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.
102 JH/02 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
103 This may cause build issues on older platforms.
105 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
106 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
107 gnutls_require_mac & gnutls_require_protocols (no longer supported).
108 Added SNI support via GnuTLS too.
110 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
116 PP/01 Solaris build fix for Oracle's LDAP libraries.
117 Bugzilla 1109, patch from Stephen Usher.
119 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
121 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
124 TF/02 Fix a couple more cases where we did not log the error message
125 when unlink() failed. See also change 4.74-TF/03.
127 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
128 lock up or crash if it happened to be inside a call to libc when it
129 got a SIGUSR1 from exiwhat.
131 The SIGUSR1 handler appends the current process status to the process
132 log which is later printed by exiwhat. It used to use the general
133 purpose logging code to do this, but several functions it calls are
134 not safe for signals.
136 The new output code in the SIGUSR1 handler is specific to the process
137 log, and simple enough that it's easy to inspect for signal safety.
138 Removing some special cases also simplifies the general logging code.
139 Removing the spurious timestamps from the process log simplifies
142 TF/04 Improved ratelimit ACL condition.
144 The /noupdate option has been deprecated in favour of /readonly which
145 has clearer semantics. The /leaky, /strict, and /readonly update modes
146 are mutually exclusive. The update mode is no longer included in the
147 database key; it just determines when the database is updated. (This
148 means that when you upgrde Exim will forget old rate measurements.)
150 Exim now checks that the per_* options are used with an update mode that
151 makes sense for the current ACL. For example, when Exim is processing a
152 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
153 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
154 must specify per_mail/readonly. If you omit the update mode it defaults to
155 /leaky where that makes sense (as before) or /readonly where required.
157 The /noupdate option is now undocumented but still supported for
158 backwards compatibility. It is equivalent to /readonly except that in
159 ACLs where /readonly is required you may specify /leaky/noupdate or
160 /strict/noupdate which are treated the same as /readonly.
162 A useful new feature is the /count= option. This is a generalization
163 of the per_byte option, so that you can measure the throughput of other
164 aggregate values. For example, the per_byte option is now equivalent
165 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
167 The per_rcpt option has been generalized using the /count= mechanism
168 (though it's more complicated than the per_byte equivalence). When it is
169 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
170 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
171 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
172 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
173 Note that using per_rcpt with a non-readonly update mode in more than
174 one ACL will cause the recipients to be double-counted. (The per_mail
175 and per_byte options don't have this problem.)
177 The handling of very low rates has changed slightly. If the computed rate
178 is less than the event's count (usually one) then this event is the first
179 after a long gap. In this case the rate is set to the same as this event's
180 count, so that the first message of a spam run is counted properly.
182 The major new feature is a mechanism for counting the rate of unique
183 events. The new per_addr option counts the number of different
184 recipients that someone has sent messages to in the last time period. It
185 behaves like per_rcpt if all the recipient addresses are different, but
186 duplicate recipient addresses do not increase the measured rate. Like
187 the /count= option this is a general mechanism, so the per_addr option
188 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
189 example, measure the rate that a client uses different sender addresses
190 with the options per_mail/unique=$sender_address. There are further
191 details in the main documentation.
193 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
195 TF/06 Removed a few PCRE remnants.
197 TF/07 Automatically extract Exim's version number from tags in the git
198 repository when doing development or release builds.
200 PP/02 Raise smtp_cmd_buffer_size to 16kB.
201 Bugzilla 879. Patch from Paul Fisher.
203 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
204 Heavily based on revision 40f9a89a from Simon Arlott's tree.
207 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
209 PP/05 Variable $av_failed, true if the AV scanner deferred.
210 Bugzilla 1078. Patch from John Horne.
212 PP/06 Stop make process more reliably on build failure.
213 Bugzilla 1087. Patch from Heiko Schlittermann.
215 PP/07 Make maildir_use_size_file an _expandable_ boolean.
216 Bugzilla 1089. Patch from Heiko Schlittermann.
218 PP/08 Handle ${run} returning more data than OS pipe buffer size.
219 Bugzilla 1131. Patch from Holger Weiß.
221 PP/09 Handle IPv6 addresses with SPF.
222 Bugzilla 860. Patch from Wolfgang Breyha.
224 PP/10 GnuTLS: support TLS 1.2 & 1.1.
226 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
229 PP/11 match_* no longer expand right-hand-side by default.
230 New compile-time build option, EXPAND_LISTMATCH_RHS.
231 New expansion conditions, "inlist", "inlisti".
233 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
235 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
237 PP/14 fix log_write() format string regression from TF/03.
238 Bugzilla 1152. Patch from Dmitry Isaikin.
244 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
246 PP/02 Harmonised TLS library version reporting; only show if debugging.
247 Layout now matches that introduced for other libraries in 4.74 PP/03.
249 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
251 PP/04 New "dns_use_edns0" global option.
253 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
256 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
257 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
259 TK/01 Updated PolarSSL code to 0.14.2.
260 Bugzilla 1097. Patch from Andreas Metzler.
262 PP/07 Catch divide-by-zero in ${eval:...}.
265 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
268 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
269 format-string attack -- SECURITY: remote arbitrary code execution.
271 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
272 time unintentionally subject to list matching rules, letting the header
273 cause arbitrary Exim lookups (of items which can occur in lists, *not*
274 arbitrary string expansion). This allowed for information disclosure.
276 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
277 INT_MIN/-1 -- value coerced to INT_MAX.
283 NM/01 Workround for PCRE version dependency in version reporting
286 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
287 This fixes portability to compilers other than gcc, notably
288 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
290 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
291 makefiles for portability to HP-UX and POSIX correctness.
293 PP/01 Permit LOOKUP_foo enabling on the make command-line.
294 Also via indented variable definition in the Makefile.
295 (Debugging by Oliver Heesakkers).
297 PP/02 Restore caching of spamd results with expanded spamd_address.
298 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
300 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
301 Improves build reliability. Fix from: Frank Elsner
303 NM/02 Fix wide character breakage in the rfc2047 coding
304 Fixes bug 1064. Patch from Andrey N. Oktyabrski
306 NM/03 Allow underscore in dnslist lookups
307 Fixes bug 1026. Patch from Graeme Fowler
309 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
310 Code patches from Adam Ciarcinski of NetBSD.
312 NM/04 Fixed exiqgrep to cope with mailq missing size issue
315 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
316 is logged, to avoid truncation. Patch from John Horne.
318 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
319 Patch from Jakob Hirsch.
321 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
322 SQL string expansion failure details.
323 Patch from Andrey Oktyabrski.
325 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
326 Patch from Simon Arlott.
328 PP/09 New lookups functionality failed to compile on old gcc which rejects
329 extern declarations in function scope.
330 Patch from Oliver Fleischmann
332 PP/10 Use sig_atomic_t for flags set from signal handlers.
333 Check getgroups() return and improve debugging.
334 Fixed developed for diagnosis in bug 927 (which turned out to be
337 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
338 Patch from Mark Zealey.
340 PP/12 Bugzilla 1056: Improved spamd server selection.
341 Patch from Mark Zealey.
343 PP/13 Bugzilla 1086: Deal with maildir quota file races.
344 Based on patch from Heiko Schlittermann.
346 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
347 Patch from Uwe Doering, sign-off by Michael Haardt.
349 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
350 variable declaration deep within a block. Bug and patch from
353 PP/15 lookups-Makefile IRIX compatibilty coercion.
355 PP/16 Make DISABLE_DKIM build knob functional.
357 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
358 Patch by Simon Arlott
360 TF/03 Fix valgrind.h portability to C89 compilers that do not support
361 variable argument macros. Our copy now differs from upstream.
367 TF/01 Failure to get a lock on a hints database can have serious
368 consequences so log it to the panic log.
370 TF/02 Log LMTP confirmation messages in the same way as SMTP,
371 controlled using the smtp_confirmation log selector.
373 TF/03 Include the error message when we fail to unlink a spool file.
375 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
376 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
377 for maintaining out-of-tree patches for some time.
379 PP/01 Bugzilla 139: Documentation and portability issues.
380 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
381 Handle per-OS dynamic-module compilation flags.
383 PP/02 Let /dev/null have normal permissions.
384 The 4.73 fixes were a little too stringent and complained about the
385 permissions on /dev/null. Exempt it from some checks.
386 Reported by Andreas M. Kirchwitz.
388 PP/03 Report version information for many libraries, including
389 Exim version information for dynamically loaded libraries. Created
390 version.h, now support a version extension string for distributors
391 who patch heavily. Dynamic module ABI change.
393 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
394 privilege escalation vulnerability whereby the Exim run-time user
395 can cause root to append content of the attacker's choosing to
398 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
401 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
402 If dropping privileges for untrusted macros, we disabled normal logging
403 on the basis that it would fail; for the Exim run-time user, this is not
404 the case, and it resulted in successful deliveries going unlogged.
405 Fixed. Reported by Andreas Metzler.
411 PP/01 Date: & Message-Id: revert to normally being appended to a message,
412 only prepend for the Resent-* case. Fixes regression introduced in
413 Exim 4.70 by NM/22 for Bugzilla 607.
415 PP/02 Include check_rfc2047_length in configure.default because we're seeing
416 increasing numbers of administrators be bitten by this.
418 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
420 PP/03 Bugzilla 994: added openssl_options main configuration option.
422 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
424 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
426 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
428 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
429 without a peer certificate, leading to a segfault because of an
430 assumption that peers always have certificates. Be a little more
431 paranoid. Problem reported by Martin Tscholak.
433 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
434 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
435 NB: ClamAV planning to remove STREAM in "middle of 2010".
436 CL also introduces -bmalware, various -d+acl logging additions and
437 more caution in buffer sizes.
439 PP/09 Implemented reverse_ip expansion operator.
441 PP/10 Bugzilla 937: provide a "debug" ACL control.
443 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
445 PP/12 Bugzilla 973: Implement --version.
447 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
449 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
451 PP/15 Bugzilla 816: support multiple condition rules on Routers.
453 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
454 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
455 ignore trailing whitespace.
457 JJ/02 prevent non-panic DKIM error from being sent to paniclog
459 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
462 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
463 Notification from Dr Andrew Aitchison.
465 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
466 ExtendedDetectionInfo response format.
467 Notification from John Horne.
469 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
472 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
473 XSL and documented dependency on system catalogs, with examples of how
476 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
479 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
480 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
481 configuration file which is writeable by the Exim user or group.
483 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
484 of configuration files to cover files specified with the -C option if
485 they are going to be used with root privileges, not just the default
488 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
489 option (effectively making it always true).
491 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
492 files to be used while preserving root privileges.
494 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
495 that rogue child processes cannot use them.
497 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
498 run-time user, instead of root.
500 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
501 Exim run-time user without dropping privileges.
503 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
504 result string, instead of calling string_vformat() twice with the same
507 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
508 for other users. Others should always drop root privileges if they use
509 -C on the command line, even for a whitelisted configure file.
511 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
513 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
519 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
520 $data_path, and $header_path variables; fixed documentation bugs and
523 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
524 exipick to access non-standard spools, including the "frozen" queue
527 NM/01 Bugzilla 965: Support mysql stored procedures.
528 Patch from Alain Williams
530 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
532 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
533 Patch from Andreas Metzler
535 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
536 Patch from Kirill Miazine
538 NM/05 Bugzilla 671: Added umask to procmail example.
540 JJ/03 installed exipick 20100323.0, fixing doc bug
542 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
543 directory. Notification and patch from Dan Rosenberg.
545 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
547 TK/02 Improve log output when DKIM signing operation fails.
549 MH/01 Treat the transport option dkim_domain as a colon separated
550 list, not as a single string, and sign the message with each element,
551 omitting multiple occurences of the same signer.
553 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
554 Bugzilla 985, 986. Patch by Simon Arlott
556 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
557 Patch by Simon Arlott
559 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
560 MBX locking. Notification from Dan Rosenberg.
566 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
568 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
570 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
572 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
574 NM/04 Bugzilla 915: Flag broken perl installation during build.
580 TK/01 Added patch by Johannes Berg that expands the main option
581 "spamd_address" if it starts with a dollar sign.
583 TK/02 Write list of recipients to X-Envelope-Sender header when building
584 the mbox-format spool file for content scanning (suggested by Jakob
587 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
588 (http://www.dcc-servers.net/) support via dccifd. Activated by
589 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
591 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
592 by Mark Daniel Reidel <mr@df.eu>.
594 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
595 When building exim an external PCRE library is now needed -
596 PCRE is a system library on the majority of modern systems.
597 See entry on PCRE_LIBS in EDITME file.
599 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
600 conversation. Added nologin parameter to request.
601 Patch contributed by Kirill Miazine.
603 TF/01 Do not log submission mode rewrites if they do not change the address.
605 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
607 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
608 log files in place. Contributed by Roberto Lima.
610 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
612 TF/03 Bugzilla 615: When checking the local_parts router precondition
613 after a local_part_suffix or local_part_prefix option, Exim now
614 does not use the address's named list lookup cache, since this
615 contains cached lookups for the whole local part.
617 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
618 Robert Millan. Documentation is in experimental-spec.txt.
620 TF/04 Bugzilla 668: Fix parallel build (make -j).
622 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
624 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
625 Patch provided by Jan Srzednicki.
627 TF/05 Leading white space used to be stripped from $spam_report which
628 wrecked the formatting. Now it is preserved.
630 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
631 that they are available at delivery time.
633 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
635 TF/08 TLS error reporting now respects the incoming_interface and
636 incoming_port log selectors.
638 TF/09 Produce a more useful error message if an SMTP transport's hosts
639 setting expands to an empty string.
641 NM/06 Bugzilla 744: EXPN did not work under TLS.
642 Patch provided by Phil Pennock.
644 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
645 Patch provided by Richard Godbee.
647 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
648 acl_smtp_notquit, added index entry.
650 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
651 Patch provided by Eugene Bujak.
653 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
654 accept(). Patch provided by Maxim Dounin.
656 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
657 Patch provided by Phil Pennock.
659 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
661 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
662 Patch provided by Brad "anomie" Jorsch.
664 NM/14 Bugzilla 622: Added timeout setting to transport filter.
665 Patch provided by Dean Brooks.
667 TK/05 Add native DKIM support (does not depend on external libraries).
669 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
670 Patch provided by Graeme Fowler.
672 NM/16 Bugzilla 851: Documentation example syntax fix.
674 NM/17 Changed NOTICE file to remove references to embedded PCRE.
676 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
679 NM/19 Bugzilla 745: TLS version reporting.
680 Patch provided by Phil Pennock.
682 NM/20 Bugzilla 167: bool: condition support.
683 Patch provided by Phil Pennock.
685 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
686 clients. Patch provided by Phil Pennock.
688 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
689 Patch provided by Brad "anomie" Jorsch.
691 NM/23 Bugzilla 687: Fix misparses in eximstats.
692 Patch provided by Heiko Schlittermann.
694 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
695 Patch provided by Heiko Schlittermann.
697 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
698 plus update to original patch.
700 NM/26 Bugzilla 799: Documentation correction for ratelimit.
702 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
703 Patch provided by David Brownlee.
705 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
707 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
709 NM/30 Bugzilla 888: TLS documentation bugfixes.
711 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
713 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
714 Unlike the original bugzilla I have changed all shell scripts in src tree.
716 NM/33 Bugzilla 898: Transport filter timeout fix.
717 Patch by Todd Rinaldo.
719 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
720 Patch by Serge Demonchaux.
722 NM/35 Bugzilla 39: Base64 decode bug fixes.
723 Patch by Jakob Hirsch.
725 NM/36 Bugzilla 909: Correct connect() call in dcc code.
727 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
729 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
731 NM/39 Bugzilla 911: Fixed MakeLinks build script.
737 TK/01 Add preliminary DKIM support. Currently requires a forked version of
738 ALT-N's libdkim that I have put here:
739 http://duncanthrax.net/exim-experimental/
741 Note to Michael Haardt: I had to rename some vars in sieve.c. They
742 were called 'true' and it seems that C99 defines that as a reserved
743 keyword to be used with 'bool' variable types. That means you could
744 not include C99-style headers which use bools without triggering
745 build errors in sieve.c.
747 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
748 as mailq or other aliases. Changed the --help handling significantly
749 to do whats expected. exim_usage() emits usage/help information.
751 SC/01 Added the -bylocaldomain option to eximstats.
753 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
755 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
757 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
763 PH/01 Another patch from the Sieve maintainer.
765 PH/02 When an IPv6 address is converted to a string for single-key lookup
766 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
767 dots are used instead of colons so that keys in lsearch files need not
768 contain colons. This was done some time before quoting was made available
769 in lsearch files. However, iplsearch files do require colons in IPv6 keys
770 (notated using the quote facility) so as to distinguish them from IPv4
771 keys. This meant that lookups for IP addresses in host lists did not work
772 for iplsearch lookups.
774 This has been fixed by arranging for IPv6 addresses to be expressed with
775 colons if the lookup type is iplsearch. This is not incompatible, because
776 previously such lookups could never work.
778 The situation is now rather anomolous, since one *can* have colons in
779 ordinary lsearch keys. However, making the change in all cases is
780 incompatible and would probably break a number of configurations.
782 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
785 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
786 conversion specification without a maximum field width, thereby enabling
787 a rogue spamd server to cause a buffer overflow. While nobody in their
788 right mind would setup Exim to query an untrusted spamd server, an
789 attacker that gains access to a server running spamd could potentially
790 exploit this vulnerability to run arbitrary code as the Exim user.
792 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
793 $primary_hostname instead of what libspf2 thinks the hosts name is.
795 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
796 a directory entry by the name of the lookup key. Previously, if a
797 symlink pointed to a non-existing file or a file in a directory that
798 Exim lacked permissions to read, a lookup for a key matching that
799 symlink would fail. Now it is enough that a matching directory entry
800 exists, symlink or not. (Bugzilla 503.)
802 PH/03 The body_linecount and body_zerocount variables are now exported in the
805 PH/04 Added the $dnslist_matched variable.
807 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
808 This means they are set thereafter only if the connection becomes
811 PH/06 Added the client_condition to authenticators so that some can be skipped
812 by clients under certain conditions.
814 PH/07 The error message for a badly-placed control=no_multiline_responses left
815 "_responses" off the end of the name.
817 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
819 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
820 (without spaces) instead of just copying the configuration text.
822 PH/10 Added the /noupdate option to the ratelimit ACL condition.
824 PH/11 Added $max_received_linelength.
826 PH/12 Added +ignore_defer and +include_defer to host lists.
828 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
829 way in which PCRE > 7.0 is built.
831 PH/14 Implemented queue_only_load_latch.
833 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
834 MAIL command. The effect was to mangle the value on 64-bit systems.
836 PH/16 Another patch from the Sieve maintainer.
838 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
840 PH/18 If a system quota error occurred while trying to create the file for
841 a maildir delivery, the message "Mailbox is full" was not appended to the
842 bounce if the delivery eventually timed out. Change 4.67/27 below applied
843 only to a quota excession during the actual writing of the file.
845 PH/19 It seems that peer DN values may contain newlines (and other non-printing
846 characters?) which causes problems in log lines. The DN values are now
847 passed through string_printing() before being added to log lines.
849 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
850 and InterBase are left for another time.)
852 PH/21 Added message_body_newlines option.
854 PH/22 Guard against possible overflow in moan_check_errorcopy().
856 PH/23 POSIX allows open() to be a macro; guard against that.
858 PH/24 If the recipient of an error message contained an @ in the local part
859 (suitably quoted, of course), incorrect values were put in $domain and
860 $local_part during the evaluation of errors_copy.
866 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
867 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
870 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
871 issue a MAIL command.
873 PH/02 In an ACL statement such as
875 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
877 if a client was not listed at all, or was listed with a value other than
878 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
879 the condition was not true (as it should be), so access was not denied.
880 The bug was that the ! inversion was incorrectly passed on to the second
881 item. This has been fixed.
883 PH/03 Added additional dnslists conditions == and =& which are different from
884 = and & when the dns lookup returns more than one IP address.
886 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
887 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
889 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
890 FSYNC, which compiles an option called disable_fsync that allows for
891 bypassing fsync(). The documentation is heavily laced with warnings.
893 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
895 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
896 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
897 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
898 including adding "make clean"; (3) Added -fPIC when compiling the test
899 dynamically loaded module, to get rid of a warning.
901 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
902 message fails, move_frozen_messages = true and ignore_bounce_errors_after
903 = 0s. The bug is otherwise harmless.
905 PH/07 There was a bug in the dovecot authenticator such that the value of
906 $auth1 could be overwritten, and so not correctly preserved, after a
907 successful authentication. This usually meant that the value preserved by
908 the server_setid option was incorrect.
910 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
912 PH/09 Installed PCRE release 7.0.
914 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
915 run for batched SMTP input. It is now run at the start of every message
916 in the batch. While fixing this I discovered that the process information
917 (output by running exiwhat) was not always getting set for -bs and -bS
918 input. This is fixed, and it now also says "batched" for BSMTP.
920 PH/11 Added control=no_pipelining.
922 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
923 patch, slightly modified), and move the expansion of helo_data till after
924 the connection is made in the smtp transport (so it can use these
927 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
929 PH/14 Added log_selector = +pid.
931 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
933 PH/16 Add ${if forany and ${if forall.
935 PH/17 Added dsn_from option to vary the From: line in DSNs.
937 PH/18 Flush SMTP output before performing a callout, unless control =
938 no_callout_flush is set.
940 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
941 was true (the default) a successful delivery failed to delete the retry
942 item, thus causing premature timeout of the address. The bug is now
945 PH/20 Added hosts_avoid_pipelining to the smtp transport.
947 PH/21 Long custom messages for fakedefer and fakereject are now split up
948 into multiline reponses in the same way that messages for "deny" and
949 other ACL rejections are.
951 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
952 with slight modification.
954 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
955 draft, changing the syntax and factoring some duplicate code".
957 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
958 for deliveries of the second and subsequent messages over the same SMTP
961 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
962 ${reduce, with only minor "tidies".
964 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
966 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
967 expansion side effects.
969 PH/27 When a message times out after an over-quota error from an Exim-imposed
970 quota, the bounce message says "mailbox is full". This message was not
971 being given when it was a system quota that was exceeded. It now should
974 MH/03 Made $recipients available in local_scan(). local_scan() already has
975 better access to the recipient list through recipients_list[], but
976 $recipients can be useful in postmaster-provided expansion strings.
978 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
979 in the case of a MAIL command with additional options following the
980 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
981 were accidentally chopped off.
983 PH/29 SMTP synchronization checks are implemented when a command is read -
984 there is a check that no more input is waiting when there shouldn't be
985 any. However, for some commands, a delay in an ACL can mean that it is
986 some time before the response is written. In this time, more input might
987 arrive, invalidly. So now there are extra checks after an ACL has run for
988 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
989 pipelining has not been advertised.
991 PH/30 MH's patch to allow iscntrl() characters to be list separators.
993 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
994 returned in the SMTP response when smtp_return_error_details was false.
997 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
998 instead of the C I/O that was originally supplied, because problems were
1001 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1002 Exim which did not show up earlier: it was assuming that a call to
1003 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1004 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1005 was a macro that became an assignment, so it seemed to work. This has
1006 changed to a proper function call with a void return, hence the compile
1007 error. Exim's code has been fixed.
1009 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1012 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1014 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1016 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1017 because it turns out to be too big or there is a timeout) there is no
1018 "Completed" line in the log. When some messages of this type were
1019 selected by exigrep, they were listed as "not completed". Others were
1020 picked up by some special patterns. I have improved the selection
1021 criteria to be more general.
1023 PH/38 The host_find_failed option in the manualroute router can now be set
1024 to "ignore", to completely ignore a host whose IP address cannot be
1025 found. If all hosts are ignored, the behaviour is controlled by the new
1026 host_all_ignored option.
1028 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1029 homing or because of multiple MX records with /mx) generated more than
1030 one IP address, and the following item turned out to be the local host,
1031 all the secondary addresses of the first item were incorrectly removed
1032 from the list, along with the local host and any following hosts (which
1033 is what is supposed to happen).
1035 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1036 whoever called Exim into the -H file. In the case of the daemon it was
1037 behaving confusingly. When first started, it used values for whoever
1038 started the daemon, but after a SIGHUP it used the Exim user (because it
1039 calls itself on a restart). I have changed the code so that it now always
1042 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1043 message are rejected with the same error (e.g. no authentication or bad
1044 sender address), and a DATA command is nevertheless sent (as can happen
1045 with PIPELINING or a stupid MUA), the error message that was given to the
1046 RCPT commands is included in the rejection of the DATA command. This is
1047 intended to be helpful for MUAs that show only the final error to their
1050 PH/42 Another patch from the Sieve maintainer.
1052 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1053 Eximstats - Fixed some broken HTML links and added missing column headers
1055 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1056 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1058 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1060 PH/43 Yet another patch from the Sieve maintainer.
1062 PH/44 I found a way to check for a TCP/IP connection going away before sending
1063 the response to the final '.' that terminates a message, but only in the
1064 case where the client has not sent further data following the '.'
1065 (unfortunately, this is allowed). However, in many cases there won't be
1066 any further data because there won't be any more messages to send. A call
1067 to select() can be used: if it shows that the input is "ready", there is
1068 either input waiting, or the socket has been closed. An attempt to read
1069 the next input character can distinguish the two cases. Previously, Exim
1070 would have sent an OK response which the client would never have see.
1071 This could lead to message repetition. This fix should cure that, at
1072 least in a lot of common cases.
1074 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1075 advertised in response to EHLO.
1081 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1082 fixed by 4.65/MH/01 (is this a record?) are fixed:
1084 (i) An empty string was always treated as zero by the numeric comparison
1085 operators. This behaviour has been restored.
1087 (ii) It is documented that the numeric comparison operators always treat
1088 their arguments as decimal numbers. This was broken in that numbers
1089 starting with 0 were being interpreted as octal.
1091 While fixing these problems I realized that there was another issue that
1092 hadn't been noticed. Values of message_size_limit (both the global option
1093 and the transport option) were treated as octal if they started with 0.
1094 The documentation was vague. These values are now always treated as
1095 decimal, and I will make that clear in the documentation.
1101 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1102 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1105 MH/01 Don't check that the operands of numeric comparison operators are
1106 integers when their expansion is in "skipping" mode (fixes bug
1107 introduced by 4.64-PH/07).
1109 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1110 child addresses, Exim now panics and dies. Previously, because the count
1111 is held in a short int, deliveries were likely to be lost. As such a
1112 large number of recipients for a single message is ridiculous
1113 (performance will be very, very poor), I have chosen to impose a limit
1114 rather than extend the field.
1120 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1121 leftover -K file (the existence of which was triggered by #402).
1122 While we were at it, introduced process PID as part of the -K
1123 filename. This should rule out race conditions when creating
1126 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1127 processing considerably. Previous code took too long for large mails,
1128 triggering a timeout which in turn triggers #401.
1130 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1131 in the DK code in transports.c. sendfile() is not really portable,
1132 hence the _LINUX specificness.
1134 TF/01 In the add_headers option to the mail command in an Exim filter,
1135 there was a bug that Exim would claim a syntax error in any
1136 header after the first one which had an odd number of characters
1139 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1140 callout verification, Exim cached a "reject" for the entire domain. This
1141 is correct for most verifications, but it is not correct for a recipient
1142 verification with use_sender or use_postmaster set, because in that case
1143 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1144 case of MAIL FROM:<> rejection from other early rejections (e.g.
1145 rejection of HELO). When verifying a recipient using a non-null MAIL
1146 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1147 Whatever the result of the callout, the value of the domain cache is
1148 left unchanged (for any other kind of callout, getting as far as trying
1149 RCPT means that the domain itself is ok).
1151 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1154 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1155 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1156 session, the daemon ignores ECONNECT errors and logs others; it now
1157 ignores EPIPE as well.
1159 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1160 (quoted-printable decoding).
1162 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1163 later the small subsequent patch to fix an introduced bug.
1165 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1167 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1169 PH/08 An error is now given if message_size_limit is specified negative.
1171 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1172 to be given (somewhat) arbitrary names.
1174 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1177 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1178 miscellaneous code fixes
1180 PH/10 Added the log_reject_target ACL modifier to specify where to log
1183 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1184 hostname. This is wrong, because it relates to the incoming message (and
1185 probably the interface on which it is arriving) and not to the outgoing
1186 callout (which could be using a different interface). This has been
1187 changed to use the value of the helo_data option from the smtp transport
1188 instead - this is what is used when a message is actually being sent. If
1189 there is no remote transport (possible with a router that sets up host
1190 addresses), $smtp_active_hostname is used.
1192 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1193 tweaks were necessary in order to get it to work (see also 21 below):
1194 (a) The code assumed that strncpy() returns a negative number on buffer
1195 overflow, which isn't the case. Replaced with Exim's string_format()
1197 (b) There were several signed/unsigned issues. I just did the minimum
1198 hacking in of casts. There is scope for a larger refactoring.
1199 (c) The code used strcasecmp() which is not a standard C function.
1200 Replaced with Exim's strcmpic() function.
1201 (d) The code set only $1; it now sets $auth1 as well.
1202 (e) A simple test gave the error "authentication client didn't specify
1203 service in request". It would seem that Dovecot has changed its
1204 interface. Fortunately there's a specification; I followed it and
1205 changed what the client sends and it appears to be working now.
1207 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1210 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1211 address A is aliased to B and C, where B exists and C does not. Without
1212 -v the output is "A verified" because verification stops after a
1213 successful redirection if more than one address is generated. However,
1214 with -v the child addresses are also verified. Exim was outputting "A
1215 failed to verify" and then showing the successful verification for C,
1216 with its parentage. It now outputs "B failed to verify", showing B's
1217 parentage before showing the successful verification of C.
1219 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1220 look up a TXT record in a specific list after matching in a combined
1223 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1224 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1225 they consult the DNS. I had assumed they would set it the way they
1226 wanted; and indeed my experiments on Linux seem to show that in some
1227 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1228 To be on the safe side, however, I have now made the interface to
1229 host_find_byname() similar to host_find_bydns(), with an argument
1230 containing the DNS resolver options. The host_find_byname() function now
1231 sets these options at its start, just as host_find_bydns() does. The smtp
1232 transport options dns_qualify_single and dns_search_parents are passed to
1233 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1234 of host_find_byname() use the default settings of RES_DEFNAMES
1235 (qualify_single) but not RES_DNSRCH (search_parents).
1237 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1238 spool_read_header() do less string testing, by means of a preliminary
1239 switch on the second character of optional "-foo" lines. (This is
1240 overdue, caused by the large number of possibilities that now exist.
1241 Originally there were few.) While I was there, I also converted the
1242 str(n)cmp tests so they don't re-test the leading "-" and the first
1243 character, in the hope this might squeeze out yet more improvement.
1245 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1246 flag allowing group syntax was set by the header_syntax check but not
1247 turned off, possible causing trouble later; (2) The flag was not being
1248 set at all for the header_verify test, causing "group"-style headers to
1249 be rejected. I have now set it in this case, and also caused header_
1250 verify to ignore an empty address taken from a group. While doing this, I
1251 came across some other cases where the code for allowing group syntax
1252 while scanning a header line wasn't quite right (mostly, not resetting
1253 the flag correctly in the right place). These bugs could have caused
1254 trouble for malformed header lines. I hope it is now all correct.
1256 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1257 with the "reply" argument non-NULL. The code, however (which originally
1258 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1259 but it didn't always do it. This confused somebody who was copying the
1260 code for some other use. I have removed all the tests.
1262 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1263 feature that was used to support insecure browsers during the U.S. crypto
1264 embargo. It requires special client support, and Exim is probably the
1265 only MTA that supported it -- and would never use it because real RSA is
1266 always available. This code has been removed, because it had the bad
1267 effect of slowing Exim down by computing (never used) parameters for the
1268 RSA_EXPORT functionality.
1270 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1271 authenticator to fail if there's a tab character in the incoming data
1272 (there should never be unless someone is messing about, as it's supposed
1273 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1274 if the connection is using TLS or if the remote IP is the same as the
1275 local IP, and the "valid-client-cert option" if a client certificate has
1278 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1279 authenticators. This can be used for authorization after authentication
1280 succeeds. (In the case of plaintext, it servers for both authentication
1283 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1284 if any retry times were supplied.
1286 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1287 connection, where there is no client IP address to check. In this
1288 situation, the verify now always succeeds.
1290 PH/25 Applied John Jetmore's -Mset patch.
1292 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1294 PH/27 In a string expansion for a processed (not raw) header when multiple
1295 headers of the same name were present, leading whitespace was being
1296 removed from all of them, but trailing whitespace was being removed only
1297 from the last one. Now trailing whitespace is removed from each header
1298 before concatenation. Completely empty headers in a concatenation (as
1299 before) are ignored.
1301 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1302 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1304 PH/29 [Removed. This was a change that I later backed out, and forgot to
1305 correct the ChangeLog entry (that I had efficiently created) before
1306 committing the later change.]
1308 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1309 address errors (4xx response to RCPT) over the same connection as other
1310 messages routed to the same hosts. Such deliveries are always "forced",
1311 so retry times are not inspected. This resulted in far too many retries
1312 for the affected addresses. The effect occurred only when there were more
1313 hosts than the hosts_max_try setting in the smtp transport when it had
1314 the 4xx errors. Those hosts that it had tried were not added to the list
1315 of hosts for which the message was waiting, so if all were tried, there
1316 was no problem. Two fixes have been applied:
1318 (i) If there are any address or message errors in an SMTP delivery, none
1319 of the hosts (tried or untried) are now added to the list of hosts
1320 for which the message is waiting, so the message should not be a
1321 candidate for sending over the same connection that was used for a
1322 successful delivery of some other message. This seems entirely
1323 reasonable: after all the message is NOT "waiting for some host".
1324 This is so "obvious" that I'm not sure why it wasn't done
1325 previously. Hope I haven't missed anything, but it can't do any
1326 harm, as the worst effect is to miss an optimization.
1328 (ii) If, despite (i), such a delivery is accidentally attempted, the
1329 routing retry time is respected, so at least it doesn't keep
1330 hammering the server.
1332 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1333 in ${readsocket because some servers need this prod.
1335 PH/32 Added some extra debug output when updating a wait-xxx database.
1337 PH/33 The hint "could be header name not terminated by colon", which has been
1338 given for certain expansion errors for a long time, was not being given
1339 for the ${if def:h_colon_omitted{... case.
1341 PH/34 The spec says: "With one important exception, whenever a domain list is
1342 being scanned, $domain contains the subject domain." There was at least
1343 one case where this was not true.
1345 PH/35 The error "getsockname() failed: connection reset by peer" was being
1346 written to the panic log as well as the main log, but it isn't really
1347 panic-worthy as it just means the connection died rather early on. I have
1348 removed the panic log writing for the ECONNRESET error when getsockname()
1351 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1352 runs only) independently of the message's sender address. This meant
1353 that, if the 4xx error was in fact related to the sender, a different
1354 message to the same recipient with a different sender could confuse
1355 things. In particualar, this can happen when sending to a greylisting
1356 server, but other circumstances could also provoke similar problems.
1357 I have changed the default so that the retry time for these errors is now
1358 based a combination of the sender and recipient addresses. This change
1359 can be overridden by setting address_retry_include_sender=false in the
1362 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1363 remote server are returned as part of bounce messages. This was not
1364 happening for LMTP over a pipe (the lmtp transport), but now it is the
1365 same for both kinds of LMTP.
1367 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1368 in header lines that were in fact CNAMEs. This is no longer the case.
1370 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1371 and queue runs started by the daemon processed all messages. This has
1372 been fixed so that -R and -S can now usefully be given with -q<time>.
1374 PH/40 Import PCRE release 6.7 (fixes some bugs).
1376 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1378 PH/42 Give an error if -q is specified more than once.
1380 PH/43 Renamed the variables $interface_address and $interface_port as
1381 $received_ip_address and $received_port, to make it clear that these
1382 values apply to message reception, and not to the outgoing interface when
1383 a message is delivered. (The old names remain recognized, of course.)
1385 PH/44 There was no timeout on the connect() call when using a Unix domain
1386 socket in the ${readsocket expansion. There now is.
1388 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1389 be meaningful with "accept".
1391 SC/01 Eximstats V1.43
1392 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1394 SC/02 Eximstats V1.44
1395 Use a glob alias rather than an array ref in the generated
1396 parser. This improves both readability and performance.
1398 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1399 Collect SpamAssassin and rejection statistics.
1400 Don't display local sender or destination tables unless
1401 there is data to show.
1402 Added average volumes into the top table text output.
1404 SC/04 Eximstats V1.46
1405 Collect data on the number of addresses (recipients)
1406 as well as the number of messages.
1408 SC/05 Eximstats V1.47
1409 Added 'Message too big' to the list of mail rejection
1410 reasons (thanks to Marco Gaiarin).
1412 SC/06 Eximstats V1.48
1413 Mainlog lines which have GMT offsets and are too short to
1414 have a flag are now skipped.
1416 SC/07 Eximstats V1.49 (Alain Williams)
1417 Added the -emptyok flag.
1419 SC/08 Eximstats V1.50
1420 Fixes for obtaining the IP address from reject messages.
1422 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1423 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1424 whitesspace changes from 4.64-PH/27
1426 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1429 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1430 are found, allow negative numbers in numeric criteria)
1432 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1434 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1435 to match changes made in 4.64-PH/43
1437 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1439 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1440 "domain" is actually an IP address, and force a failure. This locks out
1441 those revolvers/nameservers that support "A-for-A" lookups, in
1442 contravention of the specifications.
1444 PH/48 When a host name was looked up from an IP address, and the subsequent
1445 forward lookup of the name timed out, the host name was left in
1446 $sender_host_name, contrary to the specification.
1448 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1449 restricted to single-key lookups, Exim was not diagnosing an error if
1450 * or *@ was used with a query-style lookup.
1452 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1454 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1455 long ago, but noone interested enough thought of it. Let's just say that
1456 the "1.1" means that there are some new functions that weren't there at
1457 some point in the past.
1459 PH/51 Error processing for expansion failure of helo_data from an smtp
1460 transport during callout processing was broken.
1462 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1463 tested/used via the -bh/-bhc/-bs options.
1465 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1466 bug, fixed in subsequent PCRE releases).
1468 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1469 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1471 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1477 SC/01 Use a glob alias rather than an array ref in eximstats generated
1478 parser. This improves both readability and performance.
1480 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1481 Don't display local sender or destination tables in eximstats unless
1482 there is data to show.
1483 Added average volumes into the eximstats top table text output.
1485 SC/03 Collect data on the number of addresses (recipients) as well
1486 as the number of messages in eximstats.
1488 TF/01 Correct an error in the documentation for the redirect router. Exim
1489 does (usually) call initgroups() when daemonizing.
1491 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1492 with consistent privilege compared to when running as a daemon.
1494 TF/03 Note in the spec that $authenticated_id is not set for local
1495 submissions from trusted users.
1497 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1498 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1500 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1501 by adding some example configuration directives to the default
1502 configuration file. A little bit of work is required to uncomment the
1503 directives and define how usernames and passwords are checked, but
1504 there is now a framework to start from.
1506 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1507 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1508 without this. I don't know how relevant this is to other LDAP libraries.
1510 PH/02 Add the verb name to the "unknown ACL verb" error.
1512 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1514 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1516 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1517 directory not expanded when it should be if an expanded home directory
1518 was set for the address (which is overridden by the transport).
1520 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1523 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1524 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1525 because it is too late at that time, and has no effect.
1527 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1528 security issue with \' (bugzilla #107). I could not use the
1529 PQescapeStringConn() function, because it needs a PGconn value as one of
1532 PH/08 When testing addresses using -bt, indicate those final addresses that
1533 are duplicates that would not cause an additional delivery. At least one
1534 person was confused, thinking that -bt output corresponded to deliveries.
1535 (Suppressing duplicates isn't a good idea as you lose the information
1536 about possibly different redirections that led to the duplicates.)
1538 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1539 systems where poll() doesn't work, in particular OS X.
1541 PH/10 Added more information to debugging output for retry time not reached.
1543 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1544 operations in malware.c.
1546 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1549 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1550 syslog_duplication set false, log lines that would normally be written
1551 both the the main log and to the reject log were not written to syslog at
1554 PH/14 In the default configuration, change the use of "message" in ACL warn
1555 statements to "add_header".
1557 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1558 not followed by a command (e.g. "seen endif").
1560 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1561 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1564 PH/17 Added extra conditions to the default value of delay_warning_condition
1568 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1569 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1570 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1573 The Auto-Submitted: and various List- headers are standardised, whereas I
1574 don't think Precedence: ever was.
1576 PH/18 Refactored debugging code in route_finduser() to show more information,
1577 in particular, the error code if getpwnam() issues one.
1579 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1580 This is apparently needed in addition to the PH/07 change above to avoid
1581 any possible encoding problems.
1583 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1584 but not after initializing Perl.
1586 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1587 output them only if debugging. By default they are written stderr,
1588 apparently, which is not desirable.
1590 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1593 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1596 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1598 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1599 authenticated or an ident call has been made. Suppress the default
1600 values for $authenticated_id and $authenticated_sender (but permit -oMai
1601 and -oMas) when testing with -bh.
1603 PH/24 Re-jigged the order of the tests in the default configuration so that the
1604 tests for valid domains and recipients precede the DNS black list and CSA
1605 tests, on the grounds that those ones are more expensive.
1607 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1608 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1609 This bug exists in every version of Exim that I still have, right back to
1612 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1613 However, an attempt to turn on case-sensitivity in a regex key by
1614 including (?-i) didn't work because the subject string was already
1615 lowercased, and the effects were non-intuitive. It turns out that a
1616 one-line patch can be used to allow (?-i) to work as expected.
1622 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1623 other effects) broke the use of negated acl sub-conditions.
1625 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1628 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1629 "Deny" causes Exim to reject the incoming connection with a 554 error.
1630 Unfortunately, if there is a major crisis, such as a disk failure,
1631 tcp-wrappers gives "deny", whereas what one would like would be some
1632 kind of temporary error. A kludge has been added to help with this.
1633 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1634 554 error is used if errno is still zero or contains ENOENT (which occurs
1635 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1638 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1640 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1641 errors. Otherwise a message that provokes a temporary error (when other
1642 messages do not) can cause a whole host to time out.
1644 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1645 addresses were routed directly to files or pipes from a redirect router.
1646 File deliveries just didn't batch; pipe deliveries might have suffered
1649 PH/06 A failure to get a lock for a hints database would erroneously always say
1650 "Failed to get write lock", even when it was really a read lock.
1652 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1653 of 0600. This has been changed to use the value of the lockfile_mode
1654 option (which defaults to 0600).
1656 PH/08 Applied small patch from the Sieve maintainer.
1658 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1659 folder from quota calculations, a direct delivery into this folder messed
1660 up the contents of the maildirsize file. This was because the regex was
1661 used only to exclude .Trash (or whatever) when the size of the mailbox
1662 was calculated. There was no check that a delivery was happening into an
1663 excluded directory. This bug has been fixed by ignoring all quota
1664 processing for deliveries into excluded directories.
1666 PH/10 Added the maildirfolder_create_regex option to appendfile.
1672 PH/01 The code for finding all the local interface addresses on a FreeBSD
1673 system running IPv6 was broken. This may well have applied to all BSD
1674 systems, as well as to others that have similar system calls. The broken
1675 code found IPv4 interfaces correctly, but gave incorrect values for the
1676 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1677 that it would not match correctly against @[] and not recognize the IPv6
1680 PH/02 The ipliteral router was not recognizing addresses of the form user@
1681 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1683 PH/03 Added disable_ipv6.
1685 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1686 decoded form, because it is most often used to construct To: headers
1687 lines in autoreplies, and the decoded form may well be syntactically
1688 invalid. However, $reply_address has leading white space removed, and all
1689 newlines turned into spaces so that the autoreply transport does not
1692 PH/05 If group was specified without a user on a router, and no group or user
1693 was specified on a transport, the group from the router was ignored.
1695 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1696 for visible compile-time settings that can be used to change these
1697 numbers, for those that want even more. Backwards compatibility with old
1698 spool files has been maintained. However, going back to a previous Exim
1699 release will lost any variables that are in spool files.
1701 PH/07 Two small changes when running in the test harness: increase delay when
1702 passing a TCP/IP connection to a new process, in case the original
1703 process has to generate a bounce, and remove special handling of
1704 127.0.0.2 (sic), which is no longer necessary.
1706 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1707 be the same on different OS.
1709 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1712 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1713 whether --show-vars was specified or not
1715 JJ/02 exipick: Added support for new ACL variable spool format introduced
1718 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1719 syntactically invalid From: or Reply-to: line, and a filter used this to
1720 generate an autoreply, and therefore failed to obtain an address for the
1721 autoreply, Exim could try to deliver to a non-existent relative file
1722 name, causing unrelated and misleading errors. What now happens is that
1723 it logs this as a hard delivery error, but does not attempt to create a
1726 PH/11 The exinext utility has a -C option for testing purposes, but although
1727 the given file was scanned by exinext itself; it wasn't being passed on
1728 when Exim was called.
1730 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1731 an end-of-file indication when reading a command response.
1733 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1734 compiled. In many other places in Exim, IPv6 addresses are always
1735 recognized, so I have changed this. It also means that IPv4 domain
1736 literals of the form [IPV4:n.n.n.n] are now always recognized.
1738 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1739 used if the router is not running as root, for example, when verifying at
1740 ACL time, or when using -bh. The debugging output from this situation was
1741 non-existent - all you got was a failure to exec. I have made two
1744 (a) Failures to set uid/gid, the current directory, or a process leader
1745 in a subprocess such as that created by queryprogram now generate
1746 suitable debugging ouput when -d is set.
1748 (b) The queryprogram router detects when it is not running as root,
1749 outputs suitable debugging information if -d is set, and then runs
1750 the subprocess without attempting to change uid/gid.
1752 PH/15 Minor change to Makefile for building test_host (undocumented testing
1755 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1756 additional section of a DNS packet that returns MX or SRV records.
1757 Instead, it always explicitly searches for A/AAAA records. This avoids
1758 major problems that occur when a DNS server includes only records of one
1759 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1760 fixed another bug: if SRV records were looked up and the corresponding
1761 address records were *not* found in the additional section, the port
1762 values from the SRV records were lost.
1764 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1765 using the correct key (the original address) when searching the retry
1766 rules in order to find which one to use for generating the retry hint.
1768 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1769 adding the default one. Similarly, if it contains a Reply-To: header, the
1770 errors_reply_to option, if set, is not used.
1772 PH/19 When calculating a retry time, Exim used to measure the "time since
1773 failure" by looking at the "first failed" field in the retry record. Now
1774 it does not use this if it is later than than the arrival time of the
1775 message. Instead it uses the arrival time. This makes for better
1776 behaviour in cases where some deliveries succeed, thus re-setting the
1777 "first failed" field. An example is a quota failure for a huge message
1778 when small messages continue to be delivered. Without this change, the
1779 "time since failure" will always be short, possible causing more frequent
1780 delivery attempts for the huge message than are intended.
1781 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1783 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1784 $1, $2, $3) because the numerical variables can be reset during some
1785 expansion items (e.g. "match"), thereby losing the authentication data.
1787 PH/21 Make -bV show the size of off_t variables so that the test suite can
1788 decide whether to run tests for quotas > 2G.
1790 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1791 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1792 filecount value is greater than 2G or if a quota value is greater than 2G
1793 on a system where the size of off_t is not greater than 4, a panic error
1796 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1797 never match. The debug and -bh output now contains an explicit error
1798 message indicating a malformed IPv4 address or mask.
1800 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1801 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1802 PH/23 above applies.
1804 PH/25 Do not write to syslog when running in the test harness. The only
1805 occasion when this arises is a failure to open the main or panic logs
1806 (for which there is an explicit test).
1808 PH/26 Added the /no_tell option to "control=freeze".
1810 PH/27 If a host name lookup failed very early in a connection, for example, if
1811 the IP address matched host_lookup and the reverse lookup yielded a name
1812 that did not have a forward lookup, an error message of the form "no IP
1813 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1814 could be logged. Now it outputs the IP address instead of "NULL".
1816 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1817 allows the sender and the authenticated sender to be set when
1818 submitting a message from within Exim. Since child_open_exim() is
1819 documented for local_scan(), the new function should be too.
1821 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1822 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1823 results in an empty string is now treated as unset.
1825 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1827 PH/31 Added sender_verify_fail logging option.
1829 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1830 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1831 tidied the source and removed it altogether.
1833 PH/33 When a queue run was abandoned because the load average was too high, a
1834 log line was always written; now it is written only if the queue_run log
1835 selector is set. In addition, the log line for abandonment now contains
1836 information about the queue run such as the pid. This is always present
1837 in "start" and "stop" lines but was omitted from the "abandon" line.
1839 PH/34 Omit spaces between a header name and the colon in the error message that
1840 is given when verify = headers_syntax fails (if there are lots of them,
1841 the message gets confusing).
1843 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1844 names, as there are now some PTR records that contain slashes. This check
1845 is only to protect against broken name servers that fall over on strange
1846 characters, so the fact that it applies to all lookups doesn't matter.
1848 PH/36 Now that the new test suite is complete, we can remove some of the
1849 special code in Exim that was needed for the old test suite. For example,
1850 sorting DNS records because real resolvers return them in an arbitrary
1851 order. The new test suite's fake resolver always returns records in the
1854 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1855 bounces) except when queue_only is set, to avoid logging races between
1856 the different processes.
1858 PH/38 Panic-die if .include specifies a non-absolute path.
1860 PH/39 A tweak to the "H" retry rule from its user.
1862 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1863 a label. They prevented compilation on older perls.
1865 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1866 a warning to be raised on newish perls.
1868 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1869 on queue. Changes to match documented behaviour of showing count of
1870 messages matching specified criteria.
1872 PH/40 Changed the default ident timeout from 30s to 5s.
1874 PH/41 Added support for the use of login_cap features, on those BSD systems
1875 that have them, for controlling the resources used by pipe deliveries.
1877 PH/42 The content-scanning code uses fopen() to create files in which to put
1878 message data. Previously it was not paying any attention to the mode of
1879 the files. Exim runs with umask(0) because the rest of the code creates
1880 files with open(), and sets the required mode explicitly. Thus, these
1881 files were ending up world-writeable. This was not a big issue, because,
1882 being within the spool directory, they were not world-accessible. I have
1883 created a function called modefopen, which takes an additional mode
1884 argument. It sets umask(777), creates the file, chmods it to the required
1885 mode, then resets the umask. All the relevant calls to fopen() in the
1886 content scanning code have been changed to use this function.
1888 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1889 to 24 hours. This avoids potential overflow problems when processing G
1890 and H retry rules. I suspect nobody ever tinkers with this value.
1892 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1894 PH/45 When the plaintext authenticator is running as a client, the server's
1895 challenges are checked to ensure they are valid base64 strings. By
1896 default, the authentication attempt is cancelled if an invalid string is
1897 received. Setting client_ignore_invalid_base64 true ignores these errors.
1898 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1899 they are received. Thus, the responses can be made to depend on the
1900 challenges. If an invalid string is ignored, an empty string is placed in
1903 PH/46 Messages that are created by the autoreply transport now contains a
1904 References: header, in accordance with RFCs 2822 and 3834.
1906 PH/47 Added authenticated_sender_force to the smtp transport.
1908 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1910 PH/49 Installed latest patch from the Sieve maintainer.
1912 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1913 was also set, the appendfile transport was unnecessarily scanning a
1914 directory of message files (e.g. for maildir delivery) to find the count
1915 of files (along with the size), even though it did not need this
1916 information. It now does the scan only if it needs to find either the
1917 size of the count of files.
1919 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1921 PH/52 Two bugs concerned with error handling when the smtp transport is
1924 (i) Exim was not creating retry information for temporary errors given
1925 for individual recipients after the DATA command when the smtp transport
1926 was used in LMTP mode. This meant that they could be retried too
1927 frequently, and not timed out correctly.
1929 (ii) Exim was setting the flag that allows error details to be returned
1930 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1931 recipients that were returned after the DATA command.
1933 PH/53 This is related to PH/52, but is more general: for any failing address,
1934 when detailed error information was permitted to be returned to the
1935 sender, but the error was temporary, then after the final timeout, only
1936 "retry timeout exceeded" was returned. Now it returns the full error as
1937 well as "retry timeout exceeded".
1939 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1940 do this, and (what is worse) MTAs that accept it.
1942 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1943 will now be deprecated.
1945 PH/56 New os.c-cygwin from the Cygwin maintainer.
1947 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1948 formats (previously only available in exim formats via -bpr, -bpru,
1949 and -bpra. Now also available in native and exiqgrep formats)
1951 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1952 with very large, slow to parse queues
1954 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1956 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1958 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1959 responses to authentication challenges, though it was showing the
1960 challenges; (ii) I've removed the CR characters from the debug output for
1963 PH/58 Allow for the insertion of a newline as well as a space when a string
1964 is turned into more than one encoded-word during RFC 2047 encoding. The
1965 Sieve code now uses this.
1967 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1968 data_4xx, lost_connection, tls_required.
1970 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1971 message was being sent as an SMTP response.
1973 PH/61 Add -l and -k options to exicyclog.
1975 PH/62 When verifying, if an address was redirected to one new address, so that
1976 verification continued, and the new address failed or deferred after
1977 having set something in $address_data, the value of $address_data was not
1978 passed back to the ACL. This was different to the case when no
1979 redirection occurred. The value is now passed back in both cases.
1981 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1982 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1983 use login_cap.h, so on its own it isn't the distinguishing feature. The
1984 new name refers directly to the setclassresources() function.
1986 PH/65 Added configuration files for NetBSD3.
1988 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1990 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1991 is preferred over IPv4.
1993 PH/68 The bounce_return_message and bounce_return_body options were not being
1994 honoured for bounces generated during the reception of non-SMTP messages.
1995 In particular, this applied to messages rejected by the ACL. This bug has
1996 been fixed. However, if bounce_return_message is true and bounce_return_
1997 body is false, the headers that are returned for a non-SMTP message
1998 include only those that have been read before the error was detected.
1999 (In the case of an ACL rejection, they have all been read.)
2001 PH/69 The HTML version of the specification is now built in a directory called
2002 spec_html instead of spec.html, because the latter looks like a path with
2003 a MIME-type, and this confuses some software.
2005 PH/70 Catch two compiler warnings in sieve.c.
2007 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2008 function verify_get_ident() calls ip_connect() to connect a socket, but
2009 if the "connect()" function timed out, ip_connect() used to close the
2010 socket. However, verify_get_ident() also closes the socket later, and in
2011 between Exim writes to the log, which may get opened at this point. When
2012 the socket was closed in ip_connect(), the log could get the same file
2013 descriptor number as the socket. This naturally causes chaos. The fix is
2014 not to close the socket in ip_connect(); the socket should be closed by
2015 the function that creates it. There was only one place in the code where
2016 this was missing, in the iplookup router, which I don't think anybody now
2017 uses, but I've fixed it anyway.
2019 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2020 well as to direct DNS lookups. Otherwise the handling of names in host
2021 lists is inconsistent and therefore confusing.
2027 PH/01 Two changes to the default runtime configuration:
2029 (1) Move the checks for relay_from_hosts and authenticated clients from
2030 after to before the (commented out) DNS black list checks.
2032 (2) Add control=submission to the relay_from_hosts and authenticated
2033 clients checks, on the grounds that messages accepted by these
2034 statements are most likely to be submissions.
2036 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2038 (1) Generate an error if the third argument for the ${prvs expansion is
2041 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2044 (3) Reset the variables that are obtained from the first argument of
2045 ${prvscheck and used in the second argument before leaving the code,
2046 because their memory is reclaimed, so using them afterwards may do
2049 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2050 one (it's much easier than Tom thought :-).
2052 (5) Because of (4), we can now allow for the use of $prvscheck_result
2053 inside the third argument.
2055 PH/03 For some reason, the default setting of PATH when running a command from
2056 a pipe transport was just "/usr/bin". I have changed it to
2059 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2060 anything to be listed in the output from -bV.
2062 PH/05 When a filter generated an autoreply, the entire To: header line was
2063 quoted in the delivery log line, like this:
2065 => >A.N.Other <ano@some.domain> <original@ddress> ...
2067 This has been changed so that it extracts the operative address. There
2068 may be more than one such address. If so, they are comma-separated, like
2071 => >ano@some.domain,ona@other.domain <original@ddress> ...
2073 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2074 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2075 not being looked up in the rDNS to get a host name, Exim was showing the
2076 IP address twice in Received: lines, even though the IP addresses were
2077 identical. For example:
2079 Received: from [1.2.3.4] (helo=[1.2.3.4])
2081 However, if the real host name was known, it was omitting the HELO data
2082 if it matched the actual IP address. This has been tidied up so that it
2083 doesn't show the same IP address twice.
2085 PH/07 When both +timestamp and +memory debugging was on, the value given by
2086 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2087 called by the debug printing, thereby overwriting the timestamp buffer.
2088 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2090 PH/08 When the original message was included in an autoreply transport, it
2091 always said "this is a copy of the message, including all the headers",
2092 even if body_only or headers_only was set. It now gives an appropriate
2095 PH/09 Applied a patch from the Sieve maintainer which:
2097 o fixes some comments
2098 o adds the (disabled) notify extension core
2099 o adds some debug output for the result of if/elsif tests
2100 o points to the current vacation draft in the documentation
2101 and documents the missing references header update
2105 o fixes a bug in processing the envelope test (when testing
2106 multiple envelope elements, the last element determinted the
2109 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2110 Electronic Mail") by including:
2112 Auto-submitted: auto-generated
2114 in the messages that it generates (bounce messages and others, such as
2115 warnings). In the case of bounce messages for non-SMTP mesages, there was
2116 also a typo: it was using "Auto_submitted" (underscore instead of
2117 hyphen). Since every message generated by Exim is necessarily in response
2118 to another message, thes have all been changed to:
2120 Auto-Submitted: auto-replied
2122 in accordance with these statements in the RFC:
2124 The auto-replied keyword:
2126 - SHOULD be used on messages sent in direct response to another
2127 message by an automatic process,
2129 - MUST NOT be used on manually-generated messages,
2131 - MAY be used on Delivery Status Notifications (DSNs) and Message
2132 Disposition Notifications (MDNs),
2134 - MUST NOT be used on messages generated by automatic or periodic
2135 processes, except for messages which are automatic responses to
2138 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2139 to the default Received: header definition.
2141 PH/12 Added log selector acl_warn_skipped (default on).
2143 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2144 variables because (a) they are in the wrong storage pool and (b) even if
2145 they were copied, it wouldn't work properly because of the caching.
2147 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2148 checking when decoding. Apparently there are clients that generate
2149 overlong encoded strings. Why am I not surprised?
2151 PH/15 If the first argument of "${if match_address" was not empty, but did not
2152 contain an "@" character, Exim crashed. Now it writes a panic log message
2153 and treats the condition as false.
2155 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2157 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2158 extension "envelope-auth". The code is finished and in agreement with
2159 other implementations, but there is no documentation so far and in fact,
2160 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2161 not changing the active code.
2163 Print executed "if" and "elsif" statements when debugging is used. This
2164 helps a great deal to understand what a filter does.
2166 Document more things not specified clearly in RFC3028. I had all this
2167 sorted out, when out of a sudden new issues came to my mind. Oops."
2169 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2172 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2173 canonical form (as documented). However, after a host name lookup from
2174 the IP address, check_host() was doing a simple string comparison with
2175 addresses acquired from the DNS when checking that the found name did
2176 have the original IP as one of its addresses. Since any found IPv6
2177 addresses are likely to be in abbreviated form, the comparison could
2178 fail. Luckily, there already exists a function for doing the comparison
2179 by converting both addresses to binary, so now that is used instead of
2180 the text comparison.
2182 PH/20 There was another similar case to PH/19, when a complete host name was
2183 given in a host list; looking up its IP address could give an abbreviated
2184 form, whereas the current host's name might or might not be abbreviated.
2185 The same fix has been applied.
2191 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2192 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2195 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2196 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2198 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2200 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2201 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2202 Also, for "Auto-Submitted", treat anything other than "no" as
2203 non-personal, in accordance with RFC 3834. (Previously it treated
2204 anything starting "auto-" as non-personal.)
2206 TF/01 The control=submission/name=... option had a problem with syntax
2207 errors if the name included a slash character. The /name= option
2208 now slurps the rest of the string, so it can include any characters
2209 but it must come last in the list of options (after /sender_retain
2212 PH/05 Some modifications to the interface to the fake nameserver for the new
2220 TK/01 Added the "success_on_redirect" address verification option. See
2221 NewStuff for rationale and an example.
2223 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2225 PH/02 Patch to exigrep to allow it to work on syslog lines.
2227 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2228 fread() to skip over the body file's header line, because in Cygwin the
2229 header line is locked and is inaccessible.
2231 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2232 co-exist for some time) to make it clear that it is the Exim ID that is
2233 referenced, not the Message-ID: header line.
2235 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2236 string_format() function, because snprintf() does not exist on all
2239 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2240 ${stat: expansion item.
2242 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2243 protocol synchronization error", to keep the pedants happy.
2245 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2246 well as for IRIX systems, when gcc is being used. See the host.c source
2249 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2251 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2254 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2255 transport and to the smtp transport in LMTP mode.
2257 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2259 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2260 run a filter in a subprocess. This could lead to confusion in subsequent
2261 lookups in the parent process. There should also be a search_tidyup() at
2262 the end of the subprocess.
2264 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2265 only if the host matched helo_try_verify_hosts, which caused the
2266 verification to occur when the EHLO/HELO command was issued. The ACL just
2267 tested the remembered result. Now, if a previous verification attempt has
2268 not happened, "verify = helo" does it there and then.
2270 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2272 TK/03 Fix log output including CR from clamd.
2274 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2275 exist provoked a memory error which could cause a segfault.
2277 PH/15 Installed PCRE 6.2
2279 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2281 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2282 of the problem. Specifically, suggested +O2 rather than +O1 for the
2285 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2287 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2288 record, the port was not correctly logged when the outgoing_port log
2289 selector was set (it logged the transort's default port).
2291 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2292 fallback_hosts, and "hosts" in the smtp transport.
2294 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2295 host errors such as "Connection refused".
2297 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2298 authentication with radiusclient 0.4.9:
2300 - Error returned from rc_read_config was caught wrongly
2301 - Username/password not passed on to radius server due to wrong length.
2303 The presumption is that some radiusclient API changes for 4.51/PH/17
2304 were not taken care of correctly. The code is still untested by me (my
2305 Linux distribution still has 0.3.2 of radiusclient), but it was
2306 contributed by a Radius user.
2308 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2309 expanding the "port" option of the smtp transport.
2311 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2312 while reading a MIME header. Thanks to Tom Hughes for a patch.
2314 PH/24 Include config.h inside local_scan.h so that configuration settings are
2317 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2318 that in an ACL for RCPT (for example), you can examine exactly what was
2321 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2322 commands, but it was not correctly comparing the address with the actual
2323 client host address. Thus, it would show the EHLO address in Received:
2324 header lines when this was not necessary.
2326 PH/27 Added the % operator to ${eval:}.
2328 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2329 it should be ignoring failures (because with -C, for example, it has lost
2330 privilege). It wasn't ignoring creation failures other than "already
2333 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2336 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2337 because an input line was too long, either on its own, or by virtue of
2338 too many continuations, the temporary file was not being removed, and the
2339 return code was incorrect.
2341 PH/31 Missing "BOOL" in function definition in filtertest.c.
2343 PH/32 Applied Sieve patches from the maintainer.
2345 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2347 PH/33 Added "verify = not_blind".
2349 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2350 Local/Makefile (with some defaults set). These are used in built scripts
2351 such as exicyclog, but they have never been used in the exim_install
2352 script (though there are many overriding facilities there). I have
2353 arranged that the exim_install script now takes note of these two
2356 PH/35 Installed configuration files for Dragonfly.
2358 PH/36 When a locally submitted message by a trusted user did not contain a
2359 From: header, and the sender address was obtained from -f or from an SMTP
2360 MAIL command, and the trusted user did not use -F to supply a sender
2361 name, $originator_name was incorrectly used when constructing a From:
2362 header. Furthermore, $originator_name was used for submission mode
2363 messages from external hosts without From: headers in a similar way,
2364 which is clearly wrong.
2366 PH/37 Added control=suppress_local_fixups.
2368 PH/38 When log_selector = +received_sender was set, and the addition of the
2369 sender made the log line's construction buffer exactly full, or one byte
2370 less than full, an overflow happened when the terminating "\n" was
2373 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2374 when the result of a list match is failure because a DNS lookup failed.
2376 PH/40 RM_COMMAND is now used in the building process.
2378 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2379 the "build-* directories that it finds.
2381 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2382 address in a domain literal was a prefix of an interface address.
2384 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2385 when verifying a sender address, unless rewrite_headers is false.
2387 PH/44 Wrote a long comment about why errors_to addresses are verified as
2388 recipients, not senders.
2390 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2391 the ratelimit ACL was added.
2393 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2395 PH/46 Added extra information about PostgreSQL errors to the error string.
2397 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2398 avoiding the need to install special zones in a real server. This is
2399 backwards compatible; if it can't find the fake resolver, it drops back.
2400 Thus, both old and new test suites can be run.
2402 TF/02 Added util/ratelimit.pl
2404 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2405 clock is set back in time.
2407 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2408 Candler <B.Candler@pobox.com>.
2410 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2411 OK for addresses that are the result of redirections.
2413 PH/48 A number of further additions for the benefit of the new test suite,
2414 including a fake gethostbyname() that interfaces to the fake DNS resolver
2417 TF/06 The fix for widen_domains has also been applied to qualify_single and
2418 search_parents which are the other dnslookup options that can cause
2421 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2424 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2426 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2427 service" if it encountered an MX record with an empty target hostname.
2428 The message is now "an MX or SRV record indicated no SMTP service".
2430 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2431 if the DNS of the sending site is misconfigured. This is quite a
2432 common situation. This change restores the behaviour of treating a
2433 helo verification defer as a failure.
2435 PH/51 If self=fail was set on a router, the bounce message did not include the
2436 actual error message.
2442 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2444 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2445 command itself ended in error, the underlying message about the transport
2446 filter timeout was being overwritten with the pipe command error. Now the
2447 underlying error message should be appended to the second error message.
2449 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2451 PH/02 Reduce the amount of output that "make" produces by default. Full output
2452 can still be requested.
2454 PH/03 The warning log line about a condition test deferring for a "warn" verb
2455 was being output only once per connection, rather than after each
2456 occurrence (because it was using the same function as for successful
2457 "warn" verbs). This seems wrong, so I have changed it.
2459 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2460 it should not have, which might have caused a crash in the right
2461 circumstances, but probably never did.
2463 PH/04 Installed a modified version of Tony Finch's patch to make submission
2464 mode fix the return path as well as the Sender: header line, and to
2465 add a /name= option so that you can make the user's friendly name appear
2468 TF/03 Added the control = fakedefer ACL modifier.
2470 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2471 Mark Lowes for thorough testing.
2473 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2475 TK/03 Merged latest SRS patch from Miles Wilton.
2477 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2478 with the definition in sysexits.h (which is #included earlier).
2479 Fortunately, Exim does not actually use EX_OK. The code used to try to
2480 preserve the sysexits.h value, by assumimg that macro definitions were
2481 scanned for macro replacements. I have been disabused of this notion,
2482 so now the code just undefines EX_OK before #including unistd.h.
2484 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2485 in the smtp transport. When a block could not be written in a single
2486 write() function, the timeout was being re-applied to each part-write.
2487 This seems wrong - if the receiver was accepting one byte at a time it
2488 would take for ever. The timeout is now adjusted when this happens. It
2489 doesn't have to be particularly precise.
2491 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2492 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2494 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2495 without a domain if the check to <postmaster@domain> fails.
2497 SC/01 Eximstats: added -xls and the ability to specify output files
2498 (patch written by Frank Heydlauf).
2500 SC/02 Eximstats: use FileHandles for outputing results.
2502 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2504 SC/04 Eximstats: fixed display of large numbers with -nvr option
2506 SC/05 Eximstats: fixed merging of reports with empty tables.
2508 SC/06 Eximstats: added the -include_original_destination flag
2510 SC/07 Eximstats: removed tabs and trailing whitespace.
2512 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2514 TK/06 MBOX spool code: Add real "From " MBOX separator line
2515 so the .eml file is really in mbox format (even though
2516 most programs do not really care). Patch from Alex Miller.
2518 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2519 The latter is generated from $received_to and is only set if the
2520 message has one envelope recipient. SA can use these headers,
2521 obviously out-of-the-box. Patch from Alex Miller.
2523 PH/08 The ${def test on a variable was returning false if the variable's
2524 value was "0", contrary to what the specification has always said!
2525 The result should be true unless the variable is empty.
2527 PH/09 The syntax error of a character other than { following "${if
2528 def:variable_name" (after optional whitespace) was not being diagnosed.
2529 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2530 accidental colon was present, for example, could give incorrect results.
2532 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2533 result is used (not including appendfile, where other changes are about
2536 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2537 This involved changing a lot of size variables from int to off_t. It
2538 should work with maildirs and everything.
2540 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2541 spamd dying while we are connected to it.
2543 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2546 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2547 function for BDB 4.3.
2549 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2551 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2552 the sources for PCRE have been re-arranged and more files are now
2555 PH/15 The code I had for printing potentially long long variables in PH/11
2556 above was not the best (it lost precision). The length of off_t variables
2557 is now inspected at build time, and an appropriate printing format (%ld
2558 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2559 to be "long long int" or "long int". This is needed for the internal
2560 formatting function string_vformat().
2562 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2563 the configuration file to be ":syslog", then the script "guesses" where
2564 the logs files are, rather than using the compiled in default. In our
2565 case the guess is not the same as the compiled default, so the script
2566 suddenly stopped working when I started to use syslog. The patch checks
2567 to see if log_file_path is "". If so, it attempts to read it from exim
2568 with no configuration file to get the compiled in version, before it
2569 falls back to the previous guessing code."
2571 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2572 implementing BATV in an Exim configuration. See NewStuff for the gory
2575 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2576 Makefile that are specific to HP-UX.
2578 PH/18 If the "use_postmaster" option was set for a recipient callout together
2579 with the "random" option, the postmaster address was used as the MAIL
2580 FROM address for the random test, but not for the subsequent recipient
2581 test. It is now used for both.
2583 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2584 patch removes a few documentation additions to RFC 3028, because the
2585 latest draft now contains them. It adds the new en;ascii-case comparator
2586 and a new error check for 8bit text in MIME parts. Comparator and
2587 require names are now matched exactly. I enabled the subaddress
2588 extension, but it is not well tested yet (read: it works for me)."
2590 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2591 rework some of the code of TK/09 above to avoid the hardwired use of
2592 "%lld" and "long long". Replaced the call to snprintf() with a call to
2595 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2596 records point to non-existent hosts", "retry timeout exceeded", and
2597 "retry time not reached for any host after a long failure period".
2599 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2600 experimental DomainKeys support:
2602 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2603 (2) On an error such as an illegally used "control", the wrong name for
2604 the control was given.
2606 These problems did NOT occur unless DomainKeys support was compiled.
2608 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2610 PH/24 Added ${if match_ip condition.
2612 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2613 databases so that it will be absolutely obvious if a crash occurs in the
2614 DB library. This is a regular occurrence (often caused by mis-matched
2617 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2618 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2619 picked up on a user's system that detects such things. There doesn't seem
2620 to be a gcc warning option for this - only an attribute that has to be
2621 put on the function's prototype. It seems that in Fedora Core 4 they have
2622 set this on a number of new functions. No doubt there will be more in due
2625 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2626 specify a transport. However, if an address that was verified by such a
2627 router was the subject of a callout, Exim crashed because it tried to
2628 read the rcpt_include_affixes from the non-existent transport. Now it
2629 just assumes that the setting of that option is false. This bug was
2630 introduced by 4.51/PH/31.
2632 PH/28 Changed -d+all to exclude +memory, because that information is very
2633 rarely of interest, but it makes the output a lot bigger. People tend to
2634 do -d+all out of habit.
2636 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2637 code in os-type was giving problems when libc.so lives in lib64, like on
2640 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2641 aren't the modern standard, and it seems that some systems' include files
2642 don't always have them. Exim was already checking for some of the newer
2643 ones like T_AAAA, and defining it itself. I've added checks for all the
2644 record types that Exim uses.
2646 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2647 not automatically generating a new one, as it is supposed to. This
2648 prevented TLS from working. If the file did exist, but contained invalid
2649 data, a new version was generated, as expected. It was only the case of a
2650 non-existent file that was broken.
2652 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2653 with a change in libdomainkeys > 0.64.
2655 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2656 from DNS. If the selector record carries the flag, it now has
2657 precedence over the domain-wide flag.
2659 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2661 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2662 the use of a port name that isn't defined in /etc/services) occurred, the
2663 message was deferred as in a normal delivery, and thus remained on the
2664 spool, instead of being failed because of the mua_wrapper setting. This
2665 is now fixed, and I tidied up some of the mua_wrapper messages at the
2668 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2669 the messages in a hash of arrays rather than using individual hashes.
2670 This is a bit cleaner and results in dramatic memory savings, albeit
2671 at a slight CPU cost.
2673 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2674 as requested by Marc Sherman.
2676 SC/10 Eximstats: added histograms for user specified patterns as requested
2679 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2681 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2682 fopen() in the content-scanning modules that did not already have it.
2688 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2689 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2691 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2693 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2695 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2696 to test Sieve filters that use "vacation".
2698 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2699 that changes the way the GnuTLS parameters are stored in the cache file.
2700 The new format can be generated externally. For backward compatibility,
2701 if the data in the cache doesn't make sense, Exim assumes it has read an
2702 old-format file, and it generates new data and writes a new file. This
2703 means that you can't go back to an older release without removing the
2706 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2707 work if there are any delivery delays because "one_time" forces the
2708 parent to be marked "delivered", so its unseen clone is never tried
2709 again. For this reason, Exim now forbids the simultaneous setting of
2712 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2713 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2714 entry if you want to know the details. The fix, however, neglected to
2715 consider the case when local delivery batching is involved. The test for
2716 "previously delivered" was not happening when checking to see if an
2717 address could be batched with a previous (undelivered) one; under
2718 certain circumstances this could lead to multiple deliveries to the same
2721 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2722 in its include files, and this causes problems building Exim.
2724 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2725 header_syntax) but Exim was just ignoring anything given after a slash.
2726 In particular, this caused confusion with an attempt to use "verify =
2727 reverse_host_lookup/defer_ok". An error is now given when options are
2728 supplied for verify items that do not have them. (Maybe reverse_host_
2729 lookup should have a defer_ok option, but that's a different point.)
2731 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2732 defined by RFC 821) to 2048, because there were problems with some AUTH
2733 commands, and RFC 1869 says the size should be increased for extended
2734 SMTP commands that take arguments.
2736 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2739 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2740 "unknown" error; now it says that the functionality isn't in the binary.
2742 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2743 an address' error message when a string expansion fails (syntax or
2744 whatever). Otherwise the password may appear in the log. Following change
2745 PH/42 below, there is no longer a chance of it appearing in a bounce
2748 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2750 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2751 of its addresses was ever tried. (Bugzilla bug #2.)
2753 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2754 the result incorrectly in the debug output. (It correctly added a newline
2755 to what was transported.)
2757 TF/01 Added $received_time.
2759 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2760 commented out examples of how to interface to a virus scanner and to
2761 SpamAssassin. Also added commented examples of av_scanner and
2762 spamd_address settings.
2764 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2765 and controls are allowed in which ACLs. There were a couple of minor
2766 errors. Some of the entries in the conditions table (which is a table of
2767 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2768 negation of where the condition IS allowed.
2770 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2772 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2773 header file does not have a version number, so I've had to invent a new
2774 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2775 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2776 radiusclient), but it was contributed by a Radius user.
2778 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2779 files or directories, for OpenSSL.
2781 PH/19 When an Exim process that is running as root has to create an Exim log
2782 file, it does so in a subprocess that runs as exim:exim so as to get the
2783 ownership right at creation (otherwise, other Exim processes might see
2784 the file with the wrong ownership). There was no test for failure of this
2785 fork() call, which would lead to the process getting stuck as it waited
2786 for a non-existent subprocess. Forks do occasionally fail when resources
2787 run out. I reviewed all the other calls to fork(); they all seem to check
2790 PH/20 When checking for unexpected SMTP input at connect time (before writing
2791 the banner), Exim was not dealing correctly with a non-positive return
2792 from the read() function. If the client had disconnected by this time,
2793 the result was a log entry for a synchronization error with an empty
2794 string after "input=" when read() returned zero. If read() returned -1
2795 (an event I could not check), uninitialized data bytes were printed.
2796 There were reports of junk text (parts of files, etc) appearing after
2799 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2801 PH/22 Added support for macro redefinition, and (re)definition in between
2802 driver and ACL definitions.
2804 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2805 forgetting to use the resulting value; it was using the unexpanded value.
2807 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2808 hadn't been configured. The fix is from Juergen Kreileder, who
2809 understands it better than I do:
2811 "Here's what I see happening with three configured cyrus_sasl
2812 authenticators configured (plain, login, cram-md5):
2814 On startup auth_cyrus_sasl_init() gets called for each of these.
2815 This means three calls to sasl_listmech() without a specified mech_list.
2816 => SASL tests which mechs of all available mechs actually work
2817 => three warnings about OTP not working
2818 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2820 With the patch, sasl_listmech() also gets called three times. But now
2821 SASL's mech_list option is set to the server_mech specified in the the
2822 authenticator. Or in other words, the answer from sasl_listmech()
2823 gets limited to just the mech you're testing for (which is different
2825 => the return list contains just 'plain' or 'login', 'cram-md5' or
2826 nothing depending on the value of ob->server_mech.
2828 I've just tested the patch: Authentication still works fine,
2829 unavailable mechs specified in the exim configuration are still
2830 caught, and the auth.log warnings about OTP are gone."
2832 PH/25 When debugging is enabled, the contents of the command line are added
2833 to the debugging output, even when log_selector=+arguments is not
2836 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2837 answer is "GNU", and only if the return is "GNU/something" is the answer
2840 PH/27 $acl_verify_message is now set immediately after the failure of a
2841 verification in an ACL, and so is available in subsequent modifiers. In
2842 particular, the message can be preserved by coding like this:
2844 warn !verify = sender
2845 set acl_m0 = $acl_verify_message
2847 Previously, $acl_verify_message was set only while expanding "message"
2848 and "log_message" when a very denied access.
2850 PH/28 Modified OS/os.c-Linux with
2852 -#ifndef OS_LOAD_AVERAGE
2853 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2855 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2856 nomenclature these days.)
2858 PH/29 Installed patch from the Sieve maintainer that adds the options
2859 sieve_useraddress and sieve_subaddress to the redirect router.
2861 PH/30 In these circumstances:
2862 . Two addresses routed to the same list of hosts;
2863 . First host does not offer TLS;
2864 . First host accepts first address;
2865 . First host gives temporary error to second address;
2866 . Second host offers TLS and a TLS session is established;
2867 . Second host accepts second address.
2868 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2869 and peerdn, if requested) that were in fact used only for the second
2872 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2873 attention to any local part prefix or suffix that was matched by the
2874 router that accepted the address. It now behaves in the same way as it
2875 does for delivery: the affixes are removed from the local part unless
2876 rcpt_include_affixes is set on the transport.
2878 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2879 timeout during the DATA phase of an incoming message.
2881 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2882 applied a patch sanctioned by the Sieve maintainer.
2884 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2885 the uid or gid is negative. A case of a negative gid caused this to be
2886 noticed. The fix allows for either to be negative.
2888 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2889 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2892 PH/36 The change PH/12 above was broken. Fixed it.
2894 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2895 the grounds that routing the same address twice would always produce the
2896 same answer. This might have been true once, but it is certainly no
2897 longer true now. Routing a child address may depend on the previous
2898 routing that produced that child. Some complicated redirection strategies
2899 went wrong when messages had multiple recipients, and made Exim's
2900 behaviour dependent on the order in which the addresses were given.
2902 I have moved the duplicate checking until after the routing is complete.
2903 Exim scans the addresses that are assigned to local and remote
2904 transports, and removes any duplicates. This means that more work will be
2905 done, as duplicates will always all be routed, but duplicates are
2906 presumably rare, so I don't expect this is of any significance.
2908 For deliveries to pipes, files, and autoreplies, the duplicate checking
2909 still happens during the routing process, since they are not going to be
2912 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2913 It corrects a timeout issue with spamd. This is Ian's comment: "The
2914 background is that sometimes spamd either never reads data from a
2915 connection it has accepted, or it never writes response data. The exiscan
2916 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2917 blindly assumes that writes won't block so it may never time out."
2919 PH/39 Allow G after quota size as well as K and M.
2921 PH/40 The value set for $authenticated_id in an authenticator may not contain
2922 binary zeroes or newlines because the value is written to log lines and
2923 to spool files. There was no check on this. Now the value is run through
2924 the string_printing() function so that such characters are converted to
2925 printable escape sequences.
2927 PH/41 $message_linecount is a new variable that contains the total number of
2928 lines in the message. Compare $body_linecount, which is the count for the
2931 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2932 bounce and delay warning messages, except in certain special cases, which
2935 (a) An SMTP error message from a remote host;
2936 (b) A message specified in a :fail: redirection;
2937 (c) A message specified in a "fail" command in a system filter;
2938 (d) A message specified in a FAIL return from the queryprogram router;
2939 (e) A message specified by the cannot_route_message router option.
2941 In these cases only, Exim does include the error details in bounce and
2942 warning messages. There are also a few cases where bland messages such
2943 as "unrouteable address" or "local delivery error" are given.
2945 PH/43 $value is now also set for the "else" part of a ${run expansion.
2947 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2948 being worked on, but at least Exim now implements the latest version to
2951 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2952 process to complete was treated as a delivery failure, a timeout while
2953 writing the message to the pipe was logged, but erroneously treated as a
2954 successful delivery. Such timeouts include transport filter timeouts. For
2955 consistency with the overall process timeout, these timeouts are now
2956 treated as errors, giving rise to delivery failures by default. However,
2957 there is now a new Boolean option for the pipe transport called
2958 timeout_defer, which, if set TRUE, converts the failures into defers for
2959 both kinds of timeout. A transport filter timeout is now identified in
2962 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2963 systems where "make" and "gmake" are different, calling "gmake" at top
2964 level broke things. I've arranged for the value of $(MAKE) to be passed
2965 from the Makefile to this script so that it can call the same version of
2969 A note about Exim versions 4.44 and 4.50
2970 ----------------------------------------
2972 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2973 changes of various kinds. As a consequence, a big documentation update was
2974 needed. This delayed the release for rather longer than seemed good, especially
2975 in the light of a couple of (minor) security issues. Therefore, the changes
2976 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2977 release. So 4.44 and 4.50 are in effect two different branches that both start
2980 I have left the 4.50 change log unchanged; it contains all the changes since
2981 4.43. The change log for 4.44 is below; many of its items are identical to
2982 those for 4.50. This seems to be the most sensible way to preserve the
2983 historical information.
2989 1. Minor wording change to the doc/README.SIEVE file.
2991 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2992 computation of the current number of files was incorrect.
2994 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2995 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2998 4. Give more explanation in the error message when the command for a transport
2999 filter fails to execute.
3001 5. There are several places where Exim runs a non-Exim command in a
3002 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3003 was being done only for the command run by the queryprogram router. It is
3004 now done for all such subprocesses. The other cases are: ${run, transport
3005 filters, and the commands run by the lmtp and pipe transports.
3007 6. Added CONFIGURE_GROUP build-time option.
3009 7. Some older OS have a limit of 256 on the maximum number of file
3010 descriptors. Exim was using setrlimit() to set 1000 as a large value
3011 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3012 systems. I've change it so that if it can't get 1000, it tries for 256.
3014 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3015 was an oversight, and furthermore, ever since the addition of extra
3016 controls (e.g. 4.43/32), the checks on when to allow different forms of
3017 "control" were broken. There should now be diagnostics for all cases when a
3018 control that does not make sense is encountered.
3020 9. Added the /retain_sender option to "control=submission".
3022 10. $recipients is now available in the predata ACL (oversight).
3024 11. Tidy the search cache before the fork to do a delivery from a message
3025 received from the command line. Otherwise the child will trigger a lookup
3026 failure and thereby defer the delivery if it tries to use (for example) a
3027 cached ldap connection that the parent has called unbind on.
3029 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3030 of $address_data from the recipient verification was clobbered by the
3031 sender verification.
3033 13. The value of address_data from a sender verification is now available in
3034 $sender_address_data in subsequent conditions in the ACL statement.
3036 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3038 15. Added a new option "connect=<time>" to callout options, to set a different
3041 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3042 was its contents. (It was OK if the option was not defined at all.)
3044 17. A "Completed" log line is now written for messages that are removed from
3045 the spool by the -Mrm option.
3047 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3048 information about exactly what failed.
3050 19. Added -dd to debug only the daemon process.
3052 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3053 handles timeouts, both on the server side and network timeouts. Renamed the
3054 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3056 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3057 It is now set to "smtps".
3059 22. $host_address is now set to the target address during the checking of
3060 ignore_target_hosts.
3062 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3063 being passed; this would have caused $sender_host_name to have been used if
3064 matching the list had actually called for a host name (not very likely,
3065 since this list is usually IP addresses). A host name is now passed as
3068 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3069 code that specifies a non-restarting handler (typically sigaction() in
3070 modern systems) in an attempt to fix a rare and obscure crash bug.
3072 25. Narrowed the window for a race in the daemon that could cause it to ignore
3073 SIGCHLD signals. This is not a major problem, because they are used only to
3074 wake it up if nothing else does.
3076 26. A malformed maildirsize file could cause Exim to calculate negative values
3077 for the mailbox size or file count. Odd effects could occur as a result.
3078 The maildirsize information is now recalculated if the size or filecount
3081 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3082 support for a long time. Removed HAVE_SYS_VFS_H.
3084 28. Installed the latest version of exipick from John Jetmore.
3086 29. In an address list, if the pattern was not a regular expression, an empty
3087 subject address (from a bounce message) matched only if the pattern was an
3088 empty string. Non-empty patterns were not even tested. This was the wrong
3089 because it is perfectly reasonable to use an empty address as part of a
3090 database query. An empty address is now tested by patterns that are
3091 lookups. However, all the other forms of pattern expect the subject to
3092 contain a local part and a domain, and therefore, for them, an empty
3093 address still always fails if the pattern is not itself empty.
3095 30. Exim went into a mad DNS loop when attempting to do a callout where the
3096 host was specified on an smtp transport, and looking it up yielded more
3097 than one IP address.
3099 31. Re-factored the code for checking spool and log partition space into a
3100 function that finds that data and another that does the check. The former
3101 is then used to implement four new variables: $spool_space, $log_space,
3102 $spool_inodes, and $log_inodes.
3104 32. The RFC2047 encoding function was originally intended for short strings
3105 such as real names; it was not keeping to the 75-character limit for
3106 encoded words that the RFC imposes. It now respects the limit, and
3107 generates multiple encoded words if necessary. To be on the safe side, I
3108 have increased the buffer size for the ${rfc2047: expansion operator from
3111 33. It is now permitted to omit both strings after an "if" condition; if the
3112 condition is true, the result is "true". As before, when the second string
3113 is omitted, a false condition yields an empty string. This makes it less
3114 cumbersome to write custom ACL and router conditions.
3116 34. Failure to deliver a bounce message always caused it to be frozen, even if
3117 there was an errors_to setting on the router. The errors_to setting is now
3120 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3121 canonical form (fully expanded) before being placed in
3122 $sender_host_address.
3124 36. The table in the code that translates DNS record types into text (T_A to
3125 "A" for instance) was missing entries for NS and CNAME. It is just possible
3126 that this could have caused confusion if both these types were looked up
3127 for the same domain, because the text type is used as part of Exim's
3128 per-process caching. But the chance of anyone hitting this buglet seems
3131 37. The dnsdb lookup has been extended in a number of ways.
3133 (1) There is a new type, "zns", which walks up the domain tree until it
3134 finds some nameserver records. It should be used with care.
3136 (2) There is a new type, "mxh", which is like "mx" except that it returns
3137 just the host names, not the priorities.
3139 (3) It is now possible to give a list of domains (or IP addresses) to be
3140 looked up. The behaviour when one of the lookups defers can be
3141 controlled by a keyword.
3143 (4) It is now possible to specify the separator character for use when
3144 multiple records are returned.
3146 38. The dnslists ACL condition has been extended: it is now possible to supply
3147 a list of IP addresses and/or domains to be looked up in a particular DNS
3150 39. Added log_selector=+queue_time_overall.
3152 40. When running the queue in the test harness, wait just a tad after forking a
3153 delivery process, to get repeatability of debugging output.
3155 41. Include certificate and key file names in error message when GnuTLS fails
3156 to set them up, because the GnuTLS error message doesn't include the name
3157 of the failing file when there is a problem reading it.
3159 42. Allow both -bf and -bF in the same test run.
3161 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3163 44. The "Exiscan patch" is now merged into the mainline Exim source.
3165 45. Sometimes the final signoff response after QUIT could fail to get
3166 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3167 < 0 before doing a fflush(). This bug looks as though it goes back to the
3168 introduction of TLS in release 3.20, but "sometimes" must have been rare
3169 because the tests only now provoked it.
3171 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3172 (this can affect the format of dates).
3174 47. exim_tidydb, when checking for the continued existence of a message for
3175 which it has found a message-specific retry record, was not finding
3176 messages that were in split spool directories. Consequently, it was
3177 deleting retry records that should have stayed in existence.
3179 48. Steve fixed some bugs in eximstats.
3181 49. The SPA authentication driver was not abandoning authentication and moving
3182 on to the next authenticator when an expansion was forced to fail,
3183 contradicting the general specification for all authenticators. Instead it
3184 was generating a temporary error. It now behaves as specified.
3186 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3187 (the order specifies the preference for clients). The order is now AES256,
3188 AES128, 3DES, ARCFOUR128.
3190 51. Small patch to Sieve code - explicitly set From: when generating an
3193 52. Exim crashed if a remote delivery caused a very long error message to be
3194 recorded - for instance if somebody sent an entire SpamAssassin report back
3195 as a large number of 550 error lines. This bug was coincidentally fixed by
3196 increasing the size of one of Exim's internal buffers (big_buffer) that
3197 happened as part of the Exiscan merge. However, to be on the safe side, I
3198 have made the code more robust (and fixed the comments that describe what
3201 53. Now that there can be additional text after "Completed" in log lines (if
3202 the queue_time_overall log selector is set), a one-byte patch to exigrep
3203 was needed to allow it to recognize "Completed" as not the last thing in
3206 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3207 patch that reportedly fixes this has been added. I am not expert enough to
3208 create a test for it. This is what the patch creator wrote:
3210 "I found a little strange behaviour of ldap code when working with
3211 Windows 2003 AD Domain, where users was placed in more than one
3212 Organization Units. When I tried to give exim partial DN, the exit code
3213 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3214 But simultaneously result of request was absolutely normal ldap result,
3215 so I produce this patch..."
3217 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3218 REFERENCE, so I have modified the code to exclude the patch when that macro
3221 55. Some experimental protocols are using DNS PTR records for new purposes. The
3222 keys for these records are domain names, not reversed IP addresses. The
3223 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3224 leaves it alone. Component reversal etc. now happens only for IP addresses.
3227 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3229 57. Double the size of the debug message buffer (to 2048) so that more of very
3230 long debug lines gets shown.
3232 58. The exicyclog utility now does better if the number of log files to keep
3233 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3235 59. Two changes related to the smtp_active_hostname option:
3237 (1) $smtp_active_hostname is now available as a variable.
3238 (2) The default for smtp_banner uses $smtp_active_hostname instead
3239 of $primary_hostname.
3241 60. The host_aton() function is supposed to be passed a string that is known
3242 to be a valid IP address. However, in the case of IPv6 addresses, it was
3243 not checking this. This is a hostage to fortune. Exim now panics and dies
3244 if the condition is not met. A case was found where this could be provoked
3245 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3246 components; fortuitously, this particular loophole had already been fixed
3247 by change 4.50/55 above.
3249 If there are any other similar loopholes, the new check in host_aton()
3250 itself should stop them being exploited. The report I received stated that
3251 data on the command line could provoke the exploit when Exim was running as
3252 exim, but did not say which command line option was involved. All I could
3253 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3254 running as the user.
3257 61. There was a buffer overflow vulnerability in the SPA authentication code
3258 (which came originally from the Samba project). I have added a test to the
3259 spa_base64_to_bits() function which I hope fixes it.
3262 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3263 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3265 63. The daemon start-up calls getloadavg() while still root for those OS that
3266 need the first call to be done as root, but it missed one case: when
3267 deliver_queue_load_max is set with deliver_drop_privilege. This is
3268 necessary for the benefit of the queue runner, because there is no re-exec
3269 when deliver_drop_privilege is set.
3271 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3272 This has been fixed.
3274 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3275 was in use, was not putting the data itself into the right store pool;
3276 consequently, it could be overwritten for a subsequent message in the same
3277 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3280 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3282 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3283 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3284 the function were treating the return as a boolean value, which happened to
3285 work because 0=false and not-0=true, but is not correct code.
3287 68. The host_aton() function was not handling scoped IPv6 addresses (those
3288 with, for example, "%eth0" on the end) correctly.
3290 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3291 negated items (that is, ~something) in unsigned ints. Some compilers
3292 apparently mutter when there is no cast.
3294 70. If an address verification called from an ACL failed, and did not produce a
3295 user-specific message (i.e. there was only a "system" message), nothing was
3296 put in $acl_verify_message. In this situation, it now puts the system
3299 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3300 session; change 4.31/43 added the unwanted input to the log line - except
3301 that it did not do this in the start of session case. It now does.
3303 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3304 This is wrong and can cause the other end to generate a synchronization
3305 error if it is another Exim or anything else that does the synchronization
3306 check. A QUIT command is no longer sent after a timeout.
3308 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3309 during host lookups.
3311 74. The defer_ok option of callout verification was not working if it was used
3312 when verifying addresses in header lines, that is, for this case:
3314 verify = header_sender/callout=defer_ok
3316 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3317 those file descriptors could be used for SMTP connections. If anything
3318 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3319 could be sent to the SMTP client, causing chaos. The daemon now opens
3320 stdin, stdout, and stderr to /dev/null when it puts itself into the
3323 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3324 log by default. The user can override this with suitable Perl magic.
3326 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3327 the log message when discard triggers, was not working for the DATA ACL or
3328 for the non-SMTP ACL.
3330 78. Error message wording change in sieve.c.
3332 79. If smtp_accept_max_per_host was set, the number of connections could be
3333 restricted to fewer than expected, because the daemon was trying to set up
3334 a new connection before checking whether the processes handling previous
3335 connections had finished. The check for completed processes is now done
3336 earlier. On busy systems, this bug wouldn't be noticed because something
3337 else would have woken the daemon, and it would have reaped the completed
3340 80. If a message was submitted locally by a user whose login name contained one
3341 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3342 It caused a spool format error. I have fixed the spool reading code. A
3343 related problem was that the "from" clause in the Received: line became
3344 illegal because of the space(s). It is now covered by ${quote_local_part.
3346 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3349 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3351 83. Patch from Sieve maintainer to fix unterminated string problem in
3352 "vacation" handling.
3354 84. Some minor changes to the Linux configuration files to help with other
3355 OS variants using glibc.
3357 85. One more patch for Sieve to update vacation handling to latest spec.
3360 ----------------------------------------------------
3361 See the note above about the 4.44 and 4.50 releases.
3362 ----------------------------------------------------
3368 1. Change 4.43/35 introduced a bug that caused file counts to be
3369 incorrectly computed when quota_filecount was set in an appendfile
3372 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3373 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3376 3. Give more explanation in the error message when the command for a transport
3377 filter fails to execute.
3379 4. There are several places where Exim runs a non-Exim command in a
3380 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3381 was being done only for the command run by the queryprogram router. It is
3382 now done for all such subprocesses. The other cases are: ${run, transport
3383 filters, and the commands run by the lmtp and pipe transports.
3385 5. Some older OS have a limit of 256 on the maximum number of file
3386 descriptors. Exim was using setrlimit() to set 1000 as a large value
3387 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3388 systems. I've change it so that if it can't get 1000, it tries for 256.
3390 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3391 was an oversight, and furthermore, ever since the addition of extra
3392 controls (e.g. 4.43/32), the checks on when to allow different forms of
3393 "control" were broken. There should now be diagnostics for all cases when a
3394 control that does not make sense is encountered.
3396 7. $recipients is now available in the predata ACL (oversight).
3398 8. Tidy the search cache before the fork to do a delivery from a message
3399 received from the command line. Otherwise the child will trigger a lookup
3400 failure and thereby defer the delivery if it tries to use (for example) a
3401 cached ldap connection that the parent has called unbind on.
3403 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3404 of $address_data from the recipient verification was clobbered by the
3405 sender verification.
3407 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3408 was its contents. (It was OK if the option was not defined at all.)
3410 11. A "Completed" log line is now written for messages that are removed from
3411 the spool by the -Mrm option.
3413 12. $host_address is now set to the target address during the checking of
3414 ignore_target_hosts.
3416 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3417 being passed; this would have caused $sender_host_name to have been used if
3418 matching the list had actually called for a host name (not very likely,
3419 since this list is usually IP addresses). A host name is now passed as
3422 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3423 code that specifies a non-restarting handler (typically sigaction() in
3424 modern systems) in an attempt to fix a rare and obscure crash bug.
3426 15. Narrowed the window for a race in the daemon that could cause it to ignore
3427 SIGCHLD signals. This is not a major problem, because they are used only to
3428 wake it up if nothing else does.
3430 16. A malformed maildirsize file could cause Exim to calculate negative values
3431 for the mailbox size or file count. Odd effects could occur as a result.
3432 The maildirsize information is now recalculated if the size or filecount
3435 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3436 support for a long time. Removed HAVE_SYS_VFS_H.
3438 18. Updated exipick to current release from John Jetmore.
3440 19. Allow an empty sender to be matched against a lookup in an address list.
3441 Previously the only cases considered were a regular expression, or an
3444 20. Exim went into a mad DNS lookup loop when doing a callout where the
3445 host was specified on the transport, if the DNS lookup yielded more than
3448 21. The RFC2047 encoding function was originally intended for short strings
3449 such as real names; it was not keeping to the 75-character limit for
3450 encoded words that the RFC imposes. It now respects the limit, and
3451 generates multiple encoded words if necessary. To be on the safe side, I
3452 have increased the buffer size for the ${rfc2047: expansion operator from
3455 22. Failure to deliver a bounce message always caused it to be frozen, even if
3456 there was an errors_to setting on the router. The errors_to setting is now
3459 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3460 canonical form (fully expanded) before being placed in
3461 $sender_host_address.
3463 24. Updated eximstats to version 1.33
3465 25. Include certificate and key file names in error message when GnuTLS fails
3466 to set them up, because the GnuTLS error message doesn't include the name
3467 of the failing file when there is a problem reading it.
3469 26. Expand error message when OpenSSL has problems setting up cert/key files.
3472 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3473 (this can affect the format of dates).
3475 28. exim_tidydb, when checking for the continued existence of a message for
3476 which it has found a message-specific retry record, was not finding
3477 messages that were in split spool directories. Consequently, it was
3478 deleting retry records that should have stayed in existence.
3480 29. eximstats updated to version 1.35
3481 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3482 1.35 - bugfix such that pie charts by volume are generated correctly
3484 30. The SPA authentication driver was not abandoning authentication and moving
3485 on to the next authenticator when an expansion was forced to fail,
3486 contradicting the general specification for all authenticators. Instead it
3487 was generating a temporary error. It now behaves as specified.
3489 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3490 (the order specifies the preference for clients). The order is now AES256,
3491 AES128, 3DES, ARCFOUR128.
3493 31. Small patch to Sieve code - explicitly set From: when generating an
3496 32. Exim crashed if a remote delivery caused a very long error message to be
3497 recorded - for instance if somebody sent an entire SpamAssassin report back
3498 as a large number of 550 error lines. This bug was coincidentally fixed by
3499 increasing the size of one of Exim's internal buffers (big_buffer) that
3500 happened as part of the Exiscan merge. However, to be on the safe side, I
3501 have made the code more robust (and fixed the comments that describe what
3504 33. Some experimental protocols are using DNS PTR records for new purposes. The
3505 keys for these records are domain names, not reversed IP addresses. The
3506 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3507 leaves it alone. Component reversal etc. now happens only for IP addresses.
3510 34. The host_aton() function is supposed to be passed a string that is known
3511 to be a valid IP address. However, in the case of IPv6 addresses, it was
3512 not checking this. This is a hostage to fortune. Exim now panics and dies
3513 if the condition is not met. A case was found where this could be provoked
3514 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3515 components; fortuitously, this particular loophole had already been fixed
3516 by change 4.50/55 or 4.44/33 above.
3518 If there are any other similar loopholes, the new check in host_aton()
3519 itself should stop them being exploited. The report I received stated that
3520 data on the command line could provoke the exploit when Exim was running as
3521 exim, but did not say which command line option was involved. All I could
3522 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3523 running as the user.
3526 35. There was a buffer overflow vulnerability in the SPA authentication code
3527 (which came originally from the Samba project). I have added a test to the
3528 spa_base64_to_bits() function which I hope fixes it.
3531 36. The daemon start-up calls getloadavg() while still root for those OS that
3532 need the first call to be done as root, but it missed one case: when
3533 deliver_queue_load_max is set with deliver_drop_privilege. This is
3534 necessary for the benefit of the queue runner, because there is no re-exec
3535 when deliver_drop_privilege is set.
3537 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3538 was in use, was not putting the data itself into the right store pool;
3539 consequently, it could be overwritten for a subsequent message in the same
3540 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3543 38. Sometimes the final signoff response after QUIT could fail to get
3544 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3545 < 0 before doing a fflush(). This bug looks as though it goes back to the
3546 introduction of TLS in release 3.20, but "sometimes" must have been rare
3547 because the tests only now provoked it.
3553 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3554 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3555 Now it returns an integer. A number of calls were still expecting a T/F
3556 return. Fortuitously, in all cases, the tests worked in OK situations,
3557 which is the norm. However, things would have gone wrong on any write
3558 failures on the smtp file descriptor. This function is used when sending
3559 messages over SMTP and also when doing verify callouts.
3561 2. When Exim is called to do synchronous delivery of a locally submitted
3562 message (the -odf or -odi options), it no longer closes stderr before doing
3565 3. Implemented the mua_wrapper option.
3567 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3569 5. Implemented the functions header_remove(), header_testname(),
3570 header_add_at_position(), and receive_remove_recipient(), and exported them
3573 6. If an ACL "warn" statement specified the addition of headers, Exim already
3574 inserted X-ACL-Warn: at the start if there was no header name. However, it
3575 was not making this test for the second and subsequent header lines if
3576 there were newlines in the string. This meant that an invalid header could
3577 be inserted if Exim was badly configured.
3579 7. Allow an ACL "warn" statement to add header lines at the start or after all
3580 the Received: headers, as well as at the end.
3582 8. Added the rcpt_4xx retry error code.
3584 9. Added postmaster_mailfrom=xxx to callout verification option.
3586 10. Added mailfrom=xxxx to the callout verification option, for verify=
3589 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3590 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3592 12. Inserted some casts to stop certain compilers warning when using pointer
3593 differences as field lengths or precisions in printf-type calls (mostly
3594 affecting debugging statements).
3596 13. Added optional readline() support for -be (dynamically loaded).
3598 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3599 same clock tick as a message's arrival, so that its received time was the
3600 same as the "first fail" time on the retry record, and that message
3601 remained on the queue past the ultimate address timeout, every queue runner
3602 would try a delivery (because it was past the ultimate address timeout) but
3603 after another failure, the ultimate address timeout, which should have then
3604 bounced the address, did not kick in. This was a "< instead of <=" error;
3605 in most cases the first failure would have been in the next clock tick
3606 after the received time, and all would be well.
3608 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3609 being recognized when the domain list was tested by the match_domain
3610 condition in an expansion string.
3612 16. Added the ${str2b64: operator.
3614 17. Exim was always calling setrlimit() to set a large limit for the number of
3615 processes, without checking whether the existing limit was already
3616 adequate. (It did check for the limit on file descriptors.) Furthermore,
3617 errors from getrlimit() and setrlimit() were being ignored. Now they are
3618 logged to the main and panic logs, but Exim does carry on, to try to do its
3619 job under whatever limits there are.
3621 18. Imported PCRE 5.0.
3623 19. Trivial typo in log message " temporarily refused connection" (the leading
3626 20. If the log selector return_path_on_delivery was set and an address was
3627 redirected to /dev/null, the delivery process crashed because it assumed
3628 that a return path would always be set for a "successful" delivery. In this
3629 case, the whole delivery is bypassed as an optimization, and therefore no
3632 21. Internal re-arrangement: the function for sending a challenge and reading
3633 a response while authentication was assuming a zero-terminated challenge
3634 string. It's now changed to take a pointer and a length, to allow for
3635 binary data in such strings.
3637 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3639 23. Exim was not respecting finduser_retries when seeking the login of the
3640 uid under which it was called; it was always trying 10 times. (The default
3641 setting of finduser_retries is zero.) Also, it was sleeping after the final
3642 failure, which is pointless.
3644 24. Implemented tls_on_connect_ports.
3646 25. Implemented acl_smtp_predata.
3648 26. If the domain in control=submission is set empty, Exim assumes that the
3649 authenticated id is a complete email address when it generates From: or
3650 Sender: header lines.
3652 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3653 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3654 chown and chgrp in /bin and hostname in /usr/bin.
3656 28. Exim was keeping the "process log" file open after each use, just as it
3657 does for the main log. This opens the possibility of it remaining open for
3658 long periods when the USR1 signal hits a daemon. Occasional processlog
3659 errors were reported, that could have been caused by this. Anyway, it seems
3660 much more sensible not to leave this file open at all, so that is what now
3663 29. The long-running daemon process does not normally write to the log once it
3664 has entered its main loop, and it closes the log before doing so. This is
3665 so that log files can straightforwardly be renamed and moved. However,
3666 there are a couple of unusual error situations where the daemon does write
3667 log entries, and I had neglected to close the log afterwards.
3669 30. The text of an SMTP error response that was received during a remote
3670 delivery was being truncated at 512 bytes. This is too short for some of
3671 the long messages that one sometimes sees. I've increased the limit to
3674 31. It is now possible to make retry rules that apply only when a message has a
3675 specific sender, in particular, an empty sender.
3677 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3678 it possible to be selective about when SMTP synchronization is enforced.
3680 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3682 32. Implemented hosts_connection_nolog.
3684 33. Added an ACL for QUIT.
3686 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3689 35. Added mailbox_size and mailbox_filecount to appendfile.
3691 36. Added control = no_multiline_responses to ACLs.
3693 37. There was a bug in the logic of the code that waits for the clock to tick
3694 in the case where the clock went backwards by a substantial amount such
3695 that the microsecond fraction of "now" was more than the microsecond
3696 fraction of "then" (but the whole seconds number was less).
3698 38. Added support for the libradius Radius client library this is found on
3699 FreeBSD (previously only the radiusclient library was supported).
3705 1. When certain lookups returned multiple values in the form name=value, the
3706 quoting of the values was not always being done properly. Specifically:
3707 (a) If the value started with a double quote, but contained no whitespace,
3709 (b) If the value contained whitespace other than a space character (i.e.
3710 tabs or newlines or carriage returns) it was not quoted.
3711 This fix has been applied to the mysql and pgsql lookups by writing a
3712 separate quoting function and calling it from the lookup code. The fix
3713 should probably also be applied to nisplus, ibase and oracle lookups, but
3714 since I cannot test any of those, I have not disturbed their existing code.
3716 2. A hit in the callout cache for a specific address caused a log line with no
3717 reason for rejecting RCPT. Now it says "Previous (cached) callout
3718 verification failure".
3720 3. There was an off-by-one bug in the queryprogram router. An over-long
3721 return line was truncated at 256 instead of 255 characters, thereby
3722 overflowing its buffer with the terminating zero. As well as fixing this, I
3723 have increased the buffer size to 1024 (and made a note to document this).
3725 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3726 when Exim is waiting for an SMTP response from a remote server, Exim
3727 restarts its select() call on the socket, thereby resetting its timeout.
3728 This is not a problem when such interrupts are rare. Somebody set up a cron
3729 job to run exiwhat every 2 minutes, which is less than the normal select()
3730 timeout (5 or 10 minutes). This meant that the select() timeout never
3731 kicked in because it was always reset. I have fixed this by comparing the
3732 time when an interrupt arrives with the time at the start of the first call
3733 to select(). If more time than the timeout has elapsed, the interrupt is
3734 treated as a timeout.
3736 5. Some internal re-factoring in preparation for the addition of Sieve
3737 extensions (by MH). In particular, the "personal" test is moved to a
3738 separate function, and given an option for scanning Cc: and Bcc: (which is
3739 not set for Exim filters).
3741 6. When Exim created an email address using the login of the caller as the
3742 local part (e.g. when creating a From: or Sender: header line), it was not
3743 quoting the local part when it contained special characters such as @.
3745 7. Installed new OpenBSD configuration files.
3747 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3748 try to make them clearer.
3750 9. Callout options, other than the timeout value, were being ignored when
3751 verifying sender addresses in header lines. For example, when using
3753 verify = header_sender/callout=no_cache
3755 the cache was (incorrectly) being used.
3757 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3758 only the Cygwin environment.
3760 11. When return_path_on_delivery was set as a log selector, if different remote
3761 addresses in the same message used different return paths and parallel
3762 remote delivery occurred, the wrong values would sometimes be logged.
3763 (Whenever a remote delivery process finished, the return path value from
3764 the most recently started remote delivery process was logged.)
3766 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3767 lines when AUTH and/or TLS are in use. This is the "received protocol"
3768 field. Exim used to use "asmtp" for authenticated SMTP, without any
3769 indication (in the protocol name) for TLS use. Now it follows the RFC and
3770 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3771 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3772 names appear in log lines as well as in Received: header lines.
3774 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3775 extensions, and comparison tests, and to fix some bugs.
3777 14. Changes to the "personal" filter test:
3779 (1) The test was buggy in that it was just doing the equivalent of
3780 "contains" tests on header lines. For example, if a user's address was
3781 anne@some.where, the "personal" test would incorrectly be true for
3783 To: susanne@some.where
3785 This test is now done by extracting each address from the header in turn,
3786 and checking the entire address. Other tests that are part of "personal"
3787 are now done using regular expressions (for example, to check local parts
3788 of addresses in From: header lines).
3790 (2) The list of non-personal local parts in From: addresses has been
3791 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3792 taken from the Sieve specification recommendations.
3794 (3) If the message contains any header line starting with "List-" it is
3795 treated as non-personal.
3797 (4) The test for "circular" in the Subject: header line has been removed
3798 because it now seems ill-conceived.
3800 15. Minor typos in src/EDITME comments corrected.
3802 16. Installed latest exipick from John Jetmore.
3804 17. If headers_add on a router specified a text string that was too long for
3805 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3806 of string_sprintf() is now avoided.
3808 18. $message_body_size was not set (it was always zero) when running the DATA
3809 ACL and the local_scan() function.
3811 19. For the "mail" command in an Exim filter, no default was being set for
3812 the once_repeat time, causing a random time value to be used if "once" was
3813 specified. (If the value happened to be <= 0, no repeat happened.) The
3814 default is now 0s, meaning "never repeat". The "vacation" command was OK
3815 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3816 (I found it when inspecting the code).
3818 20. There is now an overall timeout for performing a callout verification. It
3819 defaults to 4 times the callout timeout, which applies to individual SMTP
3820 commands during the callout. The overall timeout applies when there is more
3821 than one host that can be tried. The timeout is checked before trying the
3822 next host. This prevents very long delays if there are a large number of
3823 hosts and all are timing out (e.g. when the network connections are timing
3824 out). The value of the overall timeout can be changed by specifying an
3825 additional sub-option for "callout", called "maxwait". For example:
3827 verify = sender/callout=5s,maxwait=20s
3829 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3830 to the end before writing, but this should make it even safer).
3832 22. Exim was forgetting that it had advertised PIPELINING for the second and
3833 subsequent messages on an SMTP connection. It was also not resetting its
3834 memory on STARTTLS and an internal HELO.
3836 23. When Exim logs an SMTP synchronization error within a session, it now
3837 records whether PIPELINING has been advertised or not.
3839 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3840 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3841 rather than long int.
3843 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3845 26. Added the never_mail option to autoreply.
3851 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3852 crash if the getsockname() call failed; this can happen if a connection is
3853 closed very soon after it is established. The problem was simply in the
3854 order in which certain operations were done, causing Exim to try to write
3855 to the SMTP stream before it had set up the file descriptor. The bug has
3856 been fixed by making things happen in the correct order.
3862 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3863 before the connection was closed, thus losing the rejection response.
3865 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3866 some early Solaris releases, but causes trouble in current releases where
3867 socklen_t is defined.
3869 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3872 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3875 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3876 that releasing the top of it at the end releases what was used for sub-
3877 expansions (unless the block got too big). However, discard this block if
3878 the first thing is a variable or header, so that we can use its block when
3879 it is dynamic (useful for very large $message_headers, for example).
3881 6. Lookups now cache *every* query, not just the most recent. A new, separate
3882 store pool is used for this. It can be recovered when all lookup caches are
3883 flushed. Lookups now release memory at the end of their result strings.
3884 This has involved some general refactoring of the lookup sources.
3886 7. Some code has been added to the store_xxx() functions to reduce the amount
3887 of flapping under certain conditions.
3889 8. log_incoming_interface used to affect only the <= reception log lines. Now
3890 it causes the local interface and port to be added to several more SMTP log
3891 lines, for example "SMTP connection from", and rejection lines.
3893 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3895 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3897 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3898 could be overwritten at the end of the current message (or the start of a
3899 new message if it was set in a HELO ACL). The value is now preserved for
3900 the duration of the SMTP connection.
3902 12. If a transport had a headers_rewrite setting, and a matching header line
3903 contained an unqualified address, that address was qualified, even if it
3904 did not match any rewriting rules. The underlying bug was that the values
3905 of the flags that permit the existence of unqualified sender and recipient
3906 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3907 non-local messages, and by -bnq for local messages) were not being
3908 preserved with the message after it was received.
3910 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3911 "next input=" as part of the text comprising the host identity instead of
3912 the correct text. The code was using the same buffer for two different
3913 strings. However, depending on which order the printing function evaluated
3914 its arguments, the bug did not always show up. Under Linux, for example, my
3915 test suite worked just fine.
3917 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3918 doesn't work with some older versions of Perl. It has been changed to "my",
3919 which in any case is probably the better facility to use.
3921 15. A really picky compiler found some instances of statements for creating
3922 error messages that either had too many or two few arguments for the format
3925 16. The size of the buffer for calls to the DNS resolver has been increased
3926 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3927 for addresses that have a lot of PTR records. This alleviates a problem; it
3928 does not fully solve it.
3930 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3931 buffer now truncates the list and logs the incident, which is the same
3932 action as happens when Exim is looking up a host name and its aliases.
3933 Previously in this situation something unpredictable would happen;
3934 sometimes it was "internal error: store_reset failed".
3936 18. If a server dropped the connection unexpectedly when an Exim client was
3937 using GnuTLS and trying to read a response, the client delivery process
3938 crashed while trying to generate an error log message.
3940 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3941 string, for example:
3943 warn message = H1: something\nH2: something
3945 the text was added as a single header line from Exim's point of view
3946 though it ended up OK in the delivered message. However, searching for the
3947 second and subsequent header lines using $h_h2: did not work. This has been
3948 fixed. Similarly, if a system filter added multiple headers in this way,
3949 the routers could not see them.
3951 20. Expanded the error message when iplsearch is called with an invalid key to
3952 suggest using net-iplsearch in a host list.
3954 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3955 ACLs are no longer actually imposed (and a message to that effect is
3958 22. If a "gecos" field in a passwd entry contained escaped characters, in
3959 particular, if it contained a \" sequence, Exim got it wrong when building
3960 a From: or a Sender: header from that name. A second bug also caused
3961 incorrect handling when an unquoted " was present following a character
3962 that needed quoting.
3964 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3965 was not being matched caselessly.
3967 24. Arranged for all hyphens in the exim.8 source to be escaped with
3970 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3971 properly. Recipient callout cache records were still being keyed to include
3972 the sender, even when use_sender was set false. This led to far more
3973 callouts that were necessary. The sender is no longer included in the key
3974 when use_sender is false.
3976 26. Added "control = submission" modifier to ACLs.
3978 27. Added the ${base62d: operator to decode base 62 numbers.
3980 28. dnsdb lookups can now access SRV records.
3982 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3983 the configuration file.
3985 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3986 (-v) mode. This makes the output for a verbose queue run more intelligible.
3988 31. Added a use_postmaster feature to recipient callouts.
3990 32. Added the $body_zerocount variable, containing the number of binary zero
3991 bytes in the message body.
3993 33. The time of last modification of the "new" subdirectory is now used as the
3994 "mailbox time last read" when there is a quota error for a maildir
3997 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3999 35. Added +ignore_unknown as a special item in host lists.
4001 36. Code for decoding IPv6 addresses in host lists is now included, even if
4002 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4003 address was recognized as an IP address, but was then not correctly decoded
4004 into binary, causing unexpected and incorrect effects when compared with
4011 1. Very minor rewording of debugging text in manualroute to say "list of
4012 hosts" instead of "hostlist".
4014 2. If verify=header_syntax was set, and a header line with an unqualified
4015 address (no domain) and a large number of spaces between the end of the
4016 name and the colon was received, the reception process suffered a buffer
4017 overflow, and (when I tested it) crashed. This was caused by some obsolete
4018 code that should have been removed. The fix is to remove it!
4020 3. When running in the test harness, delay a bit after writing a bounce
4021 message to get a bit more predictability in the log output.
4023 4. Added a call to search_tidyup() just before forking a reception process. In
4024 theory, someone could use a lookup in the expansion of smtp_accept_max_
4025 per_host which, without the tidyup, could leave open a database connection.
4027 5. Added the variables $recipient_data and $sender_data which get set from a
4028 lookup success in an ACL "recipients" or "senders" condition, or a router
4029 "senders" option, similar to $domain_data and $local_part_data.
4031 6. Moved the writing of debug_print from before to after the "senders" test
4034 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4035 problems for message scanning, either using a data ACL, or using
4036 local_scan() because the Received: header was not generated till after they
4037 were called (in order to set the time as the time of reception completion).
4038 I have revised the way this works. The header is now generated after the
4039 body is received, but before the ACL or local_scan() are called. After they
4040 are run, the timestamp in the header is updated.
4046 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4047 before starting a queue runner without re-exec. This happened only when
4048 deliver_drop_privilege was set or when the Exim user was set to root. The
4049 effect of the bug was that timeouts during subsequent deliveries caused
4050 crashes instead of being properly handled. The handler is now left at its
4051 default (and expected) setting.
4053 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4054 message, again when deliver_drop_privilege is set or Exim is run as root.
4055 The bug described in (1) was not present in this case, but the tidying up
4056 of the other signals was missing. I have made the two cases consistent.
4058 3. The ignore_target_hosts setting on a manualroute router was being ignored
4059 for hosts that were looked up using the /MX notation.
4061 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4064 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4065 operated on the sender address. After changing the $sender_address to <>
4066 for the sender address verify, Exim was re-instated it as the original
4067 (before rewriting) address, but remembering that it had rewritten it, so it
4068 wasn't rewriting it again. This bug also had the effect of breaking the
4069 sender address verification caching when the sender address was rewritten.
4071 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4072 This has been changed so that if the ip literal address matches
4073 ignore_target_hosts, the router declines.
4075 7. Added expansion conditions match_domain, match_address, and match_local_
4076 part (NOT match_host).
4078 8. The placeholder for the Received: header didn't have a length field set.
4080 9. Added code to Exim itself and to exim_lock to test for a specific race
4081 condition that could lead to file corruption when using MBX delivery. The
4082 issue is with the lockfile that is created in /tmp. If this file is removed
4083 after a process has opened it but before that process has acquired a lock,
4084 there is the potential for a second process to recreate the file and also
4085 acquire a lock. This could lead to two Exim processes writing to the file
4086 at the same time. The added code performs the same test as UW imapd; it
4087 checks after acquiring the lock that its file descriptor still refers to
4088 the same named file.
4090 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4091 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4094 11. Added the smtp_active_hostname option. If used, this will typically be made
4095 to depend on the incoming interface address. Because $interface_address is
4096 not set up until the daemon has forked a reception process, error responses
4097 that can happen earlier (such as "too many connections") no longer contain
4100 12. If an expansion in a condition on a "warn" statement fails because a lookup
4101 defers, the "warn" statement is abandoned, and the next ACL statement is
4102 processed. Previously this caused the whole ACL to be aborted.
4104 13. Added the iplsearch lookup type.
4106 14. Added ident_timeout as a log selector.
4108 15. Added tls_certificate_verified as a log selector.
4110 16. Added a global option tls_require_ciphers (compare the smtp transport
4111 option of the same name). This controls incoming TLS connections.
4113 17. I finally figured out how to make tls_require_ciphers do a similar thing
4114 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4115 before starting the TLS session.
4117 18. Tabs are now shown as \t in -bP output.
4119 19. If the log selector return_path_on_delivery was set, Exim crashed when
4120 bouncing a message because it had too many Received: header lines.
4122 20. If two routers both had headers_remove settings, and the first one included
4123 a superfluous trailing colon, the final name in the first list and the
4124 first name in the second list were incorrectly joined into one item (with a
4125 colon in the middle).
4131 1. Added -C and -D options to the exinext utility, mainly to make it easier
4132 to include in the automated testing, but these could be helpful when
4133 multiple configurations are in use.
4135 2. The exinext utility was not formatting the output nicely when there was
4136 an alternate port involved in the retry record key, nor when there was a
4137 message id as well (for retries that were specific to a specific message
4138 and a specific host). It was also confused by IPv6 addresses, because of
4139 the additional colons they contain. I have fixed the IPv4 problem, and
4140 patched it up to do a reasonable job for IPv6.
4142 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4143 delivery, the log line now contains "pipelined" if PIPELINING was used.
4145 4. An SMTP transport process used to panic and die if the bind() call to set
4146 an explicit outgoing interface failed. This has been changed; it is now
4147 treated in the same way as a connect() failure.
4149 5. A reference to $sender_host_name in the part of a conditional expansion
4150 that was being skipped was still causing a DNS lookup. This no longer
4153 6. The def: expansion condition was not recognizing references to header lines
4154 that used bh_ and bheader_.
4156 7. Added the _cache feature to named lists.
4158 8. The code for checking quota_filecount in the appendfile transport was
4159 allowing one more file than it should have been.
4161 9. For compatibility with Sendmail, the command line option
4169 and sets the incoming protocol and host name (for trusted callers). The
4170 host name and its colon can be omitted when only the protocol is to be set.
4171 Note the Exim already has two private options, -pd and -ps, that refer to
4172 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4173 "s", but I don't think that's a major issue.
4175 10. A number of refactoring changes to the code, none of which should affect
4178 (a) The number of logging options was getting close to filling up the
4179 32-bit word that was used as a bit map. I have split them into two classes:
4180 those that are passed in the argument to log_write(), and those that are
4181 only ever tested independently outside of that function. These are now in
4182 separate 32-bit words, so there is plenty of room for expansion again.
4183 There is no change in the user interface or the logging behaviour.
4185 (b) When building, for example, log lines, the code previously used a
4186 macro that called string_cat() twice, in order to add two strings. This is
4187 not really sufficiently general. Furthermore, there was one instance where
4188 it was actually wrong because one of the argument was used twice, and in
4189 one call a function was used. (As it happened, calling the function twice
4190 did not affect the overall behaviour.) The macro has been replaced by a
4191 function that can join an arbitrary number of extra strings onto a growing
4194 (c) The code for expansion conditions now uses a table and a binary chop
4195 instead of a serial search (which was left over from when there were very
4196 few conditions). Also, it now recognizes conditions like "pam" even when
4197 the relevant support is not compiled in: a suitably worded error message is
4198 given if an attempt is made to use such a condition.
4200 11. Added ${time_interval:xxxxx}.
4202 12. A bug was causing one of the ddress fields not to be passed back correctly
4203 from remote delivery subprocesses. The field in question was not being
4204 subsequently used, so this caused to problems in practice.
4206 13. Added new log selectors queue_time and deliver_time.
4208 14. Might have fixed a bug in maildirsizefile handling that threw up
4209 "unexpected character" debug warnings, and recalculated the data
4210 unnecessarily. In any case, I expanded the warning message to give more
4213 15. Added the message "Restricted characters in address" to the statements in
4214 the default ACL that block characters like @ and % in local parts.
4216 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4217 Three changes have been made:
4219 (a) There was a serious bug; a negative response to MAIL caused the whole
4220 recipient domain to be cached as invalid, thereby blocking all messages
4221 to all local parts at the same domain, from all senders. This bug has
4222 been fixed. The domain is no longer cached after a negative response to
4223 MAIL if the sender used is not empty.
4225 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4228 (c) A new callout option, "use_sender" has been added for people who want
4229 the modified behaviour.
4235 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4238 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4239 indeed breaks things for older releases.
4241 3. Added additional logging to the case where there is a problem reading data
4242 from a filter that is running in a subprocess using a pipe, in order to
4243 try to track down a specific problem.
4245 4. Testing facility fudge: when running in the test harness and attempting
4246 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4247 getting "No route to host". Convert this to a timeout.
4249 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4252 6. Some OS don't have socklen_t but use size_t instead. This affects the
4253 fifth argument of getsockopt() amongst other things. This is now
4254 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4255 can be set for individual OS. I have set it for SunOS5, OSF1, and
4256 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4257 some earlier ones do not.
4259 7. Change 4.30/15 was not doing the test caselessly.
4261 8. The standard form for an IPv6 address literal was being rejected by address
4262 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4263 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4264 this, as well as the form without the "IPv6" on the front (but only when
4265 address literals are enabled, of course).
4267 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4269 10. Exim crashed if a message with an empty sender address specified by -f
4270 encountered a router with an errors_to setting. This could be provoked only
4271 by a command such as
4275 where an empty string was supplied; "<>" did not hit this bug.
4277 11. Installed PCRE release 4.5.
4279 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4280 remained set. It is now erased.
4282 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4283 times from message ids (which are base 36 rather than the normal 62).
4285 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4286 were being counted as actual protocol errors, and logged if the log
4287 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4288 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4289 and DATA following a set of rejected RCPTs do not count as protocol errors.
4290 In other words, Exim assumes they were pipelined, though this may not
4291 actually be the case. Of course, in all cases the client gets an
4292 appropriate error code.
4294 15. If a lookup fails in an ACL condition, a message about the failure may
4295 be available; it is used if testing the ACL cannot continue, because most
4296 such messages specify what the cause of the deferral is. However, some
4297 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4298 that caused an old message to be retained and used if a later statement
4299 caused a defer, replacing the real cause of the deferral.
4301 16. If an IP address had so many PTR records that the DNS lookup buffer
4302 was not large enough to hold them, Exim could crash while trying to process
4303 the truncated data. It now detects and logs this case.
4305 17. Further to 4.21/58, another change has been made: if (and only if) the
4306 first line of a message (the first header line) ends with CRLF, a bare LF
4307 in a subsequent header line has a space inserted after it, so as not to
4308 terminate the header.
4310 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4311 unnecessarily, used atoi() instead of strtol(), and didn't check the
4312 termination when getting file sizes from file names by regex.
4314 19. Completely re-implemented the support for maildirsize files, in the light
4315 of a number of problems with the previous contributed implementation
4316 (4.30/29). In particular:
4318 . If the quota is zero, the maildirsize file is maintained, but no quota is
4321 . If the maildir directory does not exist, it is created before any attempt
4322 to write a maildirsize file.
4324 . The quota value in the file is just a cache; if the quota is changed in
4325 the transport, the new value overrides.
4327 . A regular expression is available for excluding directories from the
4330 20. The autoreply transport checks the characters in options that define the
4331 message's headers; it allows continued headers, but it was checking with
4332 isspace() after an embedded newline instead of explicitly looking for a
4335 21. If all the "regular" hosts to which an address was routed had passed their
4336 expiry times, and had not reached their retry times, the address was
4337 bounced, even if fallback hosts were defined. Now Exim should go on to try
4340 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4341 equivalent code in the SMTP transport. Some hosts send humungous responses
4342 to HELO/EHLO, more than 1024 it seems.
4344 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4345 gives compiler warnings in some environments. I've now done it "properly",
4348 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4349 (because of problems with the built-in one) was declared to return uschar *
4350 instead of char *, causing compiler failure.
4352 25. Fixed a file descriptor leak when processing alias/forward files.
4354 26. Fixed a minor format string issue in dbfn.c.
4356 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4358 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4359 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4362 29. When a sender address is verified, it is cached, to save repeating the test
4363 when there is more than one recipient in a message. However, when the
4364 verification involves a callout, it is possible for different callout
4365 options to be set for different recipients. It is too complicated to keep
4366 track of this in the cache, so now Exim always runs a verification when a
4367 callout is required, relying on the callout cache for the optimization.
4368 The overhead is duplication of the address routing, but this should not be
4371 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4372 to be verified with callout=postmaster, and the main callout worked but the
4373 postmaster check failed, the verification correctly failed. However, if a
4374 subsequent RCPT command asked for sender verification *without* the
4375 postmaster check, incorrect caching caused this verification also to fail,
4378 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4379 it was not caching the DNS options (qualify_single, search_parents) that
4380 were used when the lookup failed. A subsequent lookup with different
4381 options therefore always gave the same answer, though there were cases
4382 where it should not have. (Example: a "domains = !$mx_any" option on a
4383 dnslookup router: the "domains" option is always processed without any
4384 widening, but the router might have qualify_single set.) Now Exim uses the
4385 cached value only when the same options are set.
4387 32. Added John Jetmore's "exipick" utility to the distribution.
4389 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4390 than a timeout (e.g. a certificate is required, and is not provided), an
4391 Exim server now closes the connection immediately. Previously it waited for
4392 the client to close - but if the client is SSL, it seems that they each
4393 wait for each other, leading to a delay before one of them times out.
4395 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4396 maintained 0.8.x compatibility because I don't think many are using it, and
4397 it is clearly obsolete.
4399 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4402 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4403 client certificate was expired. A simple patch fixes this, though I don't
4404 understand the full logic of why the verify callback is called multiple
4407 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4408 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4409 which causes problems with some clients (such as the Certicom SSL Plus
4410 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4411 disables the coutermeasure allowing Eudora to connect."
4413 38. Exim was not checking that a write() to a log file succeeded. This could
4414 lead to Bad Things if a log got too big, in particular if it hit a file
4415 size limit. Exim now panics and dies if it cannot write to a log file, just
4416 as it does if it cannot open a log file.
4418 39. Modified OS/Makefile-Linux so that it now contains
4420 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4422 The two -D definitions ensure that Exim is compiled with large file
4423 support, which makes it possible to handle log files that are bigger than
4426 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4427 instance) a domain was checked against a named list that involved a lookup,
4428 causing $domain_data to be set, then another domain was checked against the
4429 same list, then the first domain was re-checked, the value of $domain_data
4430 after the final check could be wrong. In particular, if the second check
4431 failed, it could be set empty. This bug probably also applied to
4434 41. The strip_trailing_dot option was not being applied to the address given
4435 with the -f command-line option.
4437 42. The code for reading a message's header from the spool was incrementing
4438 $received_count, but never initializing it. This meant that the value was
4439 incorrect (doubled) while delivering a message in the same process in which
4440 it was received. In the most common configuration of Exim, this never
4441 happens - a fresh exec is done - but it can happen when
4442 deliver_drop_privilege is set.
4444 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4445 it now includes up to 150 characters of the unexpected data in the log
4448 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4449 and building data strings. The size of both of these buffers was 10 000
4450 bytes - far larger than anybody would *ever* want, thought I. Needless to
4451 say, somebody hit the limit. I have increased the maximum line length to
4452 20 000 and the maximum data length of concatenated lines to 100 000. I have
4453 also fixed two bugs, because there was no checking on these buffers. Tsk,
4454 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4455 buffer is too small.
4457 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4458 lsearch lookups. Now it does.
4460 46. When parsing a route_list item in a manualroute router, a fixed-length
4461 buffer was used for the list of hosts. I made this 1024 bytes long,
4462 thinking that nobody would ever have a list of hosts that long. Wrong.
4463 Somebody had a whole pile of complicated expansion conditions, and the
4464 string was silently truncated, leading to an expansion error. It turns out
4465 that it is easier to change to an unlimited length (owing to other changes
4466 that have happened since this code was originally written) than to build
4467 structure for giving a limitation error. The length of the item that
4468 expands into the list of hosts is now unlimited.
4470 47. The lsearch lookup could not handle data where the length of text line was
4471 more than 4095 characters. Such lines were truncated, leading to shortened
4472 data being returned. It should now handle lines of any length.
4474 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4475 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4478 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4479 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4480 no longer happens in comment lines. A list of replacements is now placed
4481 at the head of all of the source files, except those whose only change is
4482 to replace PERL_COMMAND in the very first #! line.
4484 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4485 messages on the queue, with a bottom-up merge sort, using code contributed
4486 by Michael Haardt. This should make operations like -bp somewhat faster on
4487 large queues. It won't affect queue runners, except when queue_run_in_order
4490 51. Installed eximstats 1.31 in the distribution.
4492 52. Added support for SRV lookups to the dnslookup router.
4494 53. If an ACL referred to $message_body or $message_body_end, the value was not
4495 reset for any messages that followed in the same SMTP session.
4497 54. The store-handling optimization for building very long strings was not
4498 differentiating between the different store pools. I don't think this
4499 actually made any difference in practice, but I've tidied it.
4501 55. While running the routers to verify a sender address, $sender_address
4502 was still set to the sender address. This is wrong, because when routing to
4503 send a bounce to the sender, it would be empty. Therefore, I have changed
4504 it so that, while verifying a sender address, $sender_address is set to <>.
4505 (There is no change to what happens when verifying a recipient address.)
4507 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4508 target A or AAAA records (if not already returned) without resetting the
4509 qualify_single or search_parents options of the DNS resolver. These are
4510 inappropriate in this case because the targets of MX and SRV records must
4511 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4512 target that, when qualified, matched something in the local domain. These
4513 two options are now turned off when doing these lookups.
4515 57. It seems that at least some releases of Reiserfs (which does not have the
4516 concept of a fixed number of inodes) returns zero and not -1 for the
4517 number of available inodes. This interacted badly with check_spool_inodes,
4518 which assumed that -1 was the "no such thing" setting. What I have done is
4519 to check that the total number of inodes is greater than zero before doing
4520 the test of how many are available.
4522 58. When a "warn" ACL statement has a log_message modifier, the message is
4523 remembered, and not repeated. This is to avoid a lot of repetition when a
4524 message has many recipients that cause the same warning to be written.
4525 Howewer, Exim was preserving the list of already written lines for an
4526 entire SMTP session, which doesn't seem right. The memory is now reset if a
4527 new message is started.
4529 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4530 debugging output unless log_rewrite was also set.
4532 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4533 of (int)(handle) when we know that handle contains (void *)(-1).
4535 61. The Exim daemon panic-logs an error return when it closes the incoming
4536 connection. However "connection reset by peer" seems to be common, and
4537 isn't really an error worthy of noting specially, so that particular error
4540 62. When Exim is trying to find all the local interfaces, it used to panic and
4541 die if the ioctl to get the interface flags failed. However, it seems that
4542 on at least one OS (Solaris 9) it is possible to have an interface that is
4543 included in the list of interfaces, but for which you get a failure error
4544 for this call. This happens when the interface is not "plumbed" into a
4545 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4546 failure of the "get flags" call assumes that the interface is down.
4548 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4549 makes life easier for people who are doing arithmetic on fields extracted
4550 from dates, where you often get leading zeros that should not be
4551 interpreted as octal.
4553 64. Added qualify_domain to the redirect router, to override the global
4556 65. If a pathologically long header line contained very many addresses (the
4557 report of this problem mentioned 10 000) and each of them was rewritten,
4558 Exim could use up a very large amount of memory. (It kept on making new
4559 copies of the header line as it rewrote, and never released the old ones.)
4560 At the expense of a bit more processing, the header rewriting function has
4561 been changed so that it no longer eats memory in this way.
4563 66. The generation of the Received: header has been moved from the time that a
4564 message starts to be received, to the time that it finishes. The timestamp
4565 in the Received: header should now be very close to that of the <= log
4566 line. There are two side-effects of this change:
4568 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4569 logged header lines no longer include the local Received: line, because
4570 it has not yet been created. The same applies to a copy of the message
4571 that is returned to a non-SMTP sender when a message is rejected.
4573 (b) When a filter file is tested using -bf, no additional Received: header
4574 is added to the test message. After some thought, I decided that this
4577 This change does not affect the value of $received_for. It is still set
4578 after address rewriting, but before local_scan() is called.
4580 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4582 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4583 gave an unhelpful panic error message, and a defer error. I have managed to
4584 change this behaviour so that it now rejects any supplied certificate,
4585 which seems right, as the list of acceptable certificates is empty.
4587 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4588 gave an unhelpful defer error. I have not managed to make this reject any
4589 supplied certificates, but the error message it gives is "no certificate
4590 supplied", which is not helpful.
4592 70. exigrep's output now also includes lines that are not associated with any
4593 message, but which match the given pattern. Implemented by a patch from
4594 Martin Sluka, which also tidied up the Perl a bit.
4596 71. Recipient callout verification, like sender verification, was using <> in
4597 the MAIL FROM command. This isn't really the right thing, since the actual
4598 sender may affect whether the remote host accepts the recipient or not. I
4599 have changed it to use the actual sender in the callout; this means that
4600 the cache record is now keyed on a recipient/sender pair, not just the
4601 recipient address. There doesn't seem to be a real danger of callout loops,
4602 since a callout by the remote host to check the sender would use <>.
4603 [SEE ABOVE: changed after hitting problems.]
4605 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4606 temporary errors. However, in the case of such a code being given after
4607 the end of a data transmission (i.e. after ".") Exim was failing to write
4608 a retry record for the message. (Yes, there was some broken host that was
4609 actually sending 8xx at this point.)
4611 73. An unknown lookup type in a host list could cause Exim to panic-die when
4612 the list was checked. (An example that provoked this was putting <; in the
4613 middle of a list instead of at the start.) If this happened during a DATA
4614 ACL check, a -D file could be left lying around. This kind of configuration
4615 error no longer causes Exim to die; instead it causes a defer errror. The
4616 incident is still logged to the main and panic logs.
4618 74. Buglet left over from Exim 3 conversion. The message "too many messages
4619 in one connection" was written to the rejectlog but not the mainlog, except
4620 when address rewriting (yes!) was being logged.
4622 75. Added write_rejectlog option.
4624 76. When a system filter was run not as root (that is, when system_filter_user
4625 was set), the values of the $n variables were not being returned to the
4626 main process; thus, they were not subsequently available in the $sn
4629 77. Added +return_path_on_delivery log selector.
4631 78. A connection timeout was being treated differently from recipients deferred
4632 when testing hosts_max_try with a message that was older than the host's
4633 retry timeout. (The host should not be counted, thus allowing all hosts to
4634 be tried at least once before bouncing.) This may have been the cause of an
4635 occasionally reported bug whereby a message would remain on the queue
4636 longer than the retry timeout, but would be bounced if a delivery was
4637 forced. I say "may" because I never totally pinned down the problem;
4638 setting up timeout/retry tests is difficult. See also the next item.
4640 79. The ultimate address timeout was not being applied to errors that involved
4641 a combination of host plus message (for example, a timeout on a MAIL
4642 command). When an address resolved to a number of possible hosts, and they
4643 were not all tried for each delivery (e.g. because of hosts_max_try), a
4644 message could remain on the queue longer than the retry timeout.
4646 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4649 81. Fixed an obscure SMTP outgoing bug which required at least the following
4650 conditions: (a) there was another message waiting for the same server;
4651 (b) the server returned 5xx to all RCPT commands in the first message so
4652 that the message was not completed; (c) the server dropped the connection
4653 or gave a negative response to the RSET that Exim sends to abort the
4654 transaction. The observed case was a dropped connection after DATA that had
4655 been sent in pipelining mode. That is, the server had advertised PIPELINING
4656 but was not implementing it correctly. The effect of the bug was incorrect
4657 behaviour, such as trying another host, and this could lead to a crash.
4663 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4664 and daemon.c were passed as pointers to ints; they should have been
4665 pointers to socklen_t variables (which are typically unsigned ints).
4667 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4670 3. Fixed a really odd bug that affected only the testing scheme; patching a
4671 certain fixed string in the binary changed the value of another string that
4672 happened to be identical to the end of the original first string.
4674 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4675 name", it returns that address as the IP address. On some operating
4676 systems (e.g. Solaris), it also passes back the IP address string as the
4677 "host name". However, on others (e.g. Linux), it passes back an empty
4678 string. Exim wasn't checking for this, and was changing the host name to an
4679 empty string, assuming it had been canonicized.
4681 5. Although rare, it is permitted to have more than one PTR record for a given
4682 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4683 all the names associated with an address, because they do in Solaris.
4684 However, it seems that they do not in Linux for data that comes from the
4685 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4686 I found this out when I moved to a new Linux workstation and tried to run
4687 the Exim test suite.
4689 To get round this problem I have changed the code so that it now does its
4690 own call to the DNS to look up PTR records when searching for a host name.
4691 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4692 addresses that are only in /etc/hosts are still found.
4694 This behaviour is, however, controlled by an option called host_lookup_
4695 order, which defaults to "bydns:byaddr". If people want to use the other
4696 order, or indeed, just use one or the other means of lookup, they can
4697 specify it in this variable.
4699 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4700 some operating systems, this comes back from gethostbyaddr() as an empty
4701 string, and this is what Exim used to test for. However, it seems that in
4702 other systems, "." is yielded. Exim now tests for this case too.
4704 7. The values of check_spool_space and check_log_space are now held internally
4705 as a number of kilobytes instead of an absolute number of bytes. If a
4706 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4707 kilobyte. This means that much larger values can be stored.
4709 8. Exim monitor: an attempt to get the action menu when not actually pointing
4710 at a message produces an empty menu entitled "No message selected". This
4711 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4712 no entries in it ("Shell widget menu has zero width and/or height"). So I
4713 have added a single, blank menu entry in this case.
4715 9. Added ${quote_local_part.
4717 10. MIME decoding is now applied to the contents of Subject: header lines when
4720 11. Now that a reference to $sender_host_address automatically causes a reverse
4721 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4722 host lookup before query-style lookups in lists that might use this
4723 variable. This has therefore been abolished, and the "net-" prefix is no
4724 longer necessary for query-style lookups.
4726 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4727 have been a typo for LFLAGS, so it has been changed.
4729 13. The install script calls Exim with "-C /dev/null" in order to find the
4730 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4731 to be output. Howeve, since Exim outputs its version number before the
4732 error, it didn't break the script. It just looked ugly. I fixed this by
4733 always allowing "-C /dev/null" if the caller is root.
4735 14. Ignore overlarge ACL variable number when reading spool file - insurance
4736 against a later release with more variables having written the file.
4738 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4739 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4740 this, as well as the form without the "IPv6" on the front.
4742 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4743 OS/Makefile-Darwin file.
4745 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4746 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4748 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4749 variable. However, some OS use ints for inodes. I've added cast to long int
4750 to get rid of the compiler warning.
4752 19. I had forgotten to lock out "/../" in configuration file names when
4753 ALT_CONFIG_PREFIX was set.
4755 20. Routers used for verification do not need to specify transports. However,
4756 if such a router generated a host list, and callout was configured, Exim
4757 crashed, because it could not find a port number from the (non-existent)
4758 transport. It now assumes port 25 in this circumstance.
4760 21. Added the -t option to exigrep.
4762 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4763 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4764 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4765 them is set, LOOKUP_LSEARCH is forced.
4767 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4769 24. Added sender and host information to the "rejected by local_scan()" log
4770 line; previously there was no indication of these.
4772 25. Added .include_if_exists.
4774 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4775 Linux. It turns out that not all file systems support this. Apparently some
4776 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4777 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4778 not supported on the file descriptor, is now ignored when Exim is trying to
4779 sync a directory. This applies only to Linux.
4781 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4783 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4784 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4787 29. Added support for maildirsize files from supplied patch (modified a bit).
4789 30. The use of :fail: followed by an empty string could lead Exim to respond to
4790 sender verification failures with (e.g.):
4792 550 Verification failed for <xxx>
4793 550 Sender verify failed
4795 where the first response line was missing the '-' that indicates it is not
4796 the final line of the response.
4798 31. The loop for finding the name of the user that called Exim had a hardwired
4799 limit of 10; it now uses the value of finduser_retries, which is used for
4800 all other user lookups.
4802 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4805 33. Exim was neglecting to zero errno before one call of strtol() when
4806 expanding a string and expecting an integer value. On some systems this
4807 resulted in spurious "integer overflow" errors. Also, it was casting the
4808 result into an int without checking.
4810 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4811 did not work. The code looks as if it has *never* worked, though it appears
4812 to have been documented since at least releast 1.62. I have made it work.
4814 35. The "timeout_DNS" error in retry rules, also documented since at least
4815 1.62, also never worked. As it isn't clear exactly what this means, and
4816 clearly it isn't a major issue, I have abolished the feature by treating it
4817 as "timeout", and writing a warning to the main and panic logs.
4819 36. The display of retry rules for -brt wasn't always showing the error code
4822 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4823 timeout_connect_A, timeout_connect_MX.
4825 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4826 to the empty sender.
4828 39. The daemon was not analysing the content of -oX till after it had closed
4829 stderr and disconnected from the controlling terminal. This meant that any
4830 syntax errors were only noted on the panic log, and the return code from
4831 the command was 0. By re-arranging the code a little, I've made the
4832 decoding happen first, so such errors now appear on stderr, and the return
4833 code is 1. However, the actual setting up of the sockets still happens in
4834 the disconnected process, so errors there are still only recorded on the
4837 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4838 connections (as happens on some IP stacks) was logged at start up time as
4839 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4840 it from "IPv6 and IPv4", which means that two separate sockets are being
4843 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4844 says whether AF_INET or AF_INET6 was passed as an argument.
4846 42. Exiwhat output was messed up when time zones were included in log
4849 43. Exiwhat now gives more information about the daemon's listening ports,
4850 and whether -tls-on-connect was used.
4852 44. The "port" option of the smtp transport is now expanded.
4854 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4855 silently ignored. Now an error message is written to the main and panic
4858 46. There's a new ACL modifier called "logwrite" which writes to a log file
4859 as soon as it is encountered.
4861 47. Added $local_user_uid and $local_user_gid at routing time.
4863 48. Exim crashed when trying to verify a sender address that was being
4866 49. Exim was recognizing only a space character after ".include". It now also
4867 recognizes a tab character.
4869 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4870 extracting the relevant information from the specification. The man page no
4871 longer contains scrambled data for the -d option, and I've added a section
4872 at the front about calling Exim under different names.
4874 51. Added "extra_headers" argument to the "mail" command in filter files.
4876 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4879 53. Installed eximstats 1.29.
4881 54. Added transport_filter_timeout as a generic transport option.
4883 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4884 Cc: header lines. This was required by RFC 822, but it not required by RFC
4887 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4888 incoming messages that did not have them. Now it does so only if the
4889 message originates locally, that is, if there is no associated remote host
4890 address. When Resent- header lines are present, this applies to the Resent-
4891 lines rather than the non-Resent- lines.
4893 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4894 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4896 58. Messages for configuration errors now include the name of the main
4897 configuration file - useful now that there may be more than one file in a
4898 list (.included file names were always shown).
4900 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4901 for those rare installations that do not start the daemon as root or run it
4902 setuid root. I've cut out the call to initgroups() if the daemon is not
4905 60. The Exim user and group can now be bound into the binary as text strings
4906 that are looked up at the start of Exim's processing.
4908 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4910 62. Added $mailstore_basename variable.
4912 63. Installed patch to sieve.c from Michael Haardt.
4914 64. When Exim failed to open the panic log after failing to open the main log,
4915 the original message it was trying to log was written to stderr and debug
4916 output, but if they were not available (the usual case in production), it
4917 was lost. Now it is written to syslog before the two lines that record the
4918 failures to open the logs.
4920 65. Users' Exim filters run in subprocesses under the user's uid. It is
4921 possible for a "deliver" command or an alias in a "personal" command to
4922 provoke an address rewrite. If logging of address rewriting is configured,
4923 this fails because the process is not running as root or exim. There may be
4924 a better way of dealing with this, but for the moment (because 4.30 needs
4925 to be released), I have disabled address rewrite logging when running a
4926 filter in a non-root, non-exim process.
4932 1. The buildconfig auxiliary program wasn't quoting the value set for
4933 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4934 not defined. This bug was masked in 4.22 by the effect that was fixed in
4937 2. Some messages that were rejected after a message id was allocated were
4938 shown as "incomplete" by exigrep. It no longer does this for messages that
4939 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4941 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4942 have allow_domain_literals set, the ID did not get logged in the <= line.
4943 Domain literals are now always recognized in Message-ID: header lines.
4945 4. The first argument for a ${extract expansion item is the key name or field
4946 number. Leading and trailing spaces in this item were not being ignored,
4947 causing some misleading effects.
4949 5. When deliver_drop_privilege was set, single queue runner processes started
4950 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4951 same command in the process it spins off) were not dropping privilege.
4953 6. When the daemon running as "exim" started a queue runner, it always
4954 re-executed Exim in the spun-off process. This is a waste of effort when
4955 deliver_drop_privilege is set. The new process now just calls the
4956 queue-runner function directly.
4962 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4965 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4966 ignored. Though the use of -oP was forcing the writing of a pid file, it
4967 was always written to the default place.
4969 3. If the message "no IP address found for host xxxx" is generated during
4970 incoming verification, it is now followed by identification of the incoming
4971 connection (so you can more easily find what provoked it).
4973 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4975 5. Added some features to "harden" Exim a bit more against certain attacks:
4977 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4978 be put in Local/Makefile. This is like the never_users runtime option,
4979 but it cannot be overridden. The default setting is "root".
4981 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4982 prefix string with which any file named in a -C command line option
4985 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4986 is retained for -C and -D only if the caller of Exim is root. Without
4987 it, the exim user may also use -C and -D and retain privilege.
4989 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4990 command line option is disabled.
4992 6. Macro names set by the -D option must start with an upper case letter, just
4993 like macro names defined in the configuration file.
4995 7. Added "dereference=" facility to LDAP.
4997 8. Two instances of the typo "uknown" in the source files are fixed.
4999 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5000 the Configure-Makefile script screwed up while processing it.
5002 10. Incorporated PCRE 4.4.
5004 11. The SMTP synchronization check was not operating right at the start of an
5005 SMTP session. For example, it could not catch a HELO sent before the client
5006 waited for the greeting. There is now a check for outstanding input at the
5007 point when the greeting is written. Because of the duplex, asynchronous
5008 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5009 way, but not yet received, when the check is performed.
5011 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5012 on TCP/IP sockets, because this apparently causes some broken clients to
5015 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5016 unchanged) from the Cygwin maintainer.
5018 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5019 was supported instead of testing for mbx. Effectively a typo.
5021 15. The spa authenticator server code was not checking that the input it
5022 received was valid base64.
5024 16. The debug output line for the "set" modifier in ACLs was not showing the
5025 name of the variable that was being set.
5027 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5029 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5030 lookup on its own. Something else was needed to trigger the lookup. For
5031 example, a match in host_lookup or the need for a host name in a host list.
5032 Now, if $sender_host_name is referenced and the host name has not yet been
5033 looked up, a lookup is performed. If the lookup fails, the variable remains
5034 empty, and $host_lookup_failed is set to "1".
5036 19. Added "eqi" as a case-independent comparison operator.
5038 20. The saslauthd authentication condition could segfault if neither service
5039 nor realm was specified.
5041 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5042 error message that was logged was misleading, and incoming SMTP
5043 connections were dropped. The message is now more accurate, and temporary
5044 errors are given to SMTP connections.
5046 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5047 (except RSET) with a 421 error, until QUIT is received. However, it was
5048 failing to send a response to QUIT.
5050 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5051 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5052 ACL is now run after the helo_try_verify_hosts code.
5054 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5055 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5056 case-independent, so other case variants are also recognized). Apparently
5057 some systems use these upper case variants.
5059 25. If more than two messages were waiting for the same host, and a transport
5060 filter was specified for the transport, Exim sent two messages over the
5061 same TCP/IP connection, and then failed with "socket operation on non-
5062 socket" when it tried to send the third.
5064 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5066 27. The extern definition of crypt16() in expand.c was not being excluded when
5067 the OS had its own crypt16() function.
5069 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5070 as a preferred synonym for return_size_limit, both as an option and as an
5073 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5075 30. Changed the default configuration ACL to relax the local part checking rule
5076 for addresses that are not in any local domains. For these addresses,
5077 slashes and pipe symbols are allowed within local parts, but the sequence
5078 /../ is explicitly forbidden.
5080 31. SPA server authentication was not clearing the challenge buffer before
5083 32. log_message in a "warn" ACL statement was writing to the reject log as
5084 well as to the main log, which contradicts the documentation and doesn't
5085 seem right (because no rejection is happening). So I have stopped it.
5087 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5088 However, I am unable to do any testing of this.
5090 34. Fixed an infelicity in the appendfile transport. When checking directories
5091 for a mailbox, to see if any needed to be created, it was accidentally
5092 using path names with one or more superfluous leading slashes; tracing
5093 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5095 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5096 contents are added to the log line that is written for every discarded
5097 recipient. (Previously a log_message setting was ignored.)
5099 36. The ${quote: operator now quotes the string if it is empty.
5101 37. The install script runs exim in order to find its version number. If for
5102 some reason other than non-existence or emptiness, which it checks, it
5103 could not run './exim', it was installing it with an empty version number,
5104 i.e. as "exim-". This error state is now caught, and the installation is
5107 38. An argument was missing from the function that creates an error message
5108 when Exim fails to connect to the socket for saslauthd authentication.
5109 This could cause Exim to crash, or give a corrupted message.
5111 39. Added isip, isip4, and isip6 to ${if conditions.
5113 40. The ACL variables $acl_xx are now saved with the message, and can be
5114 accessed later in routers, transports, and filters.
5116 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5117 strings in the file are not string-expanded.
5119 42. If a MAIL command specified a SIZE value that was too large to fit into an
5120 int variable, the check against message_size_limit failed. Such values are
5121 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5122 day this will have to be increased, but I don't think I want to be around
5123 when emails are that large.
5130 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5131 iconv() is not standard in FreeBSD.
5133 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5134 IPv6 enabled. The observed symptom was a segmentation fault on return from
5135 the function os_common_find_running_interfaces() in src/os.c.
5137 3. In the check_special_case() function in daemon.c I had used "errno" as an
5138 argument name, which causes warnings on some systems. This was basically a
5139 typo, since it was named "eno" in the comments!
5141 4. The code that waits for the clock to tick (at a resolution of some fraction
5142 of a second) so as to ensure message-id uniqueness was always waiting for
5143 at least one whole tick, when it could have waited for less. [This is
5144 almost certainly not relevant at current processor speeds, where it is
5145 unlikely to ever wait at all. But we try to future-proof.]
5147 5. The function that sleeps for a time interval that includes fractions of a
5148 second contained a race. It did not block SIGALRM between setting the
5149 timer, and suspending (a couple of lines later). If the interval was short
5150 and the sigsuspend() was delayed until after it had expired, the suspension
5151 never ended. On busy systems this could lead to processes getting stuck for
5154 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5155 process, before it forks any delivery processes. The open lookup caching
5156 mechanism meant that the open file or database connection was passed into
5157 the delivery process. The problem was that delivery processes always tidy
5158 up cached lookup data. This could cause a problem for the next delivery
5159 process started by the queue runner, because the external queue runner
5160 process does not know about the closure. So the next delivery process
5161 still has data in the lookup cache. In the case of a file lookup, there was
5162 no problem because closing a file descriptor in a subprocess doesn't affect
5163 the parent. However, if the lookup was caching a connection to a database,
5164 the connection was closed, and the second delivery process was likely to
5165 see errors such as "PGSQL: query failed: server closed the connection
5166 unexpectedly". The problem has been fixed by closing all cached lookups
5167 in a queue runner before running a delivery process.
5169 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5170 seem to have the "const" qualifier which it has on other OS. I've
5173 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5174 *of the same type* (client or server) with the same public name that an
5175 error should be diagnosed.
5177 9. When Exim looked up a host name for an IP address, but failed to find the
5178 original IP address when looking up the host name (a safety check), it
5179 output the message "<ip address> does not match any IP for NULL", which was
5180 confusing, to say the least. The bug was that the host name should have
5181 appeared instead of "NULL".
5183 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5184 user that is built into the binary, and the -C or -D options is used, root
5185 privilege is dropped before the configuration file is read. In addition,
5186 logging is switched to stderr instead of the normal log files. If the
5187 configuration then re-defines the Exim user, the unprivileged environment
5188 is probably not what is expected, so Exim logs a panic warning message (but
5191 However, if deliver_drop_privilege is set, the unprivileged state may well
5192 be exactly what is intended, so the warning has been cut out in that case,
5193 and Exim is allowed to try to write to its normal log files.
5199 1. smtp_return_error_details was not giving details for temporary sender
5200 or receiver verification errors.
5202 2. Diagnose a configuration error if two authenticators have the same public
5205 3. Exim used not to create the message log file for a message until the first
5206 delivery attempt. This could be confusing when incoming messages were held
5207 for policy or load reasons. The message log file is now created at the time
5208 the message is received, and an initial "Received" line is written to it.
5210 4. The automatically generated man page for command line options had a minor
5211 bug that caused no ill effects; however, a more serious problem was that
5212 the procedure for building the man page automatically didn't always
5213 operate. Consequently, release 4.20 contains an out-of-date version. This
5214 shouldn't happen again.
5216 5. When building Exim with embedded Perl support, the script that builds the
5217 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5218 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5220 6. The freeze_tell option was not being used for messages that were frozen on
5221 arrival, either by an ACL or by local_scan().
5223 7. Added the smtp_incomplete_transaction log selector.
5225 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5226 was accepting AUTH without a new EHLO.
5228 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5229 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5232 10. Logging of TCP/IP connections (when configured) now happens in the main
5233 daemon process instead of the child process, so that the TCP/IP connection
5234 count is more accurate (but it can never be perfect).
5236 11. The use of "drop" in a nested ACL was not being handled correctly in the
5237 outer ACL. Now, if condition failure induced by the nested "drop" causes
5238 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5239 or "require"), the connection is dropped.
5241 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5242 that yield "discard" can now be used with an "accept" or a "discard" verb,
5243 but an error is generated for any others (because I can't see a useful way
5244 to define what should happen).
5246 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5247 are now processed in the same way as lines in the Exim configuration file.
5248 In particular, continuation lines are supported.
5250 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5252 15. Added -ti meaning -t -i.
5254 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5255 domains, and warn by logging if others are found.
5257 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5258 returned from GIFCONF when Exim is trying to find the list of interfaces on
5259 a host. The code in os.c has been modified to copy each ifreq to an aligned
5260 structure in all cases.
5262 Also, in some cases, the returned ifreq's were being copied to a 'struct
5263 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5264 means the last couple of bytes of an IPv6 address could be chopped if the
5265 ifreq contained only a normal sockaddr (14 bytes storage).
5267 18. Named domain lists were not supported in the hosts_treat_as_local option.
5268 An entry such as +xxxx was not recognized, and was treated as a literal
5271 19. Ensure that header lines added by a DATA ACL are included in the reject log
5272 if the ACL subsequently rejects the message.
5274 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5275 MD5 (which is deprecated).
5277 21. When testing a filter file using -bf, Exim was writing a message when it
5278 took the sender from a "From " line in the message, but it was not doing so
5279 when it took $return_path from a Return-Path: header line. It now does.
5281 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5282 with a valid header line field name (a series of printing characters
5283 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5285 23. Changed "disc" in the source to "disk" to conform to the documentation and
5286 the book and for uniformity.
5288 24. Ignore Sendmail's -Ooption=value command line item.
5290 25. When execve() failed while trying to run a command in a pipe transport,
5291 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5292 could be confused with a return value of 69 from the command itself. This
5293 has been changed to 127, the value the shell returns if it is asked to run
5294 a non-existent command. The wording for the related log line suggests a
5295 non-existent command as the problem.
5297 26. If received_header_text expands to an empty string, do not add a Received:
5298 header line to the message. (Well, it adds a token one on the spool, but
5299 marks it "old" so that it doesn't get used or transmitted.)
5301 27. Installed eximstats 1.28 (addition of -nt option).
5303 28. There was no check for failure on the call to getsockname() in the daemon
5304 code. This can fail if there is a shortage of resources on the system, with
5305 ENOMEM, for example. A temporary error is now given on failure.
5307 29. Contrary to the C standard, it seems that in some environments, the
5308 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5309 program. Exim now does this explicitly; it affects the formatting of
5310 timestamps using strftime().
5312 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5313 complaints. I've now initialized all the variables, to avoid this.
5315 32. Header lines added by a system filter were not being "seen" during
5316 transport-time rewrites.
5318 33. The info_callback() function passed to OpenSSL is set up with type void
5319 (*)(SSL *, int, int), as described somewhere. However, when calling the
5320 function (actually a macro) that sets it up, the type void(*)() is
5321 expected. I've put in a cast to prevent warnings from picky compilers.
5323 34. If a DNS black list lookup found a CNAME record, but there were no A
5324 records associated with the domain it pointed at, Exim crashed.
5326 35. If a DNS black list lookup returned more than one A record, Exim ignored
5327 all but the first. It now scans all returned addresses if a particular IP
5328 value is being sought. In this situation, the contents of the
5329 $dnslist_value variable are a list of all the addresses, separated by a
5332 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5333 to accept a host name and all its aliases if the forward lookup for any of
5334 them yielded the IP address of the incoming connection. Now it accepts only
5335 those names whose forward lookup yields the correct IP address. Any other
5336 names are discarded. This closes a loophole whereby a rogue DNS
5337 administrator could create reverse DNS records to break through a
5338 wildcarded host restriction in an ACL.
5340 37. If a user filter or a system filter that ran in a subprocess used any of
5341 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5342 the wrong values were passed to the pipe command ($thisaddress had the
5343 value of $0, $0 had the value of $1, etc). This bug was introduced by
5344 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5346 38. Improved the line breaking for long SMTP error messages from ACLs.
5347 Previously, if there was no break point between 40 and 75 characters, Exim
5348 left the rest of the message alone. Two changes have been made: (a) I've
5349 reduced the minimum length to 35 characters; (b) if it can't find a break
5350 point between 35 and 75 characters, it looks ahead and uses the first one
5351 that it finds. This may give the occasional overlong line, but at least the
5352 remaining text gets split now.
5354 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5355 file descriptors might be low, and that setting 1000 would always raise it.
5356 It turns out that in some environments, the limit is already over 1000 and
5357 that lowering it causes trouble. So now Exim takes care not to decrease it.
5359 40. When delivering a message, the value of $return_path is set to $sender_
5360 address at the start of routing (routers may change the value). By an
5361 oversight, this default was not being set up when an address was tested by
5362 -bt or -bv, which affected the outcome if any router or filter referred to
5365 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5366 log only when "message" or "log_message" are set. However, if one of the
5367 conditions was an address verification, or a call to a nested ACL, the
5368 messages generated by the underlying test were being passed through. This
5369 no longer happens. The underlying message is available in $acl_verify_
5370 message for both "message" and "log_message" expansions, so it can be
5371 passed through if needed.
5373 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5374 new expansion $bh_ to give the encoded byte string without charset
5375 translation. Translation happens only if iconv() is available; HAVE_ICONV
5376 indicates this at build time. HEADERS_CHARSET gives the charset to
5377 translate to; headers_charset can change it in the configuration, and
5378 "headers charset" can change it in an individual filter file.
5380 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5381 that creates RFC 2047 encoded "words" labels them as that charset instead
5382 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5383 expansion operator; (ii) when Exim creates a From: line for a local
5384 message; (iii) when a header line is rewritten to include a "phrase" part.
5386 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5387 buggy, causing it to skip the first lines of messages whose message ID
5388 ended in 'D'. This would not have bitten before Exim release 4.14, because
5389 message IDs were unlikely to end in 'D' before then. The effect was to have
5390 incorrect size information for certain domains.
5392 45. #include "config.h" was missing at the start of the crypt16.c module. This
5393 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5396 46. If there was a timeout during a "random" callout check, Exim treated it as
5397 a failure of the random address, and carried on sending RSET and the real
5398 address. If the delay was just some slowness somewhere, the response to the
5399 original RCPT would be taken as a response to RSET and so on, causing
5400 mayhem of various kinds.
5402 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5403 when I implemented it. It didn't allow for the fact that some option values
5404 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5405 the right test for positive values.
5407 48. Domain names in DNS records are case-independent. Exim always looks them up
5408 in lower case. Some resolvers return domain names in exactly the case they
5409 appear in the zone file, that is, they may contain uppercase letters. Not
5410 all resolvers do this - some return always lower case. Exim was treating a
5411 change of case by a resolver as a change of domain, similar to a widening
5412 of a domain abbreviation. This triggered its re-routing code and so it was
5413 trying to route what was effectively the same domain again. This normally
5414 caused routing to fail (because the router wouldn't handle the domain
5415 twice). Now Exim checks for this case specially, and just changes the
5416 casing of the domain that it ultimately uses when it transmits the message
5419 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5422 50. If a filter generated a file delivery with a non-absolute name (possible if
5423 no home directory exists for the router), the forbid_file option was not
5426 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5427 the existing equality matching.
5429 52. Exim was using ints instead of ino_t variables in some places where it was
5430 dealing with inode numbers.
5432 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5433 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5434 called TMPDIR, and if it finds it is different, it changes its value.
5436 54. The smtp_printf() function is now made available to local_scan() so
5437 additional output lines can be written before returning. There is also an
5438 smtp_fflush() function to enable the detection of a dropped connection.
5439 The variables smtp_input and smtp_batched_input are exported to
5442 55. Changed the default runtime configuration: the message "Unknown user"
5443 has been removed from the ACL, and instead placed on the localuser router,
5444 using the cannot_route_message feature. This means that any verification
5445 failures that generate their own messages won't get overridden. Similarly,
5446 the "Unrouteable address" message that was in the ACL for unverifiable
5447 relay addresses has also been removed.
5449 56. Added hosts_avoid_esmtp to the smtp transport.
5451 57. The exicyclog script was not checking for the esoteric option
5452 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5453 will work only if exicyclog is run under the appropriate euid.
5455 58. Following a discussion on the list, the rules by which Exim recognises line
5456 endings on incoming messages have been changed. The -dropcr and drop_cr
5457 options are now no-ops, retained only for backwards compatibility. The
5458 following line terminators are recognized: LF CRLF CR. However, special
5459 processing applies to CR:
5461 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5462 nor a local message in the state where . is a terminator.
5464 (ii) If a bare CR is encountered in a header line, an extra space is added
5465 after the line terminator so as not to end the header. The reasoning
5466 behind this is that bare CRs in header lines are most likely either
5467 to be mistakes, or people trying to play silly games.
5469 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5470 was being incorrectly given as 18 bytes larger than it should have been.
5471 This is a VOB (very old bug).
5473 60. This may never have affected anything current, but just in case it has:
5474 When the local host is found other than at the start of a list of hosts,
5475 the local host, those with the same MX, and any that follow, are discarded.
5476 When the list in question was part of a longer list of hosts, the following
5477 hosts (not currently being processed) were also being discarded. This no
5478 longer happens. I'm not sure if this situation could ever has previously
5481 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5484 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5485 header. This is something that is recommended in a new Internet Draft, and
5486 is something that is documented as being done by Sendmail. There are two
5487 possible values. For messages generated by the autoreply transport, Exim
5490 Auto-Submitted: auto-replied
5492 whereas for all other generated messages (e.g. bounces) it adds
5494 Auto-Submitted: auto-generated
5496 63. The "personal" condition in filters now includes a test for the
5497 Auto-Submitted: header. If it contains the string "auto-" the message it
5498 not considered personal.
5500 64. Added rcpt_include_affixes as a generic transport option.
5502 65. Added queue_only_override (default true).
5504 66. Added the syslog_duplication option.
5506 67. If what should have been the first header line of a message consisted of
5507 a space followed by a colon, Exim was mis-interpreting it as a header line.
5508 It isn't of course - it is syntactically invalid and should therefore be
5509 treated as the start of the message body. The misbehaviour could have
5510 caused a number of strange effects, including loss of data in subsequent
5511 header lines, and spool format errors.
5513 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5514 client host had authenticated. This control can now be exercised by an ACL
5515 for more flexibility.
5517 69. By default, callouts do not happen when testing with -bh. There is now a
5518 variant, -bhc, which does actually run the callout code, including
5519 consulting and updating the callout cache.
5521 70. Added support for saslauthd authentication, courtesy of Alexander
5524 71. If statvfs() failed on the spool or log directories while checking their
5525 size for availability, Exim confusingly gave the error "space shortage".
5526 Furthermore, in debugging mode it crashed with a floating point exception.
5527 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5528 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5529 serious problem, Exim now writes to the main and panic logs when this
5530 happens, with details of the failure. It then refuses to accept the
5531 incoming message, giving the message "spool directory problem" or "log
5532 directory problem" with a 421 code for SMTP messages.
5534 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5535 0, 1, and 2 exist, because some OS complain for execs without them (see
5536 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5537 descriptors. However, the code omitted to check that the open succeeded,
5538 causing mysterious errors if for some reason the permissions on /dev/null
5539 got screwed. Now Exim writes a message to the main and panic logs, and
5540 bombs out if it can't open /dev/null.
5542 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5543 interact so that it is all more flexible. It is supposed to remain
5544 backwards compatible. Also added extra_local_interfaces.
5546 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5547 to bomb out with an assertion failure - to the client this appears as a
5548 connection drop. This problem occurs in the part of the code that was taken
5549 from the Samba project. Fortunately, the assertion is in a very simple
5550 function, so I have fixed this by reproducing the function inline in the
5551 one place where it is called, and arranging for authentication to fail
5552 instead of killing the process with assert().
5554 75. The SPA client code was not working when the server requested OEM rather
5555 than Unicode encoding.
5557 76. Added code to make require_files with a specific uid setting more usable in
5558 the case where statting the file as root fails - usually a non-root-mounted
5559 NFS file system. When this happens and the failure is EACCES, Exim now
5560 forks a subprocess and does the per-uid checking as the relevant uid.
5562 77. Added process_log_path.
5564 78. If log_file_path was not explicitly set, a setting of check_log_space or
5565 check_log_inodes was ignored.
5567 79. If a space check for the spool or log partitions fails, the incident is now
5568 logged. Of course, in the latter case the data may get lost...
5570 80. Added the %p formatting code to string_format() so that it can be used to
5571 print addresses in debug_print(). Adjusted all the address printing in the
5572 debugging in store.c to use %p rather than %d.
5574 81. There was a concern that a line of code in smtp_in.c could overflow a
5575 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5576 initially expressed, the concern was not well-founded, because trailing
5577 spaces are removed early. However, if the trailing spaces were followed by
5578 a NULL, they did not get removed, so the overflow was possible. Two fixes
5581 (a) I re-wrote the offending code in a cleaner fashion.
5582 (b) If an incoming SMTP command contains a NULL character, it is rejected
5585 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5586 runs initgroups(), so that if the Exim user is in any additional groups,
5587 they will be used during message reception.
5593 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob