1 Change log file for Exim from version 4.21
2 -------------------------------------------
7 PP/01 Handle short writes when writing local log-files.
8 In practice, only affects FreeBSD (8 onwards).
9 Bugzilla 1053, with thanks to Dmitry Isaikin.
11 NM/01 Bugzilla 949 - Documentation tweak
13 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
16 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
18 PP/02 Implemented gsasl authenticator.
20 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
22 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
23 `pkg-config foo` for cflags/libs.
25 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
26 with rest of GSASL and with heimdal_gssapi.
28 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
29 `pkg-config foo` for cflags/libs for the TLS implementation.
31 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
32 properties get this fed in as external SSF. A number of robustness
33 and debugging improvements to the cyrus_sasl authenticator.
35 PP/08 cyrus_sasl server now expands the server_realm option.
37 PP/09 Bugzilla 1214 - Log authentication information in reject log.
38 Patch by Jeremy Harris.
40 PP/10 Added dbmjz lookup type.
42 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
44 PP/12 MAIL args handles TAB as well as SP, for better interop with
45 non-compliant senders.
46 Analysis and variant patch by Todd Lyons.
48 NM/04 Bugzilla 1237 - fix cases where printf format usage no indicated
49 Bug report from Lars Müller <lars@samba.org> (via SUSE),
50 Patch from Dirk Mueller <dmueller@suse.com>
52 PP/13 tls_peerdn now print-escaped for spool files.
53 Observed some $tls_peerdn in wild which contained \n, which resulted
54 in spool file corruption.
56 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
57 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
58 or write after TLS renegotiation, which otherwise led to messages
61 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
62 as a tracking header (ie: a signed header comes before the signature).
63 Patch from Wolfgang Breyha.
65 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
66 comma-sep list; embedded commas doubled.
68 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
70 Report and patch from Dmitry Banschikov.
72 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
73 Removed SSL_clear() after SSL_new() which led to protocol negotiation
74 failures. We appear to now support TLS1.1+ with Exim.
76 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
77 lets Exim select keys and certificates based upon TLS SNI from client.
78 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
79 before an outbound SMTP session. New log_selector, +tls_sni.
81 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
82 NULL dereference. Report and patch from Alun Jones.
84 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
85 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
86 Not seeing resolver debug output on NetBSD, but suspect this is a
87 resolver implementation change.
89 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
90 Left warnings. Added "eximon gdb" invocation mode.
92 PP/21 Defaulting "accept_8bitmime" to true, not false.
94 PP/22 Added -bw for inetd wait mode support.
96 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
97 locate the relevant includes and libraries. Made this the default.
99 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
100 Bugzilla 1246, report and most of solution from Tomasz Kusy.
106 PP/01 Solaris build fix for Oracle's LDAP libraries.
107 Bugzilla 1109, patch from Stephen Usher.
109 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
111 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
114 TF/02 Fix a couple more cases where we did not log the error message
115 when unlink() failed. See also change 4.74-TF/03.
117 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
118 lock up or crash if it happened to be inside a call to libc when it
119 got a SIGUSR1 from exiwhat.
121 The SIGUSR1 handler appends the current process status to the process
122 log which is later printed by exiwhat. It used to use the general
123 purpose logging code to do this, but several functions it calls are
124 not safe for signals.
126 The new output code in the SIGUSR1 handler is specific to the process
127 log, and simple enough that it's easy to inspect for signal safety.
128 Removing some special cases also simplifies the general logging code.
129 Removing the spurious timestamps from the process log simplifies
132 TF/04 Improved ratelimit ACL condition.
134 The /noupdate option has been deprecated in favour of /readonly which
135 has clearer semantics. The /leaky, /strict, and /readonly update modes
136 are mutually exclusive. The update mode is no longer included in the
137 database key; it just determines when the database is updated. (This
138 means that when you upgrde Exim will forget old rate measurements.)
140 Exim now checks that the per_* options are used with an update mode that
141 makes sense for the current ACL. For example, when Exim is processing a
142 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
143 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
144 must specify per_mail/readonly. If you omit the update mode it defaults to
145 /leaky where that makes sense (as before) or /readonly where required.
147 The /noupdate option is now undocumented but still supported for
148 backwards compatibility. It is equivalent to /readonly except that in
149 ACLs where /readonly is required you may specify /leaky/noupdate or
150 /strict/noupdate which are treated the same as /readonly.
152 A useful new feature is the /count= option. This is a generalization
153 of the per_byte option, so that you can measure the throughput of other
154 aggregate values. For example, the per_byte option is now equivalent
155 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
157 The per_rcpt option has been generalized using the /count= mechanism
158 (though it's more complicated than the per_byte equivalence). When it is
159 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
160 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
161 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
162 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
163 Note that using per_rcpt with a non-readonly update mode in more than
164 one ACL will cause the recipients to be double-counted. (The per_mail
165 and per_byte options don't have this problem.)
167 The handling of very low rates has changed slightly. If the computed rate
168 is less than the event's count (usually one) then this event is the first
169 after a long gap. In this case the rate is set to the same as this event's
170 count, so that the first message of a spam run is counted properly.
172 The major new feature is a mechanism for counting the rate of unique
173 events. The new per_addr option counts the number of different
174 recipients that someone has sent messages to in the last time period. It
175 behaves like per_rcpt if all the recipient addresses are different, but
176 duplicate recipient addresses do not increase the measured rate. Like
177 the /count= option this is a general mechanism, so the per_addr option
178 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
179 example, measure the rate that a client uses different sender addresses
180 with the options per_mail/unique=$sender_address. There are further
181 details in the main documentation.
183 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
185 TF/06 Removed a few PCRE remnants.
187 TF/07 Automatically extract Exim's version number from tags in the git
188 repository when doing development or release builds.
190 PP/02 Raise smtp_cmd_buffer_size to 16kB.
191 Bugzilla 879. Patch from Paul Fisher.
193 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
194 Heavily based on revision 40f9a89a from Simon Arlott's tree.
197 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
199 PP/05 Variable $av_failed, true if the AV scanner deferred.
200 Bugzilla 1078. Patch from John Horne.
202 PP/06 Stop make process more reliably on build failure.
203 Bugzilla 1087. Patch from Heiko Schlittermann.
205 PP/07 Make maildir_use_size_file an _expandable_ boolean.
206 Bugzilla 1089. Patch from Heiko Schlittermann.
208 PP/08 Handle ${run} returning more data than OS pipe buffer size.
209 Bugzilla 1131. Patch from Holger Weiß.
211 PP/09 Handle IPv6 addresses with SPF.
212 Bugzilla 860. Patch from Wolfgang Breyha.
214 PP/10 GnuTLS: support TLS 1.2 & 1.1.
216 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
219 PP/11 match_* no longer expand right-hand-side by default.
220 New compile-time build option, EXPAND_LISTMATCH_RHS.
221 New expansion conditions, "inlist", "inlisti".
223 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
225 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
227 PP/14 fix log_write() format string regression from TF/03.
228 Bugzilla 1152. Patch from Dmitry Isaikin.
234 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
236 PP/02 Harmonised TLS library version reporting; only show if debugging.
237 Layout now matches that introduced for other libraries in 4.74 PP/03.
239 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
241 PP/04 New "dns_use_edns0" global option.
243 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
246 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
247 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
249 TK/01 Updated PolarSSL code to 0.14.2.
250 Bugzilla 1097. Patch from Andreas Metzler.
252 PP/07 Catch divide-by-zero in ${eval:...}.
255 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
258 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
259 format-string attack -- SECURITY: remote arbitrary code execution.
261 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
262 time unintentionally subject to list matching rules, letting the header
263 cause arbitrary Exim lookups (of items which can occur in lists, *not*
264 arbitrary string expansion). This allowed for information disclosure.
266 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
267 INT_MIN/-1 -- value coerced to INT_MAX.
273 NM/01 Workround for PCRE version dependency in version reporting
276 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
277 This fixes portability to compilers other than gcc, notably
278 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
280 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
281 makefiles for portability to HP-UX and POSIX correctness.
283 PP/01 Permit LOOKUP_foo enabling on the make command-line.
284 Also via indented variable definition in the Makefile.
285 (Debugging by Oliver Heesakkers).
287 PP/02 Restore caching of spamd results with expanded spamd_address.
288 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
290 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
291 Improves build reliability. Fix from: Frank Elsner
293 NM/02 Fix wide character breakage in the rfc2047 coding
294 Fixes bug 1064. Patch from Andrey N. Oktyabrski
296 NM/03 Allow underscore in dnslist lookups
297 Fixes bug 1026. Patch from Graeme Fowler
299 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
300 Code patches from Adam Ciarcinski of NetBSD.
302 NM/04 Fixed exiqgrep to cope with mailq missing size issue
305 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
306 is logged, to avoid truncation. Patch from John Horne.
308 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
309 Patch from Jakob Hirsch.
311 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
312 SQL string expansion failure details.
313 Patch from Andrey Oktyabrski.
315 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
316 Patch from Simon Arlott.
318 PP/09 New lookups functionality failed to compile on old gcc which rejects
319 extern declarations in function scope.
320 Patch from Oliver Fleischmann
322 PP/10 Use sig_atomic_t for flags set from signal handlers.
323 Check getgroups() return and improve debugging.
324 Fixed developed for diagnosis in bug 927 (which turned out to be
327 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
328 Patch from Mark Zealey.
330 PP/12 Bugzilla 1056: Improved spamd server selection.
331 Patch from Mark Zealey.
333 PP/13 Bugzilla 1086: Deal with maildir quota file races.
334 Based on patch from Heiko Schlittermann.
336 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
337 Patch from Uwe Doering, sign-off by Michael Haardt.
339 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
340 variable declaration deep within a block. Bug and patch from
343 PP/15 lookups-Makefile IRIX compatibilty coercion.
345 PP/16 Make DISABLE_DKIM build knob functional.
347 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
348 Patch by Simon Arlott
350 TF/03 Fix valgrind.h portability to C89 compilers that do not support
351 variable argument macros. Our copy now differs from upstream.
357 TF/01 Failure to get a lock on a hints database can have serious
358 consequences so log it to the panic log.
360 TF/02 Log LMTP confirmation messages in the same way as SMTP,
361 controlled using the smtp_confirmation log selector.
363 TF/03 Include the error message when we fail to unlink a spool file.
365 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
366 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
367 for maintaining out-of-tree patches for some time.
369 PP/01 Bugzilla 139: Documentation and portability issues.
370 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
371 Handle per-OS dynamic-module compilation flags.
373 PP/02 Let /dev/null have normal permissions.
374 The 4.73 fixes were a little too stringent and complained about the
375 permissions on /dev/null. Exempt it from some checks.
376 Reported by Andreas M. Kirchwitz.
378 PP/03 Report version information for many libraries, including
379 Exim version information for dynamically loaded libraries. Created
380 version.h, now support a version extension string for distributors
381 who patch heavily. Dynamic module ABI change.
383 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
384 privilege escalation vulnerability whereby the Exim run-time user
385 can cause root to append content of the attacker's choosing to
388 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
391 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
392 If dropping privileges for untrusted macros, we disabled normal logging
393 on the basis that it would fail; for the Exim run-time user, this is not
394 the case, and it resulted in successful deliveries going unlogged.
395 Fixed. Reported by Andreas Metzler.
401 PP/01 Date: & Message-Id: revert to normally being appended to a message,
402 only prepend for the Resent-* case. Fixes regression introduced in
403 Exim 4.70 by NM/22 for Bugzilla 607.
405 PP/02 Include check_rfc2047_length in configure.default because we're seeing
406 increasing numbers of administrators be bitten by this.
408 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
410 PP/03 Bugzilla 994: added openssl_options main configuration option.
412 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
414 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
416 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
418 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
419 without a peer certificate, leading to a segfault because of an
420 assumption that peers always have certificates. Be a little more
421 paranoid. Problem reported by Martin Tscholak.
423 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
424 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
425 NB: ClamAV planning to remove STREAM in "middle of 2010".
426 CL also introduces -bmalware, various -d+acl logging additions and
427 more caution in buffer sizes.
429 PP/09 Implemented reverse_ip expansion operator.
431 PP/10 Bugzilla 937: provide a "debug" ACL control.
433 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
435 PP/12 Bugzilla 973: Implement --version.
437 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
439 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
441 PP/15 Bugzilla 816: support multiple condition rules on Routers.
443 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
444 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
445 ignore trailing whitespace.
447 JJ/02 prevent non-panic DKIM error from being sent to paniclog
449 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
452 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
453 Notification from Dr Andrew Aitchison.
455 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
456 ExtendedDetectionInfo response format.
457 Notification from John Horne.
459 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
462 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
463 XSL and documented dependency on system catalogs, with examples of how
466 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
469 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
470 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
471 configuration file which is writeable by the Exim user or group.
473 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
474 of configuration files to cover files specified with the -C option if
475 they are going to be used with root privileges, not just the default
478 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
479 option (effectively making it always true).
481 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
482 files to be used while preserving root privileges.
484 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
485 that rogue child processes cannot use them.
487 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
488 run-time user, instead of root.
490 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
491 Exim run-time user without dropping privileges.
493 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
494 result string, instead of calling string_vformat() twice with the same
497 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
498 for other users. Others should always drop root privileges if they use
499 -C on the command line, even for a whitelisted configure file.
501 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
503 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
509 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
510 $data_path, and $header_path variables; fixed documentation bugs and
513 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
514 exipick to access non-standard spools, including the "frozen" queue
517 NM/01 Bugzilla 965: Support mysql stored procedures.
518 Patch from Alain Williams
520 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
522 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
523 Patch from Andreas Metzler
525 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
526 Patch from Kirill Miazine
528 NM/05 Bugzilla 671: Added umask to procmail example.
530 JJ/03 installed exipick 20100323.0, fixing doc bug
532 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
533 directory. Notification and patch from Dan Rosenberg.
535 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
537 TK/02 Improve log output when DKIM signing operation fails.
539 MH/01 Treat the transport option dkim_domain as a colon separated
540 list, not as a single string, and sign the message with each element,
541 omitting multiple occurences of the same signer.
543 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
544 Bugzilla 985, 986. Patch by Simon Arlott
546 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
547 Patch by Simon Arlott
549 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
550 MBX locking. Notification from Dan Rosenberg.
556 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
558 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
560 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
562 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
564 NM/04 Bugzilla 915: Flag broken perl installation during build.
570 TK/01 Added patch by Johannes Berg that expands the main option
571 "spamd_address" if it starts with a dollar sign.
573 TK/02 Write list of recipients to X-Envelope-Sender header when building
574 the mbox-format spool file for content scanning (suggested by Jakob
577 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
578 (http://www.dcc-servers.net/) support via dccifd. Activated by
579 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
581 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
582 by Mark Daniel Reidel <mr@df.eu>.
584 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
585 When building exim an external PCRE library is now needed -
586 PCRE is a system library on the majority of modern systems.
587 See entry on PCRE_LIBS in EDITME file.
589 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
590 conversation. Added nologin parameter to request.
591 Patch contributed by Kirill Miazine.
593 TF/01 Do not log submission mode rewrites if they do not change the address.
595 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
597 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
598 log files in place. Contributed by Roberto Lima.
600 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
602 TF/03 Bugzilla 615: When checking the local_parts router precondition
603 after a local_part_suffix or local_part_prefix option, Exim now
604 does not use the address's named list lookup cache, since this
605 contains cached lookups for the whole local part.
607 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
608 Robert Millan. Documentation is in experimental-spec.txt.
610 TF/04 Bugzilla 668: Fix parallel build (make -j).
612 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
614 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
615 Patch provided by Jan Srzednicki.
617 TF/05 Leading white space used to be stripped from $spam_report which
618 wrecked the formatting. Now it is preserved.
620 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
621 that they are available at delivery time.
623 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
625 TF/08 TLS error reporting now respects the incoming_interface and
626 incoming_port log selectors.
628 TF/09 Produce a more useful error message if an SMTP transport's hosts
629 setting expands to an empty string.
631 NM/06 Bugzilla 744: EXPN did not work under TLS.
632 Patch provided by Phil Pennock.
634 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
635 Patch provided by Richard Godbee.
637 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
638 acl_smtp_notquit, added index entry.
640 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
641 Patch provided by Eugene Bujak.
643 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
644 accept(). Patch provided by Maxim Dounin.
646 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
647 Patch provided by Phil Pennock.
649 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
651 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
652 Patch provided by Brad "anomie" Jorsch.
654 NM/14 Bugzilla 622: Added timeout setting to transport filter.
655 Patch provided by Dean Brooks.
657 TK/05 Add native DKIM support (does not depend on external libraries).
659 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
660 Patch provided by Graeme Fowler.
662 NM/16 Bugzilla 851: Documentation example syntax fix.
664 NM/17 Changed NOTICE file to remove references to embedded PCRE.
666 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
669 NM/19 Bugzilla 745: TLS version reporting.
670 Patch provided by Phil Pennock.
672 NM/20 Bugzilla 167: bool: condition support.
673 Patch provided by Phil Pennock.
675 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
676 clients. Patch provided by Phil Pennock.
678 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
679 Patch provided by Brad "anomie" Jorsch.
681 NM/23 Bugzilla 687: Fix misparses in eximstats.
682 Patch provided by Heiko Schlittermann.
684 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
685 Patch provided by Heiko Schlittermann.
687 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
688 plus update to original patch.
690 NM/26 Bugzilla 799: Documentation correction for ratelimit.
692 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
693 Patch provided by David Brownlee.
695 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
697 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
699 NM/30 Bugzilla 888: TLS documentation bugfixes.
701 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
703 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
704 Unlike the original bugzilla I have changed all shell scripts in src tree.
706 NM/33 Bugzilla 898: Transport filter timeout fix.
707 Patch by Todd Rinaldo.
709 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
710 Patch by Serge Demonchaux.
712 NM/35 Bugzilla 39: Base64 decode bug fixes.
713 Patch by Jakob Hirsch.
715 NM/36 Bugzilla 909: Correct connect() call in dcc code.
717 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
719 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
721 NM/39 Bugzilla 911: Fixed MakeLinks build script.
727 TK/01 Add preliminary DKIM support. Currently requires a forked version of
728 ALT-N's libdkim that I have put here:
729 http://duncanthrax.net/exim-experimental/
731 Note to Michael Haardt: I had to rename some vars in sieve.c. They
732 were called 'true' and it seems that C99 defines that as a reserved
733 keyword to be used with 'bool' variable types. That means you could
734 not include C99-style headers which use bools without triggering
735 build errors in sieve.c.
737 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
738 as mailq or other aliases. Changed the --help handling significantly
739 to do whats expected. exim_usage() emits usage/help information.
741 SC/01 Added the -bylocaldomain option to eximstats.
743 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
745 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
747 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
753 PH/01 Another patch from the Sieve maintainer.
755 PH/02 When an IPv6 address is converted to a string for single-key lookup
756 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
757 dots are used instead of colons so that keys in lsearch files need not
758 contain colons. This was done some time before quoting was made available
759 in lsearch files. However, iplsearch files do require colons in IPv6 keys
760 (notated using the quote facility) so as to distinguish them from IPv4
761 keys. This meant that lookups for IP addresses in host lists did not work
762 for iplsearch lookups.
764 This has been fixed by arranging for IPv6 addresses to be expressed with
765 colons if the lookup type is iplsearch. This is not incompatible, because
766 previously such lookups could never work.
768 The situation is now rather anomolous, since one *can* have colons in
769 ordinary lsearch keys. However, making the change in all cases is
770 incompatible and would probably break a number of configurations.
772 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
775 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
776 conversion specification without a maximum field width, thereby enabling
777 a rogue spamd server to cause a buffer overflow. While nobody in their
778 right mind would setup Exim to query an untrusted spamd server, an
779 attacker that gains access to a server running spamd could potentially
780 exploit this vulnerability to run arbitrary code as the Exim user.
782 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
783 $primary_hostname instead of what libspf2 thinks the hosts name is.
785 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
786 a directory entry by the name of the lookup key. Previously, if a
787 symlink pointed to a non-existing file or a file in a directory that
788 Exim lacked permissions to read, a lookup for a key matching that
789 symlink would fail. Now it is enough that a matching directory entry
790 exists, symlink or not. (Bugzilla 503.)
792 PH/03 The body_linecount and body_zerocount variables are now exported in the
795 PH/04 Added the $dnslist_matched variable.
797 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
798 This means they are set thereafter only if the connection becomes
801 PH/06 Added the client_condition to authenticators so that some can be skipped
802 by clients under certain conditions.
804 PH/07 The error message for a badly-placed control=no_multiline_responses left
805 "_responses" off the end of the name.
807 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
809 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
810 (without spaces) instead of just copying the configuration text.
812 PH/10 Added the /noupdate option to the ratelimit ACL condition.
814 PH/11 Added $max_received_linelength.
816 PH/12 Added +ignore_defer and +include_defer to host lists.
818 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
819 way in which PCRE > 7.0 is built.
821 PH/14 Implemented queue_only_load_latch.
823 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
824 MAIL command. The effect was to mangle the value on 64-bit systems.
826 PH/16 Another patch from the Sieve maintainer.
828 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
830 PH/18 If a system quota error occurred while trying to create the file for
831 a maildir delivery, the message "Mailbox is full" was not appended to the
832 bounce if the delivery eventually timed out. Change 4.67/27 below applied
833 only to a quota excession during the actual writing of the file.
835 PH/19 It seems that peer DN values may contain newlines (and other non-printing
836 characters?) which causes problems in log lines. The DN values are now
837 passed through string_printing() before being added to log lines.
839 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
840 and InterBase are left for another time.)
842 PH/21 Added message_body_newlines option.
844 PH/22 Guard against possible overflow in moan_check_errorcopy().
846 PH/23 POSIX allows open() to be a macro; guard against that.
848 PH/24 If the recipient of an error message contained an @ in the local part
849 (suitably quoted, of course), incorrect values were put in $domain and
850 $local_part during the evaluation of errors_copy.
856 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
857 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
860 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
861 issue a MAIL command.
863 PH/02 In an ACL statement such as
865 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
867 if a client was not listed at all, or was listed with a value other than
868 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
869 the condition was not true (as it should be), so access was not denied.
870 The bug was that the ! inversion was incorrectly passed on to the second
871 item. This has been fixed.
873 PH/03 Added additional dnslists conditions == and =& which are different from
874 = and & when the dns lookup returns more than one IP address.
876 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
877 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
879 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
880 FSYNC, which compiles an option called disable_fsync that allows for
881 bypassing fsync(). The documentation is heavily laced with warnings.
883 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
885 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
886 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
887 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
888 including adding "make clean"; (3) Added -fPIC when compiling the test
889 dynamically loaded module, to get rid of a warning.
891 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
892 message fails, move_frozen_messages = true and ignore_bounce_errors_after
893 = 0s. The bug is otherwise harmless.
895 PH/07 There was a bug in the dovecot authenticator such that the value of
896 $auth1 could be overwritten, and so not correctly preserved, after a
897 successful authentication. This usually meant that the value preserved by
898 the server_setid option was incorrect.
900 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
902 PH/09 Installed PCRE release 7.0.
904 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
905 run for batched SMTP input. It is now run at the start of every message
906 in the batch. While fixing this I discovered that the process information
907 (output by running exiwhat) was not always getting set for -bs and -bS
908 input. This is fixed, and it now also says "batched" for BSMTP.
910 PH/11 Added control=no_pipelining.
912 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
913 patch, slightly modified), and move the expansion of helo_data till after
914 the connection is made in the smtp transport (so it can use these
917 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
919 PH/14 Added log_selector = +pid.
921 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
923 PH/16 Add ${if forany and ${if forall.
925 PH/17 Added dsn_from option to vary the From: line in DSNs.
927 PH/18 Flush SMTP output before performing a callout, unless control =
928 no_callout_flush is set.
930 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
931 was true (the default) a successful delivery failed to delete the retry
932 item, thus causing premature timeout of the address. The bug is now
935 PH/20 Added hosts_avoid_pipelining to the smtp transport.
937 PH/21 Long custom messages for fakedefer and fakereject are now split up
938 into multiline reponses in the same way that messages for "deny" and
939 other ACL rejections are.
941 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
942 with slight modification.
944 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
945 draft, changing the syntax and factoring some duplicate code".
947 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
948 for deliveries of the second and subsequent messages over the same SMTP
951 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
952 ${reduce, with only minor "tidies".
954 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
956 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
957 expansion side effects.
959 PH/27 When a message times out after an over-quota error from an Exim-imposed
960 quota, the bounce message says "mailbox is full". This message was not
961 being given when it was a system quota that was exceeded. It now should
964 MH/03 Made $recipients available in local_scan(). local_scan() already has
965 better access to the recipient list through recipients_list[], but
966 $recipients can be useful in postmaster-provided expansion strings.
968 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
969 in the case of a MAIL command with additional options following the
970 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
971 were accidentally chopped off.
973 PH/29 SMTP synchronization checks are implemented when a command is read -
974 there is a check that no more input is waiting when there shouldn't be
975 any. However, for some commands, a delay in an ACL can mean that it is
976 some time before the response is written. In this time, more input might
977 arrive, invalidly. So now there are extra checks after an ACL has run for
978 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
979 pipelining has not been advertised.
981 PH/30 MH's patch to allow iscntrl() characters to be list separators.
983 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
984 returned in the SMTP response when smtp_return_error_details was false.
987 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
988 instead of the C I/O that was originally supplied, because problems were
991 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
992 Exim which did not show up earlier: it was assuming that a call to
993 SSL_CTX_set_info_callback() might give an error value. In fact, there is
994 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
995 was a macro that became an assignment, so it seemed to work. This has
996 changed to a proper function call with a void return, hence the compile
997 error. Exim's code has been fixed.
999 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1002 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1004 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1006 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1007 because it turns out to be too big or there is a timeout) there is no
1008 "Completed" line in the log. When some messages of this type were
1009 selected by exigrep, they were listed as "not completed". Others were
1010 picked up by some special patterns. I have improved the selection
1011 criteria to be more general.
1013 PH/38 The host_find_failed option in the manualroute router can now be set
1014 to "ignore", to completely ignore a host whose IP address cannot be
1015 found. If all hosts are ignored, the behaviour is controlled by the new
1016 host_all_ignored option.
1018 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1019 homing or because of multiple MX records with /mx) generated more than
1020 one IP address, and the following item turned out to be the local host,
1021 all the secondary addresses of the first item were incorrectly removed
1022 from the list, along with the local host and any following hosts (which
1023 is what is supposed to happen).
1025 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1026 whoever called Exim into the -H file. In the case of the daemon it was
1027 behaving confusingly. When first started, it used values for whoever
1028 started the daemon, but after a SIGHUP it used the Exim user (because it
1029 calls itself on a restart). I have changed the code so that it now always
1032 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1033 message are rejected with the same error (e.g. no authentication or bad
1034 sender address), and a DATA command is nevertheless sent (as can happen
1035 with PIPELINING or a stupid MUA), the error message that was given to the
1036 RCPT commands is included in the rejection of the DATA command. This is
1037 intended to be helpful for MUAs that show only the final error to their
1040 PH/42 Another patch from the Sieve maintainer.
1042 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1043 Eximstats - Fixed some broken HTML links and added missing column headers
1045 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1046 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1048 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1050 PH/43 Yet another patch from the Sieve maintainer.
1052 PH/44 I found a way to check for a TCP/IP connection going away before sending
1053 the response to the final '.' that terminates a message, but only in the
1054 case where the client has not sent further data following the '.'
1055 (unfortunately, this is allowed). However, in many cases there won't be
1056 any further data because there won't be any more messages to send. A call
1057 to select() can be used: if it shows that the input is "ready", there is
1058 either input waiting, or the socket has been closed. An attempt to read
1059 the next input character can distinguish the two cases. Previously, Exim
1060 would have sent an OK response which the client would never have see.
1061 This could lead to message repetition. This fix should cure that, at
1062 least in a lot of common cases.
1064 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1065 advertised in response to EHLO.
1071 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1072 fixed by 4.65/MH/01 (is this a record?) are fixed:
1074 (i) An empty string was always treated as zero by the numeric comparison
1075 operators. This behaviour has been restored.
1077 (ii) It is documented that the numeric comparison operators always treat
1078 their arguments as decimal numbers. This was broken in that numbers
1079 starting with 0 were being interpreted as octal.
1081 While fixing these problems I realized that there was another issue that
1082 hadn't been noticed. Values of message_size_limit (both the global option
1083 and the transport option) were treated as octal if they started with 0.
1084 The documentation was vague. These values are now always treated as
1085 decimal, and I will make that clear in the documentation.
1091 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1092 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1095 MH/01 Don't check that the operands of numeric comparison operators are
1096 integers when their expansion is in "skipping" mode (fixes bug
1097 introduced by 4.64-PH/07).
1099 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1100 child addresses, Exim now panics and dies. Previously, because the count
1101 is held in a short int, deliveries were likely to be lost. As such a
1102 large number of recipients for a single message is ridiculous
1103 (performance will be very, very poor), I have chosen to impose a limit
1104 rather than extend the field.
1110 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1111 leftover -K file (the existence of which was triggered by #402).
1112 While we were at it, introduced process PID as part of the -K
1113 filename. This should rule out race conditions when creating
1116 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1117 processing considerably. Previous code took too long for large mails,
1118 triggering a timeout which in turn triggers #401.
1120 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1121 in the DK code in transports.c. sendfile() is not really portable,
1122 hence the _LINUX specificness.
1124 TF/01 In the add_headers option to the mail command in an Exim filter,
1125 there was a bug that Exim would claim a syntax error in any
1126 header after the first one which had an odd number of characters
1129 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1130 callout verification, Exim cached a "reject" for the entire domain. This
1131 is correct for most verifications, but it is not correct for a recipient
1132 verification with use_sender or use_postmaster set, because in that case
1133 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1134 case of MAIL FROM:<> rejection from other early rejections (e.g.
1135 rejection of HELO). When verifying a recipient using a non-null MAIL
1136 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1137 Whatever the result of the callout, the value of the domain cache is
1138 left unchanged (for any other kind of callout, getting as far as trying
1139 RCPT means that the domain itself is ok).
1141 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1144 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1145 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1146 session, the daemon ignores ECONNECT errors and logs others; it now
1147 ignores EPIPE as well.
1149 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1150 (quoted-printable decoding).
1152 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1153 later the small subsequent patch to fix an introduced bug.
1155 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1157 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1159 PH/08 An error is now given if message_size_limit is specified negative.
1161 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1162 to be given (somewhat) arbitrary names.
1164 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1167 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1168 miscellaneous code fixes
1170 PH/10 Added the log_reject_target ACL modifier to specify where to log
1173 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1174 hostname. This is wrong, because it relates to the incoming message (and
1175 probably the interface on which it is arriving) and not to the outgoing
1176 callout (which could be using a different interface). This has been
1177 changed to use the value of the helo_data option from the smtp transport
1178 instead - this is what is used when a message is actually being sent. If
1179 there is no remote transport (possible with a router that sets up host
1180 addresses), $smtp_active_hostname is used.
1182 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1183 tweaks were necessary in order to get it to work (see also 21 below):
1184 (a) The code assumed that strncpy() returns a negative number on buffer
1185 overflow, which isn't the case. Replaced with Exim's string_format()
1187 (b) There were several signed/unsigned issues. I just did the minimum
1188 hacking in of casts. There is scope for a larger refactoring.
1189 (c) The code used strcasecmp() which is not a standard C function.
1190 Replaced with Exim's strcmpic() function.
1191 (d) The code set only $1; it now sets $auth1 as well.
1192 (e) A simple test gave the error "authentication client didn't specify
1193 service in request". It would seem that Dovecot has changed its
1194 interface. Fortunately there's a specification; I followed it and
1195 changed what the client sends and it appears to be working now.
1197 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1200 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1201 address A is aliased to B and C, where B exists and C does not. Without
1202 -v the output is "A verified" because verification stops after a
1203 successful redirection if more than one address is generated. However,
1204 with -v the child addresses are also verified. Exim was outputting "A
1205 failed to verify" and then showing the successful verification for C,
1206 with its parentage. It now outputs "B failed to verify", showing B's
1207 parentage before showing the successful verification of C.
1209 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1210 look up a TXT record in a specific list after matching in a combined
1213 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1214 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1215 they consult the DNS. I had assumed they would set it the way they
1216 wanted; and indeed my experiments on Linux seem to show that in some
1217 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1218 To be on the safe side, however, I have now made the interface to
1219 host_find_byname() similar to host_find_bydns(), with an argument
1220 containing the DNS resolver options. The host_find_byname() function now
1221 sets these options at its start, just as host_find_bydns() does. The smtp
1222 transport options dns_qualify_single and dns_search_parents are passed to
1223 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1224 of host_find_byname() use the default settings of RES_DEFNAMES
1225 (qualify_single) but not RES_DNSRCH (search_parents).
1227 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1228 spool_read_header() do less string testing, by means of a preliminary
1229 switch on the second character of optional "-foo" lines. (This is
1230 overdue, caused by the large number of possibilities that now exist.
1231 Originally there were few.) While I was there, I also converted the
1232 str(n)cmp tests so they don't re-test the leading "-" and the first
1233 character, in the hope this might squeeze out yet more improvement.
1235 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1236 flag allowing group syntax was set by the header_syntax check but not
1237 turned off, possible causing trouble later; (2) The flag was not being
1238 set at all for the header_verify test, causing "group"-style headers to
1239 be rejected. I have now set it in this case, and also caused header_
1240 verify to ignore an empty address taken from a group. While doing this, I
1241 came across some other cases where the code for allowing group syntax
1242 while scanning a header line wasn't quite right (mostly, not resetting
1243 the flag correctly in the right place). These bugs could have caused
1244 trouble for malformed header lines. I hope it is now all correct.
1246 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1247 with the "reply" argument non-NULL. The code, however (which originally
1248 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1249 but it didn't always do it. This confused somebody who was copying the
1250 code for some other use. I have removed all the tests.
1252 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1253 feature that was used to support insecure browsers during the U.S. crypto
1254 embargo. It requires special client support, and Exim is probably the
1255 only MTA that supported it -- and would never use it because real RSA is
1256 always available. This code has been removed, because it had the bad
1257 effect of slowing Exim down by computing (never used) parameters for the
1258 RSA_EXPORT functionality.
1260 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1261 authenticator to fail if there's a tab character in the incoming data
1262 (there should never be unless someone is messing about, as it's supposed
1263 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1264 if the connection is using TLS or if the remote IP is the same as the
1265 local IP, and the "valid-client-cert option" if a client certificate has
1268 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1269 authenticators. This can be used for authorization after authentication
1270 succeeds. (In the case of plaintext, it servers for both authentication
1273 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1274 if any retry times were supplied.
1276 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1277 connection, where there is no client IP address to check. In this
1278 situation, the verify now always succeeds.
1280 PH/25 Applied John Jetmore's -Mset patch.
1282 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1284 PH/27 In a string expansion for a processed (not raw) header when multiple
1285 headers of the same name were present, leading whitespace was being
1286 removed from all of them, but trailing whitespace was being removed only
1287 from the last one. Now trailing whitespace is removed from each header
1288 before concatenation. Completely empty headers in a concatenation (as
1289 before) are ignored.
1291 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1292 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1294 PH/29 [Removed. This was a change that I later backed out, and forgot to
1295 correct the ChangeLog entry (that I had efficiently created) before
1296 committing the later change.]
1298 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1299 address errors (4xx response to RCPT) over the same connection as other
1300 messages routed to the same hosts. Such deliveries are always "forced",
1301 so retry times are not inspected. This resulted in far too many retries
1302 for the affected addresses. The effect occurred only when there were more
1303 hosts than the hosts_max_try setting in the smtp transport when it had
1304 the 4xx errors. Those hosts that it had tried were not added to the list
1305 of hosts for which the message was waiting, so if all were tried, there
1306 was no problem. Two fixes have been applied:
1308 (i) If there are any address or message errors in an SMTP delivery, none
1309 of the hosts (tried or untried) are now added to the list of hosts
1310 for which the message is waiting, so the message should not be a
1311 candidate for sending over the same connection that was used for a
1312 successful delivery of some other message. This seems entirely
1313 reasonable: after all the message is NOT "waiting for some host".
1314 This is so "obvious" that I'm not sure why it wasn't done
1315 previously. Hope I haven't missed anything, but it can't do any
1316 harm, as the worst effect is to miss an optimization.
1318 (ii) If, despite (i), such a delivery is accidentally attempted, the
1319 routing retry time is respected, so at least it doesn't keep
1320 hammering the server.
1322 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1323 in ${readsocket because some servers need this prod.
1325 PH/32 Added some extra debug output when updating a wait-xxx database.
1327 PH/33 The hint "could be header name not terminated by colon", which has been
1328 given for certain expansion errors for a long time, was not being given
1329 for the ${if def:h_colon_omitted{... case.
1331 PH/34 The spec says: "With one important exception, whenever a domain list is
1332 being scanned, $domain contains the subject domain." There was at least
1333 one case where this was not true.
1335 PH/35 The error "getsockname() failed: connection reset by peer" was being
1336 written to the panic log as well as the main log, but it isn't really
1337 panic-worthy as it just means the connection died rather early on. I have
1338 removed the panic log writing for the ECONNRESET error when getsockname()
1341 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1342 runs only) independently of the message's sender address. This meant
1343 that, if the 4xx error was in fact related to the sender, a different
1344 message to the same recipient with a different sender could confuse
1345 things. In particualar, this can happen when sending to a greylisting
1346 server, but other circumstances could also provoke similar problems.
1347 I have changed the default so that the retry time for these errors is now
1348 based a combination of the sender and recipient addresses. This change
1349 can be overridden by setting address_retry_include_sender=false in the
1352 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1353 remote server are returned as part of bounce messages. This was not
1354 happening for LMTP over a pipe (the lmtp transport), but now it is the
1355 same for both kinds of LMTP.
1357 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1358 in header lines that were in fact CNAMEs. This is no longer the case.
1360 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1361 and queue runs started by the daemon processed all messages. This has
1362 been fixed so that -R and -S can now usefully be given with -q<time>.
1364 PH/40 Import PCRE release 6.7 (fixes some bugs).
1366 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1368 PH/42 Give an error if -q is specified more than once.
1370 PH/43 Renamed the variables $interface_address and $interface_port as
1371 $received_ip_address and $received_port, to make it clear that these
1372 values apply to message reception, and not to the outgoing interface when
1373 a message is delivered. (The old names remain recognized, of course.)
1375 PH/44 There was no timeout on the connect() call when using a Unix domain
1376 socket in the ${readsocket expansion. There now is.
1378 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1379 be meaningful with "accept".
1381 SC/01 Eximstats V1.43
1382 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1384 SC/02 Eximstats V1.44
1385 Use a glob alias rather than an array ref in the generated
1386 parser. This improves both readability and performance.
1388 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1389 Collect SpamAssassin and rejection statistics.
1390 Don't display local sender or destination tables unless
1391 there is data to show.
1392 Added average volumes into the top table text output.
1394 SC/04 Eximstats V1.46
1395 Collect data on the number of addresses (recipients)
1396 as well as the number of messages.
1398 SC/05 Eximstats V1.47
1399 Added 'Message too big' to the list of mail rejection
1400 reasons (thanks to Marco Gaiarin).
1402 SC/06 Eximstats V1.48
1403 Mainlog lines which have GMT offsets and are too short to
1404 have a flag are now skipped.
1406 SC/07 Eximstats V1.49 (Alain Williams)
1407 Added the -emptyok flag.
1409 SC/08 Eximstats V1.50
1410 Fixes for obtaining the IP address from reject messages.
1412 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1413 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1414 whitesspace changes from 4.64-PH/27
1416 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1419 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1420 are found, allow negative numbers in numeric criteria)
1422 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1424 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1425 to match changes made in 4.64-PH/43
1427 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1429 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1430 "domain" is actually an IP address, and force a failure. This locks out
1431 those revolvers/nameservers that support "A-for-A" lookups, in
1432 contravention of the specifications.
1434 PH/48 When a host name was looked up from an IP address, and the subsequent
1435 forward lookup of the name timed out, the host name was left in
1436 $sender_host_name, contrary to the specification.
1438 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1439 restricted to single-key lookups, Exim was not diagnosing an error if
1440 * or *@ was used with a query-style lookup.
1442 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1444 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1445 long ago, but noone interested enough thought of it. Let's just say that
1446 the "1.1" means that there are some new functions that weren't there at
1447 some point in the past.
1449 PH/51 Error processing for expansion failure of helo_data from an smtp
1450 transport during callout processing was broken.
1452 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1453 tested/used via the -bh/-bhc/-bs options.
1455 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1456 bug, fixed in subsequent PCRE releases).
1458 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1459 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1461 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1467 SC/01 Use a glob alias rather than an array ref in eximstats generated
1468 parser. This improves both readability and performance.
1470 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1471 Don't display local sender or destination tables in eximstats unless
1472 there is data to show.
1473 Added average volumes into the eximstats top table text output.
1475 SC/03 Collect data on the number of addresses (recipients) as well
1476 as the number of messages in eximstats.
1478 TF/01 Correct an error in the documentation for the redirect router. Exim
1479 does (usually) call initgroups() when daemonizing.
1481 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1482 with consistent privilege compared to when running as a daemon.
1484 TF/03 Note in the spec that $authenticated_id is not set for local
1485 submissions from trusted users.
1487 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1488 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1490 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1491 by adding some example configuration directives to the default
1492 configuration file. A little bit of work is required to uncomment the
1493 directives and define how usernames and passwords are checked, but
1494 there is now a framework to start from.
1496 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1497 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1498 without this. I don't know how relevant this is to other LDAP libraries.
1500 PH/02 Add the verb name to the "unknown ACL verb" error.
1502 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1504 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1506 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1507 directory not expanded when it should be if an expanded home directory
1508 was set for the address (which is overridden by the transport).
1510 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1513 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1514 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1515 because it is too late at that time, and has no effect.
1517 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1518 security issue with \' (bugzilla #107). I could not use the
1519 PQescapeStringConn() function, because it needs a PGconn value as one of
1522 PH/08 When testing addresses using -bt, indicate those final addresses that
1523 are duplicates that would not cause an additional delivery. At least one
1524 person was confused, thinking that -bt output corresponded to deliveries.
1525 (Suppressing duplicates isn't a good idea as you lose the information
1526 about possibly different redirections that led to the duplicates.)
1528 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1529 systems where poll() doesn't work, in particular OS X.
1531 PH/10 Added more information to debugging output for retry time not reached.
1533 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1534 operations in malware.c.
1536 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1539 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1540 syslog_duplication set false, log lines that would normally be written
1541 both the the main log and to the reject log were not written to syslog at
1544 PH/14 In the default configuration, change the use of "message" in ACL warn
1545 statements to "add_header".
1547 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1548 not followed by a command (e.g. "seen endif").
1550 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1551 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1554 PH/17 Added extra conditions to the default value of delay_warning_condition
1558 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1559 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1560 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1563 The Auto-Submitted: and various List- headers are standardised, whereas I
1564 don't think Precedence: ever was.
1566 PH/18 Refactored debugging code in route_finduser() to show more information,
1567 in particular, the error code if getpwnam() issues one.
1569 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1570 This is apparently needed in addition to the PH/07 change above to avoid
1571 any possible encoding problems.
1573 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1574 but not after initializing Perl.
1576 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1577 output them only if debugging. By default they are written stderr,
1578 apparently, which is not desirable.
1580 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1583 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1586 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1588 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1589 authenticated or an ident call has been made. Suppress the default
1590 values for $authenticated_id and $authenticated_sender (but permit -oMai
1591 and -oMas) when testing with -bh.
1593 PH/24 Re-jigged the order of the tests in the default configuration so that the
1594 tests for valid domains and recipients precede the DNS black list and CSA
1595 tests, on the grounds that those ones are more expensive.
1597 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1598 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1599 This bug exists in every version of Exim that I still have, right back to
1602 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1603 However, an attempt to turn on case-sensitivity in a regex key by
1604 including (?-i) didn't work because the subject string was already
1605 lowercased, and the effects were non-intuitive. It turns out that a
1606 one-line patch can be used to allow (?-i) to work as expected.
1612 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1613 other effects) broke the use of negated acl sub-conditions.
1615 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1618 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1619 "Deny" causes Exim to reject the incoming connection with a 554 error.
1620 Unfortunately, if there is a major crisis, such as a disk failure,
1621 tcp-wrappers gives "deny", whereas what one would like would be some
1622 kind of temporary error. A kludge has been added to help with this.
1623 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1624 554 error is used if errno is still zero or contains ENOENT (which occurs
1625 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1628 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1630 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1631 errors. Otherwise a message that provokes a temporary error (when other
1632 messages do not) can cause a whole host to time out.
1634 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1635 addresses were routed directly to files or pipes from a redirect router.
1636 File deliveries just didn't batch; pipe deliveries might have suffered
1639 PH/06 A failure to get a lock for a hints database would erroneously always say
1640 "Failed to get write lock", even when it was really a read lock.
1642 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1643 of 0600. This has been changed to use the value of the lockfile_mode
1644 option (which defaults to 0600).
1646 PH/08 Applied small patch from the Sieve maintainer.
1648 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1649 folder from quota calculations, a direct delivery into this folder messed
1650 up the contents of the maildirsize file. This was because the regex was
1651 used only to exclude .Trash (or whatever) when the size of the mailbox
1652 was calculated. There was no check that a delivery was happening into an
1653 excluded directory. This bug has been fixed by ignoring all quota
1654 processing for deliveries into excluded directories.
1656 PH/10 Added the maildirfolder_create_regex option to appendfile.
1662 PH/01 The code for finding all the local interface addresses on a FreeBSD
1663 system running IPv6 was broken. This may well have applied to all BSD
1664 systems, as well as to others that have similar system calls. The broken
1665 code found IPv4 interfaces correctly, but gave incorrect values for the
1666 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1667 that it would not match correctly against @[] and not recognize the IPv6
1670 PH/02 The ipliteral router was not recognizing addresses of the form user@
1671 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1673 PH/03 Added disable_ipv6.
1675 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1676 decoded form, because it is most often used to construct To: headers
1677 lines in autoreplies, and the decoded form may well be syntactically
1678 invalid. However, $reply_address has leading white space removed, and all
1679 newlines turned into spaces so that the autoreply transport does not
1682 PH/05 If group was specified without a user on a router, and no group or user
1683 was specified on a transport, the group from the router was ignored.
1685 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1686 for visible compile-time settings that can be used to change these
1687 numbers, for those that want even more. Backwards compatibility with old
1688 spool files has been maintained. However, going back to a previous Exim
1689 release will lost any variables that are in spool files.
1691 PH/07 Two small changes when running in the test harness: increase delay when
1692 passing a TCP/IP connection to a new process, in case the original
1693 process has to generate a bounce, and remove special handling of
1694 127.0.0.2 (sic), which is no longer necessary.
1696 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1697 be the same on different OS.
1699 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1702 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1703 whether --show-vars was specified or not
1705 JJ/02 exipick: Added support for new ACL variable spool format introduced
1708 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1709 syntactically invalid From: or Reply-to: line, and a filter used this to
1710 generate an autoreply, and therefore failed to obtain an address for the
1711 autoreply, Exim could try to deliver to a non-existent relative file
1712 name, causing unrelated and misleading errors. What now happens is that
1713 it logs this as a hard delivery error, but does not attempt to create a
1716 PH/11 The exinext utility has a -C option for testing purposes, but although
1717 the given file was scanned by exinext itself; it wasn't being passed on
1718 when Exim was called.
1720 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1721 an end-of-file indication when reading a command response.
1723 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1724 compiled. In many other places in Exim, IPv6 addresses are always
1725 recognized, so I have changed this. It also means that IPv4 domain
1726 literals of the form [IPV4:n.n.n.n] are now always recognized.
1728 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1729 used if the router is not running as root, for example, when verifying at
1730 ACL time, or when using -bh. The debugging output from this situation was
1731 non-existent - all you got was a failure to exec. I have made two
1734 (a) Failures to set uid/gid, the current directory, or a process leader
1735 in a subprocess such as that created by queryprogram now generate
1736 suitable debugging ouput when -d is set.
1738 (b) The queryprogram router detects when it is not running as root,
1739 outputs suitable debugging information if -d is set, and then runs
1740 the subprocess without attempting to change uid/gid.
1742 PH/15 Minor change to Makefile for building test_host (undocumented testing
1745 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1746 additional section of a DNS packet that returns MX or SRV records.
1747 Instead, it always explicitly searches for A/AAAA records. This avoids
1748 major problems that occur when a DNS server includes only records of one
1749 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1750 fixed another bug: if SRV records were looked up and the corresponding
1751 address records were *not* found in the additional section, the port
1752 values from the SRV records were lost.
1754 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1755 using the correct key (the original address) when searching the retry
1756 rules in order to find which one to use for generating the retry hint.
1758 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1759 adding the default one. Similarly, if it contains a Reply-To: header, the
1760 errors_reply_to option, if set, is not used.
1762 PH/19 When calculating a retry time, Exim used to measure the "time since
1763 failure" by looking at the "first failed" field in the retry record. Now
1764 it does not use this if it is later than than the arrival time of the
1765 message. Instead it uses the arrival time. This makes for better
1766 behaviour in cases where some deliveries succeed, thus re-setting the
1767 "first failed" field. An example is a quota failure for a huge message
1768 when small messages continue to be delivered. Without this change, the
1769 "time since failure" will always be short, possible causing more frequent
1770 delivery attempts for the huge message than are intended.
1771 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1773 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1774 $1, $2, $3) because the numerical variables can be reset during some
1775 expansion items (e.g. "match"), thereby losing the authentication data.
1777 PH/21 Make -bV show the size of off_t variables so that the test suite can
1778 decide whether to run tests for quotas > 2G.
1780 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1781 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1782 filecount value is greater than 2G or if a quota value is greater than 2G
1783 on a system where the size of off_t is not greater than 4, a panic error
1786 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1787 never match. The debug and -bh output now contains an explicit error
1788 message indicating a malformed IPv4 address or mask.
1790 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1791 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1792 PH/23 above applies.
1794 PH/25 Do not write to syslog when running in the test harness. The only
1795 occasion when this arises is a failure to open the main or panic logs
1796 (for which there is an explicit test).
1798 PH/26 Added the /no_tell option to "control=freeze".
1800 PH/27 If a host name lookup failed very early in a connection, for example, if
1801 the IP address matched host_lookup and the reverse lookup yielded a name
1802 that did not have a forward lookup, an error message of the form "no IP
1803 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1804 could be logged. Now it outputs the IP address instead of "NULL".
1806 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1807 allows the sender and the authenticated sender to be set when
1808 submitting a message from within Exim. Since child_open_exim() is
1809 documented for local_scan(), the new function should be too.
1811 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1812 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1813 results in an empty string is now treated as unset.
1815 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1817 PH/31 Added sender_verify_fail logging option.
1819 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1820 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1821 tidied the source and removed it altogether.
1823 PH/33 When a queue run was abandoned because the load average was too high, a
1824 log line was always written; now it is written only if the queue_run log
1825 selector is set. In addition, the log line for abandonment now contains
1826 information about the queue run such as the pid. This is always present
1827 in "start" and "stop" lines but was omitted from the "abandon" line.
1829 PH/34 Omit spaces between a header name and the colon in the error message that
1830 is given when verify = headers_syntax fails (if there are lots of them,
1831 the message gets confusing).
1833 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1834 names, as there are now some PTR records that contain slashes. This check
1835 is only to protect against broken name servers that fall over on strange
1836 characters, so the fact that it applies to all lookups doesn't matter.
1838 PH/36 Now that the new test suite is complete, we can remove some of the
1839 special code in Exim that was needed for the old test suite. For example,
1840 sorting DNS records because real resolvers return them in an arbitrary
1841 order. The new test suite's fake resolver always returns records in the
1844 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1845 bounces) except when queue_only is set, to avoid logging races between
1846 the different processes.
1848 PH/38 Panic-die if .include specifies a non-absolute path.
1850 PH/39 A tweak to the "H" retry rule from its user.
1852 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1853 a label. They prevented compilation on older perls.
1855 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1856 a warning to be raised on newish perls.
1858 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1859 on queue. Changes to match documented behaviour of showing count of
1860 messages matching specified criteria.
1862 PH/40 Changed the default ident timeout from 30s to 5s.
1864 PH/41 Added support for the use of login_cap features, on those BSD systems
1865 that have them, for controlling the resources used by pipe deliveries.
1867 PH/42 The content-scanning code uses fopen() to create files in which to put
1868 message data. Previously it was not paying any attention to the mode of
1869 the files. Exim runs with umask(0) because the rest of the code creates
1870 files with open(), and sets the required mode explicitly. Thus, these
1871 files were ending up world-writeable. This was not a big issue, because,
1872 being within the spool directory, they were not world-accessible. I have
1873 created a function called modefopen, which takes an additional mode
1874 argument. It sets umask(777), creates the file, chmods it to the required
1875 mode, then resets the umask. All the relevant calls to fopen() in the
1876 content scanning code have been changed to use this function.
1878 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1879 to 24 hours. This avoids potential overflow problems when processing G
1880 and H retry rules. I suspect nobody ever tinkers with this value.
1882 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1884 PH/45 When the plaintext authenticator is running as a client, the server's
1885 challenges are checked to ensure they are valid base64 strings. By
1886 default, the authentication attempt is cancelled if an invalid string is
1887 received. Setting client_ignore_invalid_base64 true ignores these errors.
1888 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1889 they are received. Thus, the responses can be made to depend on the
1890 challenges. If an invalid string is ignored, an empty string is placed in
1893 PH/46 Messages that are created by the autoreply transport now contains a
1894 References: header, in accordance with RFCs 2822 and 3834.
1896 PH/47 Added authenticated_sender_force to the smtp transport.
1898 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1900 PH/49 Installed latest patch from the Sieve maintainer.
1902 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1903 was also set, the appendfile transport was unnecessarily scanning a
1904 directory of message files (e.g. for maildir delivery) to find the count
1905 of files (along with the size), even though it did not need this
1906 information. It now does the scan only if it needs to find either the
1907 size of the count of files.
1909 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1911 PH/52 Two bugs concerned with error handling when the smtp transport is
1914 (i) Exim was not creating retry information for temporary errors given
1915 for individual recipients after the DATA command when the smtp transport
1916 was used in LMTP mode. This meant that they could be retried too
1917 frequently, and not timed out correctly.
1919 (ii) Exim was setting the flag that allows error details to be returned
1920 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1921 recipients that were returned after the DATA command.
1923 PH/53 This is related to PH/52, but is more general: for any failing address,
1924 when detailed error information was permitted to be returned to the
1925 sender, but the error was temporary, then after the final timeout, only
1926 "retry timeout exceeded" was returned. Now it returns the full error as
1927 well as "retry timeout exceeded".
1929 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1930 do this, and (what is worse) MTAs that accept it.
1932 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1933 will now be deprecated.
1935 PH/56 New os.c-cygwin from the Cygwin maintainer.
1937 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1938 formats (previously only available in exim formats via -bpr, -bpru,
1939 and -bpra. Now also available in native and exiqgrep formats)
1941 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1942 with very large, slow to parse queues
1944 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1946 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1948 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1949 responses to authentication challenges, though it was showing the
1950 challenges; (ii) I've removed the CR characters from the debug output for
1953 PH/58 Allow for the insertion of a newline as well as a space when a string
1954 is turned into more than one encoded-word during RFC 2047 encoding. The
1955 Sieve code now uses this.
1957 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1958 data_4xx, lost_connection, tls_required.
1960 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1961 message was being sent as an SMTP response.
1963 PH/61 Add -l and -k options to exicyclog.
1965 PH/62 When verifying, if an address was redirected to one new address, so that
1966 verification continued, and the new address failed or deferred after
1967 having set something in $address_data, the value of $address_data was not
1968 passed back to the ACL. This was different to the case when no
1969 redirection occurred. The value is now passed back in both cases.
1971 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1972 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1973 use login_cap.h, so on its own it isn't the distinguishing feature. The
1974 new name refers directly to the setclassresources() function.
1976 PH/65 Added configuration files for NetBSD3.
1978 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1980 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1981 is preferred over IPv4.
1983 PH/68 The bounce_return_message and bounce_return_body options were not being
1984 honoured for bounces generated during the reception of non-SMTP messages.
1985 In particular, this applied to messages rejected by the ACL. This bug has
1986 been fixed. However, if bounce_return_message is true and bounce_return_
1987 body is false, the headers that are returned for a non-SMTP message
1988 include only those that have been read before the error was detected.
1989 (In the case of an ACL rejection, they have all been read.)
1991 PH/69 The HTML version of the specification is now built in a directory called
1992 spec_html instead of spec.html, because the latter looks like a path with
1993 a MIME-type, and this confuses some software.
1995 PH/70 Catch two compiler warnings in sieve.c.
1997 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1998 function verify_get_ident() calls ip_connect() to connect a socket, but
1999 if the "connect()" function timed out, ip_connect() used to close the
2000 socket. However, verify_get_ident() also closes the socket later, and in
2001 between Exim writes to the log, which may get opened at this point. When
2002 the socket was closed in ip_connect(), the log could get the same file
2003 descriptor number as the socket. This naturally causes chaos. The fix is
2004 not to close the socket in ip_connect(); the socket should be closed by
2005 the function that creates it. There was only one place in the code where
2006 this was missing, in the iplookup router, which I don't think anybody now
2007 uses, but I've fixed it anyway.
2009 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2010 well as to direct DNS lookups. Otherwise the handling of names in host
2011 lists is inconsistent and therefore confusing.
2017 PH/01 Two changes to the default runtime configuration:
2019 (1) Move the checks for relay_from_hosts and authenticated clients from
2020 after to before the (commented out) DNS black list checks.
2022 (2) Add control=submission to the relay_from_hosts and authenticated
2023 clients checks, on the grounds that messages accepted by these
2024 statements are most likely to be submissions.
2026 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2028 (1) Generate an error if the third argument for the ${prvs expansion is
2031 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2034 (3) Reset the variables that are obtained from the first argument of
2035 ${prvscheck and used in the second argument before leaving the code,
2036 because their memory is reclaimed, so using them afterwards may do
2039 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2040 one (it's much easier than Tom thought :-).
2042 (5) Because of (4), we can now allow for the use of $prvscheck_result
2043 inside the third argument.
2045 PH/03 For some reason, the default setting of PATH when running a command from
2046 a pipe transport was just "/usr/bin". I have changed it to
2049 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2050 anything to be listed in the output from -bV.
2052 PH/05 When a filter generated an autoreply, the entire To: header line was
2053 quoted in the delivery log line, like this:
2055 => >A.N.Other <ano@some.domain> <original@ddress> ...
2057 This has been changed so that it extracts the operative address. There
2058 may be more than one such address. If so, they are comma-separated, like
2061 => >ano@some.domain,ona@other.domain <original@ddress> ...
2063 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2064 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2065 not being looked up in the rDNS to get a host name, Exim was showing the
2066 IP address twice in Received: lines, even though the IP addresses were
2067 identical. For example:
2069 Received: from [1.2.3.4] (helo=[1.2.3.4])
2071 However, if the real host name was known, it was omitting the HELO data
2072 if it matched the actual IP address. This has been tidied up so that it
2073 doesn't show the same IP address twice.
2075 PH/07 When both +timestamp and +memory debugging was on, the value given by
2076 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2077 called by the debug printing, thereby overwriting the timestamp buffer.
2078 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2080 PH/08 When the original message was included in an autoreply transport, it
2081 always said "this is a copy of the message, including all the headers",
2082 even if body_only or headers_only was set. It now gives an appropriate
2085 PH/09 Applied a patch from the Sieve maintainer which:
2087 o fixes some comments
2088 o adds the (disabled) notify extension core
2089 o adds some debug output for the result of if/elsif tests
2090 o points to the current vacation draft in the documentation
2091 and documents the missing references header update
2095 o fixes a bug in processing the envelope test (when testing
2096 multiple envelope elements, the last element determinted the
2099 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2100 Electronic Mail") by including:
2102 Auto-submitted: auto-generated
2104 in the messages that it generates (bounce messages and others, such as
2105 warnings). In the case of bounce messages for non-SMTP mesages, there was
2106 also a typo: it was using "Auto_submitted" (underscore instead of
2107 hyphen). Since every message generated by Exim is necessarily in response
2108 to another message, thes have all been changed to:
2110 Auto-Submitted: auto-replied
2112 in accordance with these statements in the RFC:
2114 The auto-replied keyword:
2116 - SHOULD be used on messages sent in direct response to another
2117 message by an automatic process,
2119 - MUST NOT be used on manually-generated messages,
2121 - MAY be used on Delivery Status Notifications (DSNs) and Message
2122 Disposition Notifications (MDNs),
2124 - MUST NOT be used on messages generated by automatic or periodic
2125 processes, except for messages which are automatic responses to
2128 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2129 to the default Received: header definition.
2131 PH/12 Added log selector acl_warn_skipped (default on).
2133 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2134 variables because (a) they are in the wrong storage pool and (b) even if
2135 they were copied, it wouldn't work properly because of the caching.
2137 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2138 checking when decoding. Apparently there are clients that generate
2139 overlong encoded strings. Why am I not surprised?
2141 PH/15 If the first argument of "${if match_address" was not empty, but did not
2142 contain an "@" character, Exim crashed. Now it writes a panic log message
2143 and treats the condition as false.
2145 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2147 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2148 extension "envelope-auth". The code is finished and in agreement with
2149 other implementations, but there is no documentation so far and in fact,
2150 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2151 not changing the active code.
2153 Print executed "if" and "elsif" statements when debugging is used. This
2154 helps a great deal to understand what a filter does.
2156 Document more things not specified clearly in RFC3028. I had all this
2157 sorted out, when out of a sudden new issues came to my mind. Oops."
2159 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2162 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2163 canonical form (as documented). However, after a host name lookup from
2164 the IP address, check_host() was doing a simple string comparison with
2165 addresses acquired from the DNS when checking that the found name did
2166 have the original IP as one of its addresses. Since any found IPv6
2167 addresses are likely to be in abbreviated form, the comparison could
2168 fail. Luckily, there already exists a function for doing the comparison
2169 by converting both addresses to binary, so now that is used instead of
2170 the text comparison.
2172 PH/20 There was another similar case to PH/19, when a complete host name was
2173 given in a host list; looking up its IP address could give an abbreviated
2174 form, whereas the current host's name might or might not be abbreviated.
2175 The same fix has been applied.
2181 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2182 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2185 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2186 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2188 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2190 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2191 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2192 Also, for "Auto-Submitted", treat anything other than "no" as
2193 non-personal, in accordance with RFC 3834. (Previously it treated
2194 anything starting "auto-" as non-personal.)
2196 TF/01 The control=submission/name=... option had a problem with syntax
2197 errors if the name included a slash character. The /name= option
2198 now slurps the rest of the string, so it can include any characters
2199 but it must come last in the list of options (after /sender_retain
2202 PH/05 Some modifications to the interface to the fake nameserver for the new
2210 TK/01 Added the "success_on_redirect" address verification option. See
2211 NewStuff for rationale and an example.
2213 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2215 PH/02 Patch to exigrep to allow it to work on syslog lines.
2217 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2218 fread() to skip over the body file's header line, because in Cygwin the
2219 header line is locked and is inaccessible.
2221 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2222 co-exist for some time) to make it clear that it is the Exim ID that is
2223 referenced, not the Message-ID: header line.
2225 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2226 string_format() function, because snprintf() does not exist on all
2229 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2230 ${stat: expansion item.
2232 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2233 protocol synchronization error", to keep the pedants happy.
2235 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2236 well as for IRIX systems, when gcc is being used. See the host.c source
2239 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2241 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2244 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2245 transport and to the smtp transport in LMTP mode.
2247 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2249 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2250 run a filter in a subprocess. This could lead to confusion in subsequent
2251 lookups in the parent process. There should also be a search_tidyup() at
2252 the end of the subprocess.
2254 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2255 only if the host matched helo_try_verify_hosts, which caused the
2256 verification to occur when the EHLO/HELO command was issued. The ACL just
2257 tested the remembered result. Now, if a previous verification attempt has
2258 not happened, "verify = helo" does it there and then.
2260 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2262 TK/03 Fix log output including CR from clamd.
2264 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2265 exist provoked a memory error which could cause a segfault.
2267 PH/15 Installed PCRE 6.2
2269 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2271 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2272 of the problem. Specifically, suggested +O2 rather than +O1 for the
2275 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2277 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2278 record, the port was not correctly logged when the outgoing_port log
2279 selector was set (it logged the transort's default port).
2281 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2282 fallback_hosts, and "hosts" in the smtp transport.
2284 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2285 host errors such as "Connection refused".
2287 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2288 authentication with radiusclient 0.4.9:
2290 - Error returned from rc_read_config was caught wrongly
2291 - Username/password not passed on to radius server due to wrong length.
2293 The presumption is that some radiusclient API changes for 4.51/PH/17
2294 were not taken care of correctly. The code is still untested by me (my
2295 Linux distribution still has 0.3.2 of radiusclient), but it was
2296 contributed by a Radius user.
2298 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2299 expanding the "port" option of the smtp transport.
2301 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2302 while reading a MIME header. Thanks to Tom Hughes for a patch.
2304 PH/24 Include config.h inside local_scan.h so that configuration settings are
2307 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2308 that in an ACL for RCPT (for example), you can examine exactly what was
2311 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2312 commands, but it was not correctly comparing the address with the actual
2313 client host address. Thus, it would show the EHLO address in Received:
2314 header lines when this was not necessary.
2316 PH/27 Added the % operator to ${eval:}.
2318 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2319 it should be ignoring failures (because with -C, for example, it has lost
2320 privilege). It wasn't ignoring creation failures other than "already
2323 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2326 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2327 because an input line was too long, either on its own, or by virtue of
2328 too many continuations, the temporary file was not being removed, and the
2329 return code was incorrect.
2331 PH/31 Missing "BOOL" in function definition in filtertest.c.
2333 PH/32 Applied Sieve patches from the maintainer.
2335 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2337 PH/33 Added "verify = not_blind".
2339 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2340 Local/Makefile (with some defaults set). These are used in built scripts
2341 such as exicyclog, but they have never been used in the exim_install
2342 script (though there are many overriding facilities there). I have
2343 arranged that the exim_install script now takes note of these two
2346 PH/35 Installed configuration files for Dragonfly.
2348 PH/36 When a locally submitted message by a trusted user did not contain a
2349 From: header, and the sender address was obtained from -f or from an SMTP
2350 MAIL command, and the trusted user did not use -F to supply a sender
2351 name, $originator_name was incorrectly used when constructing a From:
2352 header. Furthermore, $originator_name was used for submission mode
2353 messages from external hosts without From: headers in a similar way,
2354 which is clearly wrong.
2356 PH/37 Added control=suppress_local_fixups.
2358 PH/38 When log_selector = +received_sender was set, and the addition of the
2359 sender made the log line's construction buffer exactly full, or one byte
2360 less than full, an overflow happened when the terminating "\n" was
2363 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2364 when the result of a list match is failure because a DNS lookup failed.
2366 PH/40 RM_COMMAND is now used in the building process.
2368 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2369 the "build-* directories that it finds.
2371 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2372 address in a domain literal was a prefix of an interface address.
2374 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2375 when verifying a sender address, unless rewrite_headers is false.
2377 PH/44 Wrote a long comment about why errors_to addresses are verified as
2378 recipients, not senders.
2380 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2381 the ratelimit ACL was added.
2383 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2385 PH/46 Added extra information about PostgreSQL errors to the error string.
2387 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2388 avoiding the need to install special zones in a real server. This is
2389 backwards compatible; if it can't find the fake resolver, it drops back.
2390 Thus, both old and new test suites can be run.
2392 TF/02 Added util/ratelimit.pl
2394 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2395 clock is set back in time.
2397 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2398 Candler <B.Candler@pobox.com>.
2400 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2401 OK for addresses that are the result of redirections.
2403 PH/48 A number of further additions for the benefit of the new test suite,
2404 including a fake gethostbyname() that interfaces to the fake DNS resolver
2407 TF/06 The fix for widen_domains has also been applied to qualify_single and
2408 search_parents which are the other dnslookup options that can cause
2411 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2414 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2416 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2417 service" if it encountered an MX record with an empty target hostname.
2418 The message is now "an MX or SRV record indicated no SMTP service".
2420 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2421 if the DNS of the sending site is misconfigured. This is quite a
2422 common situation. This change restores the behaviour of treating a
2423 helo verification defer as a failure.
2425 PH/51 If self=fail was set on a router, the bounce message did not include the
2426 actual error message.
2432 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2434 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2435 command itself ended in error, the underlying message about the transport
2436 filter timeout was being overwritten with the pipe command error. Now the
2437 underlying error message should be appended to the second error message.
2439 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2441 PH/02 Reduce the amount of output that "make" produces by default. Full output
2442 can still be requested.
2444 PH/03 The warning log line about a condition test deferring for a "warn" verb
2445 was being output only once per connection, rather than after each
2446 occurrence (because it was using the same function as for successful
2447 "warn" verbs). This seems wrong, so I have changed it.
2449 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2450 it should not have, which might have caused a crash in the right
2451 circumstances, but probably never did.
2453 PH/04 Installed a modified version of Tony Finch's patch to make submission
2454 mode fix the return path as well as the Sender: header line, and to
2455 add a /name= option so that you can make the user's friendly name appear
2458 TF/03 Added the control = fakedefer ACL modifier.
2460 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2461 Mark Lowes for thorough testing.
2463 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2465 TK/03 Merged latest SRS patch from Miles Wilton.
2467 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2468 with the definition in sysexits.h (which is #included earlier).
2469 Fortunately, Exim does not actually use EX_OK. The code used to try to
2470 preserve the sysexits.h value, by assumimg that macro definitions were
2471 scanned for macro replacements. I have been disabused of this notion,
2472 so now the code just undefines EX_OK before #including unistd.h.
2474 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2475 in the smtp transport. When a block could not be written in a single
2476 write() function, the timeout was being re-applied to each part-write.
2477 This seems wrong - if the receiver was accepting one byte at a time it
2478 would take for ever. The timeout is now adjusted when this happens. It
2479 doesn't have to be particularly precise.
2481 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2482 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2484 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2485 without a domain if the check to <postmaster@domain> fails.
2487 SC/01 Eximstats: added -xls and the ability to specify output files
2488 (patch written by Frank Heydlauf).
2490 SC/02 Eximstats: use FileHandles for outputing results.
2492 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2494 SC/04 Eximstats: fixed display of large numbers with -nvr option
2496 SC/05 Eximstats: fixed merging of reports with empty tables.
2498 SC/06 Eximstats: added the -include_original_destination flag
2500 SC/07 Eximstats: removed tabs and trailing whitespace.
2502 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2504 TK/06 MBOX spool code: Add real "From " MBOX separator line
2505 so the .eml file is really in mbox format (even though
2506 most programs do not really care). Patch from Alex Miller.
2508 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2509 The latter is generated from $received_to and is only set if the
2510 message has one envelope recipient. SA can use these headers,
2511 obviously out-of-the-box. Patch from Alex Miller.
2513 PH/08 The ${def test on a variable was returning false if the variable's
2514 value was "0", contrary to what the specification has always said!
2515 The result should be true unless the variable is empty.
2517 PH/09 The syntax error of a character other than { following "${if
2518 def:variable_name" (after optional whitespace) was not being diagnosed.
2519 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2520 accidental colon was present, for example, could give incorrect results.
2522 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2523 result is used (not including appendfile, where other changes are about
2526 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2527 This involved changing a lot of size variables from int to off_t. It
2528 should work with maildirs and everything.
2530 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2531 spamd dying while we are connected to it.
2533 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2536 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2537 function for BDB 4.3.
2539 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2541 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2542 the sources for PCRE have been re-arranged and more files are now
2545 PH/15 The code I had for printing potentially long long variables in PH/11
2546 above was not the best (it lost precision). The length of off_t variables
2547 is now inspected at build time, and an appropriate printing format (%ld
2548 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2549 to be "long long int" or "long int". This is needed for the internal
2550 formatting function string_vformat().
2552 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2553 the configuration file to be ":syslog", then the script "guesses" where
2554 the logs files are, rather than using the compiled in default. In our
2555 case the guess is not the same as the compiled default, so the script
2556 suddenly stopped working when I started to use syslog. The patch checks
2557 to see if log_file_path is "". If so, it attempts to read it from exim
2558 with no configuration file to get the compiled in version, before it
2559 falls back to the previous guessing code."
2561 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2562 implementing BATV in an Exim configuration. See NewStuff for the gory
2565 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2566 Makefile that are specific to HP-UX.
2568 PH/18 If the "use_postmaster" option was set for a recipient callout together
2569 with the "random" option, the postmaster address was used as the MAIL
2570 FROM address for the random test, but not for the subsequent recipient
2571 test. It is now used for both.
2573 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2574 patch removes a few documentation additions to RFC 3028, because the
2575 latest draft now contains them. It adds the new en;ascii-case comparator
2576 and a new error check for 8bit text in MIME parts. Comparator and
2577 require names are now matched exactly. I enabled the subaddress
2578 extension, but it is not well tested yet (read: it works for me)."
2580 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2581 rework some of the code of TK/09 above to avoid the hardwired use of
2582 "%lld" and "long long". Replaced the call to snprintf() with a call to
2585 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2586 records point to non-existent hosts", "retry timeout exceeded", and
2587 "retry time not reached for any host after a long failure period".
2589 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2590 experimental DomainKeys support:
2592 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2593 (2) On an error such as an illegally used "control", the wrong name for
2594 the control was given.
2596 These problems did NOT occur unless DomainKeys support was compiled.
2598 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2600 PH/24 Added ${if match_ip condition.
2602 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2603 databases so that it will be absolutely obvious if a crash occurs in the
2604 DB library. This is a regular occurrence (often caused by mis-matched
2607 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2608 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2609 picked up on a user's system that detects such things. There doesn't seem
2610 to be a gcc warning option for this - only an attribute that has to be
2611 put on the function's prototype. It seems that in Fedora Core 4 they have
2612 set this on a number of new functions. No doubt there will be more in due
2615 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2616 specify a transport. However, if an address that was verified by such a
2617 router was the subject of a callout, Exim crashed because it tried to
2618 read the rcpt_include_affixes from the non-existent transport. Now it
2619 just assumes that the setting of that option is false. This bug was
2620 introduced by 4.51/PH/31.
2622 PH/28 Changed -d+all to exclude +memory, because that information is very
2623 rarely of interest, but it makes the output a lot bigger. People tend to
2624 do -d+all out of habit.
2626 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2627 code in os-type was giving problems when libc.so lives in lib64, like on
2630 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2631 aren't the modern standard, and it seems that some systems' include files
2632 don't always have them. Exim was already checking for some of the newer
2633 ones like T_AAAA, and defining it itself. I've added checks for all the
2634 record types that Exim uses.
2636 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2637 not automatically generating a new one, as it is supposed to. This
2638 prevented TLS from working. If the file did exist, but contained invalid
2639 data, a new version was generated, as expected. It was only the case of a
2640 non-existent file that was broken.
2642 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2643 with a change in libdomainkeys > 0.64.
2645 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2646 from DNS. If the selector record carries the flag, it now has
2647 precedence over the domain-wide flag.
2649 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2651 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2652 the use of a port name that isn't defined in /etc/services) occurred, the
2653 message was deferred as in a normal delivery, and thus remained on the
2654 spool, instead of being failed because of the mua_wrapper setting. This
2655 is now fixed, and I tidied up some of the mua_wrapper messages at the
2658 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2659 the messages in a hash of arrays rather than using individual hashes.
2660 This is a bit cleaner and results in dramatic memory savings, albeit
2661 at a slight CPU cost.
2663 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2664 as requested by Marc Sherman.
2666 SC/10 Eximstats: added histograms for user specified patterns as requested
2669 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2671 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2672 fopen() in the content-scanning modules that did not already have it.
2678 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2679 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2681 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2683 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2685 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2686 to test Sieve filters that use "vacation".
2688 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2689 that changes the way the GnuTLS parameters are stored in the cache file.
2690 The new format can be generated externally. For backward compatibility,
2691 if the data in the cache doesn't make sense, Exim assumes it has read an
2692 old-format file, and it generates new data and writes a new file. This
2693 means that you can't go back to an older release without removing the
2696 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2697 work if there are any delivery delays because "one_time" forces the
2698 parent to be marked "delivered", so its unseen clone is never tried
2699 again. For this reason, Exim now forbids the simultaneous setting of
2702 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2703 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2704 entry if you want to know the details. The fix, however, neglected to
2705 consider the case when local delivery batching is involved. The test for
2706 "previously delivered" was not happening when checking to see if an
2707 address could be batched with a previous (undelivered) one; under
2708 certain circumstances this could lead to multiple deliveries to the same
2711 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2712 in its include files, and this causes problems building Exim.
2714 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2715 header_syntax) but Exim was just ignoring anything given after a slash.
2716 In particular, this caused confusion with an attempt to use "verify =
2717 reverse_host_lookup/defer_ok". An error is now given when options are
2718 supplied for verify items that do not have them. (Maybe reverse_host_
2719 lookup should have a defer_ok option, but that's a different point.)
2721 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2722 defined by RFC 821) to 2048, because there were problems with some AUTH
2723 commands, and RFC 1869 says the size should be increased for extended
2724 SMTP commands that take arguments.
2726 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2729 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2730 "unknown" error; now it says that the functionality isn't in the binary.
2732 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2733 an address' error message when a string expansion fails (syntax or
2734 whatever). Otherwise the password may appear in the log. Following change
2735 PH/42 below, there is no longer a chance of it appearing in a bounce
2738 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2740 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2741 of its addresses was ever tried. (Bugzilla bug #2.)
2743 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2744 the result incorrectly in the debug output. (It correctly added a newline
2745 to what was transported.)
2747 TF/01 Added $received_time.
2749 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2750 commented out examples of how to interface to a virus scanner and to
2751 SpamAssassin. Also added commented examples of av_scanner and
2752 spamd_address settings.
2754 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2755 and controls are allowed in which ACLs. There were a couple of minor
2756 errors. Some of the entries in the conditions table (which is a table of
2757 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2758 negation of where the condition IS allowed.
2760 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2762 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2763 header file does not have a version number, so I've had to invent a new
2764 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2765 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2766 radiusclient), but it was contributed by a Radius user.
2768 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2769 files or directories, for OpenSSL.
2771 PH/19 When an Exim process that is running as root has to create an Exim log
2772 file, it does so in a subprocess that runs as exim:exim so as to get the
2773 ownership right at creation (otherwise, other Exim processes might see
2774 the file with the wrong ownership). There was no test for failure of this
2775 fork() call, which would lead to the process getting stuck as it waited
2776 for a non-existent subprocess. Forks do occasionally fail when resources
2777 run out. I reviewed all the other calls to fork(); they all seem to check
2780 PH/20 When checking for unexpected SMTP input at connect time (before writing
2781 the banner), Exim was not dealing correctly with a non-positive return
2782 from the read() function. If the client had disconnected by this time,
2783 the result was a log entry for a synchronization error with an empty
2784 string after "input=" when read() returned zero. If read() returned -1
2785 (an event I could not check), uninitialized data bytes were printed.
2786 There were reports of junk text (parts of files, etc) appearing after
2789 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2791 PH/22 Added support for macro redefinition, and (re)definition in between
2792 driver and ACL definitions.
2794 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2795 forgetting to use the resulting value; it was using the unexpanded value.
2797 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2798 hadn't been configured. The fix is from Juergen Kreileder, who
2799 understands it better than I do:
2801 "Here's what I see happening with three configured cyrus_sasl
2802 authenticators configured (plain, login, cram-md5):
2804 On startup auth_cyrus_sasl_init() gets called for each of these.
2805 This means three calls to sasl_listmech() without a specified mech_list.
2806 => SASL tests which mechs of all available mechs actually work
2807 => three warnings about OTP not working
2808 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2810 With the patch, sasl_listmech() also gets called three times. But now
2811 SASL's mech_list option is set to the server_mech specified in the the
2812 authenticator. Or in other words, the answer from sasl_listmech()
2813 gets limited to just the mech you're testing for (which is different
2815 => the return list contains just 'plain' or 'login', 'cram-md5' or
2816 nothing depending on the value of ob->server_mech.
2818 I've just tested the patch: Authentication still works fine,
2819 unavailable mechs specified in the exim configuration are still
2820 caught, and the auth.log warnings about OTP are gone."
2822 PH/25 When debugging is enabled, the contents of the command line are added
2823 to the debugging output, even when log_selector=+arguments is not
2826 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2827 answer is "GNU", and only if the return is "GNU/something" is the answer
2830 PH/27 $acl_verify_message is now set immediately after the failure of a
2831 verification in an ACL, and so is available in subsequent modifiers. In
2832 particular, the message can be preserved by coding like this:
2834 warn !verify = sender
2835 set acl_m0 = $acl_verify_message
2837 Previously, $acl_verify_message was set only while expanding "message"
2838 and "log_message" when a very denied access.
2840 PH/28 Modified OS/os.c-Linux with
2842 -#ifndef OS_LOAD_AVERAGE
2843 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2845 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2846 nomenclature these days.)
2848 PH/29 Installed patch from the Sieve maintainer that adds the options
2849 sieve_useraddress and sieve_subaddress to the redirect router.
2851 PH/30 In these circumstances:
2852 . Two addresses routed to the same list of hosts;
2853 . First host does not offer TLS;
2854 . First host accepts first address;
2855 . First host gives temporary error to second address;
2856 . Second host offers TLS and a TLS session is established;
2857 . Second host accepts second address.
2858 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2859 and peerdn, if requested) that were in fact used only for the second
2862 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2863 attention to any local part prefix or suffix that was matched by the
2864 router that accepted the address. It now behaves in the same way as it
2865 does for delivery: the affixes are removed from the local part unless
2866 rcpt_include_affixes is set on the transport.
2868 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2869 timeout during the DATA phase of an incoming message.
2871 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2872 applied a patch sanctioned by the Sieve maintainer.
2874 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2875 the uid or gid is negative. A case of a negative gid caused this to be
2876 noticed. The fix allows for either to be negative.
2878 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2879 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2882 PH/36 The change PH/12 above was broken. Fixed it.
2884 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2885 the grounds that routing the same address twice would always produce the
2886 same answer. This might have been true once, but it is certainly no
2887 longer true now. Routing a child address may depend on the previous
2888 routing that produced that child. Some complicated redirection strategies
2889 went wrong when messages had multiple recipients, and made Exim's
2890 behaviour dependent on the order in which the addresses were given.
2892 I have moved the duplicate checking until after the routing is complete.
2893 Exim scans the addresses that are assigned to local and remote
2894 transports, and removes any duplicates. This means that more work will be
2895 done, as duplicates will always all be routed, but duplicates are
2896 presumably rare, so I don't expect this is of any significance.
2898 For deliveries to pipes, files, and autoreplies, the duplicate checking
2899 still happens during the routing process, since they are not going to be
2902 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2903 It corrects a timeout issue with spamd. This is Ian's comment: "The
2904 background is that sometimes spamd either never reads data from a
2905 connection it has accepted, or it never writes response data. The exiscan
2906 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2907 blindly assumes that writes won't block so it may never time out."
2909 PH/39 Allow G after quota size as well as K and M.
2911 PH/40 The value set for $authenticated_id in an authenticator may not contain
2912 binary zeroes or newlines because the value is written to log lines and
2913 to spool files. There was no check on this. Now the value is run through
2914 the string_printing() function so that such characters are converted to
2915 printable escape sequences.
2917 PH/41 $message_linecount is a new variable that contains the total number of
2918 lines in the message. Compare $body_linecount, which is the count for the
2921 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2922 bounce and delay warning messages, except in certain special cases, which
2925 (a) An SMTP error message from a remote host;
2926 (b) A message specified in a :fail: redirection;
2927 (c) A message specified in a "fail" command in a system filter;
2928 (d) A message specified in a FAIL return from the queryprogram router;
2929 (e) A message specified by the cannot_route_message router option.
2931 In these cases only, Exim does include the error details in bounce and
2932 warning messages. There are also a few cases where bland messages such
2933 as "unrouteable address" or "local delivery error" are given.
2935 PH/43 $value is now also set for the "else" part of a ${run expansion.
2937 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2938 being worked on, but at least Exim now implements the latest version to
2941 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2942 process to complete was treated as a delivery failure, a timeout while
2943 writing the message to the pipe was logged, but erroneously treated as a
2944 successful delivery. Such timeouts include transport filter timeouts. For
2945 consistency with the overall process timeout, these timeouts are now
2946 treated as errors, giving rise to delivery failures by default. However,
2947 there is now a new Boolean option for the pipe transport called
2948 timeout_defer, which, if set TRUE, converts the failures into defers for
2949 both kinds of timeout. A transport filter timeout is now identified in
2952 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2953 systems where "make" and "gmake" are different, calling "gmake" at top
2954 level broke things. I've arranged for the value of $(MAKE) to be passed
2955 from the Makefile to this script so that it can call the same version of
2959 A note about Exim versions 4.44 and 4.50
2960 ----------------------------------------
2962 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2963 changes of various kinds. As a consequence, a big documentation update was
2964 needed. This delayed the release for rather longer than seemed good, especially
2965 in the light of a couple of (minor) security issues. Therefore, the changes
2966 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2967 release. So 4.44 and 4.50 are in effect two different branches that both start
2970 I have left the 4.50 change log unchanged; it contains all the changes since
2971 4.43. The change log for 4.44 is below; many of its items are identical to
2972 those for 4.50. This seems to be the most sensible way to preserve the
2973 historical information.
2979 1. Minor wording change to the doc/README.SIEVE file.
2981 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2982 computation of the current number of files was incorrect.
2984 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2985 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2988 4. Give more explanation in the error message when the command for a transport
2989 filter fails to execute.
2991 5. There are several places where Exim runs a non-Exim command in a
2992 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2993 was being done only for the command run by the queryprogram router. It is
2994 now done for all such subprocesses. The other cases are: ${run, transport
2995 filters, and the commands run by the lmtp and pipe transports.
2997 6. Added CONFIGURE_GROUP build-time option.
2999 7. Some older OS have a limit of 256 on the maximum number of file
3000 descriptors. Exim was using setrlimit() to set 1000 as a large value
3001 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3002 systems. I've change it so that if it can't get 1000, it tries for 256.
3004 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3005 was an oversight, and furthermore, ever since the addition of extra
3006 controls (e.g. 4.43/32), the checks on when to allow different forms of
3007 "control" were broken. There should now be diagnostics for all cases when a
3008 control that does not make sense is encountered.
3010 9. Added the /retain_sender option to "control=submission".
3012 10. $recipients is now available in the predata ACL (oversight).
3014 11. Tidy the search cache before the fork to do a delivery from a message
3015 received from the command line. Otherwise the child will trigger a lookup
3016 failure and thereby defer the delivery if it tries to use (for example) a
3017 cached ldap connection that the parent has called unbind on.
3019 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3020 of $address_data from the recipient verification was clobbered by the
3021 sender verification.
3023 13. The value of address_data from a sender verification is now available in
3024 $sender_address_data in subsequent conditions in the ACL statement.
3026 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3028 15. Added a new option "connect=<time>" to callout options, to set a different
3031 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3032 was its contents. (It was OK if the option was not defined at all.)
3034 17. A "Completed" log line is now written for messages that are removed from
3035 the spool by the -Mrm option.
3037 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3038 information about exactly what failed.
3040 19. Added -dd to debug only the daemon process.
3042 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3043 handles timeouts, both on the server side and network timeouts. Renamed the
3044 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3046 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3047 It is now set to "smtps".
3049 22. $host_address is now set to the target address during the checking of
3050 ignore_target_hosts.
3052 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3053 being passed; this would have caused $sender_host_name to have been used if
3054 matching the list had actually called for a host name (not very likely,
3055 since this list is usually IP addresses). A host name is now passed as
3058 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3059 code that specifies a non-restarting handler (typically sigaction() in
3060 modern systems) in an attempt to fix a rare and obscure crash bug.
3062 25. Narrowed the window for a race in the daemon that could cause it to ignore
3063 SIGCHLD signals. This is not a major problem, because they are used only to
3064 wake it up if nothing else does.
3066 26. A malformed maildirsize file could cause Exim to calculate negative values
3067 for the mailbox size or file count. Odd effects could occur as a result.
3068 The maildirsize information is now recalculated if the size or filecount
3071 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3072 support for a long time. Removed HAVE_SYS_VFS_H.
3074 28. Installed the latest version of exipick from John Jetmore.
3076 29. In an address list, if the pattern was not a regular expression, an empty
3077 subject address (from a bounce message) matched only if the pattern was an
3078 empty string. Non-empty patterns were not even tested. This was the wrong
3079 because it is perfectly reasonable to use an empty address as part of a
3080 database query. An empty address is now tested by patterns that are
3081 lookups. However, all the other forms of pattern expect the subject to
3082 contain a local part and a domain, and therefore, for them, an empty
3083 address still always fails if the pattern is not itself empty.
3085 30. Exim went into a mad DNS loop when attempting to do a callout where the
3086 host was specified on an smtp transport, and looking it up yielded more
3087 than one IP address.
3089 31. Re-factored the code for checking spool and log partition space into a
3090 function that finds that data and another that does the check. The former
3091 is then used to implement four new variables: $spool_space, $log_space,
3092 $spool_inodes, and $log_inodes.
3094 32. The RFC2047 encoding function was originally intended for short strings
3095 such as real names; it was not keeping to the 75-character limit for
3096 encoded words that the RFC imposes. It now respects the limit, and
3097 generates multiple encoded words if necessary. To be on the safe side, I
3098 have increased the buffer size for the ${rfc2047: expansion operator from
3101 33. It is now permitted to omit both strings after an "if" condition; if the
3102 condition is true, the result is "true". As before, when the second string
3103 is omitted, a false condition yields an empty string. This makes it less
3104 cumbersome to write custom ACL and router conditions.
3106 34. Failure to deliver a bounce message always caused it to be frozen, even if
3107 there was an errors_to setting on the router. The errors_to setting is now
3110 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3111 canonical form (fully expanded) before being placed in
3112 $sender_host_address.
3114 36. The table in the code that translates DNS record types into text (T_A to
3115 "A" for instance) was missing entries for NS and CNAME. It is just possible
3116 that this could have caused confusion if both these types were looked up
3117 for the same domain, because the text type is used as part of Exim's
3118 per-process caching. But the chance of anyone hitting this buglet seems
3121 37. The dnsdb lookup has been extended in a number of ways.
3123 (1) There is a new type, "zns", which walks up the domain tree until it
3124 finds some nameserver records. It should be used with care.
3126 (2) There is a new type, "mxh", which is like "mx" except that it returns
3127 just the host names, not the priorities.
3129 (3) It is now possible to give a list of domains (or IP addresses) to be
3130 looked up. The behaviour when one of the lookups defers can be
3131 controlled by a keyword.
3133 (4) It is now possible to specify the separator character for use when
3134 multiple records are returned.
3136 38. The dnslists ACL condition has been extended: it is now possible to supply
3137 a list of IP addresses and/or domains to be looked up in a particular DNS
3140 39. Added log_selector=+queue_time_overall.
3142 40. When running the queue in the test harness, wait just a tad after forking a
3143 delivery process, to get repeatability of debugging output.
3145 41. Include certificate and key file names in error message when GnuTLS fails
3146 to set them up, because the GnuTLS error message doesn't include the name
3147 of the failing file when there is a problem reading it.
3149 42. Allow both -bf and -bF in the same test run.
3151 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3153 44. The "Exiscan patch" is now merged into the mainline Exim source.
3155 45. Sometimes the final signoff response after QUIT could fail to get
3156 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3157 < 0 before doing a fflush(). This bug looks as though it goes back to the
3158 introduction of TLS in release 3.20, but "sometimes" must have been rare
3159 because the tests only now provoked it.
3161 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3162 (this can affect the format of dates).
3164 47. exim_tidydb, when checking for the continued existence of a message for
3165 which it has found a message-specific retry record, was not finding
3166 messages that were in split spool directories. Consequently, it was
3167 deleting retry records that should have stayed in existence.
3169 48. Steve fixed some bugs in eximstats.
3171 49. The SPA authentication driver was not abandoning authentication and moving
3172 on to the next authenticator when an expansion was forced to fail,
3173 contradicting the general specification for all authenticators. Instead it
3174 was generating a temporary error. It now behaves as specified.
3176 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3177 (the order specifies the preference for clients). The order is now AES256,
3178 AES128, 3DES, ARCFOUR128.
3180 51. Small patch to Sieve code - explicitly set From: when generating an
3183 52. Exim crashed if a remote delivery caused a very long error message to be
3184 recorded - for instance if somebody sent an entire SpamAssassin report back
3185 as a large number of 550 error lines. This bug was coincidentally fixed by
3186 increasing the size of one of Exim's internal buffers (big_buffer) that
3187 happened as part of the Exiscan merge. However, to be on the safe side, I
3188 have made the code more robust (and fixed the comments that describe what
3191 53. Now that there can be additional text after "Completed" in log lines (if
3192 the queue_time_overall log selector is set), a one-byte patch to exigrep
3193 was needed to allow it to recognize "Completed" as not the last thing in
3196 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3197 patch that reportedly fixes this has been added. I am not expert enough to
3198 create a test for it. This is what the patch creator wrote:
3200 "I found a little strange behaviour of ldap code when working with
3201 Windows 2003 AD Domain, where users was placed in more than one
3202 Organization Units. When I tried to give exim partial DN, the exit code
3203 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3204 But simultaneously result of request was absolutely normal ldap result,
3205 so I produce this patch..."
3207 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3208 REFERENCE, so I have modified the code to exclude the patch when that macro
3211 55. Some experimental protocols are using DNS PTR records for new purposes. The
3212 keys for these records are domain names, not reversed IP addresses. The
3213 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3214 leaves it alone. Component reversal etc. now happens only for IP addresses.
3217 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3219 57. Double the size of the debug message buffer (to 2048) so that more of very
3220 long debug lines gets shown.
3222 58. The exicyclog utility now does better if the number of log files to keep
3223 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3225 59. Two changes related to the smtp_active_hostname option:
3227 (1) $smtp_active_hostname is now available as a variable.
3228 (2) The default for smtp_banner uses $smtp_active_hostname instead
3229 of $primary_hostname.
3231 60. The host_aton() function is supposed to be passed a string that is known
3232 to be a valid IP address. However, in the case of IPv6 addresses, it was
3233 not checking this. This is a hostage to fortune. Exim now panics and dies
3234 if the condition is not met. A case was found where this could be provoked
3235 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3236 components; fortuitously, this particular loophole had already been fixed
3237 by change 4.50/55 above.
3239 If there are any other similar loopholes, the new check in host_aton()
3240 itself should stop them being exploited. The report I received stated that
3241 data on the command line could provoke the exploit when Exim was running as
3242 exim, but did not say which command line option was involved. All I could
3243 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3244 running as the user.
3247 61. There was a buffer overflow vulnerability in the SPA authentication code
3248 (which came originally from the Samba project). I have added a test to the
3249 spa_base64_to_bits() function which I hope fixes it.
3252 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3253 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3255 63. The daemon start-up calls getloadavg() while still root for those OS that
3256 need the first call to be done as root, but it missed one case: when
3257 deliver_queue_load_max is set with deliver_drop_privilege. This is
3258 necessary for the benefit of the queue runner, because there is no re-exec
3259 when deliver_drop_privilege is set.
3261 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3262 This has been fixed.
3264 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3265 was in use, was not putting the data itself into the right store pool;
3266 consequently, it could be overwritten for a subsequent message in the same
3267 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3270 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3272 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3273 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3274 the function were treating the return as a boolean value, which happened to
3275 work because 0=false and not-0=true, but is not correct code.
3277 68. The host_aton() function was not handling scoped IPv6 addresses (those
3278 with, for example, "%eth0" on the end) correctly.
3280 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3281 negated items (that is, ~something) in unsigned ints. Some compilers
3282 apparently mutter when there is no cast.
3284 70. If an address verification called from an ACL failed, and did not produce a
3285 user-specific message (i.e. there was only a "system" message), nothing was
3286 put in $acl_verify_message. In this situation, it now puts the system
3289 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3290 session; change 4.31/43 added the unwanted input to the log line - except
3291 that it did not do this in the start of session case. It now does.
3293 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3294 This is wrong and can cause the other end to generate a synchronization
3295 error if it is another Exim or anything else that does the synchronization
3296 check. A QUIT command is no longer sent after a timeout.
3298 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3299 during host lookups.
3301 74. The defer_ok option of callout verification was not working if it was used
3302 when verifying addresses in header lines, that is, for this case:
3304 verify = header_sender/callout=defer_ok
3306 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3307 those file descriptors could be used for SMTP connections. If anything
3308 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3309 could be sent to the SMTP client, causing chaos. The daemon now opens
3310 stdin, stdout, and stderr to /dev/null when it puts itself into the
3313 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3314 log by default. The user can override this with suitable Perl magic.
3316 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3317 the log message when discard triggers, was not working for the DATA ACL or
3318 for the non-SMTP ACL.
3320 78. Error message wording change in sieve.c.
3322 79. If smtp_accept_max_per_host was set, the number of connections could be
3323 restricted to fewer than expected, because the daemon was trying to set up
3324 a new connection before checking whether the processes handling previous
3325 connections had finished. The check for completed processes is now done
3326 earlier. On busy systems, this bug wouldn't be noticed because something
3327 else would have woken the daemon, and it would have reaped the completed
3330 80. If a message was submitted locally by a user whose login name contained one
3331 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3332 It caused a spool format error. I have fixed the spool reading code. A
3333 related problem was that the "from" clause in the Received: line became
3334 illegal because of the space(s). It is now covered by ${quote_local_part.
3336 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3339 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3341 83. Patch from Sieve maintainer to fix unterminated string problem in
3342 "vacation" handling.
3344 84. Some minor changes to the Linux configuration files to help with other
3345 OS variants using glibc.
3347 85. One more patch for Sieve to update vacation handling to latest spec.
3350 ----------------------------------------------------
3351 See the note above about the 4.44 and 4.50 releases.
3352 ----------------------------------------------------
3358 1. Change 4.43/35 introduced a bug that caused file counts to be
3359 incorrectly computed when quota_filecount was set in an appendfile
3362 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3363 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3366 3. Give more explanation in the error message when the command for a transport
3367 filter fails to execute.
3369 4. There are several places where Exim runs a non-Exim command in a
3370 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3371 was being done only for the command run by the queryprogram router. It is
3372 now done for all such subprocesses. The other cases are: ${run, transport
3373 filters, and the commands run by the lmtp and pipe transports.
3375 5. Some older OS have a limit of 256 on the maximum number of file
3376 descriptors. Exim was using setrlimit() to set 1000 as a large value
3377 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3378 systems. I've change it so that if it can't get 1000, it tries for 256.
3380 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3381 was an oversight, and furthermore, ever since the addition of extra
3382 controls (e.g. 4.43/32), the checks on when to allow different forms of
3383 "control" were broken. There should now be diagnostics for all cases when a
3384 control that does not make sense is encountered.
3386 7. $recipients is now available in the predata ACL (oversight).
3388 8. Tidy the search cache before the fork to do a delivery from a message
3389 received from the command line. Otherwise the child will trigger a lookup
3390 failure and thereby defer the delivery if it tries to use (for example) a
3391 cached ldap connection that the parent has called unbind on.
3393 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3394 of $address_data from the recipient verification was clobbered by the
3395 sender verification.
3397 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3398 was its contents. (It was OK if the option was not defined at all.)
3400 11. A "Completed" log line is now written for messages that are removed from
3401 the spool by the -Mrm option.
3403 12. $host_address is now set to the target address during the checking of
3404 ignore_target_hosts.
3406 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3407 being passed; this would have caused $sender_host_name to have been used if
3408 matching the list had actually called for a host name (not very likely,
3409 since this list is usually IP addresses). A host name is now passed as
3412 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3413 code that specifies a non-restarting handler (typically sigaction() in
3414 modern systems) in an attempt to fix a rare and obscure crash bug.
3416 15. Narrowed the window for a race in the daemon that could cause it to ignore
3417 SIGCHLD signals. This is not a major problem, because they are used only to
3418 wake it up if nothing else does.
3420 16. A malformed maildirsize file could cause Exim to calculate negative values
3421 for the mailbox size or file count. Odd effects could occur as a result.
3422 The maildirsize information is now recalculated if the size or filecount
3425 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3426 support for a long time. Removed HAVE_SYS_VFS_H.
3428 18. Updated exipick to current release from John Jetmore.
3430 19. Allow an empty sender to be matched against a lookup in an address list.
3431 Previously the only cases considered were a regular expression, or an
3434 20. Exim went into a mad DNS lookup loop when doing a callout where the
3435 host was specified on the transport, if the DNS lookup yielded more than
3438 21. The RFC2047 encoding function was originally intended for short strings
3439 such as real names; it was not keeping to the 75-character limit for
3440 encoded words that the RFC imposes. It now respects the limit, and
3441 generates multiple encoded words if necessary. To be on the safe side, I
3442 have increased the buffer size for the ${rfc2047: expansion operator from
3445 22. Failure to deliver a bounce message always caused it to be frozen, even if
3446 there was an errors_to setting on the router. The errors_to setting is now
3449 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3450 canonical form (fully expanded) before being placed in
3451 $sender_host_address.
3453 24. Updated eximstats to version 1.33
3455 25. Include certificate and key file names in error message when GnuTLS fails
3456 to set them up, because the GnuTLS error message doesn't include the name
3457 of the failing file when there is a problem reading it.
3459 26. Expand error message when OpenSSL has problems setting up cert/key files.
3462 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3463 (this can affect the format of dates).
3465 28. exim_tidydb, when checking for the continued existence of a message for
3466 which it has found a message-specific retry record, was not finding
3467 messages that were in split spool directories. Consequently, it was
3468 deleting retry records that should have stayed in existence.
3470 29. eximstats updated to version 1.35
3471 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3472 1.35 - bugfix such that pie charts by volume are generated correctly
3474 30. The SPA authentication driver was not abandoning authentication and moving
3475 on to the next authenticator when an expansion was forced to fail,
3476 contradicting the general specification for all authenticators. Instead it
3477 was generating a temporary error. It now behaves as specified.
3479 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3480 (the order specifies the preference for clients). The order is now AES256,
3481 AES128, 3DES, ARCFOUR128.
3483 31. Small patch to Sieve code - explicitly set From: when generating an
3486 32. Exim crashed if a remote delivery caused a very long error message to be
3487 recorded - for instance if somebody sent an entire SpamAssassin report back
3488 as a large number of 550 error lines. This bug was coincidentally fixed by
3489 increasing the size of one of Exim's internal buffers (big_buffer) that
3490 happened as part of the Exiscan merge. However, to be on the safe side, I
3491 have made the code more robust (and fixed the comments that describe what
3494 33. Some experimental protocols are using DNS PTR records for new purposes. The
3495 keys for these records are domain names, not reversed IP addresses. The
3496 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3497 leaves it alone. Component reversal etc. now happens only for IP addresses.
3500 34. The host_aton() function is supposed to be passed a string that is known
3501 to be a valid IP address. However, in the case of IPv6 addresses, it was
3502 not checking this. This is a hostage to fortune. Exim now panics and dies
3503 if the condition is not met. A case was found where this could be provoked
3504 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3505 components; fortuitously, this particular loophole had already been fixed
3506 by change 4.50/55 or 4.44/33 above.
3508 If there are any other similar loopholes, the new check in host_aton()
3509 itself should stop them being exploited. The report I received stated that
3510 data on the command line could provoke the exploit when Exim was running as
3511 exim, but did not say which command line option was involved. All I could
3512 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3513 running as the user.
3516 35. There was a buffer overflow vulnerability in the SPA authentication code
3517 (which came originally from the Samba project). I have added a test to the
3518 spa_base64_to_bits() function which I hope fixes it.
3521 36. The daemon start-up calls getloadavg() while still root for those OS that
3522 need the first call to be done as root, but it missed one case: when
3523 deliver_queue_load_max is set with deliver_drop_privilege. This is
3524 necessary for the benefit of the queue runner, because there is no re-exec
3525 when deliver_drop_privilege is set.
3527 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3528 was in use, was not putting the data itself into the right store pool;
3529 consequently, it could be overwritten for a subsequent message in the same
3530 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3533 38. Sometimes the final signoff response after QUIT could fail to get
3534 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3535 < 0 before doing a fflush(). This bug looks as though it goes back to the
3536 introduction of TLS in release 3.20, but "sometimes" must have been rare
3537 because the tests only now provoked it.
3543 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3544 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3545 Now it returns an integer. A number of calls were still expecting a T/F
3546 return. Fortuitously, in all cases, the tests worked in OK situations,
3547 which is the norm. However, things would have gone wrong on any write
3548 failures on the smtp file descriptor. This function is used when sending
3549 messages over SMTP and also when doing verify callouts.
3551 2. When Exim is called to do synchronous delivery of a locally submitted
3552 message (the -odf or -odi options), it no longer closes stderr before doing
3555 3. Implemented the mua_wrapper option.
3557 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3559 5. Implemented the functions header_remove(), header_testname(),
3560 header_add_at_position(), and receive_remove_recipient(), and exported them
3563 6. If an ACL "warn" statement specified the addition of headers, Exim already
3564 inserted X-ACL-Warn: at the start if there was no header name. However, it
3565 was not making this test for the second and subsequent header lines if
3566 there were newlines in the string. This meant that an invalid header could
3567 be inserted if Exim was badly configured.
3569 7. Allow an ACL "warn" statement to add header lines at the start or after all
3570 the Received: headers, as well as at the end.
3572 8. Added the rcpt_4xx retry error code.
3574 9. Added postmaster_mailfrom=xxx to callout verification option.
3576 10. Added mailfrom=xxxx to the callout verification option, for verify=
3579 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3580 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3582 12. Inserted some casts to stop certain compilers warning when using pointer
3583 differences as field lengths or precisions in printf-type calls (mostly
3584 affecting debugging statements).
3586 13. Added optional readline() support for -be (dynamically loaded).
3588 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3589 same clock tick as a message's arrival, so that its received time was the
3590 same as the "first fail" time on the retry record, and that message
3591 remained on the queue past the ultimate address timeout, every queue runner
3592 would try a delivery (because it was past the ultimate address timeout) but
3593 after another failure, the ultimate address timeout, which should have then
3594 bounced the address, did not kick in. This was a "< instead of <=" error;
3595 in most cases the first failure would have been in the next clock tick
3596 after the received time, and all would be well.
3598 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3599 being recognized when the domain list was tested by the match_domain
3600 condition in an expansion string.
3602 16. Added the ${str2b64: operator.
3604 17. Exim was always calling setrlimit() to set a large limit for the number of
3605 processes, without checking whether the existing limit was already
3606 adequate. (It did check for the limit on file descriptors.) Furthermore,
3607 errors from getrlimit() and setrlimit() were being ignored. Now they are
3608 logged to the main and panic logs, but Exim does carry on, to try to do its
3609 job under whatever limits there are.
3611 18. Imported PCRE 5.0.
3613 19. Trivial typo in log message " temporarily refused connection" (the leading
3616 20. If the log selector return_path_on_delivery was set and an address was
3617 redirected to /dev/null, the delivery process crashed because it assumed
3618 that a return path would always be set for a "successful" delivery. In this
3619 case, the whole delivery is bypassed as an optimization, and therefore no
3622 21. Internal re-arrangement: the function for sending a challenge and reading
3623 a response while authentication was assuming a zero-terminated challenge
3624 string. It's now changed to take a pointer and a length, to allow for
3625 binary data in such strings.
3627 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3629 23. Exim was not respecting finduser_retries when seeking the login of the
3630 uid under which it was called; it was always trying 10 times. (The default
3631 setting of finduser_retries is zero.) Also, it was sleeping after the final
3632 failure, which is pointless.
3634 24. Implemented tls_on_connect_ports.
3636 25. Implemented acl_smtp_predata.
3638 26. If the domain in control=submission is set empty, Exim assumes that the
3639 authenticated id is a complete email address when it generates From: or
3640 Sender: header lines.
3642 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3643 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3644 chown and chgrp in /bin and hostname in /usr/bin.
3646 28. Exim was keeping the "process log" file open after each use, just as it
3647 does for the main log. This opens the possibility of it remaining open for
3648 long periods when the USR1 signal hits a daemon. Occasional processlog
3649 errors were reported, that could have been caused by this. Anyway, it seems
3650 much more sensible not to leave this file open at all, so that is what now
3653 29. The long-running daemon process does not normally write to the log once it
3654 has entered its main loop, and it closes the log before doing so. This is
3655 so that log files can straightforwardly be renamed and moved. However,
3656 there are a couple of unusual error situations where the daemon does write
3657 log entries, and I had neglected to close the log afterwards.
3659 30. The text of an SMTP error response that was received during a remote
3660 delivery was being truncated at 512 bytes. This is too short for some of
3661 the long messages that one sometimes sees. I've increased the limit to
3664 31. It is now possible to make retry rules that apply only when a message has a
3665 specific sender, in particular, an empty sender.
3667 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3668 it possible to be selective about when SMTP synchronization is enforced.
3670 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3672 32. Implemented hosts_connection_nolog.
3674 33. Added an ACL for QUIT.
3676 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3679 35. Added mailbox_size and mailbox_filecount to appendfile.
3681 36. Added control = no_multiline_responses to ACLs.
3683 37. There was a bug in the logic of the code that waits for the clock to tick
3684 in the case where the clock went backwards by a substantial amount such
3685 that the microsecond fraction of "now" was more than the microsecond
3686 fraction of "then" (but the whole seconds number was less).
3688 38. Added support for the libradius Radius client library this is found on
3689 FreeBSD (previously only the radiusclient library was supported).
3695 1. When certain lookups returned multiple values in the form name=value, the
3696 quoting of the values was not always being done properly. Specifically:
3697 (a) If the value started with a double quote, but contained no whitespace,
3699 (b) If the value contained whitespace other than a space character (i.e.
3700 tabs or newlines or carriage returns) it was not quoted.
3701 This fix has been applied to the mysql and pgsql lookups by writing a
3702 separate quoting function and calling it from the lookup code. The fix
3703 should probably also be applied to nisplus, ibase and oracle lookups, but
3704 since I cannot test any of those, I have not disturbed their existing code.
3706 2. A hit in the callout cache for a specific address caused a log line with no
3707 reason for rejecting RCPT. Now it says "Previous (cached) callout
3708 verification failure".
3710 3. There was an off-by-one bug in the queryprogram router. An over-long
3711 return line was truncated at 256 instead of 255 characters, thereby
3712 overflowing its buffer with the terminating zero. As well as fixing this, I
3713 have increased the buffer size to 1024 (and made a note to document this).
3715 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3716 when Exim is waiting for an SMTP response from a remote server, Exim
3717 restarts its select() call on the socket, thereby resetting its timeout.
3718 This is not a problem when such interrupts are rare. Somebody set up a cron
3719 job to run exiwhat every 2 minutes, which is less than the normal select()
3720 timeout (5 or 10 minutes). This meant that the select() timeout never
3721 kicked in because it was always reset. I have fixed this by comparing the
3722 time when an interrupt arrives with the time at the start of the first call
3723 to select(). If more time than the timeout has elapsed, the interrupt is
3724 treated as a timeout.
3726 5. Some internal re-factoring in preparation for the addition of Sieve
3727 extensions (by MH). In particular, the "personal" test is moved to a
3728 separate function, and given an option for scanning Cc: and Bcc: (which is
3729 not set for Exim filters).
3731 6. When Exim created an email address using the login of the caller as the
3732 local part (e.g. when creating a From: or Sender: header line), it was not
3733 quoting the local part when it contained special characters such as @.
3735 7. Installed new OpenBSD configuration files.
3737 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3738 try to make them clearer.
3740 9. Callout options, other than the timeout value, were being ignored when
3741 verifying sender addresses in header lines. For example, when using
3743 verify = header_sender/callout=no_cache
3745 the cache was (incorrectly) being used.
3747 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3748 only the Cygwin environment.
3750 11. When return_path_on_delivery was set as a log selector, if different remote
3751 addresses in the same message used different return paths and parallel
3752 remote delivery occurred, the wrong values would sometimes be logged.
3753 (Whenever a remote delivery process finished, the return path value from
3754 the most recently started remote delivery process was logged.)
3756 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3757 lines when AUTH and/or TLS are in use. This is the "received protocol"
3758 field. Exim used to use "asmtp" for authenticated SMTP, without any
3759 indication (in the protocol name) for TLS use. Now it follows the RFC and
3760 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3761 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3762 names appear in log lines as well as in Received: header lines.
3764 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3765 extensions, and comparison tests, and to fix some bugs.
3767 14. Changes to the "personal" filter test:
3769 (1) The test was buggy in that it was just doing the equivalent of
3770 "contains" tests on header lines. For example, if a user's address was
3771 anne@some.where, the "personal" test would incorrectly be true for
3773 To: susanne@some.where
3775 This test is now done by extracting each address from the header in turn,
3776 and checking the entire address. Other tests that are part of "personal"
3777 are now done using regular expressions (for example, to check local parts
3778 of addresses in From: header lines).
3780 (2) The list of non-personal local parts in From: addresses has been
3781 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3782 taken from the Sieve specification recommendations.
3784 (3) If the message contains any header line starting with "List-" it is
3785 treated as non-personal.
3787 (4) The test for "circular" in the Subject: header line has been removed
3788 because it now seems ill-conceived.
3790 15. Minor typos in src/EDITME comments corrected.
3792 16. Installed latest exipick from John Jetmore.
3794 17. If headers_add on a router specified a text string that was too long for
3795 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3796 of string_sprintf() is now avoided.
3798 18. $message_body_size was not set (it was always zero) when running the DATA
3799 ACL and the local_scan() function.
3801 19. For the "mail" command in an Exim filter, no default was being set for
3802 the once_repeat time, causing a random time value to be used if "once" was
3803 specified. (If the value happened to be <= 0, no repeat happened.) The
3804 default is now 0s, meaning "never repeat". The "vacation" command was OK
3805 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3806 (I found it when inspecting the code).
3808 20. There is now an overall timeout for performing a callout verification. It
3809 defaults to 4 times the callout timeout, which applies to individual SMTP
3810 commands during the callout. The overall timeout applies when there is more
3811 than one host that can be tried. The timeout is checked before trying the
3812 next host. This prevents very long delays if there are a large number of
3813 hosts and all are timing out (e.g. when the network connections are timing
3814 out). The value of the overall timeout can be changed by specifying an
3815 additional sub-option for "callout", called "maxwait". For example:
3817 verify = sender/callout=5s,maxwait=20s
3819 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3820 to the end before writing, but this should make it even safer).
3822 22. Exim was forgetting that it had advertised PIPELINING for the second and
3823 subsequent messages on an SMTP connection. It was also not resetting its
3824 memory on STARTTLS and an internal HELO.
3826 23. When Exim logs an SMTP synchronization error within a session, it now
3827 records whether PIPELINING has been advertised or not.
3829 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3830 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3831 rather than long int.
3833 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3835 26. Added the never_mail option to autoreply.
3841 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3842 crash if the getsockname() call failed; this can happen if a connection is
3843 closed very soon after it is established. The problem was simply in the
3844 order in which certain operations were done, causing Exim to try to write
3845 to the SMTP stream before it had set up the file descriptor. The bug has
3846 been fixed by making things happen in the correct order.
3852 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3853 before the connection was closed, thus losing the rejection response.
3855 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3856 some early Solaris releases, but causes trouble in current releases where
3857 socklen_t is defined.
3859 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3862 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3865 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3866 that releasing the top of it at the end releases what was used for sub-
3867 expansions (unless the block got too big). However, discard this block if
3868 the first thing is a variable or header, so that we can use its block when
3869 it is dynamic (useful for very large $message_headers, for example).
3871 6. Lookups now cache *every* query, not just the most recent. A new, separate
3872 store pool is used for this. It can be recovered when all lookup caches are
3873 flushed. Lookups now release memory at the end of their result strings.
3874 This has involved some general refactoring of the lookup sources.
3876 7. Some code has been added to the store_xxx() functions to reduce the amount
3877 of flapping under certain conditions.
3879 8. log_incoming_interface used to affect only the <= reception log lines. Now
3880 it causes the local interface and port to be added to several more SMTP log
3881 lines, for example "SMTP connection from", and rejection lines.
3883 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3885 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3887 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3888 could be overwritten at the end of the current message (or the start of a
3889 new message if it was set in a HELO ACL). The value is now preserved for
3890 the duration of the SMTP connection.
3892 12. If a transport had a headers_rewrite setting, and a matching header line
3893 contained an unqualified address, that address was qualified, even if it
3894 did not match any rewriting rules. The underlying bug was that the values
3895 of the flags that permit the existence of unqualified sender and recipient
3896 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3897 non-local messages, and by -bnq for local messages) were not being
3898 preserved with the message after it was received.
3900 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3901 "next input=" as part of the text comprising the host identity instead of
3902 the correct text. The code was using the same buffer for two different
3903 strings. However, depending on which order the printing function evaluated
3904 its arguments, the bug did not always show up. Under Linux, for example, my
3905 test suite worked just fine.
3907 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3908 doesn't work with some older versions of Perl. It has been changed to "my",
3909 which in any case is probably the better facility to use.
3911 15. A really picky compiler found some instances of statements for creating
3912 error messages that either had too many or two few arguments for the format
3915 16. The size of the buffer for calls to the DNS resolver has been increased
3916 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3917 for addresses that have a lot of PTR records. This alleviates a problem; it
3918 does not fully solve it.
3920 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3921 buffer now truncates the list and logs the incident, which is the same
3922 action as happens when Exim is looking up a host name and its aliases.
3923 Previously in this situation something unpredictable would happen;
3924 sometimes it was "internal error: store_reset failed".
3926 18. If a server dropped the connection unexpectedly when an Exim client was
3927 using GnuTLS and trying to read a response, the client delivery process
3928 crashed while trying to generate an error log message.
3930 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3931 string, for example:
3933 warn message = H1: something\nH2: something
3935 the text was added as a single header line from Exim's point of view
3936 though it ended up OK in the delivered message. However, searching for the
3937 second and subsequent header lines using $h_h2: did not work. This has been
3938 fixed. Similarly, if a system filter added multiple headers in this way,
3939 the routers could not see them.
3941 20. Expanded the error message when iplsearch is called with an invalid key to
3942 suggest using net-iplsearch in a host list.
3944 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3945 ACLs are no longer actually imposed (and a message to that effect is
3948 22. If a "gecos" field in a passwd entry contained escaped characters, in
3949 particular, if it contained a \" sequence, Exim got it wrong when building
3950 a From: or a Sender: header from that name. A second bug also caused
3951 incorrect handling when an unquoted " was present following a character
3952 that needed quoting.
3954 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3955 was not being matched caselessly.
3957 24. Arranged for all hyphens in the exim.8 source to be escaped with
3960 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3961 properly. Recipient callout cache records were still being keyed to include
3962 the sender, even when use_sender was set false. This led to far more
3963 callouts that were necessary. The sender is no longer included in the key
3964 when use_sender is false.
3966 26. Added "control = submission" modifier to ACLs.
3968 27. Added the ${base62d: operator to decode base 62 numbers.
3970 28. dnsdb lookups can now access SRV records.
3972 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3973 the configuration file.
3975 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3976 (-v) mode. This makes the output for a verbose queue run more intelligible.
3978 31. Added a use_postmaster feature to recipient callouts.
3980 32. Added the $body_zerocount variable, containing the number of binary zero
3981 bytes in the message body.
3983 33. The time of last modification of the "new" subdirectory is now used as the
3984 "mailbox time last read" when there is a quota error for a maildir
3987 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3989 35. Added +ignore_unknown as a special item in host lists.
3991 36. Code for decoding IPv6 addresses in host lists is now included, even if
3992 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3993 address was recognized as an IP address, but was then not correctly decoded
3994 into binary, causing unexpected and incorrect effects when compared with
4001 1. Very minor rewording of debugging text in manualroute to say "list of
4002 hosts" instead of "hostlist".
4004 2. If verify=header_syntax was set, and a header line with an unqualified
4005 address (no domain) and a large number of spaces between the end of the
4006 name and the colon was received, the reception process suffered a buffer
4007 overflow, and (when I tested it) crashed. This was caused by some obsolete
4008 code that should have been removed. The fix is to remove it!
4010 3. When running in the test harness, delay a bit after writing a bounce
4011 message to get a bit more predictability in the log output.
4013 4. Added a call to search_tidyup() just before forking a reception process. In
4014 theory, someone could use a lookup in the expansion of smtp_accept_max_
4015 per_host which, without the tidyup, could leave open a database connection.
4017 5. Added the variables $recipient_data and $sender_data which get set from a
4018 lookup success in an ACL "recipients" or "senders" condition, or a router
4019 "senders" option, similar to $domain_data and $local_part_data.
4021 6. Moved the writing of debug_print from before to after the "senders" test
4024 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4025 problems for message scanning, either using a data ACL, or using
4026 local_scan() because the Received: header was not generated till after they
4027 were called (in order to set the time as the time of reception completion).
4028 I have revised the way this works. The header is now generated after the
4029 body is received, but before the ACL or local_scan() are called. After they
4030 are run, the timestamp in the header is updated.
4036 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4037 before starting a queue runner without re-exec. This happened only when
4038 deliver_drop_privilege was set or when the Exim user was set to root. The
4039 effect of the bug was that timeouts during subsequent deliveries caused
4040 crashes instead of being properly handled. The handler is now left at its
4041 default (and expected) setting.
4043 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4044 message, again when deliver_drop_privilege is set or Exim is run as root.
4045 The bug described in (1) was not present in this case, but the tidying up
4046 of the other signals was missing. I have made the two cases consistent.
4048 3. The ignore_target_hosts setting on a manualroute router was being ignored
4049 for hosts that were looked up using the /MX notation.
4051 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4054 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4055 operated on the sender address. After changing the $sender_address to <>
4056 for the sender address verify, Exim was re-instated it as the original
4057 (before rewriting) address, but remembering that it had rewritten it, so it
4058 wasn't rewriting it again. This bug also had the effect of breaking the
4059 sender address verification caching when the sender address was rewritten.
4061 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4062 This has been changed so that if the ip literal address matches
4063 ignore_target_hosts, the router declines.
4065 7. Added expansion conditions match_domain, match_address, and match_local_
4066 part (NOT match_host).
4068 8. The placeholder for the Received: header didn't have a length field set.
4070 9. Added code to Exim itself and to exim_lock to test for a specific race
4071 condition that could lead to file corruption when using MBX delivery. The
4072 issue is with the lockfile that is created in /tmp. If this file is removed
4073 after a process has opened it but before that process has acquired a lock,
4074 there is the potential for a second process to recreate the file and also
4075 acquire a lock. This could lead to two Exim processes writing to the file
4076 at the same time. The added code performs the same test as UW imapd; it
4077 checks after acquiring the lock that its file descriptor still refers to
4078 the same named file.
4080 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4081 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4084 11. Added the smtp_active_hostname option. If used, this will typically be made
4085 to depend on the incoming interface address. Because $interface_address is
4086 not set up until the daemon has forked a reception process, error responses
4087 that can happen earlier (such as "too many connections") no longer contain
4090 12. If an expansion in a condition on a "warn" statement fails because a lookup
4091 defers, the "warn" statement is abandoned, and the next ACL statement is
4092 processed. Previously this caused the whole ACL to be aborted.
4094 13. Added the iplsearch lookup type.
4096 14. Added ident_timeout as a log selector.
4098 15. Added tls_certificate_verified as a log selector.
4100 16. Added a global option tls_require_ciphers (compare the smtp transport
4101 option of the same name). This controls incoming TLS connections.
4103 17. I finally figured out how to make tls_require_ciphers do a similar thing
4104 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4105 before starting the TLS session.
4107 18. Tabs are now shown as \t in -bP output.
4109 19. If the log selector return_path_on_delivery was set, Exim crashed when
4110 bouncing a message because it had too many Received: header lines.
4112 20. If two routers both had headers_remove settings, and the first one included
4113 a superfluous trailing colon, the final name in the first list and the
4114 first name in the second list were incorrectly joined into one item (with a
4115 colon in the middle).
4121 1. Added -C and -D options to the exinext utility, mainly to make it easier
4122 to include in the automated testing, but these could be helpful when
4123 multiple configurations are in use.
4125 2. The exinext utility was not formatting the output nicely when there was
4126 an alternate port involved in the retry record key, nor when there was a
4127 message id as well (for retries that were specific to a specific message
4128 and a specific host). It was also confused by IPv6 addresses, because of
4129 the additional colons they contain. I have fixed the IPv4 problem, and
4130 patched it up to do a reasonable job for IPv6.
4132 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4133 delivery, the log line now contains "pipelined" if PIPELINING was used.
4135 4. An SMTP transport process used to panic and die if the bind() call to set
4136 an explicit outgoing interface failed. This has been changed; it is now
4137 treated in the same way as a connect() failure.
4139 5. A reference to $sender_host_name in the part of a conditional expansion
4140 that was being skipped was still causing a DNS lookup. This no longer
4143 6. The def: expansion condition was not recognizing references to header lines
4144 that used bh_ and bheader_.
4146 7. Added the _cache feature to named lists.
4148 8. The code for checking quota_filecount in the appendfile transport was
4149 allowing one more file than it should have been.
4151 9. For compatibility with Sendmail, the command line option
4159 and sets the incoming protocol and host name (for trusted callers). The
4160 host name and its colon can be omitted when only the protocol is to be set.
4161 Note the Exim already has two private options, -pd and -ps, that refer to
4162 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4163 "s", but I don't think that's a major issue.
4165 10. A number of refactoring changes to the code, none of which should affect
4168 (a) The number of logging options was getting close to filling up the
4169 32-bit word that was used as a bit map. I have split them into two classes:
4170 those that are passed in the argument to log_write(), and those that are
4171 only ever tested independently outside of that function. These are now in
4172 separate 32-bit words, so there is plenty of room for expansion again.
4173 There is no change in the user interface or the logging behaviour.
4175 (b) When building, for example, log lines, the code previously used a
4176 macro that called string_cat() twice, in order to add two strings. This is
4177 not really sufficiently general. Furthermore, there was one instance where
4178 it was actually wrong because one of the argument was used twice, and in
4179 one call a function was used. (As it happened, calling the function twice
4180 did not affect the overall behaviour.) The macro has been replaced by a
4181 function that can join an arbitrary number of extra strings onto a growing
4184 (c) The code for expansion conditions now uses a table and a binary chop
4185 instead of a serial search (which was left over from when there were very
4186 few conditions). Also, it now recognizes conditions like "pam" even when
4187 the relevant support is not compiled in: a suitably worded error message is
4188 given if an attempt is made to use such a condition.
4190 11. Added ${time_interval:xxxxx}.
4192 12. A bug was causing one of the ddress fields not to be passed back correctly
4193 from remote delivery subprocesses. The field in question was not being
4194 subsequently used, so this caused to problems in practice.
4196 13. Added new log selectors queue_time and deliver_time.
4198 14. Might have fixed a bug in maildirsizefile handling that threw up
4199 "unexpected character" debug warnings, and recalculated the data
4200 unnecessarily. In any case, I expanded the warning message to give more
4203 15. Added the message "Restricted characters in address" to the statements in
4204 the default ACL that block characters like @ and % in local parts.
4206 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4207 Three changes have been made:
4209 (a) There was a serious bug; a negative response to MAIL caused the whole
4210 recipient domain to be cached as invalid, thereby blocking all messages
4211 to all local parts at the same domain, from all senders. This bug has
4212 been fixed. The domain is no longer cached after a negative response to
4213 MAIL if the sender used is not empty.
4215 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4218 (c) A new callout option, "use_sender" has been added for people who want
4219 the modified behaviour.
4225 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4228 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4229 indeed breaks things for older releases.
4231 3. Added additional logging to the case where there is a problem reading data
4232 from a filter that is running in a subprocess using a pipe, in order to
4233 try to track down a specific problem.
4235 4. Testing facility fudge: when running in the test harness and attempting
4236 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4237 getting "No route to host". Convert this to a timeout.
4239 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4242 6. Some OS don't have socklen_t but use size_t instead. This affects the
4243 fifth argument of getsockopt() amongst other things. This is now
4244 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4245 can be set for individual OS. I have set it for SunOS5, OSF1, and
4246 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4247 some earlier ones do not.
4249 7. Change 4.30/15 was not doing the test caselessly.
4251 8. The standard form for an IPv6 address literal was being rejected by address
4252 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4253 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4254 this, as well as the form without the "IPv6" on the front (but only when
4255 address literals are enabled, of course).
4257 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4259 10. Exim crashed if a message with an empty sender address specified by -f
4260 encountered a router with an errors_to setting. This could be provoked only
4261 by a command such as
4265 where an empty string was supplied; "<>" did not hit this bug.
4267 11. Installed PCRE release 4.5.
4269 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4270 remained set. It is now erased.
4272 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4273 times from message ids (which are base 36 rather than the normal 62).
4275 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4276 were being counted as actual protocol errors, and logged if the log
4277 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4278 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4279 and DATA following a set of rejected RCPTs do not count as protocol errors.
4280 In other words, Exim assumes they were pipelined, though this may not
4281 actually be the case. Of course, in all cases the client gets an
4282 appropriate error code.
4284 15. If a lookup fails in an ACL condition, a message about the failure may
4285 be available; it is used if testing the ACL cannot continue, because most
4286 such messages specify what the cause of the deferral is. However, some
4287 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4288 that caused an old message to be retained and used if a later statement
4289 caused a defer, replacing the real cause of the deferral.
4291 16. If an IP address had so many PTR records that the DNS lookup buffer
4292 was not large enough to hold them, Exim could crash while trying to process
4293 the truncated data. It now detects and logs this case.
4295 17. Further to 4.21/58, another change has been made: if (and only if) the
4296 first line of a message (the first header line) ends with CRLF, a bare LF
4297 in a subsequent header line has a space inserted after it, so as not to
4298 terminate the header.
4300 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4301 unnecessarily, used atoi() instead of strtol(), and didn't check the
4302 termination when getting file sizes from file names by regex.
4304 19. Completely re-implemented the support for maildirsize files, in the light
4305 of a number of problems with the previous contributed implementation
4306 (4.30/29). In particular:
4308 . If the quota is zero, the maildirsize file is maintained, but no quota is
4311 . If the maildir directory does not exist, it is created before any attempt
4312 to write a maildirsize file.
4314 . The quota value in the file is just a cache; if the quota is changed in
4315 the transport, the new value overrides.
4317 . A regular expression is available for excluding directories from the
4320 20. The autoreply transport checks the characters in options that define the
4321 message's headers; it allows continued headers, but it was checking with
4322 isspace() after an embedded newline instead of explicitly looking for a
4325 21. If all the "regular" hosts to which an address was routed had passed their
4326 expiry times, and had not reached their retry times, the address was
4327 bounced, even if fallback hosts were defined. Now Exim should go on to try
4330 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4331 equivalent code in the SMTP transport. Some hosts send humungous responses
4332 to HELO/EHLO, more than 1024 it seems.
4334 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4335 gives compiler warnings in some environments. I've now done it "properly",
4338 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4339 (because of problems with the built-in one) was declared to return uschar *
4340 instead of char *, causing compiler failure.
4342 25. Fixed a file descriptor leak when processing alias/forward files.
4344 26. Fixed a minor format string issue in dbfn.c.
4346 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4348 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4349 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4352 29. When a sender address is verified, it is cached, to save repeating the test
4353 when there is more than one recipient in a message. However, when the
4354 verification involves a callout, it is possible for different callout
4355 options to be set for different recipients. It is too complicated to keep
4356 track of this in the cache, so now Exim always runs a verification when a
4357 callout is required, relying on the callout cache for the optimization.
4358 The overhead is duplication of the address routing, but this should not be
4361 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4362 to be verified with callout=postmaster, and the main callout worked but the
4363 postmaster check failed, the verification correctly failed. However, if a
4364 subsequent RCPT command asked for sender verification *without* the
4365 postmaster check, incorrect caching caused this verification also to fail,
4368 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4369 it was not caching the DNS options (qualify_single, search_parents) that
4370 were used when the lookup failed. A subsequent lookup with different
4371 options therefore always gave the same answer, though there were cases
4372 where it should not have. (Example: a "domains = !$mx_any" option on a
4373 dnslookup router: the "domains" option is always processed without any
4374 widening, but the router might have qualify_single set.) Now Exim uses the
4375 cached value only when the same options are set.
4377 32. Added John Jetmore's "exipick" utility to the distribution.
4379 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4380 than a timeout (e.g. a certificate is required, and is not provided), an
4381 Exim server now closes the connection immediately. Previously it waited for
4382 the client to close - but if the client is SSL, it seems that they each
4383 wait for each other, leading to a delay before one of them times out.
4385 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4386 maintained 0.8.x compatibility because I don't think many are using it, and
4387 it is clearly obsolete.
4389 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4392 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4393 client certificate was expired. A simple patch fixes this, though I don't
4394 understand the full logic of why the verify callback is called multiple
4397 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4398 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4399 which causes problems with some clients (such as the Certicom SSL Plus
4400 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4401 disables the coutermeasure allowing Eudora to connect."
4403 38. Exim was not checking that a write() to a log file succeeded. This could
4404 lead to Bad Things if a log got too big, in particular if it hit a file
4405 size limit. Exim now panics and dies if it cannot write to a log file, just
4406 as it does if it cannot open a log file.
4408 39. Modified OS/Makefile-Linux so that it now contains
4410 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4412 The two -D definitions ensure that Exim is compiled with large file
4413 support, which makes it possible to handle log files that are bigger than
4416 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4417 instance) a domain was checked against a named list that involved a lookup,
4418 causing $domain_data to be set, then another domain was checked against the
4419 same list, then the first domain was re-checked, the value of $domain_data
4420 after the final check could be wrong. In particular, if the second check
4421 failed, it could be set empty. This bug probably also applied to
4424 41. The strip_trailing_dot option was not being applied to the address given
4425 with the -f command-line option.
4427 42. The code for reading a message's header from the spool was incrementing
4428 $received_count, but never initializing it. This meant that the value was
4429 incorrect (doubled) while delivering a message in the same process in which
4430 it was received. In the most common configuration of Exim, this never
4431 happens - a fresh exec is done - but it can happen when
4432 deliver_drop_privilege is set.
4434 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4435 it now includes up to 150 characters of the unexpected data in the log
4438 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4439 and building data strings. The size of both of these buffers was 10 000
4440 bytes - far larger than anybody would *ever* want, thought I. Needless to
4441 say, somebody hit the limit. I have increased the maximum line length to
4442 20 000 and the maximum data length of concatenated lines to 100 000. I have
4443 also fixed two bugs, because there was no checking on these buffers. Tsk,
4444 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4445 buffer is too small.
4447 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4448 lsearch lookups. Now it does.
4450 46. When parsing a route_list item in a manualroute router, a fixed-length
4451 buffer was used for the list of hosts. I made this 1024 bytes long,
4452 thinking that nobody would ever have a list of hosts that long. Wrong.
4453 Somebody had a whole pile of complicated expansion conditions, and the
4454 string was silently truncated, leading to an expansion error. It turns out
4455 that it is easier to change to an unlimited length (owing to other changes
4456 that have happened since this code was originally written) than to build
4457 structure for giving a limitation error. The length of the item that
4458 expands into the list of hosts is now unlimited.
4460 47. The lsearch lookup could not handle data where the length of text line was
4461 more than 4095 characters. Such lines were truncated, leading to shortened
4462 data being returned. It should now handle lines of any length.
4464 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4465 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4468 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4469 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4470 no longer happens in comment lines. A list of replacements is now placed
4471 at the head of all of the source files, except those whose only change is
4472 to replace PERL_COMMAND in the very first #! line.
4474 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4475 messages on the queue, with a bottom-up merge sort, using code contributed
4476 by Michael Haardt. This should make operations like -bp somewhat faster on
4477 large queues. It won't affect queue runners, except when queue_run_in_order
4480 51. Installed eximstats 1.31 in the distribution.
4482 52. Added support for SRV lookups to the dnslookup router.
4484 53. If an ACL referred to $message_body or $message_body_end, the value was not
4485 reset for any messages that followed in the same SMTP session.
4487 54. The store-handling optimization for building very long strings was not
4488 differentiating between the different store pools. I don't think this
4489 actually made any difference in practice, but I've tidied it.
4491 55. While running the routers to verify a sender address, $sender_address
4492 was still set to the sender address. This is wrong, because when routing to
4493 send a bounce to the sender, it would be empty. Therefore, I have changed
4494 it so that, while verifying a sender address, $sender_address is set to <>.
4495 (There is no change to what happens when verifying a recipient address.)
4497 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4498 target A or AAAA records (if not already returned) without resetting the
4499 qualify_single or search_parents options of the DNS resolver. These are
4500 inappropriate in this case because the targets of MX and SRV records must
4501 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4502 target that, when qualified, matched something in the local domain. These
4503 two options are now turned off when doing these lookups.
4505 57. It seems that at least some releases of Reiserfs (which does not have the
4506 concept of a fixed number of inodes) returns zero and not -1 for the
4507 number of available inodes. This interacted badly with check_spool_inodes,
4508 which assumed that -1 was the "no such thing" setting. What I have done is
4509 to check that the total number of inodes is greater than zero before doing
4510 the test of how many are available.
4512 58. When a "warn" ACL statement has a log_message modifier, the message is
4513 remembered, and not repeated. This is to avoid a lot of repetition when a
4514 message has many recipients that cause the same warning to be written.
4515 Howewer, Exim was preserving the list of already written lines for an
4516 entire SMTP session, which doesn't seem right. The memory is now reset if a
4517 new message is started.
4519 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4520 debugging output unless log_rewrite was also set.
4522 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4523 of (int)(handle) when we know that handle contains (void *)(-1).
4525 61. The Exim daemon panic-logs an error return when it closes the incoming
4526 connection. However "connection reset by peer" seems to be common, and
4527 isn't really an error worthy of noting specially, so that particular error
4530 62. When Exim is trying to find all the local interfaces, it used to panic and
4531 die if the ioctl to get the interface flags failed. However, it seems that
4532 on at least one OS (Solaris 9) it is possible to have an interface that is
4533 included in the list of interfaces, but for which you get a failure error
4534 for this call. This happens when the interface is not "plumbed" into a
4535 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4536 failure of the "get flags" call assumes that the interface is down.
4538 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4539 makes life easier for people who are doing arithmetic on fields extracted
4540 from dates, where you often get leading zeros that should not be
4541 interpreted as octal.
4543 64. Added qualify_domain to the redirect router, to override the global
4546 65. If a pathologically long header line contained very many addresses (the
4547 report of this problem mentioned 10 000) and each of them was rewritten,
4548 Exim could use up a very large amount of memory. (It kept on making new
4549 copies of the header line as it rewrote, and never released the old ones.)
4550 At the expense of a bit more processing, the header rewriting function has
4551 been changed so that it no longer eats memory in this way.
4553 66. The generation of the Received: header has been moved from the time that a
4554 message starts to be received, to the time that it finishes. The timestamp
4555 in the Received: header should now be very close to that of the <= log
4556 line. There are two side-effects of this change:
4558 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4559 logged header lines no longer include the local Received: line, because
4560 it has not yet been created. The same applies to a copy of the message
4561 that is returned to a non-SMTP sender when a message is rejected.
4563 (b) When a filter file is tested using -bf, no additional Received: header
4564 is added to the test message. After some thought, I decided that this
4567 This change does not affect the value of $received_for. It is still set
4568 after address rewriting, but before local_scan() is called.
4570 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4572 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4573 gave an unhelpful panic error message, and a defer error. I have managed to
4574 change this behaviour so that it now rejects any supplied certificate,
4575 which seems right, as the list of acceptable certificates is empty.
4577 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4578 gave an unhelpful defer error. I have not managed to make this reject any
4579 supplied certificates, but the error message it gives is "no certificate
4580 supplied", which is not helpful.
4582 70. exigrep's output now also includes lines that are not associated with any
4583 message, but which match the given pattern. Implemented by a patch from
4584 Martin Sluka, which also tidied up the Perl a bit.
4586 71. Recipient callout verification, like sender verification, was using <> in
4587 the MAIL FROM command. This isn't really the right thing, since the actual
4588 sender may affect whether the remote host accepts the recipient or not. I
4589 have changed it to use the actual sender in the callout; this means that
4590 the cache record is now keyed on a recipient/sender pair, not just the
4591 recipient address. There doesn't seem to be a real danger of callout loops,
4592 since a callout by the remote host to check the sender would use <>.
4593 [SEE ABOVE: changed after hitting problems.]
4595 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4596 temporary errors. However, in the case of such a code being given after
4597 the end of a data transmission (i.e. after ".") Exim was failing to write
4598 a retry record for the message. (Yes, there was some broken host that was
4599 actually sending 8xx at this point.)
4601 73. An unknown lookup type in a host list could cause Exim to panic-die when
4602 the list was checked. (An example that provoked this was putting <; in the
4603 middle of a list instead of at the start.) If this happened during a DATA
4604 ACL check, a -D file could be left lying around. This kind of configuration
4605 error no longer causes Exim to die; instead it causes a defer errror. The
4606 incident is still logged to the main and panic logs.
4608 74. Buglet left over from Exim 3 conversion. The message "too many messages
4609 in one connection" was written to the rejectlog but not the mainlog, except
4610 when address rewriting (yes!) was being logged.
4612 75. Added write_rejectlog option.
4614 76. When a system filter was run not as root (that is, when system_filter_user
4615 was set), the values of the $n variables were not being returned to the
4616 main process; thus, they were not subsequently available in the $sn
4619 77. Added +return_path_on_delivery log selector.
4621 78. A connection timeout was being treated differently from recipients deferred
4622 when testing hosts_max_try with a message that was older than the host's
4623 retry timeout. (The host should not be counted, thus allowing all hosts to
4624 be tried at least once before bouncing.) This may have been the cause of an
4625 occasionally reported bug whereby a message would remain on the queue
4626 longer than the retry timeout, but would be bounced if a delivery was
4627 forced. I say "may" because I never totally pinned down the problem;
4628 setting up timeout/retry tests is difficult. See also the next item.
4630 79. The ultimate address timeout was not being applied to errors that involved
4631 a combination of host plus message (for example, a timeout on a MAIL
4632 command). When an address resolved to a number of possible hosts, and they
4633 were not all tried for each delivery (e.g. because of hosts_max_try), a
4634 message could remain on the queue longer than the retry timeout.
4636 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4639 81. Fixed an obscure SMTP outgoing bug which required at least the following
4640 conditions: (a) there was another message waiting for the same server;
4641 (b) the server returned 5xx to all RCPT commands in the first message so
4642 that the message was not completed; (c) the server dropped the connection
4643 or gave a negative response to the RSET that Exim sends to abort the
4644 transaction. The observed case was a dropped connection after DATA that had
4645 been sent in pipelining mode. That is, the server had advertised PIPELINING
4646 but was not implementing it correctly. The effect of the bug was incorrect
4647 behaviour, such as trying another host, and this could lead to a crash.
4653 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4654 and daemon.c were passed as pointers to ints; they should have been
4655 pointers to socklen_t variables (which are typically unsigned ints).
4657 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4660 3. Fixed a really odd bug that affected only the testing scheme; patching a
4661 certain fixed string in the binary changed the value of another string that
4662 happened to be identical to the end of the original first string.
4664 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4665 name", it returns that address as the IP address. On some operating
4666 systems (e.g. Solaris), it also passes back the IP address string as the
4667 "host name". However, on others (e.g. Linux), it passes back an empty
4668 string. Exim wasn't checking for this, and was changing the host name to an
4669 empty string, assuming it had been canonicized.
4671 5. Although rare, it is permitted to have more than one PTR record for a given
4672 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4673 all the names associated with an address, because they do in Solaris.
4674 However, it seems that they do not in Linux for data that comes from the
4675 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4676 I found this out when I moved to a new Linux workstation and tried to run
4677 the Exim test suite.
4679 To get round this problem I have changed the code so that it now does its
4680 own call to the DNS to look up PTR records when searching for a host name.
4681 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4682 addresses that are only in /etc/hosts are still found.
4684 This behaviour is, however, controlled by an option called host_lookup_
4685 order, which defaults to "bydns:byaddr". If people want to use the other
4686 order, or indeed, just use one or the other means of lookup, they can
4687 specify it in this variable.
4689 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4690 some operating systems, this comes back from gethostbyaddr() as an empty
4691 string, and this is what Exim used to test for. However, it seems that in
4692 other systems, "." is yielded. Exim now tests for this case too.
4694 7. The values of check_spool_space and check_log_space are now held internally
4695 as a number of kilobytes instead of an absolute number of bytes. If a
4696 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4697 kilobyte. This means that much larger values can be stored.
4699 8. Exim monitor: an attempt to get the action menu when not actually pointing
4700 at a message produces an empty menu entitled "No message selected". This
4701 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4702 no entries in it ("Shell widget menu has zero width and/or height"). So I
4703 have added a single, blank menu entry in this case.
4705 9. Added ${quote_local_part.
4707 10. MIME decoding is now applied to the contents of Subject: header lines when
4710 11. Now that a reference to $sender_host_address automatically causes a reverse
4711 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4712 host lookup before query-style lookups in lists that might use this
4713 variable. This has therefore been abolished, and the "net-" prefix is no
4714 longer necessary for query-style lookups.
4716 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4717 have been a typo for LFLAGS, so it has been changed.
4719 13. The install script calls Exim with "-C /dev/null" in order to find the
4720 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4721 to be output. Howeve, since Exim outputs its version number before the
4722 error, it didn't break the script. It just looked ugly. I fixed this by
4723 always allowing "-C /dev/null" if the caller is root.
4725 14. Ignore overlarge ACL variable number when reading spool file - insurance
4726 against a later release with more variables having written the file.
4728 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4729 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4730 this, as well as the form without the "IPv6" on the front.
4732 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4733 OS/Makefile-Darwin file.
4735 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4736 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4738 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4739 variable. However, some OS use ints for inodes. I've added cast to long int
4740 to get rid of the compiler warning.
4742 19. I had forgotten to lock out "/../" in configuration file names when
4743 ALT_CONFIG_PREFIX was set.
4745 20. Routers used for verification do not need to specify transports. However,
4746 if such a router generated a host list, and callout was configured, Exim
4747 crashed, because it could not find a port number from the (non-existent)
4748 transport. It now assumes port 25 in this circumstance.
4750 21. Added the -t option to exigrep.
4752 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4753 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4754 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4755 them is set, LOOKUP_LSEARCH is forced.
4757 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4759 24. Added sender and host information to the "rejected by local_scan()" log
4760 line; previously there was no indication of these.
4762 25. Added .include_if_exists.
4764 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4765 Linux. It turns out that not all file systems support this. Apparently some
4766 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4767 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4768 not supported on the file descriptor, is now ignored when Exim is trying to
4769 sync a directory. This applies only to Linux.
4771 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4773 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4774 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4777 29. Added support for maildirsize files from supplied patch (modified a bit).
4779 30. The use of :fail: followed by an empty string could lead Exim to respond to
4780 sender verification failures with (e.g.):
4782 550 Verification failed for <xxx>
4783 550 Sender verify failed
4785 where the first response line was missing the '-' that indicates it is not
4786 the final line of the response.
4788 31. The loop for finding the name of the user that called Exim had a hardwired
4789 limit of 10; it now uses the value of finduser_retries, which is used for
4790 all other user lookups.
4792 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4795 33. Exim was neglecting to zero errno before one call of strtol() when
4796 expanding a string and expecting an integer value. On some systems this
4797 resulted in spurious "integer overflow" errors. Also, it was casting the
4798 result into an int without checking.
4800 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4801 did not work. The code looks as if it has *never* worked, though it appears
4802 to have been documented since at least releast 1.62. I have made it work.
4804 35. The "timeout_DNS" error in retry rules, also documented since at least
4805 1.62, also never worked. As it isn't clear exactly what this means, and
4806 clearly it isn't a major issue, I have abolished the feature by treating it
4807 as "timeout", and writing a warning to the main and panic logs.
4809 36. The display of retry rules for -brt wasn't always showing the error code
4812 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4813 timeout_connect_A, timeout_connect_MX.
4815 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4816 to the empty sender.
4818 39. The daemon was not analysing the content of -oX till after it had closed
4819 stderr and disconnected from the controlling terminal. This meant that any
4820 syntax errors were only noted on the panic log, and the return code from
4821 the command was 0. By re-arranging the code a little, I've made the
4822 decoding happen first, so such errors now appear on stderr, and the return
4823 code is 1. However, the actual setting up of the sockets still happens in
4824 the disconnected process, so errors there are still only recorded on the
4827 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4828 connections (as happens on some IP stacks) was logged at start up time as
4829 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4830 it from "IPv6 and IPv4", which means that two separate sockets are being
4833 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4834 says whether AF_INET or AF_INET6 was passed as an argument.
4836 42. Exiwhat output was messed up when time zones were included in log
4839 43. Exiwhat now gives more information about the daemon's listening ports,
4840 and whether -tls-on-connect was used.
4842 44. The "port" option of the smtp transport is now expanded.
4844 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4845 silently ignored. Now an error message is written to the main and panic
4848 46. There's a new ACL modifier called "logwrite" which writes to a log file
4849 as soon as it is encountered.
4851 47. Added $local_user_uid and $local_user_gid at routing time.
4853 48. Exim crashed when trying to verify a sender address that was being
4856 49. Exim was recognizing only a space character after ".include". It now also
4857 recognizes a tab character.
4859 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4860 extracting the relevant information from the specification. The man page no
4861 longer contains scrambled data for the -d option, and I've added a section
4862 at the front about calling Exim under different names.
4864 51. Added "extra_headers" argument to the "mail" command in filter files.
4866 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4869 53. Installed eximstats 1.29.
4871 54. Added transport_filter_timeout as a generic transport option.
4873 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4874 Cc: header lines. This was required by RFC 822, but it not required by RFC
4877 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4878 incoming messages that did not have them. Now it does so only if the
4879 message originates locally, that is, if there is no associated remote host
4880 address. When Resent- header lines are present, this applies to the Resent-
4881 lines rather than the non-Resent- lines.
4883 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4884 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4886 58. Messages for configuration errors now include the name of the main
4887 configuration file - useful now that there may be more than one file in a
4888 list (.included file names were always shown).
4890 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4891 for those rare installations that do not start the daemon as root or run it
4892 setuid root. I've cut out the call to initgroups() if the daemon is not
4895 60. The Exim user and group can now be bound into the binary as text strings
4896 that are looked up at the start of Exim's processing.
4898 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4900 62. Added $mailstore_basename variable.
4902 63. Installed patch to sieve.c from Michael Haardt.
4904 64. When Exim failed to open the panic log after failing to open the main log,
4905 the original message it was trying to log was written to stderr and debug
4906 output, but if they were not available (the usual case in production), it
4907 was lost. Now it is written to syslog before the two lines that record the
4908 failures to open the logs.
4910 65. Users' Exim filters run in subprocesses under the user's uid. It is
4911 possible for a "deliver" command or an alias in a "personal" command to
4912 provoke an address rewrite. If logging of address rewriting is configured,
4913 this fails because the process is not running as root or exim. There may be
4914 a better way of dealing with this, but for the moment (because 4.30 needs
4915 to be released), I have disabled address rewrite logging when running a
4916 filter in a non-root, non-exim process.
4922 1. The buildconfig auxiliary program wasn't quoting the value set for
4923 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4924 not defined. This bug was masked in 4.22 by the effect that was fixed in
4927 2. Some messages that were rejected after a message id was allocated were
4928 shown as "incomplete" by exigrep. It no longer does this for messages that
4929 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4931 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4932 have allow_domain_literals set, the ID did not get logged in the <= line.
4933 Domain literals are now always recognized in Message-ID: header lines.
4935 4. The first argument for a ${extract expansion item is the key name or field
4936 number. Leading and trailing spaces in this item were not being ignored,
4937 causing some misleading effects.
4939 5. When deliver_drop_privilege was set, single queue runner processes started
4940 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4941 same command in the process it spins off) were not dropping privilege.
4943 6. When the daemon running as "exim" started a queue runner, it always
4944 re-executed Exim in the spun-off process. This is a waste of effort when
4945 deliver_drop_privilege is set. The new process now just calls the
4946 queue-runner function directly.
4952 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4955 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4956 ignored. Though the use of -oP was forcing the writing of a pid file, it
4957 was always written to the default place.
4959 3. If the message "no IP address found for host xxxx" is generated during
4960 incoming verification, it is now followed by identification of the incoming
4961 connection (so you can more easily find what provoked it).
4963 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4965 5. Added some features to "harden" Exim a bit more against certain attacks:
4967 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4968 be put in Local/Makefile. This is like the never_users runtime option,
4969 but it cannot be overridden. The default setting is "root".
4971 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4972 prefix string with which any file named in a -C command line option
4975 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4976 is retained for -C and -D only if the caller of Exim is root. Without
4977 it, the exim user may also use -C and -D and retain privilege.
4979 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4980 command line option is disabled.
4982 6. Macro names set by the -D option must start with an upper case letter, just
4983 like macro names defined in the configuration file.
4985 7. Added "dereference=" facility to LDAP.
4987 8. Two instances of the typo "uknown" in the source files are fixed.
4989 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4990 the Configure-Makefile script screwed up while processing it.
4992 10. Incorporated PCRE 4.4.
4994 11. The SMTP synchronization check was not operating right at the start of an
4995 SMTP session. For example, it could not catch a HELO sent before the client
4996 waited for the greeting. There is now a check for outstanding input at the
4997 point when the greeting is written. Because of the duplex, asynchronous
4998 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4999 way, but not yet received, when the check is performed.
5001 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5002 on TCP/IP sockets, because this apparently causes some broken clients to
5005 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5006 unchanged) from the Cygwin maintainer.
5008 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5009 was supported instead of testing for mbx. Effectively a typo.
5011 15. The spa authenticator server code was not checking that the input it
5012 received was valid base64.
5014 16. The debug output line for the "set" modifier in ACLs was not showing the
5015 name of the variable that was being set.
5017 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5019 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5020 lookup on its own. Something else was needed to trigger the lookup. For
5021 example, a match in host_lookup or the need for a host name in a host list.
5022 Now, if $sender_host_name is referenced and the host name has not yet been
5023 looked up, a lookup is performed. If the lookup fails, the variable remains
5024 empty, and $host_lookup_failed is set to "1".
5026 19. Added "eqi" as a case-independent comparison operator.
5028 20. The saslauthd authentication condition could segfault if neither service
5029 nor realm was specified.
5031 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5032 error message that was logged was misleading, and incoming SMTP
5033 connections were dropped. The message is now more accurate, and temporary
5034 errors are given to SMTP connections.
5036 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5037 (except RSET) with a 421 error, until QUIT is received. However, it was
5038 failing to send a response to QUIT.
5040 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5041 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5042 ACL is now run after the helo_try_verify_hosts code.
5044 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5045 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5046 case-independent, so other case variants are also recognized). Apparently
5047 some systems use these upper case variants.
5049 25. If more than two messages were waiting for the same host, and a transport
5050 filter was specified for the transport, Exim sent two messages over the
5051 same TCP/IP connection, and then failed with "socket operation on non-
5052 socket" when it tried to send the third.
5054 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5056 27. The extern definition of crypt16() in expand.c was not being excluded when
5057 the OS had its own crypt16() function.
5059 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5060 as a preferred synonym for return_size_limit, both as an option and as an
5063 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5065 30. Changed the default configuration ACL to relax the local part checking rule
5066 for addresses that are not in any local domains. For these addresses,
5067 slashes and pipe symbols are allowed within local parts, but the sequence
5068 /../ is explicitly forbidden.
5070 31. SPA server authentication was not clearing the challenge buffer before
5073 32. log_message in a "warn" ACL statement was writing to the reject log as
5074 well as to the main log, which contradicts the documentation and doesn't
5075 seem right (because no rejection is happening). So I have stopped it.
5077 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5078 However, I am unable to do any testing of this.
5080 34. Fixed an infelicity in the appendfile transport. When checking directories
5081 for a mailbox, to see if any needed to be created, it was accidentally
5082 using path names with one or more superfluous leading slashes; tracing
5083 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5085 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5086 contents are added to the log line that is written for every discarded
5087 recipient. (Previously a log_message setting was ignored.)
5089 36. The ${quote: operator now quotes the string if it is empty.
5091 37. The install script runs exim in order to find its version number. If for
5092 some reason other than non-existence or emptiness, which it checks, it
5093 could not run './exim', it was installing it with an empty version number,
5094 i.e. as "exim-". This error state is now caught, and the installation is
5097 38. An argument was missing from the function that creates an error message
5098 when Exim fails to connect to the socket for saslauthd authentication.
5099 This could cause Exim to crash, or give a corrupted message.
5101 39. Added isip, isip4, and isip6 to ${if conditions.
5103 40. The ACL variables $acl_xx are now saved with the message, and can be
5104 accessed later in routers, transports, and filters.
5106 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5107 strings in the file are not string-expanded.
5109 42. If a MAIL command specified a SIZE value that was too large to fit into an
5110 int variable, the check against message_size_limit failed. Such values are
5111 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5112 day this will have to be increased, but I don't think I want to be around
5113 when emails are that large.
5120 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5121 iconv() is not standard in FreeBSD.
5123 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5124 IPv6 enabled. The observed symptom was a segmentation fault on return from
5125 the function os_common_find_running_interfaces() in src/os.c.
5127 3. In the check_special_case() function in daemon.c I had used "errno" as an
5128 argument name, which causes warnings on some systems. This was basically a
5129 typo, since it was named "eno" in the comments!
5131 4. The code that waits for the clock to tick (at a resolution of some fraction
5132 of a second) so as to ensure message-id uniqueness was always waiting for
5133 at least one whole tick, when it could have waited for less. [This is
5134 almost certainly not relevant at current processor speeds, where it is
5135 unlikely to ever wait at all. But we try to future-proof.]
5137 5. The function that sleeps for a time interval that includes fractions of a
5138 second contained a race. It did not block SIGALRM between setting the
5139 timer, and suspending (a couple of lines later). If the interval was short
5140 and the sigsuspend() was delayed until after it had expired, the suspension
5141 never ended. On busy systems this could lead to processes getting stuck for
5144 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5145 process, before it forks any delivery processes. The open lookup caching
5146 mechanism meant that the open file or database connection was passed into
5147 the delivery process. The problem was that delivery processes always tidy
5148 up cached lookup data. This could cause a problem for the next delivery
5149 process started by the queue runner, because the external queue runner
5150 process does not know about the closure. So the next delivery process
5151 still has data in the lookup cache. In the case of a file lookup, there was
5152 no problem because closing a file descriptor in a subprocess doesn't affect
5153 the parent. However, if the lookup was caching a connection to a database,
5154 the connection was closed, and the second delivery process was likely to
5155 see errors such as "PGSQL: query failed: server closed the connection
5156 unexpectedly". The problem has been fixed by closing all cached lookups
5157 in a queue runner before running a delivery process.
5159 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5160 seem to have the "const" qualifier which it has on other OS. I've
5163 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5164 *of the same type* (client or server) with the same public name that an
5165 error should be diagnosed.
5167 9. When Exim looked up a host name for an IP address, but failed to find the
5168 original IP address when looking up the host name (a safety check), it
5169 output the message "<ip address> does not match any IP for NULL", which was
5170 confusing, to say the least. The bug was that the host name should have
5171 appeared instead of "NULL".
5173 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5174 user that is built into the binary, and the -C or -D options is used, root
5175 privilege is dropped before the configuration file is read. In addition,
5176 logging is switched to stderr instead of the normal log files. If the
5177 configuration then re-defines the Exim user, the unprivileged environment
5178 is probably not what is expected, so Exim logs a panic warning message (but
5181 However, if deliver_drop_privilege is set, the unprivileged state may well
5182 be exactly what is intended, so the warning has been cut out in that case,
5183 and Exim is allowed to try to write to its normal log files.
5189 1. smtp_return_error_details was not giving details for temporary sender
5190 or receiver verification errors.
5192 2. Diagnose a configuration error if two authenticators have the same public
5195 3. Exim used not to create the message log file for a message until the first
5196 delivery attempt. This could be confusing when incoming messages were held
5197 for policy or load reasons. The message log file is now created at the time
5198 the message is received, and an initial "Received" line is written to it.
5200 4. The automatically generated man page for command line options had a minor
5201 bug that caused no ill effects; however, a more serious problem was that
5202 the procedure for building the man page automatically didn't always
5203 operate. Consequently, release 4.20 contains an out-of-date version. This
5204 shouldn't happen again.
5206 5. When building Exim with embedded Perl support, the script that builds the
5207 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5208 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5210 6. The freeze_tell option was not being used for messages that were frozen on
5211 arrival, either by an ACL or by local_scan().
5213 7. Added the smtp_incomplete_transaction log selector.
5215 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5216 was accepting AUTH without a new EHLO.
5218 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5219 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5222 10. Logging of TCP/IP connections (when configured) now happens in the main
5223 daemon process instead of the child process, so that the TCP/IP connection
5224 count is more accurate (but it can never be perfect).
5226 11. The use of "drop" in a nested ACL was not being handled correctly in the
5227 outer ACL. Now, if condition failure induced by the nested "drop" causes
5228 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5229 or "require"), the connection is dropped.
5231 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5232 that yield "discard" can now be used with an "accept" or a "discard" verb,
5233 but an error is generated for any others (because I can't see a useful way
5234 to define what should happen).
5236 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5237 are now processed in the same way as lines in the Exim configuration file.
5238 In particular, continuation lines are supported.
5240 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5242 15. Added -ti meaning -t -i.
5244 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5245 domains, and warn by logging if others are found.
5247 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5248 returned from GIFCONF when Exim is trying to find the list of interfaces on
5249 a host. The code in os.c has been modified to copy each ifreq to an aligned
5250 structure in all cases.
5252 Also, in some cases, the returned ifreq's were being copied to a 'struct
5253 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5254 means the last couple of bytes of an IPv6 address could be chopped if the
5255 ifreq contained only a normal sockaddr (14 bytes storage).
5257 18. Named domain lists were not supported in the hosts_treat_as_local option.
5258 An entry such as +xxxx was not recognized, and was treated as a literal
5261 19. Ensure that header lines added by a DATA ACL are included in the reject log
5262 if the ACL subsequently rejects the message.
5264 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5265 MD5 (which is deprecated).
5267 21. When testing a filter file using -bf, Exim was writing a message when it
5268 took the sender from a "From " line in the message, but it was not doing so
5269 when it took $return_path from a Return-Path: header line. It now does.
5271 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5272 with a valid header line field name (a series of printing characters
5273 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5275 23. Changed "disc" in the source to "disk" to conform to the documentation and
5276 the book and for uniformity.
5278 24. Ignore Sendmail's -Ooption=value command line item.
5280 25. When execve() failed while trying to run a command in a pipe transport,
5281 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5282 could be confused with a return value of 69 from the command itself. This
5283 has been changed to 127, the value the shell returns if it is asked to run
5284 a non-existent command. The wording for the related log line suggests a
5285 non-existent command as the problem.
5287 26. If received_header_text expands to an empty string, do not add a Received:
5288 header line to the message. (Well, it adds a token one on the spool, but
5289 marks it "old" so that it doesn't get used or transmitted.)
5291 27. Installed eximstats 1.28 (addition of -nt option).
5293 28. There was no check for failure on the call to getsockname() in the daemon
5294 code. This can fail if there is a shortage of resources on the system, with
5295 ENOMEM, for example. A temporary error is now given on failure.
5297 29. Contrary to the C standard, it seems that in some environments, the
5298 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5299 program. Exim now does this explicitly; it affects the formatting of
5300 timestamps using strftime().
5302 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5303 complaints. I've now initialized all the variables, to avoid this.
5305 32. Header lines added by a system filter were not being "seen" during
5306 transport-time rewrites.
5308 33. The info_callback() function passed to OpenSSL is set up with type void
5309 (*)(SSL *, int, int), as described somewhere. However, when calling the
5310 function (actually a macro) that sets it up, the type void(*)() is
5311 expected. I've put in a cast to prevent warnings from picky compilers.
5313 34. If a DNS black list lookup found a CNAME record, but there were no A
5314 records associated with the domain it pointed at, Exim crashed.
5316 35. If a DNS black list lookup returned more than one A record, Exim ignored
5317 all but the first. It now scans all returned addresses if a particular IP
5318 value is being sought. In this situation, the contents of the
5319 $dnslist_value variable are a list of all the addresses, separated by a
5322 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5323 to accept a host name and all its aliases if the forward lookup for any of
5324 them yielded the IP address of the incoming connection. Now it accepts only
5325 those names whose forward lookup yields the correct IP address. Any other
5326 names are discarded. This closes a loophole whereby a rogue DNS
5327 administrator could create reverse DNS records to break through a
5328 wildcarded host restriction in an ACL.
5330 37. If a user filter or a system filter that ran in a subprocess used any of
5331 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5332 the wrong values were passed to the pipe command ($thisaddress had the
5333 value of $0, $0 had the value of $1, etc). This bug was introduced by
5334 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5336 38. Improved the line breaking for long SMTP error messages from ACLs.
5337 Previously, if there was no break point between 40 and 75 characters, Exim
5338 left the rest of the message alone. Two changes have been made: (a) I've
5339 reduced the minimum length to 35 characters; (b) if it can't find a break
5340 point between 35 and 75 characters, it looks ahead and uses the first one
5341 that it finds. This may give the occasional overlong line, but at least the
5342 remaining text gets split now.
5344 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5345 file descriptors might be low, and that setting 1000 would always raise it.
5346 It turns out that in some environments, the limit is already over 1000 and
5347 that lowering it causes trouble. So now Exim takes care not to decrease it.
5349 40. When delivering a message, the value of $return_path is set to $sender_
5350 address at the start of routing (routers may change the value). By an
5351 oversight, this default was not being set up when an address was tested by
5352 -bt or -bv, which affected the outcome if any router or filter referred to
5355 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5356 log only when "message" or "log_message" are set. However, if one of the
5357 conditions was an address verification, or a call to a nested ACL, the
5358 messages generated by the underlying test were being passed through. This
5359 no longer happens. The underlying message is available in $acl_verify_
5360 message for both "message" and "log_message" expansions, so it can be
5361 passed through if needed.
5363 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5364 new expansion $bh_ to give the encoded byte string without charset
5365 translation. Translation happens only if iconv() is available; HAVE_ICONV
5366 indicates this at build time. HEADERS_CHARSET gives the charset to
5367 translate to; headers_charset can change it in the configuration, and
5368 "headers charset" can change it in an individual filter file.
5370 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5371 that creates RFC 2047 encoded "words" labels them as that charset instead
5372 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5373 expansion operator; (ii) when Exim creates a From: line for a local
5374 message; (iii) when a header line is rewritten to include a "phrase" part.
5376 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5377 buggy, causing it to skip the first lines of messages whose message ID
5378 ended in 'D'. This would not have bitten before Exim release 4.14, because
5379 message IDs were unlikely to end in 'D' before then. The effect was to have
5380 incorrect size information for certain domains.
5382 45. #include "config.h" was missing at the start of the crypt16.c module. This
5383 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5386 46. If there was a timeout during a "random" callout check, Exim treated it as
5387 a failure of the random address, and carried on sending RSET and the real
5388 address. If the delay was just some slowness somewhere, the response to the
5389 original RCPT would be taken as a response to RSET and so on, causing
5390 mayhem of various kinds.
5392 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5393 when I implemented it. It didn't allow for the fact that some option values
5394 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5395 the right test for positive values.
5397 48. Domain names in DNS records are case-independent. Exim always looks them up
5398 in lower case. Some resolvers return domain names in exactly the case they
5399 appear in the zone file, that is, they may contain uppercase letters. Not
5400 all resolvers do this - some return always lower case. Exim was treating a
5401 change of case by a resolver as a change of domain, similar to a widening
5402 of a domain abbreviation. This triggered its re-routing code and so it was
5403 trying to route what was effectively the same domain again. This normally
5404 caused routing to fail (because the router wouldn't handle the domain
5405 twice). Now Exim checks for this case specially, and just changes the
5406 casing of the domain that it ultimately uses when it transmits the message
5409 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5412 50. If a filter generated a file delivery with a non-absolute name (possible if
5413 no home directory exists for the router), the forbid_file option was not
5416 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5417 the existing equality matching.
5419 52. Exim was using ints instead of ino_t variables in some places where it was
5420 dealing with inode numbers.
5422 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5423 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5424 called TMPDIR, and if it finds it is different, it changes its value.
5426 54. The smtp_printf() function is now made available to local_scan() so
5427 additional output lines can be written before returning. There is also an
5428 smtp_fflush() function to enable the detection of a dropped connection.
5429 The variables smtp_input and smtp_batched_input are exported to
5432 55. Changed the default runtime configuration: the message "Unknown user"
5433 has been removed from the ACL, and instead placed on the localuser router,
5434 using the cannot_route_message feature. This means that any verification
5435 failures that generate their own messages won't get overridden. Similarly,
5436 the "Unrouteable address" message that was in the ACL for unverifiable
5437 relay addresses has also been removed.
5439 56. Added hosts_avoid_esmtp to the smtp transport.
5441 57. The exicyclog script was not checking for the esoteric option
5442 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5443 will work only if exicyclog is run under the appropriate euid.
5445 58. Following a discussion on the list, the rules by which Exim recognises line
5446 endings on incoming messages have been changed. The -dropcr and drop_cr
5447 options are now no-ops, retained only for backwards compatibility. The
5448 following line terminators are recognized: LF CRLF CR. However, special
5449 processing applies to CR:
5451 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5452 nor a local message in the state where . is a terminator.
5454 (ii) If a bare CR is encountered in a header line, an extra space is added
5455 after the line terminator so as not to end the header. The reasoning
5456 behind this is that bare CRs in header lines are most likely either
5457 to be mistakes, or people trying to play silly games.
5459 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5460 was being incorrectly given as 18 bytes larger than it should have been.
5461 This is a VOB (very old bug).
5463 60. This may never have affected anything current, but just in case it has:
5464 When the local host is found other than at the start of a list of hosts,
5465 the local host, those with the same MX, and any that follow, are discarded.
5466 When the list in question was part of a longer list of hosts, the following
5467 hosts (not currently being processed) were also being discarded. This no
5468 longer happens. I'm not sure if this situation could ever has previously
5471 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5474 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5475 header. This is something that is recommended in a new Internet Draft, and
5476 is something that is documented as being done by Sendmail. There are two
5477 possible values. For messages generated by the autoreply transport, Exim
5480 Auto-Submitted: auto-replied
5482 whereas for all other generated messages (e.g. bounces) it adds
5484 Auto-Submitted: auto-generated
5486 63. The "personal" condition in filters now includes a test for the
5487 Auto-Submitted: header. If it contains the string "auto-" the message it
5488 not considered personal.
5490 64. Added rcpt_include_affixes as a generic transport option.
5492 65. Added queue_only_override (default true).
5494 66. Added the syslog_duplication option.
5496 67. If what should have been the first header line of a message consisted of
5497 a space followed by a colon, Exim was mis-interpreting it as a header line.
5498 It isn't of course - it is syntactically invalid and should therefore be
5499 treated as the start of the message body. The misbehaviour could have
5500 caused a number of strange effects, including loss of data in subsequent
5501 header lines, and spool format errors.
5503 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5504 client host had authenticated. This control can now be exercised by an ACL
5505 for more flexibility.
5507 69. By default, callouts do not happen when testing with -bh. There is now a
5508 variant, -bhc, which does actually run the callout code, including
5509 consulting and updating the callout cache.
5511 70. Added support for saslauthd authentication, courtesy of Alexander
5514 71. If statvfs() failed on the spool or log directories while checking their
5515 size for availability, Exim confusingly gave the error "space shortage".
5516 Furthermore, in debugging mode it crashed with a floating point exception.
5517 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5518 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5519 serious problem, Exim now writes to the main and panic logs when this
5520 happens, with details of the failure. It then refuses to accept the
5521 incoming message, giving the message "spool directory problem" or "log
5522 directory problem" with a 421 code for SMTP messages.
5524 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5525 0, 1, and 2 exist, because some OS complain for execs without them (see
5526 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5527 descriptors. However, the code omitted to check that the open succeeded,
5528 causing mysterious errors if for some reason the permissions on /dev/null
5529 got screwed. Now Exim writes a message to the main and panic logs, and
5530 bombs out if it can't open /dev/null.
5532 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5533 interact so that it is all more flexible. It is supposed to remain
5534 backwards compatible. Also added extra_local_interfaces.
5536 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5537 to bomb out with an assertion failure - to the client this appears as a
5538 connection drop. This problem occurs in the part of the code that was taken
5539 from the Samba project. Fortunately, the assertion is in a very simple
5540 function, so I have fixed this by reproducing the function inline in the
5541 one place where it is called, and arranging for authentication to fail
5542 instead of killing the process with assert().
5544 75. The SPA client code was not working when the server requested OEM rather
5545 than Unicode encoding.
5547 76. Added code to make require_files with a specific uid setting more usable in
5548 the case where statting the file as root fails - usually a non-root-mounted
5549 NFS file system. When this happens and the failure is EACCES, Exim now
5550 forks a subprocess and does the per-uid checking as the relevant uid.
5552 77. Added process_log_path.
5554 78. If log_file_path was not explicitly set, a setting of check_log_space or
5555 check_log_inodes was ignored.
5557 79. If a space check for the spool or log partitions fails, the incident is now
5558 logged. Of course, in the latter case the data may get lost...
5560 80. Added the %p formatting code to string_format() so that it can be used to
5561 print addresses in debug_print(). Adjusted all the address printing in the
5562 debugging in store.c to use %p rather than %d.
5564 81. There was a concern that a line of code in smtp_in.c could overflow a
5565 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5566 initially expressed, the concern was not well-founded, because trailing
5567 spaces are removed early. However, if the trailing spaces were followed by
5568 a NULL, they did not get removed, so the overflow was possible. Two fixes
5571 (a) I re-wrote the offending code in a cleaner fashion.
5572 (b) If an incoming SMTP command contains a NULL character, it is rejected
5575 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5576 runs initgroups(), so that if the Exim user is in any additional groups,
5577 they will be used during message reception.
5583 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob