAs that accept it.
1679 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1680 will now be deprecated.
1682 PH/56 New os.c-cygwin from the Cygwin maintainer.
1684 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1685 formats (previously only available in exim formats via -bpr, -bpru,
1686 and -bpra. Now also available in native and exiqgrep formats)
1688 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1689 with very large, slow to parse queues
1691 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1693 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1695 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1696 responses to authentication challenges, though it was showing the
1697 challenges; (ii) I've removed the CR characters from the debug output for
1700 PH/58 Allow for the insertion of a newline as well as a space when a string
1701 is turned into more than one encoded-word during RFC 2047 encoding. The
1702 Sieve code now uses this.
1704 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1705 data_4xx, lost_connection, tls_required.
1707 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1708 message was being sent as an SMTP response.
1710 PH/61 Add -l and -k options to exicyclog.
1712 PH/62 When verifying, if an address was redirected to one new address, so that
1713 verification continued, and the new address failed or deferred after
1714 having set something in $address_data, the value of $address_data was not
1715 passed back to the ACL. This was different to the case when no
1716 redirection occurred. The value is now passed back in both cases.
1718 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1719 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1720 use login_cap.h, so on its own it isn't the distinguishing feature. The
1721 new name refers directly to the setclassresources() function.
1723 PH/65 Added configuration files for NetBSD3.
1725 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1727 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1728 is preferred over IPv4.
1730 PH/68 The bounce_return_message and bounce_return_body options were not being
1731 honoured for bounces generated during the reception of non-SMTP messages.
1732 In particular, this applied to messages rejected by the ACL. This bug has
1733 been fixed. However, if bounce_return_message is true and bounce_return_
1734 body is false, the headers that are returned for a non-SMTP message
1735 include only those that have been read before the error was detected.
1736 (In the case of an ACL rejection, they have all been read.)
1738 PH/69 The HTML version of the specification is now built in a directory called
1739 spec_html instead of spec.html, because the latter looks like a path with
1740 a MIME-type, and this confuses some software.
1742 PH/70 Catch two compiler warnings in sieve.c.
1744 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1745 function verify_get_ident() calls ip_connect() to connect a socket, but
1746 if the "connect()" function timed out, ip_connect() used to close the
1747 socket. However, verify_get_ident() also closes the socket later, and in
1748 between Exim writes to the log, which may get opened at this point. When
1749 the socket was closed in ip_connect(), the log could get the same file
1750 descriptor number as the socket. This naturally causes chaos. The fix is
1751 not to close the socket in ip_connect(); the socket should be closed by
1752 the function that creates it. There was only one place in the code where
1753 this was missing, in the iplookup router, which I don't think anybody now
1754 uses, but I've fixed it anyway.
1756 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1757 well as to direct DNS lookups. Otherwise the handling of names in host
1758 lists is inconsistent and therefore confusing.
1764 PH/01 Two changes to the default runtime configuration:
1766 (1) Move the checks for relay_from_hosts and authenticated clients from
1767 after to before the (commented out) DNS black list checks.
1769 (2) Add control=submission to the relay_from_hosts and authenticated
1770 clients checks, on the grounds that messages accepted by these
1771 statements are most likely to be submissions.
1773 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1775 (1) Generate an error if the third argument for the ${prvs expansion is
1778 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1781 (3) Reset the variables that are obtained from the first argument of
1782 ${prvscheck and used in the second argument before leaving the code,
1783 because their memory is reclaimed, so using them afterwards may do
1786 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1787 one (it's much easier than Tom thought :-).
1789 (5) Because of (4), we can now allow for the use of $prvscheck_result
1790 inside the third argument.
1792 PH/03 For some reason, the default setting of PATH when running a command from
1793 a pipe transport was just "/usr/bin". I have changed it to
1796 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1797 anything to be listed in the output from -bV.
1799 PH/05 When a filter generated an autoreply, the entire To: header line was
1800 quoted in the delivery log line, like this:
1802 => >A.N.Other <ano@some.domain> <original@ddress> ...
1804 This has been changed so that it extracts the operative address. There
1805 may be more than one such address. If so, they are comma-separated, like
1808 => >ano@some.domain,ona@other.domain <original@ddress> ...
1810 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1811 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1812 not being looked up in the rDNS to get a host name, Exim was showing the
1813 IP address twice in Received: lines, even though the IP addresses were
1814 identical. For example:
1816 Received: from [1.2.3.4] (helo=[1.2.3.4])
1818 However, if the real host name was known, it was omitting the HELO data
1819 if it matched the actual IP address. This has been tidied up so that it
1820 doesn't show the same IP address twice.
1822 PH/07 When both +timestamp and +memory debugging was on, the value given by
1823 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1824 called by the debug printing, thereby overwriting the timestamp buffer.
1825 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1827 PH/08 When the original message was included in an autoreply transport, it
1828 always said "this is a copy of the message, including all the headers",
1829 even if body_only or headers_only was set. It now gives an appropriate
1832 PH/09 Applied a patch from the Sieve maintainer which:
1834 o fixes some comments
1835 o adds the (disabled) notify extension core
1836 o adds some debug output for the result of if/elsif tests
1837 o points to the current vacation draft in the documentation
1838 and documents the missing references header update
1842 o fixes a bug in processing the envelope test (when testing
1843 multiple envelope elements, the last element determinted the
1846 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1847 Electronic Mail") by including:
1849 Auto-submitted: auto-generated
1851 in the messages that it generates (bounce messages and others, such as
1852 warnings). In the case of bounce messages for non-SMTP mesages, there was
1853 also a typo: it was using "Auto_submitted" (underscore instead of
1854 hyphen). Since every message generated by Exim is necessarily in response
1855 to another message, thes have all been changed to:
1857 Auto-Submitted: auto-replied
1859 in accordance with these statements in the RFC:
1861 The auto-replied keyword:
1863 - SHOULD be used on messages sent in direct response to another
1864 message by an automatic process,
1866 - MUST NOT be used on manually-generated messages,
1868 - MAY be used on Delivery Status Notifications (DSNs) and Message
1869 Disposition Notifications (MDNs),
1871 - MUST NOT be used on messages generated by automatic or periodic
1872 processes, except for messages which are automatic responses to
1875 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1876 to the default Received: header definition.
1878 PH/12 Added log selector acl_warn_skipped (default on).
1880 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1881 variables because (a) they are in the wrong storage pool and (b) even if
1882 they were copied, it wouldn't work properly because of the caching.
1884 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1885 checking when decoding. Apparently there are clients that generate
1886 overlong encoded strings. Why am I not surprised?
1888 PH/15 If the first argument of "${if match_address" was not empty, but did not
1889 contain an "@" character, Exim crashed. Now it writes a panic log message
1890 and treats the condition as false.
1892 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1894 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1895 extension "envelope-auth". The code is finished and in agreement with
1896 other implementations, but there is no documentation so far and in fact,
1897 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1898 not changing the active code.
1900 Print executed "if" and "elsif" statements when debugging is used. This
1901 helps a great deal to understand what a filter does.
1903 Document more things not specified clearly in RFC3028. I had all this
1904 sorted out, when out of a sudden new issues came to my mind. Oops."
1906 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1909 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1910 canonical form (as documented). However, after a host name lookup from
1911 the IP address, check_host() was doing a simple string comparison with
1912 addresses acquired from the DNS when checking that the found name did
1913 have the original IP as one of its addresses. Since any found IPv6
1914 addresses are likely to be in abbreviated form, the comparison could
1915 fail. Luckily, there already exists a function for doing the comparison
1916 by converting both addresses to binary, so now that is used instead of
1917 the text comparison.
1919 PH/20 There was another similar case to PH/19, when a complete host name was
1920 given in a host list; looking up its IP address could give an abbreviated
1921 form, whereas the current host's name might or might not be abbreviated.
1922 The same fix has been applied.
1928 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1929 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1932 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1933 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1935 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1937 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1938 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1939 Also, for "Auto-Submitted", treat anything other than "no" as
1940 non-personal, in accordance with RFC 3834. (Previously it treated
1941 anything starting "auto-" as non-personal.)
1943 TF/01 The control=submission/name=... option had a problem with syntax
1944 errors if the name included a slash character. The /name= option
1945 now slurps the rest of the string, so it can include any characters
1946 but it must come last in the list of options (after /sender_retain
1949 PH/05 Some modifications to the interface to the fake nameserver for the new
1957 TK/01 Added the "success_on_redirect" address verification option. See
1958 NewStuff for rationale and an example.
1960 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1962 PH/02 Patch to exigrep to allow it to work on syslog lines.
1964 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1965 fread() to skip over the body file's header line, because in Cygwin the
1966 header line is locked and is inaccessible.
1968 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1969 co-exist for some time) to make it clear that it is the Exim ID that is
1970 referenced, not the Message-ID: header line.
1972 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1973 string_format() function, because snprintf() does not exist on all
1976 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1977 ${stat: expansion item.
1979 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1980 protocol synchronization error", to keep the pedants happy.
1982 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1983 well as for IRIX systems, when gcc is being used. See the host.c source
1986 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1988 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1991 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1992 transport and to the smtp transport in LMTP mode.
1994 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1996 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1997 run a filter in a subprocess. This could lead to confusion in subsequent
1998 lookups in the parent process. There should also be a search_tidyup() at
1999 the end of the subprocess.
2001 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2002 only if the host matched helo_try_verify_hosts, which caused the
2003 verification to occur when the EHLO/HELO command was issued. The ACL just
2004 tested the remembered result. Now, if a previous verification attempt has
2005 not happened, "verify = helo" does it there and then.
2007 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2009 TK/03 Fix log output including CR from clamd.
2011 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2012 exist provoked a memory error which could cause a segfault.
2014 PH/15 Installed PCRE 6.2
2016 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2018 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2019 of the problem. Specifically, suggested +O2 rather than +O1 for the
2022 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2024 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2025 record, the port was not correctly logged when the outgoing_port log
2026 selector was set (it logged the transort's default port).
2028 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2029 fallback_hosts, and "hosts" in the smtp transport.
2031 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2032 host errors such as "Connection refused".
2034 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2035 authentication with radiusclient 0.4.9:
2037 - Error returned from rc_read_config was caught wrongly
2038 - Username/password not passed on to radius server due to wrong length.
2040 The presumption is that some radiusclient API changes for 4.51/PH/17
2041 were not taken care of correctly. The code is still untested by me (my
2042 Linux distribution still has 0.3.2 of radiusclient), but it was
2043 contributed by a Radius user.
2045 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2046 expanding the "port" option of the smtp transport.
2048 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2049 while reading a MIME header. Thanks to Tom Hughes for a patch.
2051 PH/24 Include config.h inside local_scan.h so that configuration settings are
2054 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2055 that in an ACL for RCPT (for example), you can examine exactly what was
2058 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2059 commands, but it was not correctly comparing the address with the actual
2060 client host address. Thus, it would show the EHLO address in Received:
2061 header lines when this was not necessary.
2063 PH/27 Added the % operator to ${eval:}.
2065 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2066 it should be ignoring failures (because with -C, for example, it has lost
2067 privilege). It wasn't ignoring creation failures other than "already
2070 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2073 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2074 because an input line was too long, either on its own, or by virtue of
2075 too many continuations, the temporary file was not being removed, and the
2076 return code was incorrect.
2078 PH/31 Missing "BOOL" in function definition in filtertest.c.
2080 PH/32 Applied Sieve patches from the maintainer.
2082 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2084 PH/33 Added "verify = not_blind".
2086 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2087 Local/Makefile (with some defaults set). These are used in built scripts
2088 such as exicyclog, but they have never been used in the exim_install
2089 script (though there are many overriding facilities there). I have
2090 arranged that the exim_install script now takes note of these two
2093 PH/35 Installed configuration files for Dragonfly.
2095 PH/36 When a locally submitted message by a trusted user did not contain a
2096 From: header, and the sender address was obtained from -f or from an SMTP
2097 MAIL command, and the trusted user did not use -F to supply a sender
2098 name, $originator_name was incorrectly used when constructing a From:
2099 header. Furthermore, $originator_name was used for submission mode
2100 messages from external hosts without From: headers in a similar way,
2101 which is clearly wrong.
2103 PH/37 Added control=suppress_local_fixups.
2105 PH/38 When log_selector = +received_sender was set, and the addition of the
2106 sender made the log line's construction buffer exactly full, or one byte
2107 less than full, an overflow happened when the terminating "\n" was
2110 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2111 when the result of a list match is failure because a DNS lookup failed.
2113 PH/40 RM_COMMAND is now used in the building process.
2115 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2116 the "build-* directories that it finds.
2118 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2119 address in a domain literal was a prefix of an interface address.
2121 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2122 when verifying a sender address, unless rewrite_headers is false.
2124 PH/44 Wrote a long comment about why errors_to addresses are verified as
2125 recipients, not senders.
2127 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2128 the ratelimit ACL was added.
2130 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2132 PH/46 Added extra information about PostgreSQL errors to the error string.
2134 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2135 avoiding the need to install special zones in a real server. This is
2136 backwards compatible; if it can't find the fake resolver, it drops back.
2137 Thus, both old and new test suites can be run.
2139 TF/02 Added util/ratelimit.pl
2141 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2142 clock is set back in time.
2144 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2145 Candler <B.Candler@pobox.com>.
2147 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2148 OK for addresses that are the result of redirections.
2150 PH/48 A number of further additions for the benefit of the new test suite,
2151 including a fake gethostbyname() that interfaces to the fake DNS resolver
2154 TF/06 The fix for widen_domains has also been applied to qualify_single and
2155 search_parents which are the other dnslookup options that can cause
2158 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2161 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2163 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2164 service" if it encountered an MX record with an empty target hostname.
2165 The message is now "an MX or SRV record indicated no SMTP service".
2167 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2168 if the DNS of the sending site is misconfigured. This is quite a
2169 common situation. This change restores the behaviour of treating a
2170 helo verification defer as a failure.
2172 PH/51 If self=fail was set on a router, the bounce message did not include the
2173 actual error message.
2179 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2181 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2182 command itself ended in error, the underlying message about the transport
2183 filter timeout was being overwritten with the pipe command error. Now the
2184 underlying error message should be appended to the second error message.
2186 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2188 PH/02 Reduce the amount of output that "make" produces by default. Full output
2189 can still be requested.
2191 PH/03 The warning log line about a condition test deferring for a "warn" verb
2192 was being output only once per connection, rather than after each
2193 occurrence (because it was using the same function as for successful
2194 "warn" verbs). This seems wrong, so I have changed it.
2196 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2197 it should not have, which might have caused a crash in the right
2198 circumstances, but probably never did.
2200 PH/04 Installed a modified version of Tony Finch's patch to make submission
2201 mode fix the return path as well as the Sender: header line, and to
2202 add a /name= option so that you can make the user's friendly name appear
2205 TF/03 Added the control = fakedefer ACL modifier.
2207 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2208 Mark Lowes for thorough testing.
2210 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2212 TK/03 Merged latest SRS patch from Miles Wilton.
2214 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2215 with the definition in sysexits.h (which is #included earlier).
2216 Fortunately, Exim does not actually use EX_OK. The code used to try to
2217 preserve the sysexits.h value, by assumimg that macro definitions were
2218 scanned for macro replacements. I have been disabused of this notion,
2219 so now the code just undefines EX_OK before #including unistd.h.
2221 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2222 in the smtp transport. When a block could not be written in a single
2223 write() function, the timeout was being re-applied to each part-write.
2224 This seems wrong - if the receiver was accepting one byte at a time it
2225 would take for ever. The timeout is now adjusted when this happens. It
2226 doesn't have to be particularly precise.
2228 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2229 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2231 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2232 without a domain if the check to <postmaster@domain> fails.
2234 SC/01 Eximstats: added -xls and the ability to specify output files
2235 (patch written by Frank Heydlauf).
2237 SC/02 Eximstats: use FileHandles for outputing results.
2239 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2241 SC/04 Eximstats: fixed display of large numbers with -nvr option
2243 SC/05 Eximstats: fixed merging of reports with empty tables.
2245 SC/06 Eximstats: added the -include_original_destination flag
2247 SC/07 Eximstats: removed tabs and trailing whitespace.
2249 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2251 TK/06 MBOX spool code: Add real "From " MBOX separator line
2252 so the .eml file is really in mbox format (even though
2253 most programs do not really care). Patch from Alex Miller.
2255 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2256 The latter is generated from $received_to and is only set if the
2257 message has one envelope recipient. SA can use these headers,
2258 obviously out-of-the-box. Patch from Alex Miller.
2260 PH/08 The ${def test on a variable was returning false if the variable's
2261 value was "0", contrary to what the specification has always said!
2262 The result should be true unless the variable is empty.
2264 PH/09 The syntax error of a character other than { following "${if
2265 def:variable_name" (after optional whitespace) was not being diagnosed.
2266 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2267 accidental colon was present, for example, could give incorrect results.
2269 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2270 result is used (not including appendfile, where other changes are about
2273 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2274 This involved changing a lot of size variables from int to off_t. It
2275 should work with maildirs and everything.
2277 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2278 spamd dying while we are connected to it.
2280 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2283 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2284 function for BDB 4.3.
2286 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2288 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2289 the sources for PCRE have been re-arranged and more files are now
2292 PH/15 The code I had for printing potentially long long variables in PH/11
2293 above was not the best (it lost precision). The length of off_t variables
2294 is now inspected at build time, and an appropriate printing format (%ld
2295 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2296 to be "long long int" or "long int". This is needed for the internal
2297 formatting function string_vformat().
2299 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2300 the configuration file to be ":syslog", then the script "guesses" where
2301 the logs files are, rather than using the compiled in default. In our
2302 case the guess is not the same as the compiled default, so the script
2303 suddenly stopped working when I started to use syslog. The patch checks
2304 to see if log_file_path is "". If so, it attempts to read it from exim
2305 with no configuration file to get the compiled in version, before it
2306 falls back to the previous guessing code."
2308 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2309 implementing BATV in an Exim configuration. See NewStuff for the gory
2312 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2313 Makefile that are specific to HP-UX.
2315 PH/18 If the "use_postmaster" option was set for a recipient callout together
2316 with the "random" option, the postmaster address was used as the MAIL
2317 FROM address for the random test, but not for the subsequent recipient
2318 test. It is now used for both.
2320 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2321 patch removes a few documentation additions to RFC 3028, because the
2322 latest draft now contains them. It adds the new en;ascii-case comparator
2323 and a new error check for 8bit text in MIME parts. Comparator and
2324 require names are now matched exactly. I enabled the subaddress
2325 extension, but it is not well tested yet (read: it works for me)."
2327 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2328 rework some of the code of TK/09 above to avoid the hardwired use of
2329 "%lld" and "long long". Replaced the call to snprintf() with a call to
2332 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2333 records point to non-existent hosts", "retry timeout exceeded", and
2334 "retry time not reached for any host after a long failure period".
2336 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2337 experimental DomainKeys support:
2339 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2340 (2) On an error such as an illegally used "control", the wrong name for
2341 the control was given.
2343 These problems did NOT occur unless DomainKeys support was compiled.
2345 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2347 PH/24 Added ${if match_ip condition.
2349 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2350 databases so that it will be absolutely obvious if a crash occurs in the
2351 DB library. This is a regular occurrence (often caused by mis-matched
2354 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2355 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2356 picked up on a user's system that detects such things. There doesn't seem
2357 to be a gcc warning option for this - only an attribute that has to be
2358 put on the function's prototype. It seems that in Fedora Core 4 they have
2359 set this on a number of new functions. No doubt there will be more in due
2362 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2363 specify a transport. However, if an address that was verified by such a
2364 router was the subject of a callout, Exim crashed because it tried to
2365 read the rcpt_include_affixes from the non-existent transport. Now it
2366 just assumes that the setting of that option is false. This bug was
2367 introduced by 4.51/PH/31.
2369 PH/28 Changed -d+all to exclude +memory, because that information is very
2370 rarely of interest, but it makes the output a lot bigger. People tend to
2371 do -d+all out of habit.
2373 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2374 code in os-type was giving problems when libc.so lives in lib64, like on
2377 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2378 aren't the modern standard, and it seems that some systems' include files
2379 don't always have them. Exim was already checking for some of the newer
2380 ones like T_AAAA, and defining it itself. I've added checks for all the
2381 record types that Exim uses.
2383 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2384 not automatically generating a new one, as it is supposed to. This
2385 prevented TLS from working. If the file did exist, but contained invalid
2386 data, a new version was generated, as expected. It was only the case of a
2387 non-existent file that was broken.
2389 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2390 with a change in libdomainkeys > 0.64.
2392 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2393 from DNS. If the selector record carries the flag, it now has
2394 precedence over the domain-wide flag.
2396 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2398 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2399 the use of a port name that isn't defined in /etc/services) occurred, the
2400 message was deferred as in a normal delivery, and thus remained on the
2401 spool, instead of being failed because of the mua_wrapper setting. This
2402 is now fixed, and I tidied up some of the mua_wrapper messages at the
2405 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2406 the messages in a hash of arrays rather than using individual hashes.
2407 This is a bit cleaner and results in dramatic memory savings, albeit
2408 at a slight CPU cost.
2410 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2411 as requested by Marc Sherman.
2413 SC/10 Eximstats: added histograms for user specified patterns as requested
2416 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2418 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2419 fopen() in the content-scanning modules that did not already have it.
2425 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2426 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2428 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2430 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2432 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2433 to test Sieve filters that use "vacation".
2435 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2436 that changes the way the GnuTLS parameters are stored in the cache file.
2437 The new format can be generated externally. For backward compatibility,
2438 if the data in the cache doesn't make sense, Exim assumes it has read an
2439 old-format file, and it generates new data and writes a new file. This
2440 means that you can't go back to an older release without removing the
2443 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2444 work if there are any delivery delays because "one_time" forces the
2445 parent to be marked "delivered", so its unseen clone is never tried
2446 again. For this reason, Exim now forbids the simultaneous setting of
2449 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2450 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2451 entry if you want to know the details. The fix, however, neglected to
2452 consider the case when local delivery batching is involved. The test for
2453 "previously delivered" was not happening when checking to see if an
2454 address could be batched with a previous (undelivered) one; under
2455 certain circumstances this could lead to multiple deliveries to the same
2458 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2459 in its include files, and this causes problems building Exim.
2461 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2462 header_syntax) but Exim was just ignoring anything given after a slash.
2463 In particular, this caused confusion with an attempt to use "verify =
2464 reverse_host_lookup/defer_ok". An error is now given when options are
2465 supplied for verify items that do not have them. (Maybe reverse_host_
2466 lookup should have a defer_ok option, but that's a different point.)
2468 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2469 defined by RFC 821) to 2048, because there were problems with some AUTH
2470 commands, and RFC 1869 says the size should be increased for extended
2471 SMTP commands that take arguments.
2473 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2476 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2477 "unknown" error; now it says that the functionality isn't in the binary.
2479 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2480 an address' error message when a string expansion fails (syntax or
2481 whatever). Otherwise the password may appear in the log. Following change
2482 PH/42 below, there is no longer a chance of it appearing in a bounce
2485 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2487 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2488 of its addresses was ever tried. (Bugzilla bug #2.)
2490 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2491 the result incorrectly in the debug output. (It correctly added a newline
2492 to what was transported.)
2494 TF/01 Added $received_time.
2496 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2497 commented out examples of how to interface to a virus scanner and to
2498 SpamAssassin. Also added commented examples of av_scanner and
2499 spamd_address settings.
2501 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2502 and controls are allowed in which ACLs. There were a couple of minor
2503 errors. Some of the entries in the conditions table (which is a table of
2504 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2505 negation of where the condition IS allowed.
2507 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2509 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2510 header file does not have a version number, so I've had to invent a new
2511 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2512 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2513 radiusclient), but it was contributed by a Radius user.
2515 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2516 files or directories, for OpenSSL.
2518 PH/19 When an Exim process that is running as root has to create an Exim log
2519 file, it does so in a subprocess that runs as exim:exim so as to get the
2520 ownership right at creation (otherwise, other Exim processes might see
2521 the file with the wrong ownership). There was no test for failure of this
2522 fork() call, which would lead to the process getting stuck as it waited
2523 for a non-existent subprocess. Forks do occasionally fail when resources
2524 run out. I reviewed all the other calls to fork(); they all seem to check
2527 PH/20 When checking for unexpected SMTP input at connect time (before writing
2528 the banner), Exim was not dealing correctly with a non-positive return
2529 from the read() function. If the client had disconnected by this time,
2530 the result was a log entry for a synchronization error with an empty
2531 string after "input=" when read() returned zero. If read() returned -1
2532 (an event I could not check), uninitialized data bytes were printed.
2533 There were reports of junk text (parts of files, etc) appearing after
2536 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2538 PH/22 Added support for macro redefinition, and (re)definition in between
2539 driver and ACL definitions.
2541 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2542 forgetting to use the resulting value; it was using the unexpanded value.
2544 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2545 hadn't been configured. The fix is from Juergen Kreileder, who
2546 understands it better than I do:
2548 "Here's what I see happening with three configured cyrus_sasl
2549 authenticators configured (plain, login, cram-md5):
2551 On startup auth_cyrus_sasl_init() gets called for each of these.
2552 This means three calls to sasl_listmech() without a specified mech_list.
2553 => SASL tests which mechs of all available mechs actually work
2554 => three warnings about OTP not working
2555 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2557 With the patch, sasl_listmech() also gets called three times. But now
2558 SASL's mech_list option is set to the server_mech specified in the the
2559 authenticator. Or in other words, the answer from sasl_listmech()
2560 gets limited to just the mech you're testing for (which is different
2562 => the return list contains just 'plain' or 'login', 'cram-md5' or
2563 nothing depending on the value of ob->server_mech.
2565 I've just tested the patch: Authentication still works fine,
2566 unavailable mechs specified in the exim configuration are still
2567 caught, and the auth.log warnings about OTP are gone."
2569 PH/25 When debugging is enabled, the contents of the command line are added
2570 to the debugging output, even when log_selector=+arguments is not
2573 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2574 answer is "GNU", and only if the return is "GNU/something" is the answer
2577 PH/27 $acl_verify_message is now set immediately after the failure of a
2578 verification in an ACL, and so is available in subsequent modifiers. In
2579 particular, the message can be preserved by coding like this:
2581 warn !verify = sender
2582 set acl_m0 = $acl_verify_message
2584 Previously, $acl_verify_message was set only while expanding "message"
2585 and "log_message" when a very denied access.
2587 PH/28 Modified OS/os.c-Linux with
2589 -#ifndef OS_LOAD_AVERAGE
2590 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2592 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2593 nomenclature these days.)
2595 PH/29 Installed patch from the Sieve maintainer that adds the options
2596 sieve_useraddress and sieve_subaddress to the redirect router.
2598 PH/30 In these circumstances:
2599 . Two addresses routed to the same list of hosts;
2600 . First host does not offer TLS;
2601 . First host accepts first address;
2602 . First host gives temporary error to second address;
2603 . Second host offers TLS and a TLS session is established;
2604 . Second host accepts second address.
2605 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2606 and peerdn, if requested) that were in fact used only for the second
2609 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2610 attention to any local part prefix or suffix that was matched by the
2611 router that accepted the address. It now behaves in the same way as it
2612 does for delivery: the affixes are removed from the local part unless
2613 rcpt_include_affixes is set on the transport.
2615 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2616 timeout during the DATA phase of an incoming message.
2618 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2619 applied a patch sanctioned by the Sieve maintainer.
2621 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2622 the uid or gid is negative. A case of a negative gid caused this to be
2623 noticed. The fix allows for either to be negative.
2625 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2626 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2629 PH/36 The change PH/12 above was broken. Fixed it.
2631 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2632 the grounds that routing the same address twice would always produce the
2633 same answer. This might have been true once, but it is certainly no
2634 longer true now. Routing a child address may depend on the previous
2635 routing that produced that child. Some complicated redirection strategies
2636 went wrong when messages had multiple recipients, and made Exim's
2637 behaviour dependent on the order in which the addresses were given.
2639 I have moved the duplicate checking until after the routing is complete.
2640 Exim scans the addresses that are assigned to local and remote
2641 transports, and removes any duplicates. This means that more work will be
2642 done, as duplicates will always all be routed, but duplicates are
2643 presumably rare, so I don't expect this is of any significance.
2645 For deliveries to pipes, files, and autoreplies, the duplicate checking
2646 still happens during the routing process, since they are not going to be
2649 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2650 It corrects a timeout issue with spamd. This is Ian's comment: "The
2651 background is that sometimes spamd either never reads data from a
2652 connection it has accepted, or it never writes response data. The exiscan
2653 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2654 blindly assumes that writes won't block so it may never time out."
2656 PH/39 Allow G after quota size as well as K and M.
2658 PH/40 The value set for $authenticated_id in an authenticator may not contain
2659 binary zeroes or newlines because the value is written to log lines and
2660 to spool files. There was no check on this. Now the value is run through
2661 the string_printing() function so that such characters are converted to
2662 printable escape sequences.
2664 PH/41 $message_linecount is a new variable that contains the total number of
2665 lines in the message. Compare $body_linecount, which is the count for the
2668 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2669 bounce and delay warning messages, except in certain special cases, which
2672 (a) An SMTP error message from a remote host;
2673 (b) A message specified in a :fail: redirection;
2674 (c) A message specified in a "fail" command in a system filter;
2675 (d) A message specified in a FAIL return from the queryprogram router;
2676 (e) A message specified by the cannot_route_message router option.
2678 In these cases only, Exim does include the error details in bounce and
2679 warning messages. There are also a few cases where bland messages such
2680 as "unrouteable address" or "local delivery error" are given.
2682 PH/43 $value is now also set for the "else" part of a ${run expansion.
2684 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2685 being worked on, but at least Exim now implements the latest version to
2688 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2689 process to complete was treated as a delivery failure, a timeout while
2690 writing the message to the pipe was logged, but erroneously treated as a
2691 successful delivery. Such timeouts include transport filter timeouts. For
2692 consistency with the overall process timeout, these timeouts are now
2693 treated as errors, giving rise to delivery failures by default. However,
2694 there is now a new Boolean option for the pipe transport called
2695 timeout_defer, which, if set TRUE, converts the failures into defers for
2696 both kinds of timeout. A transport filter timeout is now identified in
2699 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2700 systems where "make" and "gmake" are different, calling "gmake" at top
2701 level broke things. I've arranged for the value of $(MAKE) to be passed
2702 from the Makefile to this script so that it can call the same version of
2706 A note about Exim versions 4.44 and 4.50
2707 ----------------------------------------
2709 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2710 changes of various kinds. As a consequence, a big documentation update was
2711 needed. This delayed the release for rather longer than seemed good, especially
2712 in the light of a couple of (minor) security issues. Therefore, the changes
2713 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2714 release. So 4.44 and 4.50 are in effect two different branches that both start
2717 I have left the 4.50 change log unchanged; it contains all the changes since
2718 4.43. The change log for 4.44 is below; many of its items are identical to
2719 those for 4.50. This seems to be the most sensible way to preserve the
2720 historical information.
2726 1. Minor wording change to the doc/README.SIEVE file.
2728 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2729 computation of the current number of files was incorrect.
2731 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2732 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2735 4. Give more explanation in the error message when the command for a transport
2736 filter fails to execute.
2738 5. There are several places where Exim runs a non-Exim command in a
2739 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2740 was being done only for the command run by the queryprogram router. It is
2741 now done for all such subprocesses. The other cases are: ${run, transport
2742 filters, and the commands run by the lmtp and pipe transports.
2744 6. Added CONFIGURE_GROUP build-time option.
2746 7. Some older OS have a limit of 256 on the maximum number of file
2747 descriptors. Exim was using setrlimit() to set 1000 as a large value
2748 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2749 systems. I've change it so that if it can't get 1000, it tries for 256.
2751 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2752 was an oversight, and furthermore, ever since the addition of extra
2753 controls (e.g. 4.43/32), the checks on when to allow different forms of
2754 "control" were broken. There should now be diagnostics for all cases when a
2755 control that does not make sense is encountered.
2757 9. Added the /retain_sender option to "control=submission".
2759 10. $recipients is now available in the predata ACL (oversight).
2761 11. Tidy the search cache before the fork to do a delivery from a message
2762 received from the command line. Otherwise the child will trigger a lookup
2763 failure and thereby defer the delivery if it tries to use (for example) a
2764 cached ldap connection that the parent has called unbind on.
2766 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2767 of $address_data from the recipient verification was clobbered by the
2768 sender verification.
2770 13. The value of address_data from a sender verification is now available in
2771 $sender_address_data in subsequent conditions in the ACL statement.
2773 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2775 15. Added a new option "connect=<time>" to callout options, to set a different
2778 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2779 was its contents. (It was OK if the option was not defined at all.)
2781 17. A "Completed" log line is now written for messages that are removed from
2782 the spool by the -Mrm option.
2784 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2785 information about exactly what failed.
2787 19. Added -dd to debug only the daemon process.
2789 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2790 handles timeouts, both on the server side and network timeouts. Renamed the
2791 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2793 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2794 It is now set to "smtps".
2796 22. $host_address is now set to the target address during the checking of
2797 ignore_target_hosts.
2799 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2800 being passed; this would have caused $sender_host_name to have been used if
2801 matching the list had actually called for a host name (not very likely,
2802 since this list is usually IP addresses). A host name is now passed as
2805 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2806 code that specifies a non-restarting handler (typically sigaction() in
2807 modern systems) in an attempt to fix a rare and obscure crash bug.
2809 25. Narrowed the window for a race in the daemon that could cause it to ignore
2810 SIGCHLD signals. This is not a major problem, because they are used only to
2811 wake it up if nothing else does.
2813 26. A malformed maildirsize file could cause Exim to calculate negative values
2814 for the mailbox size or file count. Odd effects could occur as a result.
2815 The maildirsize information is now recalculated if the size or filecount
2818 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2819 support for a long time. Removed HAVE_SYS_VFS_H.
2821 28. Installed the latest version of exipick from John Jetmore.
2823 29. In an address list, if the pattern was not a regular expression, an empty
2824 subject address (from a bounce message) matched only if the pattern was an
2825 empty string. Non-empty patterns were not even tested. This was the wrong
2826 because it is perfectly reasonable to use an empty address as part of a
2827 database query. An empty address is now tested by patterns that are
2828 lookups. However, all the other forms of pattern expect the subject to
2829 contain a local part and a domain, and therefore, for them, an empty
2830 address still always fails if the pattern is not itself empty.
2832 30. Exim went into a mad DNS loop when attempting to do a callout where the
2833 host was specified on an smtp transport, and looking it up yielded more
2834 than one IP address.
2836 31. Re-factored the code for checking spool and log partition space into a
2837 function that finds that data and another that does the check. The former
2838 is then used to implement four new variables: $spool_space, $log_space,
2839 $spool_inodes, and $log_inodes.
2841 32. The RFC2047 encoding function was originally intended for short strings
2842 such as real names; it was not keeping to the 75-character limit for
2843 encoded words that the RFC imposes. It now respects the limit, and
2844 generates multiple encoded words if necessary. To be on the safe side, I
2845 have increased the buffer size for the ${rfc2047: expansion operator from
2848 33. It is now permitted to omit both strings after an "if" condition; if the
2849 condition is true, the result is "true". As before, when the second string
2850 is omitted, a false condition yields an empty string. This makes it less
2851 cumbersome to write custom ACL and router conditions.
2853 34. Failure to deliver a bounce message always caused it to be frozen, even if
2854 there was an errors_to setting on the router. The errors_to setting is now
2857 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2858 canonical form (fully expanded) before being placed in
2859 $sender_host_address.
2861 36. The table in the code that translates DNS record types into text (T_A to
2862 "A" for instance) was missing entries for NS and CNAME. It is just possible
2863 that this could have caused confusion if both these types were looked up
2864 for the same domain, because the text type is used as part of Exim's
2865 per-process caching. But the chance of anyone hitting this buglet seems
2868 37. The dnsdb lookup has been extended in a number of ways.
2870 (1) There is a new type, "zns", which walks up the domain tree until it
2871 finds some nameserver records. It should be used with care.
2873 (2) There is a new type, "mxh", which is like "mx" except that it returns
2874 just the host names, not the priorities.
2876 (3) It is now possible to give a list of domains (or IP addresses) to be
2877 looked up. The behaviour when one of the lookups defers can be
2878 controlled by a keyword.
2880 (4) It is now possible to specify the separator character for use when
2881 multiple records are returned.
2883 38. The dnslists ACL condition has been extended: it is now possible to supply
2884 a list of IP addresses and/or domains to be looked up in a particular DNS
2887 39. Added log_selector=+queue_time_overall.
2889 40. When running the queue in the test harness, wait just a tad after forking a
2890 delivery process, to get repeatability of debugging output.
2892 41. Include certificate and key file names in error message when GnuTLS fails
2893 to set them up, because the GnuTLS error message doesn't include the name
2894 of the failing file when there is a problem reading it.
2896 42. Allow both -bf and -bF in the same test run.
2898 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2900 44. The "Exiscan patch" is now merged into the mainline Exim source.
2902 45. Sometimes the final signoff response after QUIT could fail to get
2903 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2904 < 0 before doing a fflush(). This bug looks as though it goes back to the
2905 introduction of TLS in release 3.20, but "sometimes" must have been rare
2906 because the tests only now provoked it.
2908 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2909 (this can affect the format of dates).
2911 47. exim_tidydb, when checking for the continued existence of a message for
2912 which it has found a message-specific retry record, was not finding
2913 messages that were in split spool directories. Consequently, it was
2914 deleting retry records that should have stayed in existence.
2916 48. Steve fixed some bugs in eximstats.
2918 49. The SPA authentication driver was not abandoning authentication and moving
2919 on to the next authenticator when an expansion was forced to fail,
2920 contradicting the general specification for all authenticators. Instead it
2921 was generating a temporary error. It now behaves as specified.
2923 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2924 (the order specifies the preference for clients). The order is now AES256,
2925 AES128, 3DES, ARCFOUR128.
2927 51. Small patch to Sieve code - explicitly set From: when generating an
2930 52. Exim crashed if a remote delivery caused a very long error message to be
2931 recorded - for instance if somebody sent an entire SpamAssassin report back
2932 as a large number of 550 error lines. This bug was coincidentally fixed by
2933 increasing the size of one of Exim's internal buffers (big_buffer) that
2934 happened as part of the Exiscan merge. However, to be on the safe side, I
2935 have made the code more robust (and fixed the comments that describe what
2938 53. Now that there can be additional text after "Completed" in log lines (if
2939 the queue_time_overall log selector is set), a one-byte patch to exigrep
2940 was needed to allow it to recognize "Completed" as not the last thing in
2943 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2944 patch that reportedly fixes this has been added. I am not expert enough to
2945 create a test for it. This is what the patch creator wrote:
2947 "I found a little strange behaviour of ldap code when working with
2948 Windows 2003 AD Domain, where users was placed in more than one
2949 Organization Units. When I tried to give exim partial DN, the exit code
2950 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2951 But simultaneously result of request was absolutely normal ldap result,
2952 so I produce this patch..."
2954 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2955 REFERENCE, so I have modified the code to exclude the patch when that macro
2958 55. Some experimental protocols are using DNS PTR records for new purposes. The
2959 keys for these records are domain names, not reversed IP addresses. The
2960 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2961 leaves it alone. Component reversal etc. now happens only for IP addresses.
2964 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2966 57. Double the size of the debug message buffer (to 2048) so that more of very
2967 long debug lines gets shown.
2969 58. The exicyclog utility now does better if the number of log files to keep
2970 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2972 59. Two changes related to the smtp_active_hostname option:
2974 (1) $smtp_active_hostname is now available as a variable.
2975 (2) The default for smtp_banner uses $smtp_active_hostname instead
2976 of $primary_hostname.
2978 60. The host_aton() function is supposed to be passed a string that is known
2979 to be a valid IP address. However, in the case of IPv6 addresses, it was
2980 not checking this. This is a hostage to fortune. Exim now panics and dies
2981 if the condition is not met. A case was found where this could be provoked
2982 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2983 components; fortuitously, this particular loophole had already been fixed
2984 by change 4.50/55 above.
2986 If there are any other similar loopholes, the new check in host_aton()
2987 itself should stop them being exploited. The report I received stated that
2988 data on the command line could provoke the exploit when Exim was running as
2989 exim, but did not say which command line option was involved. All I could
2990 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2991 running as the user.
2994 61. There was a buffer overflow vulnerability in the SPA authentication code
2995 (which came originally from the Samba project). I have added a test to the
2996 spa_base64_to_bits() function which I hope fixes it.
2999 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3000 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3002 63. The daemon start-up calls getloadavg() while still root for those OS that
3003 need the first call to be done as root, but it missed one case: when
3004 deliver_queue_load_max is set with deliver_drop_privilege. This is
3005 necessary for the benefit of the queue runner, because there is no re-exec
3006 when deliver_drop_privilege is set.
3008 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3009 This has been fixed.
3011 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3012 was in use, was not putting the data itself into the right store pool;
3013 consequently, it could be overwritten for a subsequent message in the same
3014 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3017 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3019 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3020 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3021 the function were treating the return as a boolean value, which happened to
3022 work because 0=false and not-0=true, but is not correct code.
3024 68. The host_aton() function was not handling scoped IPv6 addresses (those
3025 with, for example, "%eth0" on the end) correctly.
3027 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3028 negated items (that is, ~something) in unsigned ints. Some compilers
3029 apparently mutter when there is no cast.
3031 70. If an address verification called from an ACL failed, and did not produce a
3032 user-specific message (i.e. there was only a "system" message), nothing was
3033 put in $acl_verify_message. In this situation, it now puts the system
3036 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3037 session; change 4.31/43 added the unwanted input to the log line - except
3038 that it did not do this in the start of session case. It now does.
3040 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3041 This is wrong and can cause the other end to generate a synchronization
3042 error if it is another Exim or anything else that does the synchronization
3043 check. A QUIT command is no longer sent after a timeout.
3045 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3046 during host lookups.
3048 74. The defer_ok option of callout verification was not working if it was used
3049 when verifying addresses in header lines, that is, for this case:
3051 verify = header_sender/callout=defer_ok
3053 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3054 those file descriptors could be used for SMTP connections. If anything
3055 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3056 could be sent to the SMTP client, causing chaos. The daemon now opens
3057 stdin, stdout, and stderr to /dev/null when it puts itself into the
3060 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3061 log by default. The user can override this with suitable Perl magic.
3063 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3064 the log message when discard triggers, was not working for the DATA ACL or
3065 for the non-SMTP ACL.
3067 78. Error message wording change in sieve.c.
3069 79. If smtp_accept_max_per_host was set, the number of connections could be
3070 restricted to fewer than expected, because the daemon was trying to set up
3071 a new connection before checking whether the processes handling previous
3072 connections had finished. The check for completed processes is now done
3073 earlier. On busy systems, this bug wouldn't be noticed because something
3074 else would have woken the daemon, and it would have reaped the completed
3077 80. If a message was submitted locally by a user whose login name contained one
3078 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3079 It caused a spool format error. I have fixed the spool reading code. A
3080 related problem was that the "from" clause in the Received: line became
3081 illegal because of the space(s). It is now covered by ${quote_local_part.
3083 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3086 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3088 83. Patch from Sieve maintainer to fix unterminated string problem in
3089 "vacation" handling.
3091 84. Some minor changes to the Linux configuration files to help with other
3092 OS variants using glibc.
3094 85. One more patch for Sieve to update vacation handling to latest spec.
3097 ----------------------------------------------------
3098 See the note above about the 4.44 and 4.50 releases.
3099 ----------------------------------------------------
3105 1. Change 4.43/35 introduced a bug that caused file counts to be
3106 incorrectly computed when quota_filecount was set in an appendfile
3109 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3110 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3113 3. Give more explanation in the error message when the command for a transport
3114 filter fails to execute.
3116 4. There are several places where Exim runs a non-Exim command in a
3117 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3118 was being done only for the command run by the queryprogram router. It is
3119 now done for all such subprocesses. The other cases are: ${run, transport
3120 filters, and the commands run by the lmtp and pipe transports.
3122 5. Some older OS have a limit of 256 on the maximum number of file
3123 descriptors. Exim was using setrlimit() to set 1000 as a large value
3124 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3125 systems. I've change it so that if it can't get 1000, it tries for 256.
3127 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3128 was an oversight, and furthermore, ever since the addition of extra
3129 controls (e.g. 4.43/32), the checks on when to allow different forms of
3130 "control" were broken. There should now be diagnostics for all cases when a
3131 control that does not make sense is encountered.
3133 7. $recipients is now available in the predata ACL (oversight).
3135 8. Tidy the search cache before the fork to do a delivery from a message
3136 received from the command line. Otherwise the child will trigger a lookup
3137 failure and thereby defer the delivery if it tries to use (for example) a
3138 cached ldap connection that the parent has called unbind on.
3140 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3141 of $address_data from the recipient verification was clobbered by the
3142 sender verification.
3144 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3145 was its contents. (It was OK if the option was not defined at all.)
3147 11. A "Completed" log line is now written for messages that are removed from
3148 the spool by the -Mrm option.
3150 12. $host_address is now set to the target address during the checking of
3151 ignore_target_hosts.
3153 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3154 being passed; this would have caused $sender_host_name to have been used if
3155 matching the list had actually called for a host name (not very likely,
3156 since this list is usually IP addresses). A host name is now passed as
3159 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3160 code that specifies a non-restarting handler (typically sigaction() in
3161 modern systems) in an attempt to fix a rare and obscure crash bug.
3163 15. Narrowed the window for a race in the daemon that could cause it to ignore
3164 SIGCHLD signals. This is not a major problem, because they are used only to
3165 wake it up if nothing else does.
3167 16. A malformed maildirsize file could cause Exim to calculate negative values
3168 for the mailbox size or file count. Odd effects could occur as a result.
3169 The maildirsize information is now recalculated if the size or filecount
3172 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3173 support for a long time. Removed HAVE_SYS_VFS_H.
3175 18. Updated exipick to current release from John Jetmore.
3177 19. Allow an empty sender to be matched against a lookup in an address list.
3178 Previously the only cases considered were a regular expression, or an
3181 20. Exim went into a mad DNS lookup loop when doing a callout where the
3182 host was specified on the transport, if the DNS lookup yielded more than
3185 21. The RFC2047 encoding function was originally intended for short strings
3186 such as real names; it was not keeping to the 75-character limit for
3187 encoded words that the RFC imposes. It now respects the limit, and
3188 generates multiple encoded words if necessary. To be on the safe side, I
3189 have increased the buffer size for the ${rfc2047: expansion operator from
3192 22. Failure to deliver a bounce message always caused it to be frozen, even if
3193 there was an errors_to setting on the router. The errors_to setting is now
3196 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3197 canonical form (fully expanded) before being placed in
3198 $sender_host_address.
3200 24. Updated eximstats to version 1.33
3202 25. Include certificate and key file names in error message when GnuTLS fails
3203 to set them up, because the GnuTLS error message doesn't include the name
3204 of the failing file when there is a problem reading it.
3206 26. Expand error message when OpenSSL has problems setting up cert/key files.
3209 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3210 (this can affect the format of dates).
3212 28. exim_tidydb, when checking for the continued existence of a message for
3213 which it has found a message-specific retry record, was not finding
3214 messages that were in split spool directories. Consequently, it was
3215 deleting retry records that should have stayed in existence.
3217 29. eximstats updated to version 1.35
3218 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3219 1.35 - bugfix such that pie charts by volume are generated correctly
3221 30. The SPA authentication driver was not abandoning authentication and moving
3222 on to the next authenticator when an expansion was forced to fail,
3223 contradicting the general specification for all authenticators. Instead it
3224 was generating a temporary error. It now behaves as specified.
3226 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3227 (the order specifies the preference for clients). The order is now AES256,
3228 AES128, 3DES, ARCFOUR128.
3230 31. Small patch to Sieve code - explicitly set From: when generating an
3233 32. Exim crashed if a remote delivery caused a very long error message to be
3234 recorded - for instance if somebody sent an entire SpamAssassin report back
3235 as a large number of 550 error lines. This bug was coincidentally fixed by
3236 increasing the size of one of Exim's internal buffers (big_buffer) that
3237 happened as part of the Exiscan merge. However, to be on the safe side, I
3238 have made the code more robust (and fixed the comments that describe what
3241 33. Some experimental protocols are using DNS PTR records for new purposes. The
3242 keys for these records are domain names, not reversed IP addresses. The
3243 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3244 leaves it alone. Component reversal etc. now happens only for IP addresses.
3247 34. The host_aton() function is supposed to be passed a string that is known
3248 to be a valid IP address. However, in the case of IPv6 addresses, it was
3249 not checking this. This is a hostage to fortune. Exim now panics and dies
3250 if the condition is not met. A case was found where this could be provoked
3251 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3252 components; fortuitously, this particular loophole had already been fixed
3253 by change 4.50/55 or 4.44/33 above.
3255 If there are any other similar loopholes, the new check in host_aton()
3256 itself should stop them being exploited. The report I received stated that
3257 data on the command line could provoke the exploit when Exim was running as
3258 exim, but did not say which command line option was involved. All I could
3259 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3260 running as the user.
3263 35. There was a buffer overflow vulnerability in the SPA authentication code
3264 (which came originally from the Samba project). I have added a test to the
3265 spa_base64_to_bits() function which I hope fixes it.
3268 36. The daemon start-up calls getloadavg() while still root for those OS that
3269 need the first call to be done as root, but it missed one case: when
3270 deliver_queue_load_max is set with deliver_drop_privilege. This is
3271 necessary for the benefit of the queue runner, because there is no re-exec
3272 when deliver_drop_privilege is set.
3274 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3275 was in use, was not putting the data itself into the right store pool;
3276 consequently, it could be overwritten for a subsequent message in the same
3277 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3280 38. Sometimes the final signoff response after QUIT could fail to get
3281 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3282 < 0 before doing a fflush(). This bug looks as though it goes back to the
3283 introduction of TLS in release 3.20, but "sometimes" must have been rare
3284 because the tests only now provoked it.
3290 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3291 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3292 Now it returns an integer. A number of calls were still expecting a T/F
3293 return. Fortuitously, in all cases, the tests worked in OK situations,
3294 which is the norm. However, things would have gone wrong on any write
3295 failures on the smtp file descriptor. This function is used when sending
3296 messages over SMTP and also when doing verify callouts.
3298 2. When Exim is called to do synchronous delivery of a locally submitted
3299 message (the -odf or -odi options), it no longer closes stderr before doing
3302 3. Implemented the mua_wrapper option.
3304 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3306 5. Implemented the functions header_remove(), header_testname(),
3307 header_add_at_position(), and receive_remove_recipient(), and exported them
3310 6. If an ACL "warn" statement specified the addition of headers, Exim already
3311 inserted X-ACL-Warn: at the start if there was no header name. However, it
3312 was not making this test for the second and subsequent header lines if
3313 there were newlines in the string. This meant that an invalid header could
3314 be inserted if Exim was badly configured.
3316 7. Allow an ACL "warn" statement to add header lines at the start or after all
3317 the Received: headers, as well as at the end.
3319 8. Added the rcpt_4xx retry error code.
3321 9. Added postmaster_mailfrom=xxx to callout verification option.
3323 10. Added mailfrom=xxxx to the callout verification option, for verify=
3326 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3327 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3329 12. Inserted some casts to stop certain compilers warning when using pointer
3330 differences as field lengths or precisions in printf-type calls (mostly
3331 affecting debugging statements).
3333 13. Added optional readline() support for -be (dynamically loaded).
3335 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3336 same clock tick as a message's arrival, so that its received time was the
3337 same as the "first fail" time on the retry record, and that message
3338 remained on the queue past the ultimate address timeout, every queue runner
3339 would try a delivery (because it was past the ultimate address timeout) but
3340 after another failure, the ultimate address timeout, which should have then
3341 bounced the address, did not kick in. This was a "< instead of <=" error;
3342 in most cases the first failure would have been in the next clock tick
3343 after the received time, and all would be well.
3345 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3346 being recognized when the domain list was tested by the match_domain
3347 condition in an expansion string.
3349 16. Added the ${str2b64: operator.
3351 17. Exim was always calling setrlimit() to set a large limit for the number of
3352 processes, without checking whether the existing limit was already
3353 adequate. (It did check for the limit on file descriptors.) Furthermore,
3354 errors from getrlimit() and setrlimit() were being ignored. Now they are
3355 logged to the main and panic logs, but Exim does carry on, to try to do its
3356 job under whatever limits there are.
3358 18. Imported PCRE 5.0.
3360 19. Trivial typo in log message " temporarily refused connection" (the leading
3363 20. If the log selector return_path_on_delivery was set and an address was
3364 redirected to /dev/null, the delivery process crashed because it assumed
3365 that a return path would always be set for a "successful" delivery. In this
3366 case, the whole delivery is bypassed as an optimization, and therefore no
3369 21. Internal re-arrangement: the function for sending a challenge and reading
3370 a response while authentication was assuming a zero-terminated challenge
3371 string. It's now changed to take a pointer and a length, to allow for
3372 binary data in such strings.
3374 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3376 23. Exim was not respecting finduser_retries when seeking the login of the
3377 uid under which it was called; it was always trying 10 times. (The default
3378 setting of finduser_retries is zero.) Also, it was sleeping after the final
3379 failure, which is pointless.
3381 24. Implemented tls_on_connect_ports.
3383 25. Implemented acl_smtp_predata.
3385 26. If the domain in control=submission is set empty, Exim assumes that the
3386 authenticated id is a complete email address when it generates From: or
3387 Sender: header lines.
3389 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3390 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3391 chown and chgrp in /bin and hostname in /usr/bin.
3393 28. Exim was keeping the "process log" file open after each use, just as it
3394 does for the main log. This opens the possibility of it remaining open for
3395 long periods when the USR1 signal hits a daemon. Occasional processlog
3396 errors were reported, that could have been caused by this. Anyway, it seems
3397 much more sensible not to leave this file open at all, so that is what now
3400 29. The long-running daemon process does not normally write to the log once it
3401 has entered its main loop, and it closes the log before doing so. This is
3402 so that log files can straightforwardly be renamed and moved. However,
3403 there are a couple of unusual error situations where the daemon does write
3404 log entries, and I had neglected to close the log afterwards.
3406 30. The text of an SMTP error response that was received during a remote
3407 delivery was being truncated at 512 bytes. This is too short for some of
3408 the long messages that one sometimes sees. I've increased the limit to
3411 31. It is now possible to make retry rules that apply only when a message has a
3412 specific sender, in particular, an empty sender.
3414 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3415 it possible to be selective about when SMTP synchronization is enforced.
3417 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3419 32. Implemented hosts_connection_nolog.
3421 33. Added an ACL for QUIT.
3423 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3426 35. Added mailbox_size and mailbox_filecount to appendfile.
3428 36. Added control = no_multiline_responses to ACLs.
3430 37. There was a bug in the logic of the code that waits for the clock to tick
3431 in the case where the clock went backwards by a substantial amount such
3432 that the microsecond fraction of "now" was more than the microsecond
3433 fraction of "then" (but the whole seconds number was less).
3435 38. Added support for the libradius Radius client library this is found on
3436 FreeBSD (previously only the radiusclient library was supported).
3442 1. When certain lookups returned multiple values in the form name=value, the
3443 quoting of the values was not always being done properly. Specifically:
3444 (a) If the value started with a double quote, but contained no whitespace,
3446 (b) If the value contained whitespace other than a space character (i.e.
3447 tabs or newlines or carriage returns) it was not quoted.
3448 This fix has been applied to the mysql and pgsql lookups by writing a
3449 separate quoting function and calling it from the lookup code. The fix
3450 should probably also be applied to nisplus, ibase and oracle lookups, but
3451 since I cannot test any of those, I have not disturbed their existing code.
3453 2. A hit in the callout cache for a specific address caused a log line with no
3454 reason for rejecting RCPT. Now it says "Previous (cached) callout
3455 verification failure".
3457 3. There was an off-by-one bug in the queryprogram router. An over-long
3458 return line was truncated at 256 instead of 255 characters, thereby
3459 overflowing its buffer with the terminating zero. As well as fixing this, I
3460 have increased the buffer size to 1024 (and made a note to document this).
3462 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3463 when Exim is waiting for an SMTP response from a remote server, Exim
3464 restarts its select() call on the socket, thereby resetting its timeout.
3465 This is not a problem when such interrupts are rare. Somebody set up a cron
3466 job to run exiwhat every 2 minutes, which is less than the normal select()
3467 timeout (5 or 10 minutes). This meant that the select() timeout never
3468 kicked in because it was always reset. I have fixed this by comparing the
3469 time when an interrupt arrives with the time at the start of the first call
3470 to select(). If more time than the timeout has elapsed, the interrupt is
3471 treated as a timeout.
3473 5. Some internal re-factoring in preparation for the addition of Sieve
3474 extensions (by MH). In particular, the "personal" test is moved to a
3475 separate function, and given an option for scanning Cc: and Bcc: (which is
3476 not set for Exim filters).
3478 6. When Exim created an email address using the login of the caller as the
3479 local part (e.g. when creating a From: or Sender: header line), it was not
3480 quoting the local part when it contained special characters such as @.
3482 7. Installed new OpenBSD configuration files.
3484 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3485 try to make them clearer.
3487 9. Callout options, other than the timeout value, were being ignored when
3488 verifying sender addresses in header lines. For example, when using
3490 verify = header_sender/callout=no_cache
3492 the cache was (incorrectly) being used.
3494 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3495 only the Cygwin environment.
3497 11. When return_path_on_delivery was set as a log selector, if different remote
3498 addresses in the same message used different return paths and parallel
3499 remote delivery occurred, the wrong values would sometimes be logged.
3500 (Whenever a remote delivery process finished, the return path value from
3501 the most recently started remote delivery process was logged.)
3503 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3504 lines when AUTH and/or TLS are in use. This is the "received protocol"
3505 field. Exim used to use "asmtp" for authenticated SMTP, without any
3506 indication (in the protocol name) for TLS use. Now it follows the RFC and
3507 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3508 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3509 names appear in log lines as well as in Received: header lines.
3511 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3512 extensions, and comparison tests, and to fix some bugs.
3514 14. Changes to the "personal" filter test:
3516 (1) The test was buggy in that it was just doing the equivalent of
3517 "contains" tests on header lines. For example, if a user's address was
3518 anne@some.where, the "personal" test would incorrectly be true for
3520 To: susanne@some.where
3522 This test is now done by extracting each address from the header in turn,
3523 and checking the entire address. Other tests that are part of "personal"
3524 are now done using regular expressions (for example, to check local parts
3525 of addresses in From: header lines).
3527 (2) The list of non-personal local parts in From: addresses has been
3528 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3529 taken from the Sieve specification recommendations.
3531 (3) If the message contains any header line starting with "List-" it is
3532 treated as non-personal.
3534 (4) The test for "circular" in the Subject: header line has been removed
3535 because it now seems ill-conceived.
3537 15. Minor typos in src/EDITME comments corrected.
3539 16. Installed latest exipick from John Jetmore.
3541 17. If headers_add on a router specified a text string that was too long for
3542 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3543 of string_sprintf() is now avoided.
3545 18. $message_body_size was not set (it was always zero) when running the DATA
3546 ACL and the local_scan() function.
3548 19. For the "mail" command in an Exim filter, no default was being set for
3549 the once_repeat time, causing a random time value to be used if "once" was
3550 specified. (If the value happened to be <= 0, no repeat happened.) The
3551 default is now 0s, meaning "never repeat". The "vacation" command was OK
3552 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3553 (I found it when inspecting the code).
3555 20. There is now an overall timeout for performing a callout verification. It
3556 defaults to 4 times the callout timeout, which applies to individual SMTP
3557 commands during the callout. The overall timeout applies when there is more
3558 than one host that can be tried. The timeout is checked before trying the
3559 next host. This prevents very long delays if there are a large number of
3560 hosts and all are timing out (e.g. when the network connections are timing
3561 out). The value of the overall timeout can be changed by specifying an
3562 additional sub-option for "callout", called "maxwait". For example:
3564 verify = sender/callout=5s,maxwait=20s
3566 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3567 to the end before writing, but this should make it even safer).
3569 22. Exim was forgetting that it had advertised PIPELINING for the second and
3570 subsequent messages on an SMTP connection. It was also not resetting its
3571 memory on STARTTLS and an internal HELO.
3573 23. When Exim logs an SMTP synchronization error within a session, it now
3574 records whether PIPELINING has been advertised or not.
3576 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3577 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3578 rather than long int.
3580 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3582 26. Added the never_mail option to autoreply.
3588 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3589 crash if the getsockname() call failed; this can happen if a connection is
3590 closed very soon after it is established. The problem was simply in the
3591 order in which certain operations were done, causing Exim to try to write
3592 to the SMTP stream before it had set up the file descriptor. The bug has
3593 been fixed by making things happen in the correct order.
3599 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3600 before the connection was closed, thus losing the rejection response.
3602 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3603 some early Solaris releases, but causes trouble in current releases where
3604 socklen_t is defined.
3606 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3609 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3612 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3613 that releasing the top of it at the end releases what was used for sub-
3614 expansions (unless the block got too big). However, discard this block if
3615 the first thing is a variable or header, so that we can use its block when
3616 it is dynamic (useful for very large $message_headers, for example).
3618 6. Lookups now cache *every* query, not just the most recent. A new, separate
3619 store pool is used for this. It can be recovered when all lookup caches are
3620 flushed. Lookups now release memory at the end of their result strings.
3621 This has involved some general refactoring of the lookup sources.
3623 7. Some code has been added to the store_xxx() functions to reduce the amount
3624 of flapping under certain conditions.
3626 8. log_incoming_interface used to affect only the <= reception log lines. Now
3627 it causes the local interface and port to be added to several more SMTP log
3628 lines, for example "SMTP connection from", and rejection lines.
3630 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3632 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3634 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3635 could be overwritten at the end of the current message (or the start of a
3636 new message if it was set in a HELO ACL). The value is now preserved for
3637 the duration of the SMTP connection.
3639 12. If a transport had a headers_rewrite setting, and a matching header line
3640 contained an unqualified address, that address was qualified, even if it
3641 did not match any rewriting rules. The underlying bug was that the values
3642 of the flags that permit the existence of unqualified sender and recipient
3643 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3644 non-local messages, and by -bnq for local messages) were not being
3645 preserved with the message after it was received.
3647 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3648 "next input=" as part of the text comprising the host identity instead of
3649 the correct text. The code was using the same buffer for two different
3650 strings. However, depending on which order the printing function evaluated
3651 its arguments, the bug did not always show up. Under Linux, for example, my
3652 test suite worked just fine.
3654 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3655 doesn't work with some older versions of Perl. It has been changed to "my",
3656 which in any case is probably the better facility to use.
3658 15. A really picky compiler found some instances of statements for creating
3659 error messages that either had too many or two few arguments for the format
3662 16. The size of the buffer for calls to the DNS resolver has been increased
3663 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3664 for addresses that have a lot of PTR records. This alleviates a problem; it
3665 does not fully solve it.
3667 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3668 buffer now truncates the list and logs the incident, which is the same
3669 action as happens when Exim is looking up a host name and its aliases.
3670 Previously in this situation something unpredictable would happen;
3671 sometimes it was "internal error: store_reset failed".
3673 18. If a server dropped the connection unexpectedly when an Exim client was
3674 using GnuTLS and trying to read a response, the client delivery process
3675 crashed while trying to generate an error log message.
3677 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3678 string, for example:
3680 warn message = H1: something\nH2: something
3682 the text was added as a single header line from Exim's point of view
3683 though it ended up OK in the delivered message. However, searching for the
3684 second and subsequent header lines using $h_h2: did not work. This has been
3685 fixed. Similarly, if a system filter added multiple headers in this way,
3686 the routers could not see them.
3688 20. Expanded the error message when iplsearch is called with an invalid key to
3689 suggest using net-iplsearch in a host list.
3691 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3692 ACLs are no longer actually imposed (and a message to that effect is
3695 22. If a "gecos" field in a passwd entry contained escaped characters, in
3696 particular, if it contained a \" sequence, Exim got it wrong when building
3697 a From: or a Sender: header from that name. A second bug also caused
3698 incorrect handling when an unquoted " was present following a character
3699 that needed quoting.
3701 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3702 was not being matched caselessly.
3704 24. Arranged for all hyphens in the exim.8 source to be escaped with
3707 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3708 properly. Recipient callout cache records were still being keyed to include
3709 the sender, even when use_sender was set false. This led to far more
3710 callouts that were necessary. The sender is no longer included in the key
3711 when use_sender is false.
3713 26. Added "control = submission" modifier to ACLs.
3715 27. Added the ${base62d: operator to decode base 62 numbers.
3717 28. dnsdb lookups can now access SRV records.
3719 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3720 the configuration file.
3722 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3723 (-v) mode. This makes the output for a verbose queue run more intelligible.
3725 31. Added a use_postmaster feature to recipient callouts.
3727 32. Added the $body_zerocount variable, containing the number of binary zero
3728 bytes in the message body.
3730 33. The time of last modification of the "new" subdirectory is now used as the
3731 "mailbox time last read" when there is a quota error for a maildir
3734 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3736 35. Added +ignore_unknown as a special item in host lists.
3738 36. Code for decoding IPv6 addresses in host lists is now included, even if
3739 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3740 address was recognized as an IP address, but was then not correctly decoded
3741 into binary, causing unexpected and incorrect effects when compared with
3748 1. Very minor rewording of debugging text in manualroute to say "list of
3749 hosts" instead of "hostlist".
3751 2. If verify=header_syntax was set, and a header line with an unqualified
3752 address (no domain) and a large number of spaces between the end of the
3753 name and the colon was received, the reception process suffered a buffer
3754 overflow, and (when I tested it) crashed. This was caused by some obsolete
3755 code that should have been removed. The fix is to remove it!
3757 3. When running in the test harness, delay a bit after writing a bounce
3758 message to get a bit more predictability in the log output.
3760 4. Added a call to search_tidyup() just before forking a reception process. In
3761 theory, someone could use a lookup in the expansion of smtp_accept_max_
3762 per_host which, without the tidyup, could leave open a database connection.
3764 5. Added the variables $recipient_data and $sender_data which get set from a
3765 lookup success in an ACL "recipients" or "senders" condition, or a router
3766 "senders" option, similar to $domain_data and $local_part_data.
3768 6. Moved the writing of debug_print from before to after the "senders" test
3771 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3772 problems for message scanning, either using a data ACL, or using
3773 local_scan() because the Received: header was not generated till after they
3774 were called (in order to set the time as the time of reception completion).
3775 I have revised the way this works. The header is now generated after the
3776 body is received, but before the ACL or local_scan() are called. After they
3777 are run, the timestamp in the header is updated.
3783 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3784 before starting a queue runner without re-exec. This happened only when
3785 deliver_drop_privilege was set or when the Exim user was set to root. The
3786 effect of the bug was that timeouts during subsequent deliveries caused
3787 crashes instead of being properly handled. The handler is now left at its
3788 default (and expected) setting.
3790 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3791 message, again when deliver_drop_privilege is set or Exim is run as root.
3792 The bug described in (1) was not present in this case, but the tidying up
3793 of the other signals was missing. I have made the two cases consistent.
3795 3. The ignore_target_hosts setting on a manualroute router was being ignored
3796 for hosts that were looked up using the /MX notation.
3798 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3801 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3802 operated on the sender address. After changing the $sender_address to <>
3803 for the sender address verify, Exim was re-instated it as the original
3804 (before rewriting) address, but remembering that it had rewritten it, so it
3805 wasn't rewriting it again. This bug also had the effect of breaking the
3806 sender address verification caching when the sender address was rewritten.
3808 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3809 This has been changed so that if the ip literal address matches
3810 ignore_target_hosts, the router declines.
3812 7. Added expansion conditions match_domain, match_address, and match_local_
3813 part (NOT match_host).
3815 8. The placeholder for the Received: header didn't have a length field set.
3817 9. Added code to Exim itself and to exim_lock to test for a specific race
3818 condition that could lead to file corruption when using MBX delivery. The
3819 issue is with the lockfile that is created in /tmp. If this file is removed
3820 after a process has opened it but before that process has acquired a lock,
3821 there is the potential for a second process to recreate the file and also
3822 acquire a lock. This could lead to two Exim processes writing to the file
3823 at the same time. The added code performs the same test as UW imapd; it
3824 checks after acquiring the lock that its file descriptor still refers to
3825 the same named file.
3827 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3828 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3831 11. Added the smtp_active_hostname option. If used, this will typically be made
3832 to depend on the incoming interface address. Because $interface_address is
3833 not set up until the daemon has forked a reception process, error responses
3834 that can happen earlier (such as "too many connections") no longer contain
3837 12. If an expansion in a condition on a "warn" statement fails because a lookup
3838 defers, the "warn" statement is abandoned, and the next ACL statement is
3839 processed. Previously this caused the whole ACL to be aborted.
3841 13. Added the iplsearch lookup type.
3843 14. Added ident_timeout as a log selector.
3845 15. Added tls_certificate_verified as a log selector.
3847 16. Added a global option tls_require_ciphers (compare the smtp transport
3848 option of the same name). This controls incoming TLS connections.
3850 17. I finally figured out how to make tls_require_ciphers do a similar thing
3851 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3852 before starting the TLS session.
3854 18. Tabs are now shown as \t in -bP output.
3856 19. If the log selector return_path_on_delivery was set, Exim crashed when
3857 bouncing a message because it had too many Received: header lines.
3859 20. If two routers both had headers_remove settings, and the first one included
3860 a superfluous trailing colon, the final name in the first list and the
3861 first name in the second list were incorrectly joined into one item (with a
3862 colon in the middle).
3868 1. Added -C and -D options to the exinext utility, mainly to make it easier
3869 to include in the automated testing, but these could be helpful when
3870 multiple configurations are in use.
3872 2. The exinext utility was not formatting the output nicely when there was
3873 an alternate port involved in the retry record key, nor when there was a
3874 message id as well (for retries that were specific to a specific message
3875 and a specific host). It was also confused by IPv6 addresses, because of
3876 the additional colons they contain. I have fixed the IPv4 problem, and
3877 patched it up to do a reasonable job for IPv6.
3879 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3880 delivery, the log line now contains "pipelined" if PIPELINING was used.
3882 4. An SMTP transport process used to panic and die if the bind() call to set
3883 an explicit outgoing interface failed. This has been changed; it is now
3884 treated in the same way as a connect() failure.
3886 5. A reference to $sender_host_name in the part of a conditional expansion
3887 that was being skipped was still causing a DNS lookup. This no longer
3890 6. The def: expansion condition was not recognizing references to header lines
3891 that used bh_ and bheader_.
3893 7. Added the _cache feature to named lists.
3895 8. The code for checking quota_filecount in the appendfile transport was
3896 allowing one more file than it should have been.
3898 9. For compatibility with Sendmail, the command line option
3906 and sets the incoming protocol and host name (for trusted callers). The
3907 host name and its colon can be omitted when only the protocol is to be set.
3908 Note the Exim already has two private options, -pd and -ps, that refer to
3909 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3910 "s", but I don't think that's a major issue.
3912 10. A number of refactoring changes to the code, none of which should affect
3915 (a) The number of logging options was getting close to filling up the
3916 32-bit word that was used as a bit map. I have split them into two classes:
3917 those that are passed in the argument to log_write(), and those that are
3918 only ever tested independently outside of that function. These are now in
3919 separate 32-bit words, so there is plenty of room for expansion again.
3920 There is no change in the user interface or the logging behaviour.
3922 (b) When building, for example, log lines, the code previously used a
3923 macro that called string_cat() twice, in order to add two strings. This is
3924 not really sufficiently general. Furthermore, there was one instance where
3925 it was actually wrong because one of the argument was used twice, and in
3926 one call a function was used. (As it happened, calling the function twice
3927 did not affect the overall behaviour.) The macro has been replaced by a
3928 function that can join an arbitrary number of extra strings onto a growing
3931 (c) The code for expansion conditions now uses a table and a binary chop
3932 instead of a serial search (which was left over from when there were very
3933 few conditions). Also, it now recognizes conditions like "pam" even when
3934 the relevant support is not compiled in: a suitably worded error message is
3935 given if an attempt is made to use such a condition.
3937 11. Added ${time_interval:xxxxx}.
3939 12. A bug was causing one of the ddress fields not to be passed back correctly
3940 from remote delivery subprocesses. The field in question was not being
3941 subsequently used, so this caused to problems in practice.
3943 13. Added new log selectors queue_time and deliver_time.
3945 14. Might have fixed a bug in maildirsizefile handling that threw up
3946 "unexpected character" debug warnings, and recalculated the data
3947 unnecessarily. In any case, I expanded the warning message to give more
3950 15. Added the message "Restricted characters in address" to the statements in
3951 the default ACL that block characters like @ and % in local parts.
3953 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3954 Three changes have been made:
3956 (a) There was a serious bug; a negative response to MAIL caused the whole
3957 recipient domain to be cached as invalid, thereby blocking all messages
3958 to all local parts at the same domain, from all senders. This bug has
3959 been fixed. The domain is no longer cached after a negative response to
3960 MAIL if the sender used is not empty.
3962 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3965 (c) A new callout option, "use_sender" has been added for people who want
3966 the modified behaviour.
3972 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3975 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3976 indeed breaks things for older releases.
3978 3. Added additional logging to the case where there is a problem reading data
3979 from a filter that is running in a subprocess using a pipe, in order to
3980 try to track down a specific problem.
3982 4. Testing facility fudge: when running in the test harness and attempting
3983 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3984 getting "No route to host". Convert this to a timeout.
3986 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3989 6. Some OS don't have socklen_t but use size_t instead. This affects the
3990 fifth argument of getsockopt() amongst other things. This is now
3991 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3992 can be set for individual OS. I have set it for SunOS5, OSF1, and
3993 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3994 some earlier ones do not.
3996 7. Change 4.30/15 was not doing the test caselessly.
3998 8. The standard form for an IPv6 address literal was being rejected by address
3999 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4000 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4001 this, as well as the form without the "IPv6" on the front (but only when
4002 address literals are enabled, of course).
4004 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4006 10. Exim crashed if a message with an empty sender address specified by -f
4007 encountered a router with an errors_to setting. This could be provoked only
4008 by a command such as
4012 where an empty string was supplied; "<>" did not hit this bug.
4014 11. Installed PCRE release 4.5.
4016 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4017 remained set. It is now erased.
4019 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4020 times from message ids (which are base 36 rather than the normal 62).
4022 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4023 were being counted as actual protocol errors, and logged if the log
4024 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4025 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4026 and DATA following a set of rejected RCPTs do not count as protocol errors.
4027 In other words, Exim assumes they were pipelined, though this may not
4028 actually be the case. Of course, in all cases the client gets an
4029 appropriate error code.
4031 15. If a lookup fails in an ACL condition, a message about the failure may
4032 be available; it is used if testing the ACL cannot continue, because most
4033 such messages specify what the cause of the deferral is. However, some
4034 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4035 that caused an old message to be retained and used if a later statement
4036 caused a defer, replacing the real cause of the deferral.
4038 16. If an IP address had so many PTR records that the DNS lookup buffer
4039 was not large enough to hold them, Exim could crash while trying to process
4040 the truncated data. It now detects and logs this case.
4042 17. Further to 4.21/58, another change has been made: if (and only if) the
4043 first line of a message (the first header line) ends with CRLF, a bare LF
4044 in a subsequent header line has a space inserted after it, so as not to
4045 terminate the header.
4047 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4048 unnecessarily, used atoi() instead of strtol(), and didn't check the
4049 termination when getting file sizes from file names by regex.
4051 19. Completely re-implemented the support for maildirsize files, in the light
4052 of a number of problems with the previous contributed implementation
4053 (4.30/29). In particular:
4055 . If the quota is zero, the maildirsize file is maintained, but no quota is
4058 . If the maildir directory does not exist, it is created before any attempt
4059 to write a maildirsize file.
4061 . The quota value in the file is just a cache; if the quota is changed in
4062 the transport, the new value overrides.
4064 . A regular expression is available for excluding directories from the
4067 20. The autoreply transport checks the characters in options that define the
4068 message's headers; it allows continued headers, but it was checking with
4069 isspace() after an embedded newline instead of explicitly looking for a
4072 21. If all the "regular" hosts to which an address was routed had passed their
4073 expiry times, and had not reached their retry times, the address was
4074 bounced, even if fallback hosts were defined. Now Exim should go on to try
4077 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4078 equivalent code in the SMTP transport. Some hosts send humungous responses
4079 to HELO/EHLO, more than 1024 it seems.
4081 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4082 gives compiler warnings in some environments. I've now done it "properly",
4085 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4086 (because of problems with the built-in one) was declared to return uschar *
4087 instead of char *, causing compiler failure.
4089 25. Fixed a file descriptor leak when processing alias/forward files.
4091 26. Fixed a minor format string issue in dbfn.c.
4093 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4095 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4096 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4099 29. When a sender address is verified, it is cached, to save repeating the test
4100 when there is more than one recipient in a message. However, when the
4101 verification involves a callout, it is possible for different callout
4102 options to be set for different recipients. It is too complicated to keep
4103 track of this in the cache, so now Exim always runs a verification when a
4104 callout is required, relying on the callout cache for the optimization.
4105 The overhead is duplication of the address routing, but this should not be
4108 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4109 to be verified with callout=postmaster, and the main callout worked but the
4110 postmaster check failed, the verification correctly failed. However, if a
4111 subsequent RCPT command asked for sender verification *without* the
4112 postmaster check, incorrect caching caused this verification also to fail,
4115 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4116 it was not caching the DNS options (qualify_single, search_parents) that
4117 were used when the lookup failed. A subsequent lookup with different
4118 options therefore always gave the same answer, though there were cases
4119 where it should not have. (Example: a "domains = !$mx_any" option on a
4120 dnslookup router: the "domains" option is always processed without any
4121 widening, but the router might have qualify_single set.) Now Exim uses the
4122 cached value only when the same options are set.
4124 32. Added John Jetmore's "exipick" utility to the distribution.
4126 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4127 than a timeout (e.g. a certificate is required, and is not provided), an
4128 Exim server now closes the connection immediately. Previously it waited for
4129 the client to close - but if the client is SSL, it seems that they each
4130 wait for each other, leading to a delay before one of them times out.
4132 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4133 maintained 0.8.x compatibility because I don't think many are using it, and
4134 it is clearly obsolete.
4136 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4139 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4140 client certificate was expired. A simple patch fixes this, though I don't
4141 understand the full logic of why the verify callback is called multiple
4144 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4145 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4146 which causes problems with some clients (such as the Certicom SSL Plus
4147 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4148 disables the coutermeasure allowing Eudora to connect."
4150 38. Exim was not checking that a write() to a log file succeeded. This could
4151 lead to Bad Things if a log got too big, in particular if it hit a file
4152 size limit. Exim now panics and dies if it cannot write to a log file, just
4153 as it does if it cannot open a log file.
4155 39. Modified OS/Makefile-Linux so that it now contains
4157 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4159 The two -D definitions ensure that Exim is compiled with large file
4160 support, which makes it possible to handle log files that are bigger than
4163 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4164 instance) a domain was checked against a named list that involved a lookup,
4165 causing $domain_data to be set, then another domain was checked against the
4166 same list, then the first domain was re-checked, the value of $domain_data
4167 after the final check could be wrong. In particular, if the second check
4168 failed, it could be set empty. This bug probably also applied to
4171 41. The strip_trailing_dot option was not being applied to the address given
4172 with the -f command-line option.
4174 42. The code for reading a message's header from the spool was incrementing
4175 $received_count, but never initializing it. This meant that the value was
4176 incorrect (doubled) while delivering a message in the same process in which
4177 it was received. In the most common configuration of Exim, this never
4178 happens - a fresh exec is done - but it can happen when
4179 deliver_drop_privilege is set.
4181 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4182 it now includes up to 150 characters of the unexpected data in the log
4185 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4186 and building data strings. The size of both of these buffers was 10 000
4187 bytes - far larger than anybody would *ever* want, thought I. Needless to
4188 say, somebody hit the limit. I have increased the maximum line length to
4189 20 000 and the maximum data length of concatenated lines to 100 000. I have
4190 also fixed two bugs, because there was no checking on these buffers. Tsk,
4191 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4192 buffer is too small.
4194 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4195 lsearch lookups. Now it does.
4197 46. When parsing a route_list item in a manualroute router, a fixed-length
4198 buffer was used for the list of hosts. I made this 1024 bytes long,
4199 thinking that nobody would ever have a list of hosts that long. Wrong.
4200 Somebody had a whole pile of complicated expansion conditions, and the
4201 string was silently truncated, leading to an expansion error. It turns out
4202 that it is easier to change to an unlimited length (owing to other changes
4203 that have happened since this code was originally written) than to build
4204 structure for giving a limitation error. The length of the item that
4205 expands into the list of hosts is now unlimited.
4207 47. The lsearch lookup could not handle data where the length of text line was
4208 more than 4095 characters. Such lines were truncated, leading to shortened
4209 data being returned. It should now handle lines of any length.
4211 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4212 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4215 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4216 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4217 no longer happens in comment lines. A list of replacements is now placed
4218 at the head of all of the source files, except those whose only change is
4219 to replace PERL_COMMAND in the very first #! line.
4221 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4222 messages on the queue, with a bottom-up merge sort, using code contributed
4223 by Michael Haardt. This should make operations like -bp somewhat faster on
4224 large queues. It won't affect queue runners, except when queue_run_in_order
4227 51. Installed eximstats 1.31 in the distribution.
4229 52. Added support for SRV lookups to the dnslookup router.
4231 53. If an ACL referred to $message_body or $message_body_end, the value was not
4232 reset for any messages that followed in the same SMTP session.
4234 54. The store-handling optimization for building very long strings was not
4235 differentiating between the different store pools. I don't think this
4236 actually made any difference in practice, but I've tidied it.
4238 55. While running the routers to verify a sender address, $sender_address
4239 was still set to the sender address. This is wrong, because when routing to
4240 send a bounce to the sender, it would be empty. Therefore, I have changed
4241 it so that, while verifying a sender address, $sender_address is set to <>.
4242 (There is no change to what happens when verifying a recipient address.)
4244 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4245 target A or AAAA records (if not already returned) without resetting the
4246 qualify_single or search_parents options of the DNS resolver. These are
4247 inappropriate in this case because the targets of MX and SRV records must
4248 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4249 target that, when qualified, matched something in the local domain. These
4250 two options are now turned off when doing these lookups.
4252 57. It seems that at least some releases of Reiserfs (which does not have the
4253 concept of a fixed number of inodes) returns zero and not -1 for the
4254 number of available inodes. This interacted badly with check_spool_inodes,
4255 which assumed that -1 was the "no such thing" setting. What I have done is
4256 to check that the total number of inodes is greater than zero before doing
4257 the test of how many are available.
4259 58. When a "warn" ACL statement has a log_message modifier, the message is
4260 remembered, and not repeated. This is to avoid a lot of repetition when a
4261 message has many recipients that cause the same warning to be written.
4262 Howewer, Exim was preserving the list of already written lines for an
4263 entire SMTP session, which doesn't seem right. The memory is now reset if a
4264 new message is started.
4266 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4267 debugging output unless log_rewrite was also set.
4269 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4270 of (int)(handle) when we know that handle contains (void *)(-1).
4272 61. The Exim daemon panic-logs an error return when it closes the incoming
4273 connection. However "connection reset by peer" seems to be common, and
4274 isn't really an error worthy of noting specially, so that particular error
4277 62. When Exim is trying to find all the local interfaces, it used to panic and
4278 die if the ioctl to get the interface flags failed. However, it seems that
4279 on at least one OS (Solaris 9) it is possible to have an interface that is
4280 included in the list of interfaces, but for which you get a failure error
4281 for this call. This happens when the interface is not "plumbed" into a
4282 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4283 failure of the "get flags" call assumes that the interface is down.
4285 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4286 makes life easier for people who are doing arithmetic on fields extracted
4287 from dates, where you often get leading zeros that should not be
4288 interpreted as octal.
4290 64. Added qualify_domain to the redirect router, to override the global
4293 65. If a pathologically long header line contained very many addresses (the
4294 report of this problem mentioned 10 000) and each of them was rewritten,
4295 Exim could use up a very large amount of memory. (It kept on making new
4296 copies of the header line as it rewrote, and never released the old ones.)
4297 At the expense of a bit more processing, the header rewriting function has
4298 been changed so that it no longer eats memory in this way.
4300 66. The generation of the Received: header has been moved from the time that a
4301 message starts to be received, to the time that it finishes. The timestamp
4302 in the Received: header should now be very close to that of the <= log
4303 line. There are two side-effects of this change:
4305 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4306 logged header lines no longer include the local Received: line, because
4307 it has not yet been created. The same applies to a copy of the message
4308 that is returned to a non-SMTP sender when a message is rejected.
4310 (b) When a filter file is tested using -bf, no additional Received: header
4311 is added to the test message. After some thought, I decided that this
4314 This change does not affect the value of $received_for. It is still set
4315 after address rewriting, but before local_scan() is called.
4317 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4319 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4320 gave an unhelpful panic error message, and a defer error. I have managed to
4321 change this behaviour so that it now rejects any supplied certificate,
4322 which seems right, as the list of acceptable certificates is empty.
4324 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4325 gave an unhelpful defer error. I have not managed to make this reject any
4326 supplied certificates, but the error message it gives is "no certificate
4327 supplied", which is not helpful.
4329 70. exigrep's output now also includes lines that are not associated with any
4330 message, but which match the given pattern. Implemented by a patch from
4331 Martin Sluka, which also tidied up the Perl a bit.
4333 71. Recipient callout verification, like sender verification, was using <> in
4334 the MAIL FROM command. This isn't really the right thing, since the actual
4335 sender may affect whether the remote host accepts the recipient or not. I
4336 have changed it to use the actual sender in the callout; this means that
4337 the cache record is now keyed on a recipient/sender pair, not just the
4338 recipient address. There doesn't seem to be a real danger of callout loops,
4339 since a callout by the remote host to check the sender would use <>.
4340 [SEE ABOVE: changed after hitting problems.]
4342 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4343 temporary errors. However, in the case of such a code being given after
4344 the end of a data transmission (i.e. after ".") Exim was failing to write
4345 a retry record for the message. (Yes, there was some broken host that was
4346 actually sending 8xx at this point.)
4348 73. An unknown lookup type in a host list could cause Exim to panic-die when
4349 the list was checked. (An example that provoked this was putting <; in the
4350 middle of a list instead of at the start.) If this happened during a DATA
4351 ACL check, a -D file could be left lying around. This kind of configuration
4352 error no longer causes Exim to die; instead it causes a defer errror. The
4353 incident is still logged to the main and panic logs.
4355 74. Buglet left over from Exim 3 conversion. The message "too many messages
4356 in one connection" was written to the rejectlog but not the mainlog, except
4357 when address rewriting (yes!) was being logged.
4359 75. Added write_rejectlog option.
4361 76. When a system filter was run not as root (that is, when system_filter_user
4362 was set), the values of the $n variables were not being returned to the
4363 main process; thus, they were not subsequently available in the $sn
4366 77. Added +return_path_on_delivery log selector.
4368 78. A connection timeout was being treated differently from recipients deferred
4369 when testing hosts_max_try with a message that was older than the host's
4370 retry timeout. (The host should not be counted, thus allowing all hosts to
4371 be tried at least once before bouncing.) This may have been the cause of an
4372 occasionally reported bug whereby a message would remain on the queue
4373 longer than the retry timeout, but would be bounced if a delivery was
4374 forced. I say "may" because I never totally pinned down the problem;
4375 setting up timeout/retry tests is difficult. See also the next item.
4377 79. The ultimate address timeout was not being applied to errors that involved
4378 a combination of host plus message (for example, a timeout on a MAIL
4379 command). When an address resolved to a number of possible hosts, and they
4380 were not all tried for each delivery (e.g. because of hosts_max_try), a
4381 message could remain on the queue longer than the retry timeout.
4383 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4386 81. Fixed an obscure SMTP outgoing bug which required at least the following
4387 conditions: (a) there was another message waiting for the same server;
4388 (b) the server returned 5xx to all RCPT commands in the first message so
4389 that the message was not completed; (c) the server dropped the connection
4390 or gave a negative response to the RSET that Exim sends to abort the
4391 transaction. The observed case was a dropped connection after DATA that had
4392 been sent in pipelining mode. That is, the server had advertised PIPELINING
4393 but was not implementing it correctly. The effect of the bug was incorrect
4394 behaviour, such as trying another host, and this could lead to a crash.
4400 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4401 and daemon.c were passed as pointers to ints; they should have been
4402 pointers to socklen_t variables (which are typically unsigned ints).
4404 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4407 3. Fixed a really odd bug that affected only the testing scheme; patching a
4408 certain fixed string in the binary changed the value of another string that
4409 happened to be identical to the end of the original first string.
4411 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4412 name", it returns that address as the IP address. On some operating
4413 systems (e.g. Solaris), it also passes back the IP address string as the
4414 "host name". However, on others (e.g. Linux), it passes back an empty
4415 string. Exim wasn't checking for this, and was changing the host name to an
4416 empty string, assuming it had been canonicized.
4418 5. Although rare, it is permitted to have more than one PTR record for a given
4419 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4420 all the names associated with an address, because they do in Solaris.
4421 However, it seems that they do not in Linux for data that comes from the
4422 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4423 I found this out when I moved to a new Linux workstation and tried to run
4424 the Exim test suite.
4426 To get round this problem I have changed the code so that it now does its
4427 own call to the DNS to look up PTR records when searching for a host name.
4428 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4429 addresses that are only in /etc/hosts are still found.
4431 This behaviour is, however, controlled by an option called host_lookup_
4432 order, which defaults to "bydns:byaddr". If people want to use the other
4433 order, or indeed, just use one or the other means of lookup, they can
4434 specify it in this variable.
4436 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4437 some operating systems, this comes back from gethostbyaddr() as an empty
4438 string, and this is what Exim used to test for. However, it seems that in
4439 other systems, "." is yielded. Exim now tests for this case too.
4441 7. The values of check_spool_space and check_log_space are now held internally
4442 as a number of kilobytes instead of an absolute number of bytes. If a
4443 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4444 kilobyte. This means that much larger values can be stored.
4446 8. Exim monitor: an attempt to get the action menu when not actually pointing
4447 at a message produces an empty menu entitled "No message selected". This
4448 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4449 no entries in it ("Shell widget menu has zero width and/or height"). So I
4450 have added a single, blank menu entry in this case.
4452 9. Added ${quote_local_part.
4454 10. MIME decoding is now applied to the contents of Subject: header lines when
4457 11. Now that a reference to $sender_host_address automatically causes a reverse
4458 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4459 host lookup before query-style lookups in lists that might use this
4460 variable. This has therefore been abolished, and the "net-" prefix is no
4461 longer necessary for query-style lookups.
4463 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4464 have been a typo for LFLAGS, so it has been changed.
4466 13. The install script calls Exim with "-C /dev/null" in order to find the
4467 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4468 to be output. Howeve, since Exim outputs its version number before the
4469 error, it didn't break the script. It just looked ugly. I fixed this by
4470 always allowing "-C /dev/null" if the caller is root.
4472 14. Ignore overlarge ACL variable number when reading spool file - insurance
4473 against a later release with more variables having written the file.
4475 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4476 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4477 this, as well as the form without the "IPv6" on the front.
4479 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4480 OS/Makefile-Darwin file.
4482 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4483 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4485 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4486 variable. However, some OS use ints for inodes. I've added cast to long int
4487 to get rid of the compiler warning.
4489 19. I had forgotten to lock out "/../" in configuration file names when
4490 ALT_CONFIG_PREFIX was set.
4492 20. Routers used for verification do not need to specify transports. However,
4493 if such a router generated a host list, and callout was configured, Exim
4494 crashed, because it could not find a port number from the (non-existent)
4495 transport. It now assumes port 25 in this circumstance.
4497 21. Added the -t option to exigrep.
4499 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4500 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4501 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4502 them is set, LOOKUP_LSEARCH is forced.
4504 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4506 24. Added sender and host information to the "rejected by local_scan()" log
4507 line; previously there was no indication of these.
4509 25. Added .include_if_exists.
4511 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4512 Linux. It turns out that not all file systems support this. Apparently some
4513 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4514 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4515 not supported on the file descriptor, is now ignored when Exim is trying to
4516 sync a directory. This applies only to Linux.
4518 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4520 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4521 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4524 29. Added support for maildirsize files from supplied patch (modified a bit).
4526 30. The use of :fail: followed by an empty string could lead Exim to respond to
4527 sender verification failures with (e.g.):
4529 550 Verification failed for <xxx>
4530 550 Sender verify failed
4532 where the first response line was missing the '-' that indicates it is not
4533 the final line of the response.
4535 31. The loop for finding the name of the user that called Exim had a hardwired
4536 limit of 10; it now uses the value of finduser_retries, which is used for
4537 all other user lookups.
4539 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4542 33. Exim was neglecting to zero errno before one call of strtol() when
4543 expanding a string and expecting an integer value. On some systems this
4544 resulted in spurious "integer overflow" errors. Also, it was casting the
4545 result into an int without checking.
4547 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4548 did not work. The code looks as if it has *never* worked, though it appears
4549 to have been documented since at least releast 1.62. I have made it work.
4551 35. The "timeout_DNS" error in retry rules, also documented since at least
4552 1.62, also never worked. As it isn't clear exactly what this means, and
4553 clearly it isn't a major issue, I have abolished the feature by treating it
4554 as "timeout", and writing a warning to the main and panic logs.
4556 36. The display of retry rules for -brt wasn't always showing the error code
4559 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4560 timeout_connect_A, timeout_connect_MX.
4562 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4563 to the empty sender.
4565 39. The daemon was not analysing the content of -oX till after it had closed
4566 stderr and disconnected from the controlling terminal. This meant that any
4567 syntax errors were only noted on the panic log, and the return code from
4568 the command was 0. By re-arranging the code a little, I've made the
4569 decoding happen first, so such errors now appear on stderr, and the return
4570 code is 1. However, the actual setting up of the sockets still happens in
4571 the disconnected process, so errors there are still only recorded on the
4574 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4575 connections (as happens on some IP stacks) was logged at start up time as
4576 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4577 it from "IPv6 and IPv4", which means that two separate sockets are being
4580 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4581 says whether AF_INET or AF_INET6 was passed as an argument.
4583 42. Exiwhat output was messed up when time zones were included in log
4586 43. Exiwhat now gives more information about the daemon's listening ports,
4587 and whether -tls-on-connect was used.
4589 44. The "port" option of the smtp transport is now expanded.
4591 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4592 silently ignored. Now an error message is written to the main and panic
4595 46. There's a new ACL modifier called "logwrite" which writes to a log file
4596 as soon as it is encountered.
4598 47. Added $local_user_uid and $local_user_gid at routing time.
4600 48. Exim crashed when trying to verify a sender address that was being
4603 49. Exim was recognizing only a space character after ".include". It now also
4604 recognizes a tab character.
4606 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4607 extracting the relevant information from the specification. The man page no
4608 longer contains scrambled data for the -d option, and I've added a section
4609 at the front about calling Exim under different names.
4611 51. Added "extra_headers" argument to the "mail" command in filter files.
4613 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4616 53. Installed eximstats 1.29.
4618 54. Added transport_filter_timeout as a generic transport option.
4620 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4621 Cc: header lines. This was required by RFC 822, but it not required by RFC
4624 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4625 incoming messages that did not have them. Now it does so only if the
4626 message originates locally, that is, if there is no associated remote host
4627 address. When Resent- header lines are present, this applies to the Resent-
4628 lines rather than the non-Resent- lines.
4630 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4631 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4633 58. Messages for configuration errors now include the name of the main
4634 configuration file - useful now that there may be more than one file in a
4635 list (.included file names were always shown).
4637 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4638 for those rare installations that do not start the daemon as root or run it
4639 setuid root. I've cut out the call to initgroups() if the daemon is not
4642 60. The Exim user and group can now be bound into the binary as text strings
4643 that are looked up at the start of Exim's processing.
4645 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4647 62. Added $mailstore_basename variable.
4649 63. Installed patch to sieve.c from Michael Haardt.
4651 64. When Exim failed to open the panic log after failing to open the main log,
4652 the original message it was trying to log was written to stderr and debug
4653 output, but if they were not available (the usual case in production), it
4654 was lost. Now it is written to syslog before the two lines that record the
4655 failures to open the logs.
4657 65. Users' Exim filters run in subprocesses under the user's uid. It is
4658 possible for a "deliver" command or an alias in a "personal" command to
4659 provoke an address rewrite. If logging of address rewriting is configured,
4660 this fails because the process is not running as root or exim. There may be
4661 a better way of dealing with this, but for the moment (because 4.30 needs
4662 to be released), I have disabled address rewrite logging when running a
4663 filter in a non-root, non-exim process.
4669 1. The buildconfig auxiliary program wasn't quoting the value set for
4670 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4671 not defined. This bug was masked in 4.22 by the effect that was fixed in
4674 2. Some messages that were rejected after a message id was allocated were
4675 shown as "incomplete" by exigrep. It no longer does this for messages that
4676 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4678 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4679 have allow_domain_literals set, the ID did not get logged in the <= line.
4680 Domain literals are now always recognized in Message-ID: header lines.
4682 4. The first argument for a ${extract expansion item is the key name or field
4683 number. Leading and trailing spaces in this item were not being ignored,
4684 causing some misleading effects.
4686 5. When deliver_drop_privilege was set, single queue runner processes started
4687 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4688 same command in the process it spins off) were not dropping privilege.
4690 6. When the daemon running as "exim" started a queue runner, it always
4691 re-executed Exim in the spun-off process. This is a waste of effort when
4692 deliver_drop_privilege is set. The new process now just calls the
4693 queue-runner function directly.
4699 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4702 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4703 ignored. Though the use of -oP was forcing the writing of a pid file, it
4704 was always written to the default place.
4706 3. If the message "no IP address found for host xxxx" is generated during
4707 incoming verification, it is now followed by identification of the incoming
4708 connection (so you can more easily find what provoked it).
4710 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4712 5. Added some features to "harden" Exim a bit more against certain attacks:
4714 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4715 be put in Local/Makefile. This is like the never_users runtime option,
4716 but it cannot be overridden. The default setting is "root".
4718 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4719 prefix string with which any file named in a -C command line option
4722 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4723 is retained for -C and -D only if the caller of Exim is root. Without
4724 it, the exim user may also use -C and -D and retain privilege.
4726 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4727 command line option is disabled.
4729 6. Macro names set by the -D option must start with an upper case letter, just
4730 like macro names defined in the configuration file.
4732 7. Added "dereference=" facility to LDAP.
4734 8. Two instances of the typo "uknown" in the source files are fixed.
4736 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4737 the Configure-Makefile script screwed up while processing it.
4739 10. Incorporated PCRE 4.4.
4741 11. The SMTP synchronization check was not operating right at the start of an
4742 SMTP session. For example, it could not catch a HELO sent before the client
4743 waited for the greeting. There is now a check for outstanding input at the
4744 point when the greeting is written. Because of the duplex, asynchronous
4745 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4746 way, but not yet received, when the check is performed.
4748 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4749 on TCP/IP sockets, because this apparently causes some broken clients to
4752 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4753 unchanged) from the Cygwin maintainer.
4755 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4756 was supported instead of testing for mbx. Effectively a typo.
4758 15. The spa authenticator server code was not checking that the input it
4759 received was valid base64.
4761 16. The debug output line for the "set" modifier in ACLs was not showing the
4762 name of the variable that was being set.
4764 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4766 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4767 lookup on its own. Something else was needed to trigger the lookup. For
4768 example, a match in host_lookup or the need for a host name in a host list.
4769 Now, if $sender_host_name is referenced and the host name has not yet been
4770 looked up, a lookup is performed. If the lookup fails, the variable remains
4771 empty, and $host_lookup_failed is set to "1".
4773 19. Added "eqi" as a case-independent comparison operator.
4775 20. The saslauthd authentication condition could segfault if neither service
4776 nor realm was specified.
4778 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4779 error message that was logged was misleading, and incoming SMTP
4780 connections were dropped. The message is now more accurate, and temporary
4781 errors are given to SMTP connections.
4783 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4784 (except RSET) with a 421 error, until QUIT is received. However, it was
4785 failing to send a response to QUIT.
4787 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4788 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4789 ACL is now run after the helo_try_verify_hosts code.
4791 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4792 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4793 case-independent, so other case variants are also recognized). Apparently
4794 some systems use these upper case variants.
4796 25. If more than two messages were waiting for the same host, and a transport
4797 filter was specified for the transport, Exim sent two messages over the
4798 same TCP/IP connection, and then failed with "socket operation on non-
4799 socket" when it tried to send the third.
4801 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4803 27. The extern definition of crypt16() in expand.c was not being excluded when
4804 the OS had its own crypt16() function.
4806 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4807 as a preferred synonym for return_size_limit, both as an option and as an
4810 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4812 30. Changed the default configuration ACL to relax the local part checking rule
4813 for addresses that are not in any local domains. For these addresses,
4814 slashes and pipe symbols are allowed within local parts, but the sequence
4815 /../ is explicitly forbidden.
4817 31. SPA server authentication was not clearing the challenge buffer before
4820 32. log_message in a "warn" ACL statement was writing to the reject log as
4821 well as to the main log, which contradicts the documentation and doesn't
4822 seem right (because no rejection is happening). So I have stopped it.
4824 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4825 However, I am unable to do any testing of this.
4827 34. Fixed an infelicity in the appendfile transport. When checking directories
4828 for a mailbox, to see if any needed to be created, it was accidentally
4829 using path names with one or more superfluous leading slashes; tracing
4830 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4832 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4833 contents are added to the log line that is written for every discarded
4834 recipient. (Previously a log_message setting was ignored.)
4836 36. The ${quote: operator now quotes the string if it is empty.
4838 37. The install script runs exim in order to find its version number. If for
4839 some reason other than non-existence or emptiness, which it checks, it
4840 could not run './exim', it was installing it with an empty version number,
4841 i.e. as "exim-". This error state is now caught, and the installation is
4844 38. An argument was missing from the function that creates an error message
4845 when Exim fails to connect to the socket for saslauthd authentication.
4846 This could cause Exim to crash, or give a corrupted message.
4848 39. Added isip, isip4, and isip6 to ${if conditions.
4850 40. The ACL variables $acl_xx are now saved with the message, and can be
4851 accessed later in routers, transports, and filters.
4853 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4854 strings in the file are not string-expanded.
4856 42. If a MAIL command specified a SIZE value that was too large to fit into an
4857 int variable, the check against message_size_limit failed. Such values are
4858 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4859 day this will have to be increased, but I don't think I want to be around
4860 when emails are that large.
4867 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4868 iconv() is not standard in FreeBSD.
4870 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4871 IPv6 enabled. The observed symptom was a segmentation fault on return from
4872 the function os_common_find_running_interfaces() in src/os.c.
4874 3. In the check_special_case() function in daemon.c I had used "errno" as an
4875 argument name, which causes warnings on some systems. This was basically a
4876 typo, since it was named "eno" in the comments!
4878 4. The code that waits for the clock to tick (at a resolution of some fraction
4879 of a second) so as to ensure message-id uniqueness was always waiting for
4880 at least one whole tick, when it could have waited for less. [This is
4881 almost certainly not relevant at current processor speeds, where it is
4882 unlikely to ever wait at all. But we try to future-proof.]
4884 5. The function that sleeps for a time interval that includes fractions of a
4885 second contained a race. It did not block SIGALRM between setting the
4886 timer, and suspending (a couple of lines later). If the interval was short
4887 and the sigsuspend() was delayed until after it had expired, the suspension
4888 never ended. On busy systems this could lead to processes getting stuck for
4891 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4892 process, before it forks any delivery processes. The open lookup caching
4893 mechanism meant that the open file or database connection was passed into
4894 the delivery process. The problem was that delivery processes always tidy
4895 up cached lookup data. This could cause a problem for the next delivery
4896 process started by the queue runner, because the external queue runner
4897 process does not know about the closure. So the next delivery process
4898 still has data in the lookup cache. In the case of a file lookup, there was
4899 no problem because closing a file descriptor in a subprocess doesn't affect
4900 the parent. However, if the lookup was caching a connection to a database,
4901 the connection was closed, and the second delivery process was likely to
4902 see errors such as "PGSQL: query failed: server closed the connection
4903 unexpectedly". The problem has been fixed by closing all cached lookups
4904 in a queue runner before running a delivery process.
4906 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4907 seem to have the "const" qualifier which it has on other OS. I've
4910 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4911 *of the same type* (client or server) with the same public name that an
4912 error should be diagnosed.
4914 9. When Exim looked up a host name for an IP address, but failed to find the
4915 original IP address when looking up the host name (a safety check), it
4916 output the message "<ip address> does not match any IP for NULL", which was
4917 confusing, to say the least. The bug was that the host name should have
4918 appeared instead of "NULL".
4920 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4921 user that is built into the binary, and the -C or -D options is used, root
4922 privilege is dropped before the configuration file is read. In addition,
4923 logging is switched to stderr instead of the normal log files. If the
4924 configuration then re-defines the Exim user, the unprivileged environment
4925 is probably not what is expected, so Exim logs a panic warning message (but
4928 However, if deliver_drop_privilege is set, the unprivileged state may well
4929 be exactly what is intended, so the warning has been cut out in that case,
4930 and Exim is allowed to try to write to its normal log files.
4936 1. smtp_return_error_details was not giving details for temporary sender
4937 or receiver verification errors.
4939 2. Diagnose a configuration error if two authenticators have the same public
4942 3. Exim used not to create the message log file for a message until the first
4943 delivery attempt. This could be confusing when incoming messages were held
4944 for policy or load reasons. The message log file is now created at the time
4945 the message is received, and an initial "Received" line is written to it.
4947 4. The automatically generated man page for command line options had a minor
4948 bug that caused no ill effects; however, a more serious problem was that
4949 the procedure for building the man page automatically didn't always
4950 operate. Consequently, release 4.20 contains an out-of-date version. This
4951 shouldn't happen again.
4953 5. When building Exim with embedded Perl support, the script that builds the
4954 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4955 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4957 6. The freeze_tell option was not being used for messages that were frozen on
4958 arrival, either by an ACL or by local_scan().
4960 7. Added the smtp_incomplete_transaction log selector.
4962 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4963 was accepting AUTH without a new EHLO.
4965 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4966 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4969 10. Logging of TCP/IP connections (when configured) now happens in the main
4970 daemon process instead of the child process, so that the TCP/IP connection
4971 count is more accurate (but it can never be perfect).
4973 11. The use of "drop" in a nested ACL was not being handled correctly in the
4974 outer ACL. Now, if condition failure induced by the nested "drop" causes
4975 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4976 or "require"), the connection is dropped.
4978 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4979 that yield "discard" can now be used with an "accept" or a "discard" verb,
4980 but an error is generated for any others (because I can't see a useful way
4981 to define what should happen).
4983 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4984 are now processed in the same way as lines in the Exim configuration file.
4985 In particular, continuation lines are supported.
4987 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4989 15. Added -ti meaning -t -i.
4991 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4992 domains, and warn by logging if others are found.
4994 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4995 returned from GIFCONF when Exim is trying to find the list of interfaces on
4996 a host. The code in os.c has been modified to copy each ifreq to an aligned
4997 structure in all cases.
4999 Also, in some cases, the returned ifreq's were being copied to a 'struct
5000 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5001 means the last couple of bytes of an IPv6 address could be chopped if the
5002 ifreq contained only a normal sockaddr (14 bytes storage).
5004 18. Named domain lists were not supported in the hosts_treat_as_local option.
5005 An entry such as +xxxx was not recognized, and was treated as a literal
5008 19. Ensure that header lines added by a DATA ACL are included in the reject log
5009 if the ACL subsequently rejects the message.
5011 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5012 MD5 (which is deprecated).
5014 21. When testing a filter file using -bf, Exim was writing a message when it
5015 took the sender from a "From " line in the message, but it was not doing so
5016 when it took $return_path from a Return-Path: header line. It now does.
5018 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5019 with a valid header line field name (a series of printing characters
5020 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5022 23. Changed "disc" in the source to "disk" to conform to the documentation and
5023 the book and for uniformity.
5025 24. Ignore Sendmail's -Ooption=value command line item.
5027 25. When execve() failed while trying to run a command in a pipe transport,
5028 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5029 could be confused with a return value of 69 from the command itself. This
5030 has been changed to 127, the value the shell returns if it is asked to run
5031 a non-existent command. The wording for the related log line suggests a
5032 non-existent command as the problem.
5034 26. If received_header_text expands to an empty string, do not add a Received:
5035 header line to the message. (Well, it adds a token one on the spool, but
5036 marks it "old" so that it doesn't get used or transmitted.)
5038 27. Installed eximstats 1.28 (addition of -nt option).
5040 28. There was no check for failure on the call to getsockname() in the daemon
5041 code. This can fail if there is a shortage of resources on the system, with
5042 ENOMEM, for example. A temporary error is now given on failure.
5044 29. Contrary to the C standard, it seems that in some environments, the
5045 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5046 program. Exim now does this explicitly; it affects the formatting of
5047 timestamps using strftime().
5049 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5050 complaints. I've now initialized all the variables, to avoid this.
5052 32. Header lines added by a system filter were not being "seen" during
5053 transport-time rewrites.
5055 33. The info_callback() function passed to OpenSSL is set up with type void
5056 (*)(SSL *, int, int), as described somewhere. However, when calling the
5057 function (actually a macro) that sets it up, the type void(*)() is
5058 expected. I've put in a cast to prevent warnings from picky compilers.
5060 34. If a DNS black list lookup found a CNAME record, but there were no A
5061 records associated with the domain it pointed at, Exim crashed.
5063 35. If a DNS black list lookup returned more than one A record, Exim ignored
5064 all but the first. It now scans all returned addresses if a particular IP
5065 value is being sought. In this situation, the contents of the
5066 $dnslist_value variable are a list of all the addresses, separated by a
5069 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5070 to accept a host name and all its aliases if the forward lookup for any of
5071 them yielded the IP address of the incoming connection. Now it accepts only
5072 those names whose forward lookup yields the correct IP address. Any other
5073 names are discarded. This closes a loophole whereby a rogue DNS
5074 administrator could create reverse DNS records to break through a
5075 wildcarded host restriction in an ACL.
5077 37. If a user filter or a system filter that ran in a subprocess used any of
5078 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5079 the wrong values were passed to the pipe command ($thisaddress had the
5080 value of $0, $0 had the value of $1, etc). This bug was introduced by
5081 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5083 38. Improved the line breaking for long SMTP error messages from ACLs.
5084 Previously, if there was no break point between 40 and 75 characters, Exim
5085 left the rest of the message alone. Two changes have been made: (a) I've
5086 reduced the minimum length to 35 characters; (b) if it can't find a break
5087 point between 35 and 75 characters, it looks ahead and uses the first one
5088 that it finds. This may give the occasional overlong line, but at least the
5089 remaining text gets split now.
5091 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5092 file descriptors might be low, and that setting 1000 would always raise it.
5093 It turns out that in some environments, the limit is already over 1000 and
5094 that lowering it causes trouble. So now Exim takes care not to decrease it.
5096 40. When delivering a message, the value of $return_path is set to $sender_
5097 address at the start of routing (routers may change the value). By an
5098 oversight, this default was not being set up when an address was tested by
5099 -bt or -bv, which affected the outcome if any router or filter referred to
5102 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5103 log only when "message" or "log_message" are set. However, if one of the
5104 conditions was an address verification, or a call to a nested ACL, the
5105 messages generated by the underlying test were being passed through. This
5106 no longer happens. The underlying message is available in $acl_verify_
5107 message for both "message" and "log_message" expansions, so it can be
5108 passed through if needed.
5110 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5111 new expansion $bh_ to give the encoded byte string without charset
5112 translation. Translation happens only if iconv() is available; HAVE_ICONV
5113 indicates this at build time. HEADERS_CHARSET gives the charset to
5114 translate to; headers_charset can change it in the configuration, and
5115 "headers charset" can change it in an individual filter file.
5117 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5118 that creates RFC 2047 encoded "words" labels them as that charset instead
5119 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5120 expansion operator; (ii) when Exim creates a From: line for a local
5121 message; (iii) when a header line is rewritten to include a "phrase" part.
5123 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5124 buggy, causing it to skip the first lines of messages whose message ID
5125 ended in 'D'. This would not have bitten before Exim release 4.14, because
5126 message IDs were unlikely to end in 'D' before then. The effect was to have
5127 incorrect size information for certain domains.
5129 45. #include "config.h" was missing at the start of the crypt16.c module. This
5130 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5133 46. If there was a timeout during a "random" callout check, Exim treated it as
5134 a failure of the random address, and carried on sending RSET and the real
5135 address. If the delay was just some slowness somewhere, the response to the
5136 original RCPT would be taken as a response to RSET and so on, causing
5137 mayhem of various kinds.
5139 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5140 when I implemented it. It didn't allow for the fact that some option values
5141 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5142 the right test for positive values.
5144 48. Domain names in DNS records are case-independent. Exim always looks them up
5145 in lower case. Some resolvers return domain names in exactly the case they
5146 appear in the zone file, that is, they may contain uppercase letters. Not
5147 all resolvers do this - some return always lower case. Exim was treating a
5148 change of case by a resolver as a change of domain, similar to a widening
5149 of a domain abbreviation. This triggered its re-routing code and so it was
5150 trying to route what was effectively the same domain again. This normally
5151 caused routing to fail (because the router wouldn't handle the domain
5152 twice). Now Exim checks for this case specially, and just changes the
5153 casing of the domain that it ultimately uses when it transmits the message
5156 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5159 50. If a filter generated a file delivery with a non-absolute name (possible if
5160 no home directory exists for the router), the forbid_file option was not
5163 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5164 the existing equality matching.
5166 52. Exim was using ints instead of ino_t variables in some places where it was
5167 dealing with inode numbers.
5169 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5170 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5171 called TMPDIR, and if it finds it is different, it changes its value.
5173 54. The smtp_printf() function is now made available to local_scan() so
5174 additional output lines can be written before returning. There is also an
5175 smtp_fflush() function to enable the detection of a dropped connection.
5176 The variables smtp_input and smtp_batched_input are exported to
5179 55. Changed the default runtime configuration: the message "Unknown user"
5180 has been removed from the ACL, and instead placed on the localuser router,
5181 using the cannot_route_message feature. This means that any verification
5182 failures that generate their own messages won't get overridden. Similarly,
5183 the "Unrouteable address" message that was in the ACL for unverifiable
5184 relay addresses has also been removed.
5186 56. Added hosts_avoid_esmtp to the smtp transport.
5188 57. The exicyclog script was not checking for the esoteric option
5189 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5190 will work only if exicyclog is run under the appropriate euid.
5192 58. Following a discussion on the list, the rules by which Exim recognises line
5193 endings on incoming messages have been changed. The -dropcr and drop_cr
5194 options are now no-ops, retained only for backwards compatibility. The
5195 following line terminators are recognized: LF CRLF CR. However, special
5196 processing applies to CR:
5198 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5199 nor a local message in the state where . is a terminator.
5201 (ii) If a bare CR is encountered in a header line, an extra space is added
5202 after the line terminator so as not to end the header. The reasoning
5203 behind this is that bare CRs in header lines are most likely either
5204 to be mistakes, or people trying to play silly games.
5206 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5207 was being incorrectly given as 18 bytes larger than it should have been.
5208 This is a VOB (very old bug).
5210 60. This may never have affected anything current, but just in case it has:
5211 When the local host is found other than at the start of a list of hosts,
5212 the local host, those with the same MX, and any that follow, are discarded.
5213 When the list in question was part of a longer list of hosts, the following
5214 hosts (not currently being processed) were also being discarded. This no
5215 longer happens. I'm not sure if this situation could ever has previously
5218 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5221 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5222 header. This is something that is recommended in a new Internet Draft, and
5223 is something that is documented as being done by Sendmail. There are two
5224 possible values. For messages generated by the autoreply transport, Exim
5227 Auto-Submitted: auto-replied
5229 whereas for all other generated messages (e.g. bounces) it adds
5231 Auto-Submitted: auto-generated
5233 63. The "personal" condition in filters now includes a test for the
5234 Auto-Submitted: header. If it contains the string "auto-" the message it
5235 not considered personal.
5237 64. Added rcpt_include_affixes as a generic transport option.
5239 65. Added queue_only_override (default true).
5241 66. Added the syslog_duplication option.
5243 67. If what should have been the first header line of a message consisted of
5244 a space followed by a colon, Exim was mis-interpreting it as a header line.
5245 It isn't of course - it is syntactically invalid and should therefore be
5246 treated as the start of the message body. The misbehaviour could have
5247 caused a number of strange effects, including loss of data in subsequent
5248 header lines, and spool format errors.
5250 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5251 client host had authenticated. This control can now be exercised by an ACL
5252 for more flexibility.
5254 69. By default, callouts do not happen when testing with -bh. There is now a
5255 variant, -bhc, which does actually run the callout code, including
5256 consulting and updating the callout cache.
5258 70. Added support for saslauthd authentication, courtesy of Alexander
5261 71. If statvfs() failed on the spool or log directories while checking their
5262 size for availability, Exim confusingly gave the error "space shortage".
5263 Furthermore, in debugging mode it crashed with a floating point exception.
5264 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5265 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5266 serious problem, Exim now writes to the main and panic logs when this
5267 happens, with details of the failure. It then refuses to accept the
5268 incoming message, giving the message "spool directory problem" or "log
5269 directory problem" with a 421 code for SMTP messages.
5271 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5272 0, 1, and 2 exist, because some OS complain for execs without them (see
5273 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5274 descriptors. However, the code omitted to check that the open succeeded,
5275 causing mysterious errors if for some reason the permissions on /dev/null
5276 got screwed. Now Exim writes a message to the main and panic logs, and
5277 bombs out if it can't open /dev/null.
5279 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5280 interact so that it is all more flexible. It is supposed to remain
5281 backwards compatible. Also added extra_local_interfaces.
5283 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5284 to bomb out with an assertion failure - to the client this appears as a
5285 connection drop. This problem occurs in the part of the code that was taken
5286 from the Samba project. Fortunately, the assertion is in a very simple
5287 function, so I have fixed this by reproducing the function inline in the
5288 one place where it is called, and arranging for authentication to fail
5289 instead of killing the process with assert().
5291 75. The SPA client code was not working when the server requested OEM rather
5292 than Unicode encoding.
5294 76. Added code to make require_files with a specific uid setting more usable in
5295 the case where statting the file as root fails - usually a non-root-mounted
5296 NFS file system. When this happens and the failure is EACCES, Exim now
5297 forks a subprocess and does the per-uid checking as the relevant uid.
5299 77. Added process_log_path.
5301 78. If log_file_path was not explicitly set, a setting of check_log_space or
5302 check_log_inodes was ignored.
5304 79. If a space check for the spool or log partitions fails, the incident is now
5305 logged. Of course, in the latter case the data may get lost...
5307 80. Added the %p formatting code to string_format() so that it can be used to
5308 print addresses in debug_print(). Adjusted all the address printing in the
5309 debugging in store.c to use %p rather than %d.
5311 81. There was a concern that a line of code in smtp_in.c could overflow a
5312 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5313 initially expressed, the concern was not well-founded, because trailing
5314 spaces are removed early. However, if the trailing spaces were followed by
5315 a NULL, they did not get removed, so the overflow was possible. Two fixes
5318 (a) I re-wrote the offending code in a cleaner fashion.
5319 (b) If an incoming SMTP command contains a NULL character, it is rejected
5322 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5323 runs initgroups(), so that if the Exim user is in any additional groups,
5324 they will be used during message reception.
5330 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob