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 not 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.
109 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
111 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
113 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
119 PP/01 Solaris build fix for Oracle's LDAP libraries.
120 Bugzilla 1109, patch from Stephen Usher.
122 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
124 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
127 TF/02 Fix a couple more cases where we did not log the error message
128 when unlink() failed. See also change 4.74-TF/03.
130 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
131 lock up or crash if it happened to be inside a call to libc when it
132 got a SIGUSR1 from exiwhat.
134 The SIGUSR1 handler appends the current process status to the process
135 log which is later printed by exiwhat. It used to use the general
136 purpose logging code to do this, but several functions it calls are
137 not safe for signals.
139 The new output code in the SIGUSR1 handler is specific to the process
140 log, and simple enough that it's easy to inspect for signal safety.
141 Removing some special cases also simplifies the general logging code.
142 Removing the spurious timestamps from the process log simplifies
145 TF/04 Improved ratelimit ACL condition.
147 The /noupdate option has been deprecated in favour of /readonly which
148 has clearer semantics. The /leaky, /strict, and /readonly update modes
149 are mutually exclusive. The update mode is no longer included in the
150 database key; it just determines when the database is updated. (This
151 means that when you upgrde Exim will forget old rate measurements.)
153 Exim now checks that the per_* options are used with an update mode that
154 makes sense for the current ACL. For example, when Exim is processing a
155 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
156 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
157 must specify per_mail/readonly. If you omit the update mode it defaults to
158 /leaky where that makes sense (as before) or /readonly where required.
160 The /noupdate option is now undocumented but still supported for
161 backwards compatibility. It is equivalent to /readonly except that in
162 ACLs where /readonly is required you may specify /leaky/noupdate or
163 /strict/noupdate which are treated the same as /readonly.
165 A useful new feature is the /count= option. This is a generalization
166 of the per_byte option, so that you can measure the throughput of other
167 aggregate values. For example, the per_byte option is now equivalent
168 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
170 The per_rcpt option has been generalized using the /count= mechanism
171 (though it's more complicated than the per_byte equivalence). When it is
172 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
173 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
174 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
175 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
176 Note that using per_rcpt with a non-readonly update mode in more than
177 one ACL will cause the recipients to be double-counted. (The per_mail
178 and per_byte options don't have this problem.)
180 The handling of very low rates has changed slightly. If the computed rate
181 is less than the event's count (usually one) then this event is the first
182 after a long gap. In this case the rate is set to the same as this event's
183 count, so that the first message of a spam run is counted properly.
185 The major new feature is a mechanism for counting the rate of unique
186 events. The new per_addr option counts the number of different
187 recipients that someone has sent messages to in the last time period. It
188 behaves like per_rcpt if all the recipient addresses are different, but
189 duplicate recipient addresses do not increase the measured rate. Like
190 the /count= option this is a general mechanism, so the per_addr option
191 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
192 example, measure the rate that a client uses different sender addresses
193 with the options per_mail/unique=$sender_address. There are further
194 details in the main documentation.
196 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
198 TF/06 Removed a few PCRE remnants.
200 TF/07 Automatically extract Exim's version number from tags in the git
201 repository when doing development or release builds.
203 PP/02 Raise smtp_cmd_buffer_size to 16kB.
204 Bugzilla 879. Patch from Paul Fisher.
206 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
207 Heavily based on revision 40f9a89a from Simon Arlott's tree.
210 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
212 PP/05 Variable $av_failed, true if the AV scanner deferred.
213 Bugzilla 1078. Patch from John Horne.
215 PP/06 Stop make process more reliably on build failure.
216 Bugzilla 1087. Patch from Heiko Schlittermann.
218 PP/07 Make maildir_use_size_file an _expandable_ boolean.
219 Bugzilla 1089. Patch from Heiko Schlittermann.
221 PP/08 Handle ${run} returning more data than OS pipe buffer size.
222 Bugzilla 1131. Patch from Holger Weiß.
224 PP/09 Handle IPv6 addresses with SPF.
225 Bugzilla 860. Patch from Wolfgang Breyha.
227 PP/10 GnuTLS: support TLS 1.2 & 1.1.
229 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
232 PP/11 match_* no longer expand right-hand-side by default.
233 New compile-time build option, EXPAND_LISTMATCH_RHS.
234 New expansion conditions, "inlist", "inlisti".
236 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
238 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
240 PP/14 fix log_write() format string regression from TF/03.
241 Bugzilla 1152. Patch from Dmitry Isaikin.
247 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
249 PP/02 Harmonised TLS library version reporting; only show if debugging.
250 Layout now matches that introduced for other libraries in 4.74 PP/03.
252 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
254 PP/04 New "dns_use_edns0" global option.
256 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
259 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
260 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
262 TK/01 Updated PolarSSL code to 0.14.2.
263 Bugzilla 1097. Patch from Andreas Metzler.
265 PP/07 Catch divide-by-zero in ${eval:...}.
268 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
271 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
272 format-string attack -- SECURITY: remote arbitrary code execution.
274 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
275 time unintentionally subject to list matching rules, letting the header
276 cause arbitrary Exim lookups (of items which can occur in lists, *not*
277 arbitrary string expansion). This allowed for information disclosure.
279 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
280 INT_MIN/-1 -- value coerced to INT_MAX.
286 NM/01 Workround for PCRE version dependency in version reporting
289 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
290 This fixes portability to compilers other than gcc, notably
291 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
293 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
294 makefiles for portability to HP-UX and POSIX correctness.
296 PP/01 Permit LOOKUP_foo enabling on the make command-line.
297 Also via indented variable definition in the Makefile.
298 (Debugging by Oliver Heesakkers).
300 PP/02 Restore caching of spamd results with expanded spamd_address.
301 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
303 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
304 Improves build reliability. Fix from: Frank Elsner
306 NM/02 Fix wide character breakage in the rfc2047 coding
307 Fixes bug 1064. Patch from Andrey N. Oktyabrski
309 NM/03 Allow underscore in dnslist lookups
310 Fixes bug 1026. Patch from Graeme Fowler
312 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
313 Code patches from Adam Ciarcinski of NetBSD.
315 NM/04 Fixed exiqgrep to cope with mailq missing size issue
318 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
319 is logged, to avoid truncation. Patch from John Horne.
321 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
322 Patch from Jakob Hirsch.
324 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
325 SQL string expansion failure details.
326 Patch from Andrey Oktyabrski.
328 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
329 Patch from Simon Arlott.
331 PP/09 New lookups functionality failed to compile on old gcc which rejects
332 extern declarations in function scope.
333 Patch from Oliver Fleischmann
335 PP/10 Use sig_atomic_t for flags set from signal handlers.
336 Check getgroups() return and improve debugging.
337 Fixed developed for diagnosis in bug 927 (which turned out to be
340 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
341 Patch from Mark Zealey.
343 PP/12 Bugzilla 1056: Improved spamd server selection.
344 Patch from Mark Zealey.
346 PP/13 Bugzilla 1086: Deal with maildir quota file races.
347 Based on patch from Heiko Schlittermann.
349 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
350 Patch from Uwe Doering, sign-off by Michael Haardt.
352 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
353 variable declaration deep within a block. Bug and patch from
356 PP/15 lookups-Makefile IRIX compatibilty coercion.
358 PP/16 Make DISABLE_DKIM build knob functional.
360 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
361 Patch by Simon Arlott
363 TF/03 Fix valgrind.h portability to C89 compilers that do not support
364 variable argument macros. Our copy now differs from upstream.
370 TF/01 Failure to get a lock on a hints database can have serious
371 consequences so log it to the panic log.
373 TF/02 Log LMTP confirmation messages in the same way as SMTP,
374 controlled using the smtp_confirmation log selector.
376 TF/03 Include the error message when we fail to unlink a spool file.
378 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
379 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
380 for maintaining out-of-tree patches for some time.
382 PP/01 Bugzilla 139: Documentation and portability issues.
383 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
384 Handle per-OS dynamic-module compilation flags.
386 PP/02 Let /dev/null have normal permissions.
387 The 4.73 fixes were a little too stringent and complained about the
388 permissions on /dev/null. Exempt it from some checks.
389 Reported by Andreas M. Kirchwitz.
391 PP/03 Report version information for many libraries, including
392 Exim version information for dynamically loaded libraries. Created
393 version.h, now support a version extension string for distributors
394 who patch heavily. Dynamic module ABI change.
396 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
397 privilege escalation vulnerability whereby the Exim run-time user
398 can cause root to append content of the attacker's choosing to
401 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
404 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
405 If dropping privileges for untrusted macros, we disabled normal logging
406 on the basis that it would fail; for the Exim run-time user, this is not
407 the case, and it resulted in successful deliveries going unlogged.
408 Fixed. Reported by Andreas Metzler.
414 PP/01 Date: & Message-Id: revert to normally being appended to a message,
415 only prepend for the Resent-* case. Fixes regression introduced in
416 Exim 4.70 by NM/22 for Bugzilla 607.
418 PP/02 Include check_rfc2047_length in configure.default because we're seeing
419 increasing numbers of administrators be bitten by this.
421 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
423 PP/03 Bugzilla 994: added openssl_options main configuration option.
425 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
427 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
429 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
431 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
432 without a peer certificate, leading to a segfault because of an
433 assumption that peers always have certificates. Be a little more
434 paranoid. Problem reported by Martin Tscholak.
436 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
437 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
438 NB: ClamAV planning to remove STREAM in "middle of 2010".
439 CL also introduces -bmalware, various -d+acl logging additions and
440 more caution in buffer sizes.
442 PP/09 Implemented reverse_ip expansion operator.
444 PP/10 Bugzilla 937: provide a "debug" ACL control.
446 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
448 PP/12 Bugzilla 973: Implement --version.
450 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
452 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
454 PP/15 Bugzilla 816: support multiple condition rules on Routers.
456 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
457 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
458 ignore trailing whitespace.
460 JJ/02 prevent non-panic DKIM error from being sent to paniclog
462 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
465 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
466 Notification from Dr Andrew Aitchison.
468 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
469 ExtendedDetectionInfo response format.
470 Notification from John Horne.
472 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
475 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
476 XSL and documented dependency on system catalogs, with examples of how
479 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
482 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
483 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
484 configuration file which is writeable by the Exim user or group.
486 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
487 of configuration files to cover files specified with the -C option if
488 they are going to be used with root privileges, not just the default
491 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
492 option (effectively making it always true).
494 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
495 files to be used while preserving root privileges.
497 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
498 that rogue child processes cannot use them.
500 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
501 run-time user, instead of root.
503 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
504 Exim run-time user without dropping privileges.
506 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
507 result string, instead of calling string_vformat() twice with the same
510 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
511 for other users. Others should always drop root privileges if they use
512 -C on the command line, even for a whitelisted configure file.
514 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
516 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
522 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
523 $data_path, and $header_path variables; fixed documentation bugs and
526 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
527 exipick to access non-standard spools, including the "frozen" queue
530 NM/01 Bugzilla 965: Support mysql stored procedures.
531 Patch from Alain Williams
533 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
535 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
536 Patch from Andreas Metzler
538 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
539 Patch from Kirill Miazine
541 NM/05 Bugzilla 671: Added umask to procmail example.
543 JJ/03 installed exipick 20100323.0, fixing doc bug
545 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
546 directory. Notification and patch from Dan Rosenberg.
548 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
550 TK/02 Improve log output when DKIM signing operation fails.
552 MH/01 Treat the transport option dkim_domain as a colon separated
553 list, not as a single string, and sign the message with each element,
554 omitting multiple occurences of the same signer.
556 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
557 Bugzilla 985, 986. Patch by Simon Arlott
559 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
560 Patch by Simon Arlott
562 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
563 MBX locking. Notification from Dan Rosenberg.
569 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
571 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
573 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
575 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
577 NM/04 Bugzilla 915: Flag broken perl installation during build.
583 TK/01 Added patch by Johannes Berg that expands the main option
584 "spamd_address" if it starts with a dollar sign.
586 TK/02 Write list of recipients to X-Envelope-Sender header when building
587 the mbox-format spool file for content scanning (suggested by Jakob
590 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
591 (http://www.dcc-servers.net/) support via dccifd. Activated by
592 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
594 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
595 by Mark Daniel Reidel <mr@df.eu>.
597 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
598 When building exim an external PCRE library is now needed -
599 PCRE is a system library on the majority of modern systems.
600 See entry on PCRE_LIBS in EDITME file.
602 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
603 conversation. Added nologin parameter to request.
604 Patch contributed by Kirill Miazine.
606 TF/01 Do not log submission mode rewrites if they do not change the address.
608 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
610 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
611 log files in place. Contributed by Roberto Lima.
613 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
615 TF/03 Bugzilla 615: When checking the local_parts router precondition
616 after a local_part_suffix or local_part_prefix option, Exim now
617 does not use the address's named list lookup cache, since this
618 contains cached lookups for the whole local part.
620 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
621 Robert Millan. Documentation is in experimental-spec.txt.
623 TF/04 Bugzilla 668: Fix parallel build (make -j).
625 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
627 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
628 Patch provided by Jan Srzednicki.
630 TF/05 Leading white space used to be stripped from $spam_report which
631 wrecked the formatting. Now it is preserved.
633 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
634 that they are available at delivery time.
636 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
638 TF/08 TLS error reporting now respects the incoming_interface and
639 incoming_port log selectors.
641 TF/09 Produce a more useful error message if an SMTP transport's hosts
642 setting expands to an empty string.
644 NM/06 Bugzilla 744: EXPN did not work under TLS.
645 Patch provided by Phil Pennock.
647 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
648 Patch provided by Richard Godbee.
650 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
651 acl_smtp_notquit, added index entry.
653 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
654 Patch provided by Eugene Bujak.
656 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
657 accept(). Patch provided by Maxim Dounin.
659 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
660 Patch provided by Phil Pennock.
662 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
664 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
665 Patch provided by Brad "anomie" Jorsch.
667 NM/14 Bugzilla 622: Added timeout setting to transport filter.
668 Patch provided by Dean Brooks.
670 TK/05 Add native DKIM support (does not depend on external libraries).
672 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
673 Patch provided by Graeme Fowler.
675 NM/16 Bugzilla 851: Documentation example syntax fix.
677 NM/17 Changed NOTICE file to remove references to embedded PCRE.
679 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
682 NM/19 Bugzilla 745: TLS version reporting.
683 Patch provided by Phil Pennock.
685 NM/20 Bugzilla 167: bool: condition support.
686 Patch provided by Phil Pennock.
688 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
689 clients. Patch provided by Phil Pennock.
691 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
692 Patch provided by Brad "anomie" Jorsch.
694 NM/23 Bugzilla 687: Fix misparses in eximstats.
695 Patch provided by Heiko Schlittermann.
697 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
698 Patch provided by Heiko Schlittermann.
700 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
701 plus update to original patch.
703 NM/26 Bugzilla 799: Documentation correction for ratelimit.
705 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
706 Patch provided by David Brownlee.
708 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
710 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
712 NM/30 Bugzilla 888: TLS documentation bugfixes.
714 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
716 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
717 Unlike the original bugzilla I have changed all shell scripts in src tree.
719 NM/33 Bugzilla 898: Transport filter timeout fix.
720 Patch by Todd Rinaldo.
722 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
723 Patch by Serge Demonchaux.
725 NM/35 Bugzilla 39: Base64 decode bug fixes.
726 Patch by Jakob Hirsch.
728 NM/36 Bugzilla 909: Correct connect() call in dcc code.
730 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
732 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
734 NM/39 Bugzilla 911: Fixed MakeLinks build script.
740 TK/01 Add preliminary DKIM support. Currently requires a forked version of
741 ALT-N's libdkim that I have put here:
742 http://duncanthrax.net/exim-experimental/
744 Note to Michael Haardt: I had to rename some vars in sieve.c. They
745 were called 'true' and it seems that C99 defines that as a reserved
746 keyword to be used with 'bool' variable types. That means you could
747 not include C99-style headers which use bools without triggering
748 build errors in sieve.c.
750 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
751 as mailq or other aliases. Changed the --help handling significantly
752 to do whats expected. exim_usage() emits usage/help information.
754 SC/01 Added the -bylocaldomain option to eximstats.
756 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
758 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
760 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
766 PH/01 Another patch from the Sieve maintainer.
768 PH/02 When an IPv6 address is converted to a string for single-key lookup
769 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
770 dots are used instead of colons so that keys in lsearch files need not
771 contain colons. This was done some time before quoting was made available
772 in lsearch files. However, iplsearch files do require colons in IPv6 keys
773 (notated using the quote facility) so as to distinguish them from IPv4
774 keys. This meant that lookups for IP addresses in host lists did not work
775 for iplsearch lookups.
777 This has been fixed by arranging for IPv6 addresses to be expressed with
778 colons if the lookup type is iplsearch. This is not incompatible, because
779 previously such lookups could never work.
781 The situation is now rather anomolous, since one *can* have colons in
782 ordinary lsearch keys. However, making the change in all cases is
783 incompatible and would probably break a number of configurations.
785 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
788 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
789 conversion specification without a maximum field width, thereby enabling
790 a rogue spamd server to cause a buffer overflow. While nobody in their
791 right mind would setup Exim to query an untrusted spamd server, an
792 attacker that gains access to a server running spamd could potentially
793 exploit this vulnerability to run arbitrary code as the Exim user.
795 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
796 $primary_hostname instead of what libspf2 thinks the hosts name is.
798 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
799 a directory entry by the name of the lookup key. Previously, if a
800 symlink pointed to a non-existing file or a file in a directory that
801 Exim lacked permissions to read, a lookup for a key matching that
802 symlink would fail. Now it is enough that a matching directory entry
803 exists, symlink or not. (Bugzilla 503.)
805 PH/03 The body_linecount and body_zerocount variables are now exported in the
808 PH/04 Added the $dnslist_matched variable.
810 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
811 This means they are set thereafter only if the connection becomes
814 PH/06 Added the client_condition to authenticators so that some can be skipped
815 by clients under certain conditions.
817 PH/07 The error message for a badly-placed control=no_multiline_responses left
818 "_responses" off the end of the name.
820 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
822 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
823 (without spaces) instead of just copying the configuration text.
825 PH/10 Added the /noupdate option to the ratelimit ACL condition.
827 PH/11 Added $max_received_linelength.
829 PH/12 Added +ignore_defer and +include_defer to host lists.
831 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
832 way in which PCRE > 7.0 is built.
834 PH/14 Implemented queue_only_load_latch.
836 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
837 MAIL command. The effect was to mangle the value on 64-bit systems.
839 PH/16 Another patch from the Sieve maintainer.
841 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
843 PH/18 If a system quota error occurred while trying to create the file for
844 a maildir delivery, the message "Mailbox is full" was not appended to the
845 bounce if the delivery eventually timed out. Change 4.67/27 below applied
846 only to a quota excession during the actual writing of the file.
848 PH/19 It seems that peer DN values may contain newlines (and other non-printing
849 characters?) which causes problems in log lines. The DN values are now
850 passed through string_printing() before being added to log lines.
852 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
853 and InterBase are left for another time.)
855 PH/21 Added message_body_newlines option.
857 PH/22 Guard against possible overflow in moan_check_errorcopy().
859 PH/23 POSIX allows open() to be a macro; guard against that.
861 PH/24 If the recipient of an error message contained an @ in the local part
862 (suitably quoted, of course), incorrect values were put in $domain and
863 $local_part during the evaluation of errors_copy.
869 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
870 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
873 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
874 issue a MAIL command.
876 PH/02 In an ACL statement such as
878 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
880 if a client was not listed at all, or was listed with a value other than
881 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
882 the condition was not true (as it should be), so access was not denied.
883 The bug was that the ! inversion was incorrectly passed on to the second
884 item. This has been fixed.
886 PH/03 Added additional dnslists conditions == and =& which are different from
887 = and & when the dns lookup returns more than one IP address.
889 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
890 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
892 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
893 FSYNC, which compiles an option called disable_fsync that allows for
894 bypassing fsync(). The documentation is heavily laced with warnings.
896 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
898 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
899 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
900 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
901 including adding "make clean"; (3) Added -fPIC when compiling the test
902 dynamically loaded module, to get rid of a warning.
904 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
905 message fails, move_frozen_messages = true and ignore_bounce_errors_after
906 = 0s. The bug is otherwise harmless.
908 PH/07 There was a bug in the dovecot authenticator such that the value of
909 $auth1 could be overwritten, and so not correctly preserved, after a
910 successful authentication. This usually meant that the value preserved by
911 the server_setid option was incorrect.
913 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
915 PH/09 Installed PCRE release 7.0.
917 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
918 run for batched SMTP input. It is now run at the start of every message
919 in the batch. While fixing this I discovered that the process information
920 (output by running exiwhat) was not always getting set for -bs and -bS
921 input. This is fixed, and it now also says "batched" for BSMTP.
923 PH/11 Added control=no_pipelining.
925 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
926 patch, slightly modified), and move the expansion of helo_data till after
927 the connection is made in the smtp transport (so it can use these
930 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
932 PH/14 Added log_selector = +pid.
934 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
936 PH/16 Add ${if forany and ${if forall.
938 PH/17 Added dsn_from option to vary the From: line in DSNs.
940 PH/18 Flush SMTP output before performing a callout, unless control =
941 no_callout_flush is set.
943 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
944 was true (the default) a successful delivery failed to delete the retry
945 item, thus causing premature timeout of the address. The bug is now
948 PH/20 Added hosts_avoid_pipelining to the smtp transport.
950 PH/21 Long custom messages for fakedefer and fakereject are now split up
951 into multiline reponses in the same way that messages for "deny" and
952 other ACL rejections are.
954 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
955 with slight modification.
957 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
958 draft, changing the syntax and factoring some duplicate code".
960 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
961 for deliveries of the second and subsequent messages over the same SMTP
964 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
965 ${reduce, with only minor "tidies".
967 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
969 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
970 expansion side effects.
972 PH/27 When a message times out after an over-quota error from an Exim-imposed
973 quota, the bounce message says "mailbox is full". This message was not
974 being given when it was a system quota that was exceeded. It now should
977 MH/03 Made $recipients available in local_scan(). local_scan() already has
978 better access to the recipient list through recipients_list[], but
979 $recipients can be useful in postmaster-provided expansion strings.
981 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
982 in the case of a MAIL command with additional options following the
983 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
984 were accidentally chopped off.
986 PH/29 SMTP synchronization checks are implemented when a command is read -
987 there is a check that no more input is waiting when there shouldn't be
988 any. However, for some commands, a delay in an ACL can mean that it is
989 some time before the response is written. In this time, more input might
990 arrive, invalidly. So now there are extra checks after an ACL has run for
991 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
992 pipelining has not been advertised.
994 PH/30 MH's patch to allow iscntrl() characters to be list separators.
996 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
997 returned in the SMTP response when smtp_return_error_details was false.
1000 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1001 instead of the C I/O that was originally supplied, because problems were
1002 reported on Solaris.
1004 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1005 Exim which did not show up earlier: it was assuming that a call to
1006 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1007 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1008 was a macro that became an assignment, so it seemed to work. This has
1009 changed to a proper function call with a void return, hence the compile
1010 error. Exim's code has been fixed.
1012 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1015 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1017 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1019 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1020 because it turns out to be too big or there is a timeout) there is no
1021 "Completed" line in the log. When some messages of this type were
1022 selected by exigrep, they were listed as "not completed". Others were
1023 picked up by some special patterns. I have improved the selection
1024 criteria to be more general.
1026 PH/38 The host_find_failed option in the manualroute router can now be set
1027 to "ignore", to completely ignore a host whose IP address cannot be
1028 found. If all hosts are ignored, the behaviour is controlled by the new
1029 host_all_ignored option.
1031 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1032 homing or because of multiple MX records with /mx) generated more than
1033 one IP address, and the following item turned out to be the local host,
1034 all the secondary addresses of the first item were incorrectly removed
1035 from the list, along with the local host and any following hosts (which
1036 is what is supposed to happen).
1038 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1039 whoever called Exim into the -H file. In the case of the daemon it was
1040 behaving confusingly. When first started, it used values for whoever
1041 started the daemon, but after a SIGHUP it used the Exim user (because it
1042 calls itself on a restart). I have changed the code so that it now always
1045 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1046 message are rejected with the same error (e.g. no authentication or bad
1047 sender address), and a DATA command is nevertheless sent (as can happen
1048 with PIPELINING or a stupid MUA), the error message that was given to the
1049 RCPT commands is included in the rejection of the DATA command. This is
1050 intended to be helpful for MUAs that show only the final error to their
1053 PH/42 Another patch from the Sieve maintainer.
1055 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1056 Eximstats - Fixed some broken HTML links and added missing column headers
1058 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1059 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1061 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1063 PH/43 Yet another patch from the Sieve maintainer.
1065 PH/44 I found a way to check for a TCP/IP connection going away before sending
1066 the response to the final '.' that terminates a message, but only in the
1067 case where the client has not sent further data following the '.'
1068 (unfortunately, this is allowed). However, in many cases there won't be
1069 any further data because there won't be any more messages to send. A call
1070 to select() can be used: if it shows that the input is "ready", there is
1071 either input waiting, or the socket has been closed. An attempt to read
1072 the next input character can distinguish the two cases. Previously, Exim
1073 would have sent an OK response which the client would never have see.
1074 This could lead to message repetition. This fix should cure that, at
1075 least in a lot of common cases.
1077 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1078 advertised in response to EHLO.
1084 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1085 fixed by 4.65/MH/01 (is this a record?) are fixed:
1087 (i) An empty string was always treated as zero by the numeric comparison
1088 operators. This behaviour has been restored.
1090 (ii) It is documented that the numeric comparison operators always treat
1091 their arguments as decimal numbers. This was broken in that numbers
1092 starting with 0 were being interpreted as octal.
1094 While fixing these problems I realized that there was another issue that
1095 hadn't been noticed. Values of message_size_limit (both the global option
1096 and the transport option) were treated as octal if they started with 0.
1097 The documentation was vague. These values are now always treated as
1098 decimal, and I will make that clear in the documentation.
1104 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1105 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1108 MH/01 Don't check that the operands of numeric comparison operators are
1109 integers when their expansion is in "skipping" mode (fixes bug
1110 introduced by 4.64-PH/07).
1112 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1113 child addresses, Exim now panics and dies. Previously, because the count
1114 is held in a short int, deliveries were likely to be lost. As such a
1115 large number of recipients for a single message is ridiculous
1116 (performance will be very, very poor), I have chosen to impose a limit
1117 rather than extend the field.
1123 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1124 leftover -K file (the existence of which was triggered by #402).
1125 While we were at it, introduced process PID as part of the -K
1126 filename. This should rule out race conditions when creating
1129 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1130 processing considerably. Previous code took too long for large mails,
1131 triggering a timeout which in turn triggers #401.
1133 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1134 in the DK code in transports.c. sendfile() is not really portable,
1135 hence the _LINUX specificness.
1137 TF/01 In the add_headers option to the mail command in an Exim filter,
1138 there was a bug that Exim would claim a syntax error in any
1139 header after the first one which had an odd number of characters
1142 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1143 callout verification, Exim cached a "reject" for the entire domain. This
1144 is correct for most verifications, but it is not correct for a recipient
1145 verification with use_sender or use_postmaster set, because in that case
1146 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1147 case of MAIL FROM:<> rejection from other early rejections (e.g.
1148 rejection of HELO). When verifying a recipient using a non-null MAIL
1149 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1150 Whatever the result of the callout, the value of the domain cache is
1151 left unchanged (for any other kind of callout, getting as far as trying
1152 RCPT means that the domain itself is ok).
1154 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1157 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1158 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1159 session, the daemon ignores ECONNECT errors and logs others; it now
1160 ignores EPIPE as well.
1162 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1163 (quoted-printable decoding).
1165 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1166 later the small subsequent patch to fix an introduced bug.
1168 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1170 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1172 PH/08 An error is now given if message_size_limit is specified negative.
1174 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1175 to be given (somewhat) arbitrary names.
1177 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1180 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1181 miscellaneous code fixes
1183 PH/10 Added the log_reject_target ACL modifier to specify where to log
1186 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1187 hostname. This is wrong, because it relates to the incoming message (and
1188 probably the interface on which it is arriving) and not to the outgoing
1189 callout (which could be using a different interface). This has been
1190 changed to use the value of the helo_data option from the smtp transport
1191 instead - this is what is used when a message is actually being sent. If
1192 there is no remote transport (possible with a router that sets up host
1193 addresses), $smtp_active_hostname is used.
1195 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1196 tweaks were necessary in order to get it to work (see also 21 below):
1197 (a) The code assumed that strncpy() returns a negative number on buffer
1198 overflow, which isn't the case. Replaced with Exim's string_format()
1200 (b) There were several signed/unsigned issues. I just did the minimum
1201 hacking in of casts. There is scope for a larger refactoring.
1202 (c) The code used strcasecmp() which is not a standard C function.
1203 Replaced with Exim's strcmpic() function.
1204 (d) The code set only $1; it now sets $auth1 as well.
1205 (e) A simple test gave the error "authentication client didn't specify
1206 service in request". It would seem that Dovecot has changed its
1207 interface. Fortunately there's a specification; I followed it and
1208 changed what the client sends and it appears to be working now.
1210 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1213 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1214 address A is aliased to B and C, where B exists and C does not. Without
1215 -v the output is "A verified" because verification stops after a
1216 successful redirection if more than one address is generated. However,
1217 with -v the child addresses are also verified. Exim was outputting "A
1218 failed to verify" and then showing the successful verification for C,
1219 with its parentage. It now outputs "B failed to verify", showing B's
1220 parentage before showing the successful verification of C.
1222 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1223 look up a TXT record in a specific list after matching in a combined
1226 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1227 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1228 they consult the DNS. I had assumed they would set it the way they
1229 wanted; and indeed my experiments on Linux seem to show that in some
1230 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1231 To be on the safe side, however, I have now made the interface to
1232 host_find_byname() similar to host_find_bydns(), with an argument
1233 containing the DNS resolver options. The host_find_byname() function now
1234 sets these options at its start, just as host_find_bydns() does. The smtp
1235 transport options dns_qualify_single and dns_search_parents are passed to
1236 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1237 of host_find_byname() use the default settings of RES_DEFNAMES
1238 (qualify_single) but not RES_DNSRCH (search_parents).
1240 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1241 spool_read_header() do less string testing, by means of a preliminary
1242 switch on the second character of optional "-foo" lines. (This is
1243 overdue, caused by the large number of possibilities that now exist.
1244 Originally there were few.) While I was there, I also converted the
1245 str(n)cmp tests so they don't re-test the leading "-" and the first
1246 character, in the hope this might squeeze out yet more improvement.
1248 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1249 flag allowing group syntax was set by the header_syntax check but not
1250 turned off, possible causing trouble later; (2) The flag was not being
1251 set at all for the header_verify test, causing "group"-style headers to
1252 be rejected. I have now set it in this case, and also caused header_
1253 verify to ignore an empty address taken from a group. While doing this, I
1254 came across some other cases where the code for allowing group syntax
1255 while scanning a header line wasn't quite right (mostly, not resetting
1256 the flag correctly in the right place). These bugs could have caused
1257 trouble for malformed header lines. I hope it is now all correct.
1259 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1260 with the "reply" argument non-NULL. The code, however (which originally
1261 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1262 but it didn't always do it. This confused somebody who was copying the
1263 code for some other use. I have removed all the tests.
1265 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1266 feature that was used to support insecure browsers during the U.S. crypto
1267 embargo. It requires special client support, and Exim is probably the
1268 only MTA that supported it -- and would never use it because real RSA is
1269 always available. This code has been removed, because it had the bad
1270 effect of slowing Exim down by computing (never used) parameters for the
1271 RSA_EXPORT functionality.
1273 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1274 authenticator to fail if there's a tab character in the incoming data
1275 (there should never be unless someone is messing about, as it's supposed
1276 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1277 if the connection is using TLS or if the remote IP is the same as the
1278 local IP, and the "valid-client-cert option" if a client certificate has
1281 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1282 authenticators. This can be used for authorization after authentication
1283 succeeds. (In the case of plaintext, it servers for both authentication
1286 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1287 if any retry times were supplied.
1289 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1290 connection, where there is no client IP address to check. In this
1291 situation, the verify now always succeeds.
1293 PH/25 Applied John Jetmore's -Mset patch.
1295 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1297 PH/27 In a string expansion for a processed (not raw) header when multiple
1298 headers of the same name were present, leading whitespace was being
1299 removed from all of them, but trailing whitespace was being removed only
1300 from the last one. Now trailing whitespace is removed from each header
1301 before concatenation. Completely empty headers in a concatenation (as
1302 before) are ignored.
1304 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1305 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1307 PH/29 [Removed. This was a change that I later backed out, and forgot to
1308 correct the ChangeLog entry (that I had efficiently created) before
1309 committing the later change.]
1311 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1312 address errors (4xx response to RCPT) over the same connection as other
1313 messages routed to the same hosts. Such deliveries are always "forced",
1314 so retry times are not inspected. This resulted in far too many retries
1315 for the affected addresses. The effect occurred only when there were more
1316 hosts than the hosts_max_try setting in the smtp transport when it had
1317 the 4xx errors. Those hosts that it had tried were not added to the list
1318 of hosts for which the message was waiting, so if all were tried, there
1319 was no problem. Two fixes have been applied:
1321 (i) If there are any address or message errors in an SMTP delivery, none
1322 of the hosts (tried or untried) are now added to the list of hosts
1323 for which the message is waiting, so the message should not be a
1324 candidate for sending over the same connection that was used for a
1325 successful delivery of some other message. This seems entirely
1326 reasonable: after all the message is NOT "waiting for some host".
1327 This is so "obvious" that I'm not sure why it wasn't done
1328 previously. Hope I haven't missed anything, but it can't do any
1329 harm, as the worst effect is to miss an optimization.
1331 (ii) If, despite (i), such a delivery is accidentally attempted, the
1332 routing retry time is respected, so at least it doesn't keep
1333 hammering the server.
1335 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1336 in ${readsocket because some servers need this prod.
1338 PH/32 Added some extra debug output when updating a wait-xxx database.
1340 PH/33 The hint "could be header name not terminated by colon", which has been
1341 given for certain expansion errors for a long time, was not being given
1342 for the ${if def:h_colon_omitted{... case.
1344 PH/34 The spec says: "With one important exception, whenever a domain list is
1345 being scanned, $domain contains the subject domain." There was at least
1346 one case where this was not true.
1348 PH/35 The error "getsockname() failed: connection reset by peer" was being
1349 written to the panic log as well as the main log, but it isn't really
1350 panic-worthy as it just means the connection died rather early on. I have
1351 removed the panic log writing for the ECONNRESET error when getsockname()
1354 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1355 runs only) independently of the message's sender address. This meant
1356 that, if the 4xx error was in fact related to the sender, a different
1357 message to the same recipient with a different sender could confuse
1358 things. In particualar, this can happen when sending to a greylisting
1359 server, but other circumstances could also provoke similar problems.
1360 I have changed the default so that the retry time for these errors is now
1361 based a combination of the sender and recipient addresses. This change
1362 can be overridden by setting address_retry_include_sender=false in the
1365 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1366 remote server are returned as part of bounce messages. This was not
1367 happening for LMTP over a pipe (the lmtp transport), but now it is the
1368 same for both kinds of LMTP.
1370 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1371 in header lines that were in fact CNAMEs. This is no longer the case.
1373 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1374 and queue runs started by the daemon processed all messages. This has
1375 been fixed so that -R and -S can now usefully be given with -q<time>.
1377 PH/40 Import PCRE release 6.7 (fixes some bugs).
1379 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1381 PH/42 Give an error if -q is specified more than once.
1383 PH/43 Renamed the variables $interface_address and $interface_port as
1384 $received_ip_address and $received_port, to make it clear that these
1385 values apply to message reception, and not to the outgoing interface when
1386 a message is delivered. (The old names remain recognized, of course.)
1388 PH/44 There was no timeout on the connect() call when using a Unix domain
1389 socket in the ${readsocket expansion. There now is.
1391 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1392 be meaningful with "accept".
1394 SC/01 Eximstats V1.43
1395 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1397 SC/02 Eximstats V1.44
1398 Use a glob alias rather than an array ref in the generated
1399 parser. This improves both readability and performance.
1401 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1402 Collect SpamAssassin and rejection statistics.
1403 Don't display local sender or destination tables unless
1404 there is data to show.
1405 Added average volumes into the top table text output.
1407 SC/04 Eximstats V1.46
1408 Collect data on the number of addresses (recipients)
1409 as well as the number of messages.
1411 SC/05 Eximstats V1.47
1412 Added 'Message too big' to the list of mail rejection
1413 reasons (thanks to Marco Gaiarin).
1415 SC/06 Eximstats V1.48
1416 Mainlog lines which have GMT offsets and are too short to
1417 have a flag are now skipped.
1419 SC/07 Eximstats V1.49 (Alain Williams)
1420 Added the -emptyok flag.
1422 SC/08 Eximstats V1.50
1423 Fixes for obtaining the IP address from reject messages.
1425 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1426 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1427 whitesspace changes from 4.64-PH/27
1429 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1432 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1433 are found, allow negative numbers in numeric criteria)
1435 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1437 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1438 to match changes made in 4.64-PH/43
1440 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1442 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1443 "domain" is actually an IP address, and force a failure. This locks out
1444 those revolvers/nameservers that support "A-for-A" lookups, in
1445 contravention of the specifications.
1447 PH/48 When a host name was looked up from an IP address, and the subsequent
1448 forward lookup of the name timed out, the host name was left in
1449 $sender_host_name, contrary to the specification.
1451 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1452 restricted to single-key lookups, Exim was not diagnosing an error if
1453 * or *@ was used with a query-style lookup.
1455 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1457 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1458 long ago, but noone interested enough thought of it. Let's just say that
1459 the "1.1" means that there are some new functions that weren't there at
1460 some point in the past.
1462 PH/51 Error processing for expansion failure of helo_data from an smtp
1463 transport during callout processing was broken.
1465 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1466 tested/used via the -bh/-bhc/-bs options.
1468 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1469 bug, fixed in subsequent PCRE releases).
1471 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1472 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1474 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1480 SC/01 Use a glob alias rather than an array ref in eximstats generated
1481 parser. This improves both readability and performance.
1483 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1484 Don't display local sender or destination tables in eximstats unless
1485 there is data to show.
1486 Added average volumes into the eximstats top table text output.
1488 SC/03 Collect data on the number of addresses (recipients) as well
1489 as the number of messages in eximstats.
1491 TF/01 Correct an error in the documentation for the redirect router. Exim
1492 does (usually) call initgroups() when daemonizing.
1494 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1495 with consistent privilege compared to when running as a daemon.
1497 TF/03 Note in the spec that $authenticated_id is not set for local
1498 submissions from trusted users.
1500 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1501 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1503 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1504 by adding some example configuration directives to the default
1505 configuration file. A little bit of work is required to uncomment the
1506 directives and define how usernames and passwords are checked, but
1507 there is now a framework to start from.
1509 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1510 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1511 without this. I don't know how relevant this is to other LDAP libraries.
1513 PH/02 Add the verb name to the "unknown ACL verb" error.
1515 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1517 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1519 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1520 directory not expanded when it should be if an expanded home directory
1521 was set for the address (which is overridden by the transport).
1523 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1526 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1527 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1528 because it is too late at that time, and has no effect.
1530 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1531 security issue with \' (bugzilla #107). I could not use the
1532 PQescapeStringConn() function, because it needs a PGconn value as one of
1535 PH/08 When testing addresses using -bt, indicate those final addresses that
1536 are duplicates that would not cause an additional delivery. At least one
1537 person was confused, thinking that -bt output corresponded to deliveries.
1538 (Suppressing duplicates isn't a good idea as you lose the information
1539 about possibly different redirections that led to the duplicates.)
1541 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1542 systems where poll() doesn't work, in particular OS X.
1544 PH/10 Added more information to debugging output for retry time not reached.
1546 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1547 operations in malware.c.
1549 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1552 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1553 syslog_duplication set false, log lines that would normally be written
1554 both the the main log and to the reject log were not written to syslog at
1557 PH/14 In the default configuration, change the use of "message" in ACL warn
1558 statements to "add_header".
1560 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1561 not followed by a command (e.g. "seen endif").
1563 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1564 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1567 PH/17 Added extra conditions to the default value of delay_warning_condition
1571 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1572 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1573 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1576 The Auto-Submitted: and various List- headers are standardised, whereas I
1577 don't think Precedence: ever was.
1579 PH/18 Refactored debugging code in route_finduser() to show more information,
1580 in particular, the error code if getpwnam() issues one.
1582 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1583 This is apparently needed in addition to the PH/07 change above to avoid
1584 any possible encoding problems.
1586 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1587 but not after initializing Perl.
1589 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1590 output them only if debugging. By default they are written stderr,
1591 apparently, which is not desirable.
1593 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1596 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1599 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1601 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1602 authenticated or an ident call has been made. Suppress the default
1603 values for $authenticated_id and $authenticated_sender (but permit -oMai
1604 and -oMas) when testing with -bh.
1606 PH/24 Re-jigged the order of the tests in the default configuration so that the
1607 tests for valid domains and recipients precede the DNS black list and CSA
1608 tests, on the grounds that those ones are more expensive.
1610 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1611 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1612 This bug exists in every version of Exim that I still have, right back to
1615 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1616 However, an attempt to turn on case-sensitivity in a regex key by
1617 including (?-i) didn't work because the subject string was already
1618 lowercased, and the effects were non-intuitive. It turns out that a
1619 one-line patch can be used to allow (?-i) to work as expected.
1625 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1626 other effects) broke the use of negated acl sub-conditions.
1628 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1631 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1632 "Deny" causes Exim to reject the incoming connection with a 554 error.
1633 Unfortunately, if there is a major crisis, such as a disk failure,
1634 tcp-wrappers gives "deny", whereas what one would like would be some
1635 kind of temporary error. A kludge has been added to help with this.
1636 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1637 554 error is used if errno is still zero or contains ENOENT (which occurs
1638 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1641 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1643 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1644 errors. Otherwise a message that provokes a temporary error (when other
1645 messages do not) can cause a whole host to time out.
1647 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1648 addresses were routed directly to files or pipes from a redirect router.
1649 File deliveries just didn't batch; pipe deliveries might have suffered
1652 PH/06 A failure to get a lock for a hints database would erroneously always say
1653 "Failed to get write lock", even when it was really a read lock.
1655 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1656 of 0600. This has been changed to use the value of the lockfile_mode
1657 option (which defaults to 0600).
1659 PH/08 Applied small patch from the Sieve maintainer.
1661 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1662 folder from quota calculations, a direct delivery into this folder messed
1663 up the contents of the maildirsize file. This was because the regex was
1664 used only to exclude .Trash (or whatever) when the size of the mailbox
1665 was calculated. There was no check that a delivery was happening into an
1666 excluded directory. This bug has been fixed by ignoring all quota
1667 processing for deliveries into excluded directories.
1669 PH/10 Added the maildirfolder_create_regex option to appendfile.
1675 PH/01 The code for finding all the local interface addresses on a FreeBSD
1676 system running IPv6 was broken. This may well have applied to all BSD
1677 systems, as well as to others that have similar system calls. The broken
1678 code found IPv4 interfaces correctly, but gave incorrect values for the
1679 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1680 that it would not match correctly against @[] and not recognize the IPv6
1683 PH/02 The ipliteral router was not recognizing addresses of the form user@
1684 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1686 PH/03 Added disable_ipv6.
1688 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1689 decoded form, because it is most often used to construct To: headers
1690 lines in autoreplies, and the decoded form may well be syntactically
1691 invalid. However, $reply_address has leading white space removed, and all
1692 newlines turned into spaces so that the autoreply transport does not
1695 PH/05 If group was specified without a user on a router, and no group or user
1696 was specified on a transport, the group from the router was ignored.
1698 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1699 for visible compile-time settings that can be used to change these
1700 numbers, for those that want even more. Backwards compatibility with old
1701 spool files has been maintained. However, going back to a previous Exim
1702 release will lost any variables that are in spool files.
1704 PH/07 Two small changes when running in the test harness: increase delay when
1705 passing a TCP/IP connection to a new process, in case the original
1706 process has to generate a bounce, and remove special handling of
1707 127.0.0.2 (sic), which is no longer necessary.
1709 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1710 be the same on different OS.
1712 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1715 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1716 whether --show-vars was specified or not
1718 JJ/02 exipick: Added support for new ACL variable spool format introduced
1721 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1722 syntactically invalid From: or Reply-to: line, and a filter used this to
1723 generate an autoreply, and therefore failed to obtain an address for the
1724 autoreply, Exim could try to deliver to a non-existent relative file
1725 name, causing unrelated and misleading errors. What now happens is that
1726 it logs this as a hard delivery error, but does not attempt to create a
1729 PH/11 The exinext utility has a -C option for testing purposes, but although
1730 the given file was scanned by exinext itself; it wasn't being passed on
1731 when Exim was called.
1733 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1734 an end-of-file indication when reading a command response.
1736 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1737 compiled. In many other places in Exim, IPv6 addresses are always
1738 recognized, so I have changed this. It also means that IPv4 domain
1739 literals of the form [IPV4:n.n.n.n] are now always recognized.
1741 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1742 used if the router is not running as root, for example, when verifying at
1743 ACL time, or when using -bh. The debugging output from this situation was
1744 non-existent - all you got was a failure to exec. I have made two
1747 (a) Failures to set uid/gid, the current directory, or a process leader
1748 in a subprocess such as that created by queryprogram now generate
1749 suitable debugging ouput when -d is set.
1751 (b) The queryprogram router detects when it is not running as root,
1752 outputs suitable debugging information if -d is set, and then runs
1753 the subprocess without attempting to change uid/gid.
1755 PH/15 Minor change to Makefile for building test_host (undocumented testing
1758 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1759 additional section of a DNS packet that returns MX or SRV records.
1760 Instead, it always explicitly searches for A/AAAA records. This avoids
1761 major problems that occur when a DNS server includes only records of one
1762 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1763 fixed another bug: if SRV records were looked up and the corresponding
1764 address records were *not* found in the additional section, the port
1765 values from the SRV records were lost.
1767 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1768 using the correct key (the original address) when searching the retry
1769 rules in order to find which one to use for generating the retry hint.
1771 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1772 adding the default one. Similarly, if it contains a Reply-To: header, the
1773 errors_reply_to option, if set, is not used.
1775 PH/19 When calculating a retry time, Exim used to measure the "time since
1776 failure" by looking at the "first failed" field in the retry record. Now
1777 it does not use this if it is later than than the arrival time of the
1778 message. Instead it uses the arrival time. This makes for better
1779 behaviour in cases where some deliveries succeed, thus re-setting the
1780 "first failed" field. An example is a quota failure for a huge message
1781 when small messages continue to be delivered. Without this change, the
1782 "time since failure" will always be short, possible causing more frequent
1783 delivery attempts for the huge message than are intended.
1784 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1786 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1787 $1, $2, $3) because the numerical variables can be reset during some
1788 expansion items (e.g. "match"), thereby losing the authentication data.
1790 PH/21 Make -bV show the size of off_t variables so that the test suite can
1791 decide whether to run tests for quotas > 2G.
1793 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1794 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1795 filecount value is greater than 2G or if a quota value is greater than 2G
1796 on a system where the size of off_t is not greater than 4, a panic error
1799 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1800 never match. The debug and -bh output now contains an explicit error
1801 message indicating a malformed IPv4 address or mask.
1803 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1804 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1805 PH/23 above applies.
1807 PH/25 Do not write to syslog when running in the test harness. The only
1808 occasion when this arises is a failure to open the main or panic logs
1809 (for which there is an explicit test).
1811 PH/26 Added the /no_tell option to "control=freeze".
1813 PH/27 If a host name lookup failed very early in a connection, for example, if
1814 the IP address matched host_lookup and the reverse lookup yielded a name
1815 that did not have a forward lookup, an error message of the form "no IP
1816 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1817 could be logged. Now it outputs the IP address instead of "NULL".
1819 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1820 allows the sender and the authenticated sender to be set when
1821 submitting a message from within Exim. Since child_open_exim() is
1822 documented for local_scan(), the new function should be too.
1824 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1825 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1826 results in an empty string is now treated as unset.
1828 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1830 PH/31 Added sender_verify_fail logging option.
1832 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1833 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1834 tidied the source and removed it altogether.
1836 PH/33 When a queue run was abandoned because the load average was too high, a
1837 log line was always written; now it is written only if the queue_run log
1838 selector is set. In addition, the log line for abandonment now contains
1839 information about the queue run such as the pid. This is always present
1840 in "start" and "stop" lines but was omitted from the "abandon" line.
1842 PH/34 Omit spaces between a header name and the colon in the error message that
1843 is given when verify = headers_syntax fails (if there are lots of them,
1844 the message gets confusing).
1846 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1847 names, as there are now some PTR records that contain slashes. This check
1848 is only to protect against broken name servers that fall over on strange
1849 characters, so the fact that it applies to all lookups doesn't matter.
1851 PH/36 Now that the new test suite is complete, we can remove some of the
1852 special code in Exim that was needed for the old test suite. For example,
1853 sorting DNS records because real resolvers return them in an arbitrary
1854 order. The new test suite's fake resolver always returns records in the
1857 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1858 bounces) except when queue_only is set, to avoid logging races between
1859 the different processes.
1861 PH/38 Panic-die if .include specifies a non-absolute path.
1863 PH/39 A tweak to the "H" retry rule from its user.
1865 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1866 a label. They prevented compilation on older perls.
1868 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1869 a warning to be raised on newish perls.
1871 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1872 on queue. Changes to match documented behaviour of showing count of
1873 messages matching specified criteria.
1875 PH/40 Changed the default ident timeout from 30s to 5s.
1877 PH/41 Added support for the use of login_cap features, on those BSD systems
1878 that have them, for controlling the resources used by pipe deliveries.
1880 PH/42 The content-scanning code uses fopen() to create files in which to put
1881 message data. Previously it was not paying any attention to the mode of
1882 the files. Exim runs with umask(0) because the rest of the code creates
1883 files with open(), and sets the required mode explicitly. Thus, these
1884 files were ending up world-writeable. This was not a big issue, because,
1885 being within the spool directory, they were not world-accessible. I have
1886 created a function called modefopen, which takes an additional mode
1887 argument. It sets umask(777), creates the file, chmods it to the required
1888 mode, then resets the umask. All the relevant calls to fopen() in the
1889 content scanning code have been changed to use this function.
1891 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1892 to 24 hours. This avoids potential overflow problems when processing G
1893 and H retry rules. I suspect nobody ever tinkers with this value.
1895 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1897 PH/45 When the plaintext authenticator is running as a client, the server's
1898 challenges are checked to ensure they are valid base64 strings. By
1899 default, the authentication attempt is cancelled if an invalid string is
1900 received. Setting client_ignore_invalid_base64 true ignores these errors.
1901 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1902 they are received. Thus, the responses can be made to depend on the
1903 challenges. If an invalid string is ignored, an empty string is placed in
1906 PH/46 Messages that are created by the autoreply transport now contains a
1907 References: header, in accordance with RFCs 2822 and 3834.
1909 PH/47 Added authenticated_sender_force to the smtp transport.
1911 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1913 PH/49 Installed latest patch from the Sieve maintainer.
1915 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1916 was also set, the appendfile transport was unnecessarily scanning a
1917 directory of message files (e.g. for maildir delivery) to find the count
1918 of files (along with the size), even though it did not need this
1919 information. It now does the scan only if it needs to find either the
1920 size of the count of files.
1922 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1924 PH/52 Two bugs concerned with error handling when the smtp transport is
1927 (i) Exim was not creating retry information for temporary errors given
1928 for individual recipients after the DATA command when the smtp transport
1929 was used in LMTP mode. This meant that they could be retried too
1930 frequently, and not timed out correctly.
1932 (ii) Exim was setting the flag that allows error details to be returned
1933 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1934 recipients that were returned after the DATA command.
1936 PH/53 This is related to PH/52, but is more general: for any failing address,
1937 when detailed error information was permitted to be returned to the
1938 sender, but the error was temporary, then after the final timeout, only
1939 "retry timeout exceeded" was returned. Now it returns the full error as
1940 well as "retry timeout exceeded".
1942 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1943 do this, and (what is worse) MTAs that accept it.
1945 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1946 will now be deprecated.
1948 PH/56 New os.c-cygwin from the Cygwin maintainer.
1950 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1951 formats (previously only available in exim formats via -bpr, -bpru,
1952 and -bpra. Now also available in native and exiqgrep formats)
1954 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1955 with very large, slow to parse queues
1957 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1959 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1961 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1962 responses to authentication challenges, though it was showing the
1963 challenges; (ii) I've removed the CR characters from the debug output for
1966 PH/58 Allow for the insertion of a newline as well as a space when a string
1967 is turned into more than one encoded-word during RFC 2047 encoding. The
1968 Sieve code now uses this.
1970 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1971 data_4xx, lost_connection, tls_required.
1973 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1974 message was being sent as an SMTP response.
1976 PH/61 Add -l and -k options to exicyclog.
1978 PH/62 When verifying, if an address was redirected to one new address, so that
1979 verification continued, and the new address failed or deferred after
1980 having set something in $address_data, the value of $address_data was not
1981 passed back to the ACL. This was different to the case when no
1982 redirection occurred. The value is now passed back in both cases.
1984 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1985 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1986 use login_cap.h, so on its own it isn't the distinguishing feature. The
1987 new name refers directly to the setclassresources() function.
1989 PH/65 Added configuration files for NetBSD3.
1991 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1993 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1994 is preferred over IPv4.
1996 PH/68 The bounce_return_message and bounce_return_body options were not being
1997 honoured for bounces generated during the reception of non-SMTP messages.
1998 In particular, this applied to messages rejected by the ACL. This bug has
1999 been fixed. However, if bounce_return_message is true and bounce_return_
2000 body is false, the headers that are returned for a non-SMTP message
2001 include only those that have been read before the error was detected.
2002 (In the case of an ACL rejection, they have all been read.)
2004 PH/69 The HTML version of the specification is now built in a directory called
2005 spec_html instead of spec.html, because the latter looks like a path with
2006 a MIME-type, and this confuses some software.
2008 PH/70 Catch two compiler warnings in sieve.c.
2010 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2011 function verify_get_ident() calls ip_connect() to connect a socket, but
2012 if the "connect()" function timed out, ip_connect() used to close the
2013 socket. However, verify_get_ident() also closes the socket later, and in
2014 between Exim writes to the log, which may get opened at this point. When
2015 the socket was closed in ip_connect(), the log could get the same file
2016 descriptor number as the socket. This naturally causes chaos. The fix is
2017 not to close the socket in ip_connect(); the socket should be closed by
2018 the function that creates it. There was only one place in the code where
2019 this was missing, in the iplookup router, which I don't think anybody now
2020 uses, but I've fixed it anyway.
2022 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2023 well as to direct DNS lookups. Otherwise the handling of names in host
2024 lists is inconsistent and therefore confusing.
2030 PH/01 Two changes to the default runtime configuration:
2032 (1) Move the checks for relay_from_hosts and authenticated clients from
2033 after to before the (commented out) DNS black list checks.
2035 (2) Add control=submission to the relay_from_hosts and authenticated
2036 clients checks, on the grounds that messages accepted by these
2037 statements are most likely to be submissions.
2039 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2041 (1) Generate an error if the third argument for the ${prvs expansion is
2044 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2047 (3) Reset the variables that are obtained from the first argument of
2048 ${prvscheck and used in the second argument before leaving the code,
2049 because their memory is reclaimed, so using them afterwards may do
2052 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2053 one (it's much easier than Tom thought :-).
2055 (5) Because of (4), we can now allow for the use of $prvscheck_result
2056 inside the third argument.
2058 PH/03 For some reason, the default setting of PATH when running a command from
2059 a pipe transport was just "/usr/bin". I have changed it to
2062 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2063 anything to be listed in the output from -bV.
2065 PH/05 When a filter generated an autoreply, the entire To: header line was
2066 quoted in the delivery log line, like this:
2068 => >A.N.Other <ano@some.domain> <original@ddress> ...
2070 This has been changed so that it extracts the operative address. There
2071 may be more than one such address. If so, they are comma-separated, like
2074 => >ano@some.domain,ona@other.domain <original@ddress> ...
2076 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2077 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2078 not being looked up in the rDNS to get a host name, Exim was showing the
2079 IP address twice in Received: lines, even though the IP addresses were
2080 identical. For example:
2082 Received: from [1.2.3.4] (helo=[1.2.3.4])
2084 However, if the real host name was known, it was omitting the HELO data
2085 if it matched the actual IP address. This has been tidied up so that it
2086 doesn't show the same IP address twice.
2088 PH/07 When both +timestamp and +memory debugging was on, the value given by
2089 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2090 called by the debug printing, thereby overwriting the timestamp buffer.
2091 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2093 PH/08 When the original message was included in an autoreply transport, it
2094 always said "this is a copy of the message, including all the headers",
2095 even if body_only or headers_only was set. It now gives an appropriate
2098 PH/09 Applied a patch from the Sieve maintainer which:
2100 o fixes some comments
2101 o adds the (disabled) notify extension core
2102 o adds some debug output for the result of if/elsif tests
2103 o points to the current vacation draft in the documentation
2104 and documents the missing references header update
2108 o fixes a bug in processing the envelope test (when testing
2109 multiple envelope elements, the last element determinted the
2112 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2113 Electronic Mail") by including:
2115 Auto-submitted: auto-generated
2117 in the messages that it generates (bounce messages and others, such as
2118 warnings). In the case of bounce messages for non-SMTP mesages, there was
2119 also a typo: it was using "Auto_submitted" (underscore instead of
2120 hyphen). Since every message generated by Exim is necessarily in response
2121 to another message, thes have all been changed to:
2123 Auto-Submitted: auto-replied
2125 in accordance with these statements in the RFC:
2127 The auto-replied keyword:
2129 - SHOULD be used on messages sent in direct response to another
2130 message by an automatic process,
2132 - MUST NOT be used on manually-generated messages,
2134 - MAY be used on Delivery Status Notifications (DSNs) and Message
2135 Disposition Notifications (MDNs),
2137 - MUST NOT be used on messages generated by automatic or periodic
2138 processes, except for messages which are automatic responses to
2141 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2142 to the default Received: header definition.
2144 PH/12 Added log selector acl_warn_skipped (default on).
2146 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2147 variables because (a) they are in the wrong storage pool and (b) even if
2148 they were copied, it wouldn't work properly because of the caching.
2150 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2151 checking when decoding. Apparently there are clients that generate
2152 overlong encoded strings. Why am I not surprised?
2154 PH/15 If the first argument of "${if match_address" was not empty, but did not
2155 contain an "@" character, Exim crashed. Now it writes a panic log message
2156 and treats the condition as false.
2158 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2160 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2161 extension "envelope-auth". The code is finished and in agreement with
2162 other implementations, but there is no documentation so far and in fact,
2163 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2164 not changing the active code.
2166 Print executed "if" and "elsif" statements when debugging is used. This
2167 helps a great deal to understand what a filter does.
2169 Document more things not specified clearly in RFC3028. I had all this
2170 sorted out, when out of a sudden new issues came to my mind. Oops."
2172 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2175 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2176 canonical form (as documented). However, after a host name lookup from
2177 the IP address, check_host() was doing a simple string comparison with
2178 addresses acquired from the DNS when checking that the found name did
2179 have the original IP as one of its addresses. Since any found IPv6
2180 addresses are likely to be in abbreviated form, the comparison could
2181 fail. Luckily, there already exists a function for doing the comparison
2182 by converting both addresses to binary, so now that is used instead of
2183 the text comparison.
2185 PH/20 There was another similar case to PH/19, when a complete host name was
2186 given in a host list; looking up its IP address could give an abbreviated
2187 form, whereas the current host's name might or might not be abbreviated.
2188 The same fix has been applied.
2194 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2195 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2198 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2199 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2201 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2203 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2204 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2205 Also, for "Auto-Submitted", treat anything other than "no" as
2206 non-personal, in accordance with RFC 3834. (Previously it treated
2207 anything starting "auto-" as non-personal.)
2209 TF/01 The control=submission/name=... option had a problem with syntax
2210 errors if the name included a slash character. The /name= option
2211 now slurps the rest of the string, so it can include any characters
2212 but it must come last in the list of options (after /sender_retain
2215 PH/05 Some modifications to the interface to the fake nameserver for the new
2223 TK/01 Added the "success_on_redirect" address verification option. See
2224 NewStuff for rationale and an example.
2226 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2228 PH/02 Patch to exigrep to allow it to work on syslog lines.
2230 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2231 fread() to skip over the body file's header line, because in Cygwin the
2232 header line is locked and is inaccessible.
2234 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2235 co-exist for some time) to make it clear that it is the Exim ID that is
2236 referenced, not the Message-ID: header line.
2238 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2239 string_format() function, because snprintf() does not exist on all
2242 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2243 ${stat: expansion item.
2245 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2246 protocol synchronization error", to keep the pedants happy.
2248 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2249 well as for IRIX systems, when gcc is being used. See the host.c source
2252 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2254 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2257 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2258 transport and to the smtp transport in LMTP mode.
2260 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2262 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2263 run a filter in a subprocess. This could lead to confusion in subsequent
2264 lookups in the parent process. There should also be a search_tidyup() at
2265 the end of the subprocess.
2267 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2268 only if the host matched helo_try_verify_hosts, which caused the
2269 verification to occur when the EHLO/HELO command was issued. The ACL just
2270 tested the remembered result. Now, if a previous verification attempt has
2271 not happened, "verify = helo" does it there and then.
2273 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2275 TK/03 Fix log output including CR from clamd.
2277 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2278 exist provoked a memory error which could cause a segfault.
2280 PH/15 Installed PCRE 6.2
2282 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2284 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2285 of the problem. Specifically, suggested +O2 rather than +O1 for the
2288 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2290 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2291 record, the port was not correctly logged when the outgoing_port log
2292 selector was set (it logged the transort's default port).
2294 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2295 fallback_hosts, and "hosts" in the smtp transport.
2297 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2298 host errors such as "Connection refused".
2300 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2301 authentication with radiusclient 0.4.9:
2303 - Error returned from rc_read_config was caught wrongly
2304 - Username/password not passed on to radius server due to wrong length.
2306 The presumption is that some radiusclient API changes for 4.51/PH/17
2307 were not taken care of correctly. The code is still untested by me (my
2308 Linux distribution still has 0.3.2 of radiusclient), but it was
2309 contributed by a Radius user.
2311 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2312 expanding the "port" option of the smtp transport.
2314 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2315 while reading a MIME header. Thanks to Tom Hughes for a patch.
2317 PH/24 Include config.h inside local_scan.h so that configuration settings are
2320 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2321 that in an ACL for RCPT (for example), you can examine exactly what was
2324 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2325 commands, but it was not correctly comparing the address with the actual
2326 client host address. Thus, it would show the EHLO address in Received:
2327 header lines when this was not necessary.
2329 PH/27 Added the % operator to ${eval:}.
2331 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2332 it should be ignoring failures (because with -C, for example, it has lost
2333 privilege). It wasn't ignoring creation failures other than "already
2336 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2339 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2340 because an input line was too long, either on its own, or by virtue of
2341 too many continuations, the temporary file was not being removed, and the
2342 return code was incorrect.
2344 PH/31 Missing "BOOL" in function definition in filtertest.c.
2346 PH/32 Applied Sieve patches from the maintainer.
2348 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2350 PH/33 Added "verify = not_blind".
2352 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2353 Local/Makefile (with some defaults set). These are used in built scripts
2354 such as exicyclog, but they have never been used in the exim_install
2355 script (though there are many overriding facilities there). I have
2356 arranged that the exim_install script now takes note of these two
2359 PH/35 Installed configuration files for Dragonfly.
2361 PH/36 When a locally submitted message by a trusted user did not contain a
2362 From: header, and the sender address was obtained from -f or from an SMTP
2363 MAIL command, and the trusted user did not use -F to supply a sender
2364 name, $originator_name was incorrectly used when constructing a From:
2365 header. Furthermore, $originator_name was used for submission mode
2366 messages from external hosts without From: headers in a similar way,
2367 which is clearly wrong.
2369 PH/37 Added control=suppress_local_fixups.
2371 PH/38 When log_selector = +received_sender was set, and the addition of the
2372 sender made the log line's construction buffer exactly full, or one byte
2373 less than full, an overflow happened when the terminating "\n" was
2376 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2377 when the result of a list match is failure because a DNS lookup failed.
2379 PH/40 RM_COMMAND is now used in the building process.
2381 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2382 the "build-* directories that it finds.
2384 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2385 address in a domain literal was a prefix of an interface address.
2387 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2388 when verifying a sender address, unless rewrite_headers is false.
2390 PH/44 Wrote a long comment about why errors_to addresses are verified as
2391 recipients, not senders.
2393 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2394 the ratelimit ACL was added.
2396 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2398 PH/46 Added extra information about PostgreSQL errors to the error string.
2400 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2401 avoiding the need to install special zones in a real server. This is
2402 backwards compatible; if it can't find the fake resolver, it drops back.
2403 Thus, both old and new test suites can be run.
2405 TF/02 Added util/ratelimit.pl
2407 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2408 clock is set back in time.
2410 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2411 Candler <B.Candler@pobox.com>.
2413 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2414 OK for addresses that are the result of redirections.
2416 PH/48 A number of further additions for the benefit of the new test suite,
2417 including a fake gethostbyname() that interfaces to the fake DNS resolver
2420 TF/06 The fix for widen_domains has also been applied to qualify_single and
2421 search_parents which are the other dnslookup options that can cause
2424 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2427 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2429 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2430 service" if it encountered an MX record with an empty target hostname.
2431 The message is now "an MX or SRV record indicated no SMTP service".
2433 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2434 if the DNS of the sending site is misconfigured. This is quite a
2435 common situation. This change restores the behaviour of treating a
2436 helo verification defer as a failure.
2438 PH/51 If self=fail was set on a router, the bounce message did not include the
2439 actual error message.
2445 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2447 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2448 command itself ended in error, the underlying message about the transport
2449 filter timeout was being overwritten with the pipe command error. Now the
2450 underlying error message should be appended to the second error message.
2452 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2454 PH/02 Reduce the amount of output that "make" produces by default. Full output
2455 can still be requested.
2457 PH/03 The warning log line about a condition test deferring for a "warn" verb
2458 was being output only once per connection, rather than after each
2459 occurrence (because it was using the same function as for successful
2460 "warn" verbs). This seems wrong, so I have changed it.
2462 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2463 it should not have, which might have caused a crash in the right
2464 circumstances, but probably never did.
2466 PH/04 Installed a modified version of Tony Finch's patch to make submission
2467 mode fix the return path as well as the Sender: header line, and to
2468 add a /name= option so that you can make the user's friendly name appear
2471 TF/03 Added the control = fakedefer ACL modifier.
2473 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2474 Mark Lowes for thorough testing.
2476 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2478 TK/03 Merged latest SRS patch from Miles Wilton.
2480 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2481 with the definition in sysexits.h (which is #included earlier).
2482 Fortunately, Exim does not actually use EX_OK. The code used to try to
2483 preserve the sysexits.h value, by assumimg that macro definitions were
2484 scanned for macro replacements. I have been disabused of this notion,
2485 so now the code just undefines EX_OK before #including unistd.h.
2487 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2488 in the smtp transport. When a block could not be written in a single
2489 write() function, the timeout was being re-applied to each part-write.
2490 This seems wrong - if the receiver was accepting one byte at a time it
2491 would take for ever. The timeout is now adjusted when this happens. It
2492 doesn't have to be particularly precise.
2494 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2495 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2497 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2498 without a domain if the check to <postmaster@domain> fails.
2500 SC/01 Eximstats: added -xls and the ability to specify output files
2501 (patch written by Frank Heydlauf).
2503 SC/02 Eximstats: use FileHandles for outputing results.
2505 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2507 SC/04 Eximstats: fixed display of large numbers with -nvr option
2509 SC/05 Eximstats: fixed merging of reports with empty tables.
2511 SC/06 Eximstats: added the -include_original_destination flag
2513 SC/07 Eximstats: removed tabs and trailing whitespace.
2515 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2517 TK/06 MBOX spool code: Add real "From " MBOX separator line
2518 so the .eml file is really in mbox format (even though
2519 most programs do not really care). Patch from Alex Miller.
2521 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2522 The latter is generated from $received_to and is only set if the
2523 message has one envelope recipient. SA can use these headers,
2524 obviously out-of-the-box. Patch from Alex Miller.
2526 PH/08 The ${def test on a variable was returning false if the variable's
2527 value was "0", contrary to what the specification has always said!
2528 The result should be true unless the variable is empty.
2530 PH/09 The syntax error of a character other than { following "${if
2531 def:variable_name" (after optional whitespace) was not being diagnosed.
2532 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2533 accidental colon was present, for example, could give incorrect results.
2535 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2536 result is used (not including appendfile, where other changes are about
2539 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2540 This involved changing a lot of size variables from int to off_t. It
2541 should work with maildirs and everything.
2543 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2544 spamd dying while we are connected to it.
2546 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2549 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2550 function for BDB 4.3.
2552 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2554 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2555 the sources for PCRE have been re-arranged and more files are now
2558 PH/15 The code I had for printing potentially long long variables in PH/11
2559 above was not the best (it lost precision). The length of off_t variables
2560 is now inspected at build time, and an appropriate printing format (%ld
2561 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2562 to be "long long int" or "long int". This is needed for the internal
2563 formatting function string_vformat().
2565 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2566 the configuration file to be ":syslog", then the script "guesses" where
2567 the logs files are, rather than using the compiled in default. In our
2568 case the guess is not the same as the compiled default, so the script
2569 suddenly stopped working when I started to use syslog. The patch checks
2570 to see if log_file_path is "". If so, it attempts to read it from exim
2571 with no configuration file to get the compiled in version, before it
2572 falls back to the previous guessing code."
2574 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2575 implementing BATV in an Exim configuration. See NewStuff for the gory
2578 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2579 Makefile that are specific to HP-UX.
2581 PH/18 If the "use_postmaster" option was set for a recipient callout together
2582 with the "random" option, the postmaster address was used as the MAIL
2583 FROM address for the random test, but not for the subsequent recipient
2584 test. It is now used for both.
2586 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2587 patch removes a few documentation additions to RFC 3028, because the
2588 latest draft now contains them. It adds the new en;ascii-case comparator
2589 and a new error check for 8bit text in MIME parts. Comparator and
2590 require names are now matched exactly. I enabled the subaddress
2591 extension, but it is not well tested yet (read: it works for me)."
2593 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2594 rework some of the code of TK/09 above to avoid the hardwired use of
2595 "%lld" and "long long". Replaced the call to snprintf() with a call to
2598 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2599 records point to non-existent hosts", "retry timeout exceeded", and
2600 "retry time not reached for any host after a long failure period".
2602 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2603 experimental DomainKeys support:
2605 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2606 (2) On an error such as an illegally used "control", the wrong name for
2607 the control was given.
2609 These problems did NOT occur unless DomainKeys support was compiled.
2611 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2613 PH/24 Added ${if match_ip condition.
2615 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2616 databases so that it will be absolutely obvious if a crash occurs in the
2617 DB library. This is a regular occurrence (often caused by mis-matched
2620 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2621 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2622 picked up on a user's system that detects such things. There doesn't seem
2623 to be a gcc warning option for this - only an attribute that has to be
2624 put on the function's prototype. It seems that in Fedora Core 4 they have
2625 set this on a number of new functions. No doubt there will be more in due
2628 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2629 specify a transport. However, if an address that was verified by such a
2630 router was the subject of a callout, Exim crashed because it tried to
2631 read the rcpt_include_affixes from the non-existent transport. Now it
2632 just assumes that the setting of that option is false. This bug was
2633 introduced by 4.51/PH/31.
2635 PH/28 Changed -d+all to exclude +memory, because that information is very
2636 rarely of interest, but it makes the output a lot bigger. People tend to
2637 do -d+all out of habit.
2639 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2640 code in os-type was giving problems when libc.so lives in lib64, like on
2643 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2644 aren't the modern standard, and it seems that some systems' include files
2645 don't always have them. Exim was already checking for some of the newer
2646 ones like T_AAAA, and defining it itself. I've added checks for all the
2647 record types that Exim uses.
2649 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2650 not automatically generating a new one, as it is supposed to. This
2651 prevented TLS from working. If the file did exist, but contained invalid
2652 data, a new version was generated, as expected. It was only the case of a
2653 non-existent file that was broken.
2655 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2656 with a change in libdomainkeys > 0.64.
2658 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2659 from DNS. If the selector record carries the flag, it now has
2660 precedence over the domain-wide flag.
2662 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2664 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2665 the use of a port name that isn't defined in /etc/services) occurred, the
2666 message was deferred as in a normal delivery, and thus remained on the
2667 spool, instead of being failed because of the mua_wrapper setting. This
2668 is now fixed, and I tidied up some of the mua_wrapper messages at the
2671 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2672 the messages in a hash of arrays rather than using individual hashes.
2673 This is a bit cleaner and results in dramatic memory savings, albeit
2674 at a slight CPU cost.
2676 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2677 as requested by Marc Sherman.
2679 SC/10 Eximstats: added histograms for user specified patterns as requested
2682 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2684 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2685 fopen() in the content-scanning modules that did not already have it.
2691 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2692 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2694 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2696 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2698 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2699 to test Sieve filters that use "vacation".
2701 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2702 that changes the way the GnuTLS parameters are stored in the cache file.
2703 The new format can be generated externally. For backward compatibility,
2704 if the data in the cache doesn't make sense, Exim assumes it has read an
2705 old-format file, and it generates new data and writes a new file. This
2706 means that you can't go back to an older release without removing the
2709 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2710 work if there are any delivery delays because "one_time" forces the
2711 parent to be marked "delivered", so its unseen clone is never tried
2712 again. For this reason, Exim now forbids the simultaneous setting of
2715 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2716 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2717 entry if you want to know the details. The fix, however, neglected to
2718 consider the case when local delivery batching is involved. The test for
2719 "previously delivered" was not happening when checking to see if an
2720 address could be batched with a previous (undelivered) one; under
2721 certain circumstances this could lead to multiple deliveries to the same
2724 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2725 in its include files, and this causes problems building Exim.
2727 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2728 header_syntax) but Exim was just ignoring anything given after a slash.
2729 In particular, this caused confusion with an attempt to use "verify =
2730 reverse_host_lookup/defer_ok". An error is now given when options are
2731 supplied for verify items that do not have them. (Maybe reverse_host_
2732 lookup should have a defer_ok option, but that's a different point.)
2734 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2735 defined by RFC 821) to 2048, because there were problems with some AUTH
2736 commands, and RFC 1869 says the size should be increased for extended
2737 SMTP commands that take arguments.
2739 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2742 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2743 "unknown" error; now it says that the functionality isn't in the binary.
2745 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2746 an address' error message when a string expansion fails (syntax or
2747 whatever). Otherwise the password may appear in the log. Following change
2748 PH/42 below, there is no longer a chance of it appearing in a bounce
2751 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2753 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2754 of its addresses was ever tried. (Bugzilla bug #2.)
2756 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2757 the result incorrectly in the debug output. (It correctly added a newline
2758 to what was transported.)
2760 TF/01 Added $received_time.
2762 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2763 commented out examples of how to interface to a virus scanner and to
2764 SpamAssassin. Also added commented examples of av_scanner and
2765 spamd_address settings.
2767 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2768 and controls are allowed in which ACLs. There were a couple of minor
2769 errors. Some of the entries in the conditions table (which is a table of
2770 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2771 negation of where the condition IS allowed.
2773 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2775 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2776 header file does not have a version number, so I've had to invent a new
2777 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2778 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2779 radiusclient), but it was contributed by a Radius user.
2781 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2782 files or directories, for OpenSSL.
2784 PH/19 When an Exim process that is running as root has to create an Exim log
2785 file, it does so in a subprocess that runs as exim:exim so as to get the
2786 ownership right at creation (otherwise, other Exim processes might see
2787 the file with the wrong ownership). There was no test for failure of this
2788 fork() call, which would lead to the process getting stuck as it waited
2789 for a non-existent subprocess. Forks do occasionally fail when resources
2790 run out. I reviewed all the other calls to fork(); they all seem to check
2793 PH/20 When checking for unexpected SMTP input at connect time (before writing
2794 the banner), Exim was not dealing correctly with a non-positive return
2795 from the read() function. If the client had disconnected by this time,
2796 the result was a log entry for a synchronization error with an empty
2797 string after "input=" when read() returned zero. If read() returned -1
2798 (an event I could not check), uninitialized data bytes were printed.
2799 There were reports of junk text (parts of files, etc) appearing after
2802 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2804 PH/22 Added support for macro redefinition, and (re)definition in between
2805 driver and ACL definitions.
2807 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2808 forgetting to use the resulting value; it was using the unexpanded value.
2810 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2811 hadn't been configured. The fix is from Juergen Kreileder, who
2812 understands it better than I do:
2814 "Here's what I see happening with three configured cyrus_sasl
2815 authenticators configured (plain, login, cram-md5):
2817 On startup auth_cyrus_sasl_init() gets called for each of these.
2818 This means three calls to sasl_listmech() without a specified mech_list.
2819 => SASL tests which mechs of all available mechs actually work
2820 => three warnings about OTP not working
2821 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2823 With the patch, sasl_listmech() also gets called three times. But now
2824 SASL's mech_list option is set to the server_mech specified in the the
2825 authenticator. Or in other words, the answer from sasl_listmech()
2826 gets limited to just the mech you're testing for (which is different
2828 => the return list contains just 'plain' or 'login', 'cram-md5' or
2829 nothing depending on the value of ob->server_mech.
2831 I've just tested the patch: Authentication still works fine,
2832 unavailable mechs specified in the exim configuration are still
2833 caught, and the auth.log warnings about OTP are gone."
2835 PH/25 When debugging is enabled, the contents of the command line are added
2836 to the debugging output, even when log_selector=+arguments is not
2839 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2840 answer is "GNU", and only if the return is "GNU/something" is the answer
2843 PH/27 $acl_verify_message is now set immediately after the failure of a
2844 verification in an ACL, and so is available in subsequent modifiers. In
2845 particular, the message can be preserved by coding like this:
2847 warn !verify = sender
2848 set acl_m0 = $acl_verify_message
2850 Previously, $acl_verify_message was set only while expanding "message"
2851 and "log_message" when a very denied access.
2853 PH/28 Modified OS/os.c-Linux with
2855 -#ifndef OS_LOAD_AVERAGE
2856 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2858 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2859 nomenclature these days.)
2861 PH/29 Installed patch from the Sieve maintainer that adds the options
2862 sieve_useraddress and sieve_subaddress to the redirect router.
2864 PH/30 In these circumstances:
2865 . Two addresses routed to the same list of hosts;
2866 . First host does not offer TLS;
2867 . First host accepts first address;
2868 . First host gives temporary error to second address;
2869 . Second host offers TLS and a TLS session is established;
2870 . Second host accepts second address.
2871 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2872 and peerdn, if requested) that were in fact used only for the second
2875 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2876 attention to any local part prefix or suffix that was matched by the
2877 router that accepted the address. It now behaves in the same way as it
2878 does for delivery: the affixes are removed from the local part unless
2879 rcpt_include_affixes is set on the transport.
2881 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2882 timeout during the DATA phase of an incoming message.
2884 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2885 applied a patch sanctioned by the Sieve maintainer.
2887 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2888 the uid or gid is negative. A case of a negative gid caused this to be
2889 noticed. The fix allows for either to be negative.
2891 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2892 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2895 PH/36 The change PH/12 above was broken. Fixed it.
2897 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2898 the grounds that routing the same address twice would always produce the
2899 same answer. This might have been true once, but it is certainly no
2900 longer true now. Routing a child address may depend on the previous
2901 routing that produced that child. Some complicated redirection strategies
2902 went wrong when messages had multiple recipients, and made Exim's
2903 behaviour dependent on the order in which the addresses were given.
2905 I have moved the duplicate checking until after the routing is complete.
2906 Exim scans the addresses that are assigned to local and remote
2907 transports, and removes any duplicates. This means that more work will be
2908 done, as duplicates will always all be routed, but duplicates are
2909 presumably rare, so I don't expect this is of any significance.
2911 For deliveries to pipes, files, and autoreplies, the duplicate checking
2912 still happens during the routing process, since they are not going to be
2915 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2916 It corrects a timeout issue with spamd. This is Ian's comment: "The
2917 background is that sometimes spamd either never reads data from a
2918 connection it has accepted, or it never writes response data. The exiscan
2919 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2920 blindly assumes that writes won't block so it may never time out."
2922 PH/39 Allow G after quota size as well as K and M.
2924 PH/40 The value set for $authenticated_id in an authenticator may not contain
2925 binary zeroes or newlines because the value is written to log lines and
2926 to spool files. There was no check on this. Now the value is run through
2927 the string_printing() function so that such characters are converted to
2928 printable escape sequences.
2930 PH/41 $message_linecount is a new variable that contains the total number of
2931 lines in the message. Compare $body_linecount, which is the count for the
2934 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2935 bounce and delay warning messages, except in certain special cases, which
2938 (a) An SMTP error message from a remote host;
2939 (b) A message specified in a :fail: redirection;
2940 (c) A message specified in a "fail" command in a system filter;
2941 (d) A message specified in a FAIL return from the queryprogram router;
2942 (e) A message specified by the cannot_route_message router option.
2944 In these cases only, Exim does include the error details in bounce and
2945 warning messages. There are also a few cases where bland messages such
2946 as "unrouteable address" or "local delivery error" are given.
2948 PH/43 $value is now also set for the "else" part of a ${run expansion.
2950 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2951 being worked on, but at least Exim now implements the latest version to
2954 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2955 process to complete was treated as a delivery failure, a timeout while
2956 writing the message to the pipe was logged, but erroneously treated as a
2957 successful delivery. Such timeouts include transport filter timeouts. For
2958 consistency with the overall process timeout, these timeouts are now
2959 treated as errors, giving rise to delivery failures by default. However,
2960 there is now a new Boolean option for the pipe transport called
2961 timeout_defer, which, if set TRUE, converts the failures into defers for
2962 both kinds of timeout. A transport filter timeout is now identified in
2965 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2966 systems where "make" and "gmake" are different, calling "gmake" at top
2967 level broke things. I've arranged for the value of $(MAKE) to be passed
2968 from the Makefile to this script so that it can call the same version of
2972 A note about Exim versions 4.44 and 4.50
2973 ----------------------------------------
2975 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2976 changes of various kinds. As a consequence, a big documentation update was
2977 needed. This delayed the release for rather longer than seemed good, especially
2978 in the light of a couple of (minor) security issues. Therefore, the changes
2979 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2980 release. So 4.44 and 4.50 are in effect two different branches that both start
2983 I have left the 4.50 change log unchanged; it contains all the changes since
2984 4.43. The change log for 4.44 is below; many of its items are identical to
2985 those for 4.50. This seems to be the most sensible way to preserve the
2986 historical information.
2992 1. Minor wording change to the doc/README.SIEVE file.
2994 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2995 computation of the current number of files was incorrect.
2997 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2998 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3001 4. Give more explanation in the error message when the command for a transport
3002 filter fails to execute.
3004 5. There are several places where Exim runs a non-Exim command in a
3005 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3006 was being done only for the command run by the queryprogram router. It is
3007 now done for all such subprocesses. The other cases are: ${run, transport
3008 filters, and the commands run by the lmtp and pipe transports.
3010 6. Added CONFIGURE_GROUP build-time option.
3012 7. Some older OS have a limit of 256 on the maximum number of file
3013 descriptors. Exim was using setrlimit() to set 1000 as a large value
3014 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3015 systems. I've change it so that if it can't get 1000, it tries for 256.
3017 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3018 was an oversight, and furthermore, ever since the addition of extra
3019 controls (e.g. 4.43/32), the checks on when to allow different forms of
3020 "control" were broken. There should now be diagnostics for all cases when a
3021 control that does not make sense is encountered.
3023 9. Added the /retain_sender option to "control=submission".
3025 10. $recipients is now available in the predata ACL (oversight).
3027 11. Tidy the search cache before the fork to do a delivery from a message
3028 received from the command line. Otherwise the child will trigger a lookup
3029 failure and thereby defer the delivery if it tries to use (for example) a
3030 cached ldap connection that the parent has called unbind on.
3032 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3033 of $address_data from the recipient verification was clobbered by the
3034 sender verification.
3036 13. The value of address_data from a sender verification is now available in
3037 $sender_address_data in subsequent conditions in the ACL statement.
3039 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3041 15. Added a new option "connect=<time>" to callout options, to set a different
3044 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3045 was its contents. (It was OK if the option was not defined at all.)
3047 17. A "Completed" log line is now written for messages that are removed from
3048 the spool by the -Mrm option.
3050 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3051 information about exactly what failed.
3053 19. Added -dd to debug only the daemon process.
3055 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3056 handles timeouts, both on the server side and network timeouts. Renamed the
3057 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3059 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3060 It is now set to "smtps".
3062 22. $host_address is now set to the target address during the checking of
3063 ignore_target_hosts.
3065 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3066 being passed; this would have caused $sender_host_name to have been used if
3067 matching the list had actually called for a host name (not very likely,
3068 since this list is usually IP addresses). A host name is now passed as
3071 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3072 code that specifies a non-restarting handler (typically sigaction() in
3073 modern systems) in an attempt to fix a rare and obscure crash bug.
3075 25. Narrowed the window for a race in the daemon that could cause it to ignore
3076 SIGCHLD signals. This is not a major problem, because they are used only to
3077 wake it up if nothing else does.
3079 26. A malformed maildirsize file could cause Exim to calculate negative values
3080 for the mailbox size or file count. Odd effects could occur as a result.
3081 The maildirsize information is now recalculated if the size or filecount
3084 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3085 support for a long time. Removed HAVE_SYS_VFS_H.
3087 28. Installed the latest version of exipick from John Jetmore.
3089 29. In an address list, if the pattern was not a regular expression, an empty
3090 subject address (from a bounce message) matched only if the pattern was an
3091 empty string. Non-empty patterns were not even tested. This was the wrong
3092 because it is perfectly reasonable to use an empty address as part of a
3093 database query. An empty address is now tested by patterns that are
3094 lookups. However, all the other forms of pattern expect the subject to
3095 contain a local part and a domain, and therefore, for them, an empty
3096 address still always fails if the pattern is not itself empty.
3098 30. Exim went into a mad DNS loop when attempting to do a callout where the
3099 host was specified on an smtp transport, and looking it up yielded more
3100 than one IP address.
3102 31. Re-factored the code for checking spool and log partition space into a
3103 function that finds that data and another that does the check. The former
3104 is then used to implement four new variables: $spool_space, $log_space,
3105 $spool_inodes, and $log_inodes.
3107 32. The RFC2047 encoding function was originally intended for short strings
3108 such as real names; it was not keeping to the 75-character limit for
3109 encoded words that the RFC imposes. It now respects the limit, and
3110 generates multiple encoded words if necessary. To be on the safe side, I
3111 have increased the buffer size for the ${rfc2047: expansion operator from
3114 33. It is now permitted to omit both strings after an "if" condition; if the
3115 condition is true, the result is "true". As before, when the second string
3116 is omitted, a false condition yields an empty string. This makes it less
3117 cumbersome to write custom ACL and router conditions.
3119 34. Failure to deliver a bounce message always caused it to be frozen, even if
3120 there was an errors_to setting on the router. The errors_to setting is now
3123 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3124 canonical form (fully expanded) before being placed in
3125 $sender_host_address.
3127 36. The table in the code that translates DNS record types into text (T_A to
3128 "A" for instance) was missing entries for NS and CNAME. It is just possible
3129 that this could have caused confusion if both these types were looked up
3130 for the same domain, because the text type is used as part of Exim's
3131 per-process caching. But the chance of anyone hitting this buglet seems
3134 37. The dnsdb lookup has been extended in a number of ways.
3136 (1) There is a new type, "zns", which walks up the domain tree until it
3137 finds some nameserver records. It should be used with care.
3139 (2) There is a new type, "mxh", which is like "mx" except that it returns
3140 just the host names, not the priorities.
3142 (3) It is now possible to give a list of domains (or IP addresses) to be
3143 looked up. The behaviour when one of the lookups defers can be
3144 controlled by a keyword.
3146 (4) It is now possible to specify the separator character for use when
3147 multiple records are returned.
3149 38. The dnslists ACL condition has been extended: it is now possible to supply
3150 a list of IP addresses and/or domains to be looked up in a particular DNS
3153 39. Added log_selector=+queue_time_overall.
3155 40. When running the queue in the test harness, wait just a tad after forking a
3156 delivery process, to get repeatability of debugging output.
3158 41. Include certificate and key file names in error message when GnuTLS fails
3159 to set them up, because the GnuTLS error message doesn't include the name
3160 of the failing file when there is a problem reading it.
3162 42. Allow both -bf and -bF in the same test run.
3164 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3166 44. The "Exiscan patch" is now merged into the mainline Exim source.
3168 45. Sometimes the final signoff response after QUIT could fail to get
3169 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3170 < 0 before doing a fflush(). This bug looks as though it goes back to the
3171 introduction of TLS in release 3.20, but "sometimes" must have been rare
3172 because the tests only now provoked it.
3174 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3175 (this can affect the format of dates).
3177 47. exim_tidydb, when checking for the continued existence of a message for
3178 which it has found a message-specific retry record, was not finding
3179 messages that were in split spool directories. Consequently, it was
3180 deleting retry records that should have stayed in existence.
3182 48. Steve fixed some bugs in eximstats.
3184 49. The SPA authentication driver was not abandoning authentication and moving
3185 on to the next authenticator when an expansion was forced to fail,
3186 contradicting the general specification for all authenticators. Instead it
3187 was generating a temporary error. It now behaves as specified.
3189 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3190 (the order specifies the preference for clients). The order is now AES256,
3191 AES128, 3DES, ARCFOUR128.
3193 51. Small patch to Sieve code - explicitly set From: when generating an
3196 52. Exim crashed if a remote delivery caused a very long error message to be
3197 recorded - for instance if somebody sent an entire SpamAssassin report back
3198 as a large number of 550 error lines. This bug was coincidentally fixed by
3199 increasing the size of one of Exim's internal buffers (big_buffer) that
3200 happened as part of the Exiscan merge. However, to be on the safe side, I
3201 have made the code more robust (and fixed the comments that describe what
3204 53. Now that there can be additional text after "Completed" in log lines (if
3205 the queue_time_overall log selector is set), a one-byte patch to exigrep
3206 was needed to allow it to recognize "Completed" as not the last thing in
3209 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3210 patch that reportedly fixes this has been added. I am not expert enough to
3211 create a test for it. This is what the patch creator wrote:
3213 "I found a little strange behaviour of ldap code when working with
3214 Windows 2003 AD Domain, where users was placed in more than one
3215 Organization Units. When I tried to give exim partial DN, the exit code
3216 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3217 But simultaneously result of request was absolutely normal ldap result,
3218 so I produce this patch..."
3220 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3221 REFERENCE, so I have modified the code to exclude the patch when that macro
3224 55. Some experimental protocols are using DNS PTR records for new purposes. The
3225 keys for these records are domain names, not reversed IP addresses. The
3226 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3227 leaves it alone. Component reversal etc. now happens only for IP addresses.
3230 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3232 57. Double the size of the debug message buffer (to 2048) so that more of very
3233 long debug lines gets shown.
3235 58. The exicyclog utility now does better if the number of log files to keep
3236 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3238 59. Two changes related to the smtp_active_hostname option:
3240 (1) $smtp_active_hostname is now available as a variable.
3241 (2) The default for smtp_banner uses $smtp_active_hostname instead
3242 of $primary_hostname.
3244 60. The host_aton() function is supposed to be passed a string that is known
3245 to be a valid IP address. However, in the case of IPv6 addresses, it was
3246 not checking this. This is a hostage to fortune. Exim now panics and dies
3247 if the condition is not met. A case was found where this could be provoked
3248 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3249 components; fortuitously, this particular loophole had already been fixed
3250 by change 4.50/55 above.
3252 If there are any other similar loopholes, the new check in host_aton()
3253 itself should stop them being exploited. The report I received stated that
3254 data on the command line could provoke the exploit when Exim was running as
3255 exim, but did not say which command line option was involved. All I could
3256 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3257 running as the user.
3260 61. There was a buffer overflow vulnerability in the SPA authentication code
3261 (which came originally from the Samba project). I have added a test to the
3262 spa_base64_to_bits() function which I hope fixes it.
3265 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3266 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3268 63. The daemon start-up calls getloadavg() while still root for those OS that
3269 need the first call to be done as root, but it missed one case: when
3270 deliver_queue_load_max is set with deliver_drop_privilege. This is
3271 necessary for the benefit of the queue runner, because there is no re-exec
3272 when deliver_drop_privilege is set.
3274 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3275 This has been fixed.
3277 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3278 was in use, was not putting the data itself into the right store pool;
3279 consequently, it could be overwritten for a subsequent message in the same
3280 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3283 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3285 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3286 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3287 the function were treating the return as a boolean value, which happened to
3288 work because 0=false and not-0=true, but is not correct code.
3290 68. The host_aton() function was not handling scoped IPv6 addresses (those
3291 with, for example, "%eth0" on the end) correctly.
3293 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3294 negated items (that is, ~something) in unsigned ints. Some compilers
3295 apparently mutter when there is no cast.
3297 70. If an address verification called from an ACL failed, and did not produce a
3298 user-specific message (i.e. there was only a "system" message), nothing was
3299 put in $acl_verify_message. In this situation, it now puts the system
3302 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3303 session; change 4.31/43 added the unwanted input to the log line - except
3304 that it did not do this in the start of session case. It now does.
3306 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3307 This is wrong and can cause the other end to generate a synchronization
3308 error if it is another Exim or anything else that does the synchronization
3309 check. A QUIT command is no longer sent after a timeout.
3311 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3312 during host lookups.
3314 74. The defer_ok option of callout verification was not working if it was used
3315 when verifying addresses in header lines, that is, for this case:
3317 verify = header_sender/callout=defer_ok
3319 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3320 those file descriptors could be used for SMTP connections. If anything
3321 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3322 could be sent to the SMTP client, causing chaos. The daemon now opens
3323 stdin, stdout, and stderr to /dev/null when it puts itself into the
3326 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3327 log by default. The user can override this with suitable Perl magic.
3329 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3330 the log message when discard triggers, was not working for the DATA ACL or
3331 for the non-SMTP ACL.
3333 78. Error message wording change in sieve.c.
3335 79. If smtp_accept_max_per_host was set, the number of connections could be
3336 restricted to fewer than expected, because the daemon was trying to set up
3337 a new connection before checking whether the processes handling previous
3338 connections had finished. The check for completed processes is now done
3339 earlier. On busy systems, this bug wouldn't be noticed because something
3340 else would have woken the daemon, and it would have reaped the completed
3343 80. If a message was submitted locally by a user whose login name contained one
3344 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3345 It caused a spool format error. I have fixed the spool reading code. A
3346 related problem was that the "from" clause in the Received: line became
3347 illegal because of the space(s). It is now covered by ${quote_local_part.
3349 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3352 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3354 83. Patch from Sieve maintainer to fix unterminated string problem in
3355 "vacation" handling.
3357 84. Some minor changes to the Linux configuration files to help with other
3358 OS variants using glibc.
3360 85. One more patch for Sieve to update vacation handling to latest spec.
3363 ----------------------------------------------------
3364 See the note above about the 4.44 and 4.50 releases.
3365 ----------------------------------------------------
3371 1. Change 4.43/35 introduced a bug that caused file counts to be
3372 incorrectly computed when quota_filecount was set in an appendfile
3375 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3376 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3379 3. Give more explanation in the error message when the command for a transport
3380 filter fails to execute.
3382 4. There are several places where Exim runs a non-Exim command in a
3383 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3384 was being done only for the command run by the queryprogram router. It is
3385 now done for all such subprocesses. The other cases are: ${run, transport
3386 filters, and the commands run by the lmtp and pipe transports.
3388 5. Some older OS have a limit of 256 on the maximum number of file
3389 descriptors. Exim was using setrlimit() to set 1000 as a large value
3390 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3391 systems. I've change it so that if it can't get 1000, it tries for 256.
3393 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3394 was an oversight, and furthermore, ever since the addition of extra
3395 controls (e.g. 4.43/32), the checks on when to allow different forms of
3396 "control" were broken. There should now be diagnostics for all cases when a
3397 control that does not make sense is encountered.
3399 7. $recipients is now available in the predata ACL (oversight).
3401 8. Tidy the search cache before the fork to do a delivery from a message
3402 received from the command line. Otherwise the child will trigger a lookup
3403 failure and thereby defer the delivery if it tries to use (for example) a
3404 cached ldap connection that the parent has called unbind on.
3406 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3407 of $address_data from the recipient verification was clobbered by the
3408 sender verification.
3410 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3411 was its contents. (It was OK if the option was not defined at all.)
3413 11. A "Completed" log line is now written for messages that are removed from
3414 the spool by the -Mrm option.
3416 12. $host_address is now set to the target address during the checking of
3417 ignore_target_hosts.
3419 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3420 being passed; this would have caused $sender_host_name to have been used if
3421 matching the list had actually called for a host name (not very likely,
3422 since this list is usually IP addresses). A host name is now passed as
3425 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3426 code that specifies a non-restarting handler (typically sigaction() in
3427 modern systems) in an attempt to fix a rare and obscure crash bug.
3429 15. Narrowed the window for a race in the daemon that could cause it to ignore
3430 SIGCHLD signals. This is not a major problem, because they are used only to
3431 wake it up if nothing else does.
3433 16. A malformed maildirsize file could cause Exim to calculate negative values
3434 for the mailbox size or file count. Odd effects could occur as a result.
3435 The maildirsize information is now recalculated if the size or filecount
3438 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3439 support for a long time. Removed HAVE_SYS_VFS_H.
3441 18. Updated exipick to current release from John Jetmore.
3443 19. Allow an empty sender to be matched against a lookup in an address list.
3444 Previously the only cases considered were a regular expression, or an
3447 20. Exim went into a mad DNS lookup loop when doing a callout where the
3448 host was specified on the transport, if the DNS lookup yielded more than
3451 21. The RFC2047 encoding function was originally intended for short strings
3452 such as real names; it was not keeping to the 75-character limit for
3453 encoded words that the RFC imposes. It now respects the limit, and
3454 generates multiple encoded words if necessary. To be on the safe side, I
3455 have increased the buffer size for the ${rfc2047: expansion operator from
3458 22. Failure to deliver a bounce message always caused it to be frozen, even if
3459 there was an errors_to setting on the router. The errors_to setting is now
3462 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3463 canonical form (fully expanded) before being placed in
3464 $sender_host_address.
3466 24. Updated eximstats to version 1.33
3468 25. Include certificate and key file names in error message when GnuTLS fails
3469 to set them up, because the GnuTLS error message doesn't include the name
3470 of the failing file when there is a problem reading it.
3472 26. Expand error message when OpenSSL has problems setting up cert/key files.
3475 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3476 (this can affect the format of dates).
3478 28. exim_tidydb, when checking for the continued existence of a message for
3479 which it has found a message-specific retry record, was not finding
3480 messages that were in split spool directories. Consequently, it was
3481 deleting retry records that should have stayed in existence.
3483 29. eximstats updated to version 1.35
3484 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3485 1.35 - bugfix such that pie charts by volume are generated correctly
3487 30. The SPA authentication driver was not abandoning authentication and moving
3488 on to the next authenticator when an expansion was forced to fail,
3489 contradicting the general specification for all authenticators. Instead it
3490 was generating a temporary error. It now behaves as specified.
3492 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3493 (the order specifies the preference for clients). The order is now AES256,
3494 AES128, 3DES, ARCFOUR128.
3496 31. Small patch to Sieve code - explicitly set From: when generating an
3499 32. Exim crashed if a remote delivery caused a very long error message to be
3500 recorded - for instance if somebody sent an entire SpamAssassin report back
3501 as a large number of 550 error lines. This bug was coincidentally fixed by
3502 increasing the size of one of Exim's internal buffers (big_buffer) that
3503 happened as part of the Exiscan merge. However, to be on the safe side, I
3504 have made the code more robust (and fixed the comments that describe what
3507 33. Some experimental protocols are using DNS PTR records for new purposes. The
3508 keys for these records are domain names, not reversed IP addresses. The
3509 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3510 leaves it alone. Component reversal etc. now happens only for IP addresses.
3513 34. The host_aton() function is supposed to be passed a string that is known
3514 to be a valid IP address. However, in the case of IPv6 addresses, it was
3515 not checking this. This is a hostage to fortune. Exim now panics and dies
3516 if the condition is not met. A case was found where this could be provoked
3517 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3518 components; fortuitously, this particular loophole had already been fixed
3519 by change 4.50/55 or 4.44/33 above.
3521 If there are any other similar loopholes, the new check in host_aton()
3522 itself should stop them being exploited. The report I received stated that
3523 data on the command line could provoke the exploit when Exim was running as
3524 exim, but did not say which command line option was involved. All I could
3525 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3526 running as the user.
3529 35. There was a buffer overflow vulnerability in the SPA authentication code
3530 (which came originally from the Samba project). I have added a test to the
3531 spa_base64_to_bits() function which I hope fixes it.
3534 36. The daemon start-up calls getloadavg() while still root for those OS that
3535 need the first call to be done as root, but it missed one case: when
3536 deliver_queue_load_max is set with deliver_drop_privilege. This is
3537 necessary for the benefit of the queue runner, because there is no re-exec
3538 when deliver_drop_privilege is set.
3540 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3541 was in use, was not putting the data itself into the right store pool;
3542 consequently, it could be overwritten for a subsequent message in the same
3543 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3546 38. Sometimes the final signoff response after QUIT could fail to get
3547 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3548 < 0 before doing a fflush(). This bug looks as though it goes back to the
3549 introduction of TLS in release 3.20, but "sometimes" must have been rare
3550 because the tests only now provoked it.
3556 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3557 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3558 Now it returns an integer. A number of calls were still expecting a T/F
3559 return. Fortuitously, in all cases, the tests worked in OK situations,
3560 which is the norm. However, things would have gone wrong on any write
3561 failures on the smtp file descriptor. This function is used when sending
3562 messages over SMTP and also when doing verify callouts.
3564 2. When Exim is called to do synchronous delivery of a locally submitted
3565 message (the -odf or -odi options), it no longer closes stderr before doing
3568 3. Implemented the mua_wrapper option.
3570 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3572 5. Implemented the functions header_remove(), header_testname(),
3573 header_add_at_position(), and receive_remove_recipient(), and exported them
3576 6. If an ACL "warn" statement specified the addition of headers, Exim already
3577 inserted X-ACL-Warn: at the start if there was no header name. However, it
3578 was not making this test for the second and subsequent header lines if
3579 there were newlines in the string. This meant that an invalid header could
3580 be inserted if Exim was badly configured.
3582 7. Allow an ACL "warn" statement to add header lines at the start or after all
3583 the Received: headers, as well as at the end.
3585 8. Added the rcpt_4xx retry error code.
3587 9. Added postmaster_mailfrom=xxx to callout verification option.
3589 10. Added mailfrom=xxxx to the callout verification option, for verify=
3592 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3593 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3595 12. Inserted some casts to stop certain compilers warning when using pointer
3596 differences as field lengths or precisions in printf-type calls (mostly
3597 affecting debugging statements).
3599 13. Added optional readline() support for -be (dynamically loaded).
3601 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3602 same clock tick as a message's arrival, so that its received time was the
3603 same as the "first fail" time on the retry record, and that message
3604 remained on the queue past the ultimate address timeout, every queue runner
3605 would try a delivery (because it was past the ultimate address timeout) but
3606 after another failure, the ultimate address timeout, which should have then
3607 bounced the address, did not kick in. This was a "< instead of <=" error;
3608 in most cases the first failure would have been in the next clock tick
3609 after the received time, and all would be well.
3611 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3612 being recognized when the domain list was tested by the match_domain
3613 condition in an expansion string.
3615 16. Added the ${str2b64: operator.
3617 17. Exim was always calling setrlimit() to set a large limit for the number of
3618 processes, without checking whether the existing limit was already
3619 adequate. (It did check for the limit on file descriptors.) Furthermore,
3620 errors from getrlimit() and setrlimit() were being ignored. Now they are
3621 logged to the main and panic logs, but Exim does carry on, to try to do its
3622 job under whatever limits there are.
3624 18. Imported PCRE 5.0.
3626 19. Trivial typo in log message " temporarily refused connection" (the leading
3629 20. If the log selector return_path_on_delivery was set and an address was
3630 redirected to /dev/null, the delivery process crashed because it assumed
3631 that a return path would always be set for a "successful" delivery. In this
3632 case, the whole delivery is bypassed as an optimization, and therefore no
3635 21. Internal re-arrangement: the function for sending a challenge and reading
3636 a response while authentication was assuming a zero-terminated challenge
3637 string. It's now changed to take a pointer and a length, to allow for
3638 binary data in such strings.
3640 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3642 23. Exim was not respecting finduser_retries when seeking the login of the
3643 uid under which it was called; it was always trying 10 times. (The default
3644 setting of finduser_retries is zero.) Also, it was sleeping after the final
3645 failure, which is pointless.
3647 24. Implemented tls_on_connect_ports.
3649 25. Implemented acl_smtp_predata.
3651 26. If the domain in control=submission is set empty, Exim assumes that the
3652 authenticated id is a complete email address when it generates From: or
3653 Sender: header lines.
3655 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3656 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3657 chown and chgrp in /bin and hostname in /usr/bin.
3659 28. Exim was keeping the "process log" file open after each use, just as it
3660 does for the main log. This opens the possibility of it remaining open for
3661 long periods when the USR1 signal hits a daemon. Occasional processlog
3662 errors were reported, that could have been caused by this. Anyway, it seems
3663 much more sensible not to leave this file open at all, so that is what now
3666 29. The long-running daemon process does not normally write to the log once it
3667 has entered its main loop, and it closes the log before doing so. This is
3668 so that log files can straightforwardly be renamed and moved. However,
3669 there are a couple of unusual error situations where the daemon does write
3670 log entries, and I had neglected to close the log afterwards.
3672 30. The text of an SMTP error response that was received during a remote
3673 delivery was being truncated at 512 bytes. This is too short for some of
3674 the long messages that one sometimes sees. I've increased the limit to
3677 31. It is now possible to make retry rules that apply only when a message has a
3678 specific sender, in particular, an empty sender.
3680 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3681 it possible to be selective about when SMTP synchronization is enforced.
3683 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3685 32. Implemented hosts_connection_nolog.
3687 33. Added an ACL for QUIT.
3689 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3692 35. Added mailbox_size and mailbox_filecount to appendfile.
3694 36. Added control = no_multiline_responses to ACLs.
3696 37. There was a bug in the logic of the code that waits for the clock to tick
3697 in the case where the clock went backwards by a substantial amount such
3698 that the microsecond fraction of "now" was more than the microsecond
3699 fraction of "then" (but the whole seconds number was less).
3701 38. Added support for the libradius Radius client library this is found on
3702 FreeBSD (previously only the radiusclient library was supported).
3708 1. When certain lookups returned multiple values in the form name=value, the
3709 quoting of the values was not always being done properly. Specifically:
3710 (a) If the value started with a double quote, but contained no whitespace,
3712 (b) If the value contained whitespace other than a space character (i.e.
3713 tabs or newlines or carriage returns) it was not quoted.
3714 This fix has been applied to the mysql and pgsql lookups by writing a
3715 separate quoting function and calling it from the lookup code. The fix
3716 should probably also be applied to nisplus, ibase and oracle lookups, but
3717 since I cannot test any of those, I have not disturbed their existing code.
3719 2. A hit in the callout cache for a specific address caused a log line with no
3720 reason for rejecting RCPT. Now it says "Previous (cached) callout
3721 verification failure".
3723 3. There was an off-by-one bug in the queryprogram router. An over-long
3724 return line was truncated at 256 instead of 255 characters, thereby
3725 overflowing its buffer with the terminating zero. As well as fixing this, I
3726 have increased the buffer size to 1024 (and made a note to document this).
3728 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3729 when Exim is waiting for an SMTP response from a remote server, Exim
3730 restarts its select() call on the socket, thereby resetting its timeout.
3731 This is not a problem when such interrupts are rare. Somebody set up a cron
3732 job to run exiwhat every 2 minutes, which is less than the normal select()
3733 timeout (5 or 10 minutes). This meant that the select() timeout never
3734 kicked in because it was always reset. I have fixed this by comparing the
3735 time when an interrupt arrives with the time at the start of the first call
3736 to select(). If more time than the timeout has elapsed, the interrupt is
3737 treated as a timeout.
3739 5. Some internal re-factoring in preparation for the addition of Sieve
3740 extensions (by MH). In particular, the "personal" test is moved to a
3741 separate function, and given an option for scanning Cc: and Bcc: (which is
3742 not set for Exim filters).
3744 6. When Exim created an email address using the login of the caller as the
3745 local part (e.g. when creating a From: or Sender: header line), it was not
3746 quoting the local part when it contained special characters such as @.
3748 7. Installed new OpenBSD configuration files.
3750 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3751 try to make them clearer.
3753 9. Callout options, other than the timeout value, were being ignored when
3754 verifying sender addresses in header lines. For example, when using
3756 verify = header_sender/callout=no_cache
3758 the cache was (incorrectly) being used.
3760 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3761 only the Cygwin environment.
3763 11. When return_path_on_delivery was set as a log selector, if different remote
3764 addresses in the same message used different return paths and parallel
3765 remote delivery occurred, the wrong values would sometimes be logged.
3766 (Whenever a remote delivery process finished, the return path value from
3767 the most recently started remote delivery process was logged.)
3769 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3770 lines when AUTH and/or TLS are in use. This is the "received protocol"
3771 field. Exim used to use "asmtp" for authenticated SMTP, without any
3772 indication (in the protocol name) for TLS use. Now it follows the RFC and
3773 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3774 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3775 names appear in log lines as well as in Received: header lines.
3777 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3778 extensions, and comparison tests, and to fix some bugs.
3780 14. Changes to the "personal" filter test:
3782 (1) The test was buggy in that it was just doing the equivalent of
3783 "contains" tests on header lines. For example, if a user's address was
3784 anne@some.where, the "personal" test would incorrectly be true for
3786 To: susanne@some.where
3788 This test is now done by extracting each address from the header in turn,
3789 and checking the entire address. Other tests that are part of "personal"
3790 are now done using regular expressions (for example, to check local parts
3791 of addresses in From: header lines).
3793 (2) The list of non-personal local parts in From: addresses has been
3794 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3795 taken from the Sieve specification recommendations.
3797 (3) If the message contains any header line starting with "List-" it is
3798 treated as non-personal.
3800 (4) The test for "circular" in the Subject: header line has been removed
3801 because it now seems ill-conceived.
3803 15. Minor typos in src/EDITME comments corrected.
3805 16. Installed latest exipick from John Jetmore.
3807 17. If headers_add on a router specified a text string that was too long for
3808 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3809 of string_sprintf() is now avoided.
3811 18. $message_body_size was not set (it was always zero) when running the DATA
3812 ACL and the local_scan() function.
3814 19. For the "mail" command in an Exim filter, no default was being set for
3815 the once_repeat time, causing a random time value to be used if "once" was
3816 specified. (If the value happened to be <= 0, no repeat happened.) The
3817 default is now 0s, meaning "never repeat". The "vacation" command was OK
3818 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3819 (I found it when inspecting the code).
3821 20. There is now an overall timeout for performing a callout verification. It
3822 defaults to 4 times the callout timeout, which applies to individual SMTP
3823 commands during the callout. The overall timeout applies when there is more
3824 than one host that can be tried. The timeout is checked before trying the
3825 next host. This prevents very long delays if there are a large number of
3826 hosts and all are timing out (e.g. when the network connections are timing
3827 out). The value of the overall timeout can be changed by specifying an
3828 additional sub-option for "callout", called "maxwait". For example:
3830 verify = sender/callout=5s,maxwait=20s
3832 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3833 to the end before writing, but this should make it even safer).
3835 22. Exim was forgetting that it had advertised PIPELINING for the second and
3836 subsequent messages on an SMTP connection. It was also not resetting its
3837 memory on STARTTLS and an internal HELO.
3839 23. When Exim logs an SMTP synchronization error within a session, it now
3840 records whether PIPELINING has been advertised or not.
3842 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3843 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3844 rather than long int.
3846 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3848 26. Added the never_mail option to autoreply.
3854 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3855 crash if the getsockname() call failed; this can happen if a connection is
3856 closed very soon after it is established. The problem was simply in the
3857 order in which certain operations were done, causing Exim to try to write
3858 to the SMTP stream before it had set up the file descriptor. The bug has
3859 been fixed by making things happen in the correct order.
3865 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3866 before the connection was closed, thus losing the rejection response.
3868 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3869 some early Solaris releases, but causes trouble in current releases where
3870 socklen_t is defined.
3872 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3875 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3878 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3879 that releasing the top of it at the end releases what was used for sub-
3880 expansions (unless the block got too big). However, discard this block if
3881 the first thing is a variable or header, so that we can use its block when
3882 it is dynamic (useful for very large $message_headers, for example).
3884 6. Lookups now cache *every* query, not just the most recent. A new, separate
3885 store pool is used for this. It can be recovered when all lookup caches are
3886 flushed. Lookups now release memory at the end of their result strings.
3887 This has involved some general refactoring of the lookup sources.
3889 7. Some code has been added to the store_xxx() functions to reduce the amount
3890 of flapping under certain conditions.
3892 8. log_incoming_interface used to affect only the <= reception log lines. Now
3893 it causes the local interface and port to be added to several more SMTP log
3894 lines, for example "SMTP connection from", and rejection lines.
3896 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3898 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3900 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3901 could be overwritten at the end of the current message (or the start of a
3902 new message if it was set in a HELO ACL). The value is now preserved for
3903 the duration of the SMTP connection.
3905 12. If a transport had a headers_rewrite setting, and a matching header line
3906 contained an unqualified address, that address was qualified, even if it
3907 did not match any rewriting rules. The underlying bug was that the values
3908 of the flags that permit the existence of unqualified sender and recipient
3909 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3910 non-local messages, and by -bnq for local messages) were not being
3911 preserved with the message after it was received.
3913 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3914 "next input=" as part of the text comprising the host identity instead of
3915 the correct text. The code was using the same buffer for two different
3916 strings. However, depending on which order the printing function evaluated
3917 its arguments, the bug did not always show up. Under Linux, for example, my
3918 test suite worked just fine.
3920 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3921 doesn't work with some older versions of Perl. It has been changed to "my",
3922 which in any case is probably the better facility to use.
3924 15. A really picky compiler found some instances of statements for creating
3925 error messages that either had too many or two few arguments for the format
3928 16. The size of the buffer for calls to the DNS resolver has been increased
3929 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3930 for addresses that have a lot of PTR records. This alleviates a problem; it
3931 does not fully solve it.
3933 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3934 buffer now truncates the list and logs the incident, which is the same
3935 action as happens when Exim is looking up a host name and its aliases.
3936 Previously in this situation something unpredictable would happen;
3937 sometimes it was "internal error: store_reset failed".
3939 18. If a server dropped the connection unexpectedly when an Exim client was
3940 using GnuTLS and trying to read a response, the client delivery process
3941 crashed while trying to generate an error log message.
3943 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3944 string, for example:
3946 warn message = H1: something\nH2: something
3948 the text was added as a single header line from Exim's point of view
3949 though it ended up OK in the delivered message. However, searching for the
3950 second and subsequent header lines using $h_h2: did not work. This has been
3951 fixed. Similarly, if a system filter added multiple headers in this way,
3952 the routers could not see them.
3954 20. Expanded the error message when iplsearch is called with an invalid key to
3955 suggest using net-iplsearch in a host list.
3957 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3958 ACLs are no longer actually imposed (and a message to that effect is
3961 22. If a "gecos" field in a passwd entry contained escaped characters, in
3962 particular, if it contained a \" sequence, Exim got it wrong when building
3963 a From: or a Sender: header from that name. A second bug also caused
3964 incorrect handling when an unquoted " was present following a character
3965 that needed quoting.
3967 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3968 was not being matched caselessly.
3970 24. Arranged for all hyphens in the exim.8 source to be escaped with
3973 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3974 properly. Recipient callout cache records were still being keyed to include
3975 the sender, even when use_sender was set false. This led to far more
3976 callouts that were necessary. The sender is no longer included in the key
3977 when use_sender is false.
3979 26. Added "control = submission" modifier to ACLs.
3981 27. Added the ${base62d: operator to decode base 62 numbers.
3983 28. dnsdb lookups can now access SRV records.
3985 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3986 the configuration file.
3988 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3989 (-v) mode. This makes the output for a verbose queue run more intelligible.
3991 31. Added a use_postmaster feature to recipient callouts.
3993 32. Added the $body_zerocount variable, containing the number of binary zero
3994 bytes in the message body.
3996 33. The time of last modification of the "new" subdirectory is now used as the
3997 "mailbox time last read" when there is a quota error for a maildir
4000 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4002 35. Added +ignore_unknown as a special item in host lists.
4004 36. Code for decoding IPv6 addresses in host lists is now included, even if
4005 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4006 address was recognized as an IP address, but was then not correctly decoded
4007 into binary, causing unexpected and incorrect effects when compared with
4014 1. Very minor rewording of debugging text in manualroute to say "list of
4015 hosts" instead of "hostlist".
4017 2. If verify=header_syntax was set, and a header line with an unqualified
4018 address (no domain) and a large number of spaces between the end of the
4019 name and the colon was received, the reception process suffered a buffer
4020 overflow, and (when I tested it) crashed. This was caused by some obsolete
4021 code that should have been removed. The fix is to remove it!
4023 3. When running in the test harness, delay a bit after writing a bounce
4024 message to get a bit more predictability in the log output.
4026 4. Added a call to search_tidyup() just before forking a reception process. In
4027 theory, someone could use a lookup in the expansion of smtp_accept_max_
4028 per_host which, without the tidyup, could leave open a database connection.
4030 5. Added the variables $recipient_data and $sender_data which get set from a
4031 lookup success in an ACL "recipients" or "senders" condition, or a router
4032 "senders" option, similar to $domain_data and $local_part_data.
4034 6. Moved the writing of debug_print from before to after the "senders" test
4037 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4038 problems for message scanning, either using a data ACL, or using
4039 local_scan() because the Received: header was not generated till after they
4040 were called (in order to set the time as the time of reception completion).
4041 I have revised the way this works. The header is now generated after the
4042 body is received, but before the ACL or local_scan() are called. After they
4043 are run, the timestamp in the header is updated.
4049 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4050 before starting a queue runner without re-exec. This happened only when
4051 deliver_drop_privilege was set or when the Exim user was set to root. The
4052 effect of the bug was that timeouts during subsequent deliveries caused
4053 crashes instead of being properly handled. The handler is now left at its
4054 default (and expected) setting.
4056 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4057 message, again when deliver_drop_privilege is set or Exim is run as root.
4058 The bug described in (1) was not present in this case, but the tidying up
4059 of the other signals was missing. I have made the two cases consistent.
4061 3. The ignore_target_hosts setting on a manualroute router was being ignored
4062 for hosts that were looked up using the /MX notation.
4064 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4067 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4068 operated on the sender address. After changing the $sender_address to <>
4069 for the sender address verify, Exim was re-instated it as the original
4070 (before rewriting) address, but remembering that it had rewritten it, so it
4071 wasn't rewriting it again. This bug also had the effect of breaking the
4072 sender address verification caching when the sender address was rewritten.
4074 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4075 This has been changed so that if the ip literal address matches
4076 ignore_target_hosts, the router declines.
4078 7. Added expansion conditions match_domain, match_address, and match_local_
4079 part (NOT match_host).
4081 8. The placeholder for the Received: header didn't have a length field set.
4083 9. Added code to Exim itself and to exim_lock to test for a specific race
4084 condition that could lead to file corruption when using MBX delivery. The
4085 issue is with the lockfile that is created in /tmp. If this file is removed
4086 after a process has opened it but before that process has acquired a lock,
4087 there is the potential for a second process to recreate the file and also
4088 acquire a lock. This could lead to two Exim processes writing to the file
4089 at the same time. The added code performs the same test as UW imapd; it
4090 checks after acquiring the lock that its file descriptor still refers to
4091 the same named file.
4093 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4094 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4097 11. Added the smtp_active_hostname option. If used, this will typically be made
4098 to depend on the incoming interface address. Because $interface_address is
4099 not set up until the daemon has forked a reception process, error responses
4100 that can happen earlier (such as "too many connections") no longer contain
4103 12. If an expansion in a condition on a "warn" statement fails because a lookup
4104 defers, the "warn" statement is abandoned, and the next ACL statement is
4105 processed. Previously this caused the whole ACL to be aborted.
4107 13. Added the iplsearch lookup type.
4109 14. Added ident_timeout as a log selector.
4111 15. Added tls_certificate_verified as a log selector.
4113 16. Added a global option tls_require_ciphers (compare the smtp transport
4114 option of the same name). This controls incoming TLS connections.
4116 17. I finally figured out how to make tls_require_ciphers do a similar thing
4117 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4118 before starting the TLS session.
4120 18. Tabs are now shown as \t in -bP output.
4122 19. If the log selector return_path_on_delivery was set, Exim crashed when
4123 bouncing a message because it had too many Received: header lines.
4125 20. If two routers both had headers_remove settings, and the first one included
4126 a superfluous trailing colon, the final name in the first list and the
4127 first name in the second list were incorrectly joined into one item (with a
4128 colon in the middle).
4134 1. Added -C and -D options to the exinext utility, mainly to make it easier
4135 to include in the automated testing, but these could be helpful when
4136 multiple configurations are in use.
4138 2. The exinext utility was not formatting the output nicely when there was
4139 an alternate port involved in the retry record key, nor when there was a
4140 message id as well (for retries that were specific to a specific message
4141 and a specific host). It was also confused by IPv6 addresses, because of
4142 the additional colons they contain. I have fixed the IPv4 problem, and
4143 patched it up to do a reasonable job for IPv6.
4145 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4146 delivery, the log line now contains "pipelined" if PIPELINING was used.
4148 4. An SMTP transport process used to panic and die if the bind() call to set
4149 an explicit outgoing interface failed. This has been changed; it is now
4150 treated in the same way as a connect() failure.
4152 5. A reference to $sender_host_name in the part of a conditional expansion
4153 that was being skipped was still causing a DNS lookup. This no longer
4156 6. The def: expansion condition was not recognizing references to header lines
4157 that used bh_ and bheader_.
4159 7. Added the _cache feature to named lists.
4161 8. The code for checking quota_filecount in the appendfile transport was
4162 allowing one more file than it should have been.
4164 9. For compatibility with Sendmail, the command line option
4172 and sets the incoming protocol and host name (for trusted callers). The
4173 host name and its colon can be omitted when only the protocol is to be set.
4174 Note the Exim already has two private options, -pd and -ps, that refer to
4175 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4176 "s", but I don't think that's a major issue.
4178 10. A number of refactoring changes to the code, none of which should affect
4181 (a) The number of logging options was getting close to filling up the
4182 32-bit word that was used as a bit map. I have split them into two classes:
4183 those that are passed in the argument to log_write(), and those that are
4184 only ever tested independently outside of that function. These are now in
4185 separate 32-bit words, so there is plenty of room for expansion again.
4186 There is no change in the user interface or the logging behaviour.
4188 (b) When building, for example, log lines, the code previously used a
4189 macro that called string_cat() twice, in order to add two strings. This is
4190 not really sufficiently general. Furthermore, there was one instance where
4191 it was actually wrong because one of the argument was used twice, and in
4192 one call a function was used. (As it happened, calling the function twice
4193 did not affect the overall behaviour.) The macro has been replaced by a
4194 function that can join an arbitrary number of extra strings onto a growing
4197 (c) The code for expansion conditions now uses a table and a binary chop
4198 instead of a serial search (which was left over from when there were very
4199 few conditions). Also, it now recognizes conditions like "pam" even when
4200 the relevant support is not compiled in: a suitably worded error message is
4201 given if an attempt is made to use such a condition.
4203 11. Added ${time_interval:xxxxx}.
4205 12. A bug was causing one of the ddress fields not to be passed back correctly
4206 from remote delivery subprocesses. The field in question was not being
4207 subsequently used, so this caused to problems in practice.
4209 13. Added new log selectors queue_time and deliver_time.
4211 14. Might have fixed a bug in maildirsizefile handling that threw up
4212 "unexpected character" debug warnings, and recalculated the data
4213 unnecessarily. In any case, I expanded the warning message to give more
4216 15. Added the message "Restricted characters in address" to the statements in
4217 the default ACL that block characters like @ and % in local parts.
4219 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4220 Three changes have been made:
4222 (a) There was a serious bug; a negative response to MAIL caused the whole
4223 recipient domain to be cached as invalid, thereby blocking all messages
4224 to all local parts at the same domain, from all senders. This bug has
4225 been fixed. The domain is no longer cached after a negative response to
4226 MAIL if the sender used is not empty.
4228 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4231 (c) A new callout option, "use_sender" has been added for people who want
4232 the modified behaviour.
4238 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4241 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4242 indeed breaks things for older releases.
4244 3. Added additional logging to the case where there is a problem reading data
4245 from a filter that is running in a subprocess using a pipe, in order to
4246 try to track down a specific problem.
4248 4. Testing facility fudge: when running in the test harness and attempting
4249 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4250 getting "No route to host". Convert this to a timeout.
4252 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4255 6. Some OS don't have socklen_t but use size_t instead. This affects the
4256 fifth argument of getsockopt() amongst other things. This is now
4257 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4258 can be set for individual OS. I have set it for SunOS5, OSF1, and
4259 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4260 some earlier ones do not.
4262 7. Change 4.30/15 was not doing the test caselessly.
4264 8. The standard form for an IPv6 address literal was being rejected by address
4265 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4266 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4267 this, as well as the form without the "IPv6" on the front (but only when
4268 address literals are enabled, of course).
4270 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4272 10. Exim crashed if a message with an empty sender address specified by -f
4273 encountered a router with an errors_to setting. This could be provoked only
4274 by a command such as
4278 where an empty string was supplied; "<>" did not hit this bug.
4280 11. Installed PCRE release 4.5.
4282 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4283 remained set. It is now erased.
4285 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4286 times from message ids (which are base 36 rather than the normal 62).
4288 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4289 were being counted as actual protocol errors, and logged if the log
4290 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4291 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4292 and DATA following a set of rejected RCPTs do not count as protocol errors.
4293 In other words, Exim assumes they were pipelined, though this may not
4294 actually be the case. Of course, in all cases the client gets an
4295 appropriate error code.
4297 15. If a lookup fails in an ACL condition, a message about the failure may
4298 be available; it is used if testing the ACL cannot continue, because most
4299 such messages specify what the cause of the deferral is. However, some
4300 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4301 that caused an old message to be retained and used if a later statement
4302 caused a defer, replacing the real cause of the deferral.
4304 16. If an IP address had so many PTR records that the DNS lookup buffer
4305 was not large enough to hold them, Exim could crash while trying to process
4306 the truncated data. It now detects and logs this case.
4308 17. Further to 4.21/58, another change has been made: if (and only if) the
4309 first line of a message (the first header line) ends with CRLF, a bare LF
4310 in a subsequent header line has a space inserted after it, so as not to
4311 terminate the header.
4313 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4314 unnecessarily, used atoi() instead of strtol(), and didn't check the
4315 termination when getting file sizes from file names by regex.
4317 19. Completely re-implemented the support for maildirsize files, in the light
4318 of a number of problems with the previous contributed implementation
4319 (4.30/29). In particular:
4321 . If the quota is zero, the maildirsize file is maintained, but no quota is
4324 . If the maildir directory does not exist, it is created before any attempt
4325 to write a maildirsize file.
4327 . The quota value in the file is just a cache; if the quota is changed in
4328 the transport, the new value overrides.
4330 . A regular expression is available for excluding directories from the
4333 20. The autoreply transport checks the characters in options that define the
4334 message's headers; it allows continued headers, but it was checking with
4335 isspace() after an embedded newline instead of explicitly looking for a
4338 21. If all the "regular" hosts to which an address was routed had passed their
4339 expiry times, and had not reached their retry times, the address was
4340 bounced, even if fallback hosts were defined. Now Exim should go on to try
4343 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4344 equivalent code in the SMTP transport. Some hosts send humungous responses
4345 to HELO/EHLO, more than 1024 it seems.
4347 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4348 gives compiler warnings in some environments. I've now done it "properly",
4351 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4352 (because of problems with the built-in one) was declared to return uschar *
4353 instead of char *, causing compiler failure.
4355 25. Fixed a file descriptor leak when processing alias/forward files.
4357 26. Fixed a minor format string issue in dbfn.c.
4359 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4361 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4362 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4365 29. When a sender address is verified, it is cached, to save repeating the test
4366 when there is more than one recipient in a message. However, when the
4367 verification involves a callout, it is possible for different callout
4368 options to be set for different recipients. It is too complicated to keep
4369 track of this in the cache, so now Exim always runs a verification when a
4370 callout is required, relying on the callout cache for the optimization.
4371 The overhead is duplication of the address routing, but this should not be
4374 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4375 to be verified with callout=postmaster, and the main callout worked but the
4376 postmaster check failed, the verification correctly failed. However, if a
4377 subsequent RCPT command asked for sender verification *without* the
4378 postmaster check, incorrect caching caused this verification also to fail,
4381 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4382 it was not caching the DNS options (qualify_single, search_parents) that
4383 were used when the lookup failed. A subsequent lookup with different
4384 options therefore always gave the same answer, though there were cases
4385 where it should not have. (Example: a "domains = !$mx_any" option on a
4386 dnslookup router: the "domains" option is always processed without any
4387 widening, but the router might have qualify_single set.) Now Exim uses the
4388 cached value only when the same options are set.
4390 32. Added John Jetmore's "exipick" utility to the distribution.
4392 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4393 than a timeout (e.g. a certificate is required, and is not provided), an
4394 Exim server now closes the connection immediately. Previously it waited for
4395 the client to close - but if the client is SSL, it seems that they each
4396 wait for each other, leading to a delay before one of them times out.
4398 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4399 maintained 0.8.x compatibility because I don't think many are using it, and
4400 it is clearly obsolete.
4402 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4405 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4406 client certificate was expired. A simple patch fixes this, though I don't
4407 understand the full logic of why the verify callback is called multiple
4410 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4411 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4412 which causes problems with some clients (such as the Certicom SSL Plus
4413 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4414 disables the coutermeasure allowing Eudora to connect."
4416 38. Exim was not checking that a write() to a log file succeeded. This could
4417 lead to Bad Things if a log got too big, in particular if it hit a file
4418 size limit. Exim now panics and dies if it cannot write to a log file, just
4419 as it does if it cannot open a log file.
4421 39. Modified OS/Makefile-Linux so that it now contains
4423 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4425 The two -D definitions ensure that Exim is compiled with large file
4426 support, which makes it possible to handle log files that are bigger than
4429 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4430 instance) a domain was checked against a named list that involved a lookup,
4431 causing $domain_data to be set, then another domain was checked against the
4432 same list, then the first domain was re-checked, the value of $domain_data
4433 after the final check could be wrong. In particular, if the second check
4434 failed, it could be set empty. This bug probably also applied to
4437 41. The strip_trailing_dot option was not being applied to the address given
4438 with the -f command-line option.
4440 42. The code for reading a message's header from the spool was incrementing
4441 $received_count, but never initializing it. This meant that the value was
4442 incorrect (doubled) while delivering a message in the same process in which
4443 it was received. In the most common configuration of Exim, this never
4444 happens - a fresh exec is done - but it can happen when
4445 deliver_drop_privilege is set.
4447 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4448 it now includes up to 150 characters of the unexpected data in the log
4451 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4452 and building data strings. The size of both of these buffers was 10 000
4453 bytes - far larger than anybody would *ever* want, thought I. Needless to
4454 say, somebody hit the limit. I have increased the maximum line length to
4455 20 000 and the maximum data length of concatenated lines to 100 000. I have
4456 also fixed two bugs, because there was no checking on these buffers. Tsk,
4457 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4458 buffer is too small.
4460 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4461 lsearch lookups. Now it does.
4463 46. When parsing a route_list item in a manualroute router, a fixed-length
4464 buffer was used for the list of hosts. I made this 1024 bytes long,
4465 thinking that nobody would ever have a list of hosts that long. Wrong.
4466 Somebody had a whole pile of complicated expansion conditions, and the
4467 string was silently truncated, leading to an expansion error. It turns out
4468 that it is easier to change to an unlimited length (owing to other changes
4469 that have happened since this code was originally written) than to build
4470 structure for giving a limitation error. The length of the item that
4471 expands into the list of hosts is now unlimited.
4473 47. The lsearch lookup could not handle data where the length of text line was
4474 more than 4095 characters. Such lines were truncated, leading to shortened
4475 data being returned. It should now handle lines of any length.
4477 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4478 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4481 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4482 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4483 no longer happens in comment lines. A list of replacements is now placed
4484 at the head of all of the source files, except those whose only change is
4485 to replace PERL_COMMAND in the very first #! line.
4487 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4488 messages on the queue, with a bottom-up merge sort, using code contributed
4489 by Michael Haardt. This should make operations like -bp somewhat faster on
4490 large queues. It won't affect queue runners, except when queue_run_in_order
4493 51. Installed eximstats 1.31 in the distribution.
4495 52. Added support for SRV lookups to the dnslookup router.
4497 53. If an ACL referred to $message_body or $message_body_end, the value was not
4498 reset for any messages that followed in the same SMTP session.
4500 54. The store-handling optimization for building very long strings was not
4501 differentiating between the different store pools. I don't think this
4502 actually made any difference in practice, but I've tidied it.
4504 55. While running the routers to verify a sender address, $sender_address
4505 was still set to the sender address. This is wrong, because when routing to
4506 send a bounce to the sender, it would be empty. Therefore, I have changed
4507 it so that, while verifying a sender address, $sender_address is set to <>.
4508 (There is no change to what happens when verifying a recipient address.)
4510 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4511 target A or AAAA records (if not already returned) without resetting the
4512 qualify_single or search_parents options of the DNS resolver. These are
4513 inappropriate in this case because the targets of MX and SRV records must
4514 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4515 target that, when qualified, matched something in the local domain. These
4516 two options are now turned off when doing these lookups.
4518 57. It seems that at least some releases of Reiserfs (which does not have the
4519 concept of a fixed number of inodes) returns zero and not -1 for the
4520 number of available inodes. This interacted badly with check_spool_inodes,
4521 which assumed that -1 was the "no such thing" setting. What I have done is
4522 to check that the total number of inodes is greater than zero before doing
4523 the test of how many are available.
4525 58. When a "warn" ACL statement has a log_message modifier, the message is
4526 remembered, and not repeated. This is to avoid a lot of repetition when a
4527 message has many recipients that cause the same warning to be written.
4528 Howewer, Exim was preserving the list of already written lines for an
4529 entire SMTP session, which doesn't seem right. The memory is now reset if a
4530 new message is started.
4532 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4533 debugging output unless log_rewrite was also set.
4535 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4536 of (int)(handle) when we know that handle contains (void *)(-1).
4538 61. The Exim daemon panic-logs an error return when it closes the incoming
4539 connection. However "connection reset by peer" seems to be common, and
4540 isn't really an error worthy of noting specially, so that particular error
4543 62. When Exim is trying to find all the local interfaces, it used to panic and
4544 die if the ioctl to get the interface flags failed. However, it seems that
4545 on at least one OS (Solaris 9) it is possible to have an interface that is
4546 included in the list of interfaces, but for which you get a failure error
4547 for this call. This happens when the interface is not "plumbed" into a
4548 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4549 failure of the "get flags" call assumes that the interface is down.
4551 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4552 makes life easier for people who are doing arithmetic on fields extracted
4553 from dates, where you often get leading zeros that should not be
4554 interpreted as octal.
4556 64. Added qualify_domain to the redirect router, to override the global
4559 65. If a pathologically long header line contained very many addresses (the
4560 report of this problem mentioned 10 000) and each of them was rewritten,
4561 Exim could use up a very large amount of memory. (It kept on making new
4562 copies of the header line as it rewrote, and never released the old ones.)
4563 At the expense of a bit more processing, the header rewriting function has
4564 been changed so that it no longer eats memory in this way.
4566 66. The generation of the Received: header has been moved from the time that a
4567 message starts to be received, to the time that it finishes. The timestamp
4568 in the Received: header should now be very close to that of the <= log
4569 line. There are two side-effects of this change:
4571 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4572 logged header lines no longer include the local Received: line, because
4573 it has not yet been created. The same applies to a copy of the message
4574 that is returned to a non-SMTP sender when a message is rejected.
4576 (b) When a filter file is tested using -bf, no additional Received: header
4577 is added to the test message. After some thought, I decided that this
4580 This change does not affect the value of $received_for. It is still set
4581 after address rewriting, but before local_scan() is called.
4583 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4585 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4586 gave an unhelpful panic error message, and a defer error. I have managed to
4587 change this behaviour so that it now rejects any supplied certificate,
4588 which seems right, as the list of acceptable certificates is empty.
4590 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4591 gave an unhelpful defer error. I have not managed to make this reject any
4592 supplied certificates, but the error message it gives is "no certificate
4593 supplied", which is not helpful.
4595 70. exigrep's output now also includes lines that are not associated with any
4596 message, but which match the given pattern. Implemented by a patch from
4597 Martin Sluka, which also tidied up the Perl a bit.
4599 71. Recipient callout verification, like sender verification, was using <> in
4600 the MAIL FROM command. This isn't really the right thing, since the actual
4601 sender may affect whether the remote host accepts the recipient or not. I
4602 have changed it to use the actual sender in the callout; this means that
4603 the cache record is now keyed on a recipient/sender pair, not just the
4604 recipient address. There doesn't seem to be a real danger of callout loops,
4605 since a callout by the remote host to check the sender would use <>.
4606 [SEE ABOVE: changed after hitting problems.]
4608 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4609 temporary errors. However, in the case of such a code being given after
4610 the end of a data transmission (i.e. after ".") Exim was failing to write
4611 a retry record for the message. (Yes, there was some broken host that was
4612 actually sending 8xx at this point.)
4614 73. An unknown lookup type in a host list could cause Exim to panic-die when
4615 the list was checked. (An example that provoked this was putting <; in the
4616 middle of a list instead of at the start.) If this happened during a DATA
4617 ACL check, a -D file could be left lying around. This kind of configuration
4618 error no longer causes Exim to die; instead it causes a defer errror. The
4619 incident is still logged to the main and panic logs.
4621 74. Buglet left over from Exim 3 conversion. The message "too many messages
4622 in one connection" was written to the rejectlog but not the mainlog, except
4623 when address rewriting (yes!) was being logged.
4625 75. Added write_rejectlog option.
4627 76. When a system filter was run not as root (that is, when system_filter_user
4628 was set), the values of the $n variables were not being returned to the
4629 main process; thus, they were not subsequently available in the $sn
4632 77. Added +return_path_on_delivery log selector.
4634 78. A connection timeout was being treated differently from recipients deferred
4635 when testing hosts_max_try with a message that was older than the host's
4636 retry timeout. (The host should not be counted, thus allowing all hosts to
4637 be tried at least once before bouncing.) This may have been the cause of an
4638 occasionally reported bug whereby a message would remain on the queue
4639 longer than the retry timeout, but would be bounced if a delivery was
4640 forced. I say "may" because I never totally pinned down the problem;
4641 setting up timeout/retry tests is difficult. See also the next item.
4643 79. The ultimate address timeout was not being applied to errors that involved
4644 a combination of host plus message (for example, a timeout on a MAIL
4645 command). When an address resolved to a number of possible hosts, and they
4646 were not all tried for each delivery (e.g. because of hosts_max_try), a
4647 message could remain on the queue longer than the retry timeout.
4649 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4652 81. Fixed an obscure SMTP outgoing bug which required at least the following
4653 conditions: (a) there was another message waiting for the same server;
4654 (b) the server returned 5xx to all RCPT commands in the first message so
4655 that the message was not completed; (c) the server dropped the connection
4656 or gave a negative response to the RSET that Exim sends to abort the
4657 transaction. The observed case was a dropped connection after DATA that had
4658 been sent in pipelining mode. That is, the server had advertised PIPELINING
4659 but was not implementing it correctly. The effect of the bug was incorrect
4660 behaviour, such as trying another host, and this could lead to a crash.
4666 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4667 and daemon.c were passed as pointers to ints; they should have been
4668 pointers to socklen_t variables (which are typically unsigned ints).
4670 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4673 3. Fixed a really odd bug that affected only the testing scheme; patching a
4674 certain fixed string in the binary changed the value of another string that
4675 happened to be identical to the end of the original first string.
4677 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4678 name", it returns that address as the IP address. On some operating
4679 systems (e.g. Solaris), it also passes back the IP address string as the
4680 "host name". However, on others (e.g. Linux), it passes back an empty
4681 string. Exim wasn't checking for this, and was changing the host name to an
4682 empty string, assuming it had been canonicized.
4684 5. Although rare, it is permitted to have more than one PTR record for a given
4685 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4686 all the names associated with an address, because they do in Solaris.
4687 However, it seems that they do not in Linux for data that comes from the
4688 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4689 I found this out when I moved to a new Linux workstation and tried to run
4690 the Exim test suite.
4692 To get round this problem I have changed the code so that it now does its
4693 own call to the DNS to look up PTR records when searching for a host name.
4694 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4695 addresses that are only in /etc/hosts are still found.
4697 This behaviour is, however, controlled by an option called host_lookup_
4698 order, which defaults to "bydns:byaddr". If people want to use the other
4699 order, or indeed, just use one or the other means of lookup, they can
4700 specify it in this variable.
4702 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4703 some operating systems, this comes back from gethostbyaddr() as an empty
4704 string, and this is what Exim used to test for. However, it seems that in
4705 other systems, "." is yielded. Exim now tests for this case too.
4707 7. The values of check_spool_space and check_log_space are now held internally
4708 as a number of kilobytes instead of an absolute number of bytes. If a
4709 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4710 kilobyte. This means that much larger values can be stored.
4712 8. Exim monitor: an attempt to get the action menu when not actually pointing
4713 at a message produces an empty menu entitled "No message selected". This
4714 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4715 no entries in it ("Shell widget menu has zero width and/or height"). So I
4716 have added a single, blank menu entry in this case.
4718 9. Added ${quote_local_part.
4720 10. MIME decoding is now applied to the contents of Subject: header lines when
4723 11. Now that a reference to $sender_host_address automatically causes a reverse
4724 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4725 host lookup before query-style lookups in lists that might use this
4726 variable. This has therefore been abolished, and the "net-" prefix is no
4727 longer necessary for query-style lookups.
4729 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4730 have been a typo for LFLAGS, so it has been changed.
4732 13. The install script calls Exim with "-C /dev/null" in order to find the
4733 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4734 to be output. Howeve, since Exim outputs its version number before the
4735 error, it didn't break the script. It just looked ugly. I fixed this by
4736 always allowing "-C /dev/null" if the caller is root.
4738 14. Ignore overlarge ACL variable number when reading spool file - insurance
4739 against a later release with more variables having written the file.
4741 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4742 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4743 this, as well as the form without the "IPv6" on the front.
4745 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4746 OS/Makefile-Darwin file.
4748 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4749 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4751 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4752 variable. However, some OS use ints for inodes. I've added cast to long int
4753 to get rid of the compiler warning.
4755 19. I had forgotten to lock out "/../" in configuration file names when
4756 ALT_CONFIG_PREFIX was set.
4758 20. Routers used for verification do not need to specify transports. However,
4759 if such a router generated a host list, and callout was configured, Exim
4760 crashed, because it could not find a port number from the (non-existent)
4761 transport. It now assumes port 25 in this circumstance.
4763 21. Added the -t option to exigrep.
4765 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4766 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4767 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4768 them is set, LOOKUP_LSEARCH is forced.
4770 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4772 24. Added sender and host information to the "rejected by local_scan()" log
4773 line; previously there was no indication of these.
4775 25. Added .include_if_exists.
4777 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4778 Linux. It turns out that not all file systems support this. Apparently some
4779 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4780 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4781 not supported on the file descriptor, is now ignored when Exim is trying to
4782 sync a directory. This applies only to Linux.
4784 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4786 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4787 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4790 29. Added support for maildirsize files from supplied patch (modified a bit).
4792 30. The use of :fail: followed by an empty string could lead Exim to respond to
4793 sender verification failures with (e.g.):
4795 550 Verification failed for <xxx>
4796 550 Sender verify failed
4798 where the first response line was missing the '-' that indicates it is not
4799 the final line of the response.
4801 31. The loop for finding the name of the user that called Exim had a hardwired
4802 limit of 10; it now uses the value of finduser_retries, which is used for
4803 all other user lookups.
4805 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4808 33. Exim was neglecting to zero errno before one call of strtol() when
4809 expanding a string and expecting an integer value. On some systems this
4810 resulted in spurious "integer overflow" errors. Also, it was casting the
4811 result into an int without checking.
4813 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4814 did not work. The code looks as if it has *never* worked, though it appears
4815 to have been documented since at least releast 1.62. I have made it work.
4817 35. The "timeout_DNS" error in retry rules, also documented since at least
4818 1.62, also never worked. As it isn't clear exactly what this means, and
4819 clearly it isn't a major issue, I have abolished the feature by treating it
4820 as "timeout", and writing a warning to the main and panic logs.
4822 36. The display of retry rules for -brt wasn't always showing the error code
4825 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4826 timeout_connect_A, timeout_connect_MX.
4828 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4829 to the empty sender.
4831 39. The daemon was not analysing the content of -oX till after it had closed
4832 stderr and disconnected from the controlling terminal. This meant that any
4833 syntax errors were only noted on the panic log, and the return code from
4834 the command was 0. By re-arranging the code a little, I've made the
4835 decoding happen first, so such errors now appear on stderr, and the return
4836 code is 1. However, the actual setting up of the sockets still happens in
4837 the disconnected process, so errors there are still only recorded on the
4840 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4841 connections (as happens on some IP stacks) was logged at start up time as
4842 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4843 it from "IPv6 and IPv4", which means that two separate sockets are being
4846 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4847 says whether AF_INET or AF_INET6 was passed as an argument.
4849 42. Exiwhat output was messed up when time zones were included in log
4852 43. Exiwhat now gives more information about the daemon's listening ports,
4853 and whether -tls-on-connect was used.
4855 44. The "port" option of the smtp transport is now expanded.
4857 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4858 silently ignored. Now an error message is written to the main and panic
4861 46. There's a new ACL modifier called "logwrite" which writes to a log file
4862 as soon as it is encountered.
4864 47. Added $local_user_uid and $local_user_gid at routing time.
4866 48. Exim crashed when trying to verify a sender address that was being
4869 49. Exim was recognizing only a space character after ".include". It now also
4870 recognizes a tab character.
4872 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4873 extracting the relevant information from the specification. The man page no
4874 longer contains scrambled data for the -d option, and I've added a section
4875 at the front about calling Exim under different names.
4877 51. Added "extra_headers" argument to the "mail" command in filter files.
4879 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4882 53. Installed eximstats 1.29.
4884 54. Added transport_filter_timeout as a generic transport option.
4886 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4887 Cc: header lines. This was required by RFC 822, but it not required by RFC
4890 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4891 incoming messages that did not have them. Now it does so only if the
4892 message originates locally, that is, if there is no associated remote host
4893 address. When Resent- header lines are present, this applies to the Resent-
4894 lines rather than the non-Resent- lines.
4896 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4897 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4899 58. Messages for configuration errors now include the name of the main
4900 configuration file - useful now that there may be more than one file in a
4901 list (.included file names were always shown).
4903 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4904 for those rare installations that do not start the daemon as root or run it
4905 setuid root. I've cut out the call to initgroups() if the daemon is not
4908 60. The Exim user and group can now be bound into the binary as text strings
4909 that are looked up at the start of Exim's processing.
4911 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4913 62. Added $mailstore_basename variable.
4915 63. Installed patch to sieve.c from Michael Haardt.
4917 64. When Exim failed to open the panic log after failing to open the main log,
4918 the original message it was trying to log was written to stderr and debug
4919 output, but if they were not available (the usual case in production), it
4920 was lost. Now it is written to syslog before the two lines that record the
4921 failures to open the logs.
4923 65. Users' Exim filters run in subprocesses under the user's uid. It is
4924 possible for a "deliver" command or an alias in a "personal" command to
4925 provoke an address rewrite. If logging of address rewriting is configured,
4926 this fails because the process is not running as root or exim. There may be
4927 a better way of dealing with this, but for the moment (because 4.30 needs
4928 to be released), I have disabled address rewrite logging when running a
4929 filter in a non-root, non-exim process.
4935 1. The buildconfig auxiliary program wasn't quoting the value set for
4936 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4937 not defined. This bug was masked in 4.22 by the effect that was fixed in
4940 2. Some messages that were rejected after a message id was allocated were
4941 shown as "incomplete" by exigrep. It no longer does this for messages that
4942 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4944 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4945 have allow_domain_literals set, the ID did not get logged in the <= line.
4946 Domain literals are now always recognized in Message-ID: header lines.
4948 4. The first argument for a ${extract expansion item is the key name or field
4949 number. Leading and trailing spaces in this item were not being ignored,
4950 causing some misleading effects.
4952 5. When deliver_drop_privilege was set, single queue runner processes started
4953 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4954 same command in the process it spins off) were not dropping privilege.
4956 6. When the daemon running as "exim" started a queue runner, it always
4957 re-executed Exim in the spun-off process. This is a waste of effort when
4958 deliver_drop_privilege is set. The new process now just calls the
4959 queue-runner function directly.
4965 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4968 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4969 ignored. Though the use of -oP was forcing the writing of a pid file, it
4970 was always written to the default place.
4972 3. If the message "no IP address found for host xxxx" is generated during
4973 incoming verification, it is now followed by identification of the incoming
4974 connection (so you can more easily find what provoked it).
4976 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4978 5. Added some features to "harden" Exim a bit more against certain attacks:
4980 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4981 be put in Local/Makefile. This is like the never_users runtime option,
4982 but it cannot be overridden. The default setting is "root".
4984 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4985 prefix string with which any file named in a -C command line option
4988 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4989 is retained for -C and -D only if the caller of Exim is root. Without
4990 it, the exim user may also use -C and -D and retain privilege.
4992 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4993 command line option is disabled.
4995 6. Macro names set by the -D option must start with an upper case letter, just
4996 like macro names defined in the configuration file.
4998 7. Added "dereference=" facility to LDAP.
5000 8. Two instances of the typo "uknown" in the source files are fixed.
5002 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5003 the Configure-Makefile script screwed up while processing it.
5005 10. Incorporated PCRE 4.4.
5007 11. The SMTP synchronization check was not operating right at the start of an
5008 SMTP session. For example, it could not catch a HELO sent before the client
5009 waited for the greeting. There is now a check for outstanding input at the
5010 point when the greeting is written. Because of the duplex, asynchronous
5011 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5012 way, but not yet received, when the check is performed.
5014 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5015 on TCP/IP sockets, because this apparently causes some broken clients to
5018 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5019 unchanged) from the Cygwin maintainer.
5021 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5022 was supported instead of testing for mbx. Effectively a typo.
5024 15. The spa authenticator server code was not checking that the input it
5025 received was valid base64.
5027 16. The debug output line for the "set" modifier in ACLs was not showing the
5028 name of the variable that was being set.
5030 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5032 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5033 lookup on its own. Something else was needed to trigger the lookup. For
5034 example, a match in host_lookup or the need for a host name in a host list.
5035 Now, if $sender_host_name is referenced and the host name has not yet been
5036 looked up, a lookup is performed. If the lookup fails, the variable remains
5037 empty, and $host_lookup_failed is set to "1".
5039 19. Added "eqi" as a case-independent comparison operator.
5041 20. The saslauthd authentication condition could segfault if neither service
5042 nor realm was specified.
5044 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5045 error message that was logged was misleading, and incoming SMTP
5046 connections were dropped. The message is now more accurate, and temporary
5047 errors are given to SMTP connections.
5049 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5050 (except RSET) with a 421 error, until QUIT is received. However, it was
5051 failing to send a response to QUIT.
5053 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5054 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5055 ACL is now run after the helo_try_verify_hosts code.
5057 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5058 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5059 case-independent, so other case variants are also recognized). Apparently
5060 some systems use these upper case variants.
5062 25. If more than two messages were waiting for the same host, and a transport
5063 filter was specified for the transport, Exim sent two messages over the
5064 same TCP/IP connection, and then failed with "socket operation on non-
5065 socket" when it tried to send the third.
5067 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5069 27. The extern definition of crypt16() in expand.c was not being excluded when
5070 the OS had its own crypt16() function.
5072 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5073 as a preferred synonym for return_size_limit, both as an option and as an
5076 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5078 30. Changed the default configuration ACL to relax the local part checking rule
5079 for addresses that are not in any local domains. For these addresses,
5080 slashes and pipe symbols are allowed within local parts, but the sequence
5081 /../ is explicitly forbidden.
5083 31. SPA server authentication was not clearing the challenge buffer before
5086 32. log_message in a "warn" ACL statement was writing to the reject log as
5087 well as to the main log, which contradicts the documentation and doesn't
5088 seem right (because no rejection is happening). So I have stopped it.
5090 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5091 However, I am unable to do any testing of this.
5093 34. Fixed an infelicity in the appendfile transport. When checking directories
5094 for a mailbox, to see if any needed to be created, it was accidentally
5095 using path names with one or more superfluous leading slashes; tracing
5096 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5098 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5099 contents are added to the log line that is written for every discarded
5100 recipient. (Previously a log_message setting was ignored.)
5102 36. The ${quote: operator now quotes the string if it is empty.
5104 37. The install script runs exim in order to find its version number. If for
5105 some reason other than non-existence or emptiness, which it checks, it
5106 could not run './exim', it was installing it with an empty version number,
5107 i.e. as "exim-". This error state is now caught, and the installation is
5110 38. An argument was missing from the function that creates an error message
5111 when Exim fails to connect to the socket for saslauthd authentication.
5112 This could cause Exim to crash, or give a corrupted message.
5114 39. Added isip, isip4, and isip6 to ${if conditions.
5116 40. The ACL variables $acl_xx are now saved with the message, and can be
5117 accessed later in routers, transports, and filters.
5119 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5120 strings in the file are not string-expanded.
5122 42. If a MAIL command specified a SIZE value that was too large to fit into an
5123 int variable, the check against message_size_limit failed. Such values are
5124 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5125 day this will have to be increased, but I don't think I want to be around
5126 when emails are that large.
5133 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5134 iconv() is not standard in FreeBSD.
5136 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5137 IPv6 enabled. The observed symptom was a segmentation fault on return from
5138 the function os_common_find_running_interfaces() in src/os.c.
5140 3. In the check_special_case() function in daemon.c I had used "errno" as an
5141 argument name, which causes warnings on some systems. This was basically a
5142 typo, since it was named "eno" in the comments!
5144 4. The code that waits for the clock to tick (at a resolution of some fraction
5145 of a second) so as to ensure message-id uniqueness was always waiting for
5146 at least one whole tick, when it could have waited for less. [This is
5147 almost certainly not relevant at current processor speeds, where it is
5148 unlikely to ever wait at all. But we try to future-proof.]
5150 5. The function that sleeps for a time interval that includes fractions of a
5151 second contained a race. It did not block SIGALRM between setting the
5152 timer, and suspending (a couple of lines later). If the interval was short
5153 and the sigsuspend() was delayed until after it had expired, the suspension
5154 never ended. On busy systems this could lead to processes getting stuck for
5157 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5158 process, before it forks any delivery processes. The open lookup caching
5159 mechanism meant that the open file or database connection was passed into
5160 the delivery process. The problem was that delivery processes always tidy
5161 up cached lookup data. This could cause a problem for the next delivery
5162 process started by the queue runner, because the external queue runner
5163 process does not know about the closure. So the next delivery process
5164 still has data in the lookup cache. In the case of a file lookup, there was
5165 no problem because closing a file descriptor in a subprocess doesn't affect
5166 the parent. However, if the lookup was caching a connection to a database,
5167 the connection was closed, and the second delivery process was likely to
5168 see errors such as "PGSQL: query failed: server closed the connection
5169 unexpectedly". The problem has been fixed by closing all cached lookups
5170 in a queue runner before running a delivery process.
5172 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5173 seem to have the "const" qualifier which it has on other OS. I've
5176 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5177 *of the same type* (client or server) with the same public name that an
5178 error should be diagnosed.
5180 9. When Exim looked up a host name for an IP address, but failed to find the
5181 original IP address when looking up the host name (a safety check), it
5182 output the message "<ip address> does not match any IP for NULL", which was
5183 confusing, to say the least. The bug was that the host name should have
5184 appeared instead of "NULL".
5186 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5187 user that is built into the binary, and the -C or -D options is used, root
5188 privilege is dropped before the configuration file is read. In addition,
5189 logging is switched to stderr instead of the normal log files. If the
5190 configuration then re-defines the Exim user, the unprivileged environment
5191 is probably not what is expected, so Exim logs a panic warning message (but
5194 However, if deliver_drop_privilege is set, the unprivileged state may well
5195 be exactly what is intended, so the warning has been cut out in that case,
5196 and Exim is allowed to try to write to its normal log files.
5202 1. smtp_return_error_details was not giving details for temporary sender
5203 or receiver verification errors.
5205 2. Diagnose a configuration error if two authenticators have the same public
5208 3. Exim used not to create the message log file for a message until the first
5209 delivery attempt. This could be confusing when incoming messages were held
5210 for policy or load reasons. The message log file is now created at the time
5211 the message is received, and an initial "Received" line is written to it.
5213 4. The automatically generated man page for command line options had a minor
5214 bug that caused no ill effects; however, a more serious problem was that
5215 the procedure for building the man page automatically didn't always
5216 operate. Consequently, release 4.20 contains an out-of-date version. This
5217 shouldn't happen again.
5219 5. When building Exim with embedded Perl support, the script that builds the
5220 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5221 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5223 6. The freeze_tell option was not being used for messages that were frozen on
5224 arrival, either by an ACL or by local_scan().
5226 7. Added the smtp_incomplete_transaction log selector.
5228 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5229 was accepting AUTH without a new EHLO.
5231 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5232 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5235 10. Logging of TCP/IP connections (when configured) now happens in the main
5236 daemon process instead of the child process, so that the TCP/IP connection
5237 count is more accurate (but it can never be perfect).
5239 11. The use of "drop" in a nested ACL was not being handled correctly in the
5240 outer ACL. Now, if condition failure induced by the nested "drop" causes
5241 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5242 or "require"), the connection is dropped.
5244 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5245 that yield "discard" can now be used with an "accept" or a "discard" verb,
5246 but an error is generated for any others (because I can't see a useful way
5247 to define what should happen).
5249 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5250 are now processed in the same way as lines in the Exim configuration file.
5251 In particular, continuation lines are supported.
5253 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5255 15. Added -ti meaning -t -i.
5257 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5258 domains, and warn by logging if others are found.
5260 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5261 returned from GIFCONF when Exim is trying to find the list of interfaces on
5262 a host. The code in os.c has been modified to copy each ifreq to an aligned
5263 structure in all cases.
5265 Also, in some cases, the returned ifreq's were being copied to a 'struct
5266 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5267 means the last couple of bytes of an IPv6 address could be chopped if the
5268 ifreq contained only a normal sockaddr (14 bytes storage).
5270 18. Named domain lists were not supported in the hosts_treat_as_local option.
5271 An entry such as +xxxx was not recognized, and was treated as a literal
5274 19. Ensure that header lines added by a DATA ACL are included in the reject log
5275 if the ACL subsequently rejects the message.
5277 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5278 MD5 (which is deprecated).
5280 21. When testing a filter file using -bf, Exim was writing a message when it
5281 took the sender from a "From " line in the message, but it was not doing so
5282 when it took $return_path from a Return-Path: header line. It now does.
5284 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5285 with a valid header line field name (a series of printing characters
5286 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5288 23. Changed "disc" in the source to "disk" to conform to the documentation and
5289 the book and for uniformity.
5291 24. Ignore Sendmail's -Ooption=value command line item.
5293 25. When execve() failed while trying to run a command in a pipe transport,
5294 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5295 could be confused with a return value of 69 from the command itself. This
5296 has been changed to 127, the value the shell returns if it is asked to run
5297 a non-existent command. The wording for the related log line suggests a
5298 non-existent command as the problem.
5300 26. If received_header_text expands to an empty string, do not add a Received:
5301 header line to the message. (Well, it adds a token one on the spool, but
5302 marks it "old" so that it doesn't get used or transmitted.)
5304 27. Installed eximstats 1.28 (addition of -nt option).
5306 28. There was no check for failure on the call to getsockname() in the daemon
5307 code. This can fail if there is a shortage of resources on the system, with
5308 ENOMEM, for example. A temporary error is now given on failure.
5310 29. Contrary to the C standard, it seems that in some environments, the
5311 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5312 program. Exim now does this explicitly; it affects the formatting of
5313 timestamps using strftime().
5315 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5316 complaints. I've now initialized all the variables, to avoid this.
5318 32. Header lines added by a system filter were not being "seen" during
5319 transport-time rewrites.
5321 33. The info_callback() function passed to OpenSSL is set up with type void
5322 (*)(SSL *, int, int), as described somewhere. However, when calling the
5323 function (actually a macro) that sets it up, the type void(*)() is
5324 expected. I've put in a cast to prevent warnings from picky compilers.
5326 34. If a DNS black list lookup found a CNAME record, but there were no A
5327 records associated with the domain it pointed at, Exim crashed.
5329 35. If a DNS black list lookup returned more than one A record, Exim ignored
5330 all but the first. It now scans all returned addresses if a particular IP
5331 value is being sought. In this situation, the contents of the
5332 $dnslist_value variable are a list of all the addresses, separated by a
5335 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5336 to accept a host name and all its aliases if the forward lookup for any of
5337 them yielded the IP address of the incoming connection. Now it accepts only
5338 those names whose forward lookup yields the correct IP address. Any other
5339 names are discarded. This closes a loophole whereby a rogue DNS
5340 administrator could create reverse DNS records to break through a
5341 wildcarded host restriction in an ACL.
5343 37. If a user filter or a system filter that ran in a subprocess used any of
5344 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5345 the wrong values were passed to the pipe command ($thisaddress had the
5346 value of $0, $0 had the value of $1, etc). This bug was introduced by
5347 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5349 38. Improved the line breaking for long SMTP error messages from ACLs.
5350 Previously, if there was no break point between 40 and 75 characters, Exim
5351 left the rest of the message alone. Two changes have been made: (a) I've
5352 reduced the minimum length to 35 characters; (b) if it can't find a break
5353 point between 35 and 75 characters, it looks ahead and uses the first one
5354 that it finds. This may give the occasional overlong line, but at least the
5355 remaining text gets split now.
5357 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5358 file descriptors might be low, and that setting 1000 would always raise it.
5359 It turns out that in some environments, the limit is already over 1000 and
5360 that lowering it causes trouble. So now Exim takes care not to decrease it.
5362 40. When delivering a message, the value of $return_path is set to $sender_
5363 address at the start of routing (routers may change the value). By an
5364 oversight, this default was not being set up when an address was tested by
5365 -bt or -bv, which affected the outcome if any router or filter referred to
5368 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5369 log only when "message" or "log_message" are set. However, if one of the
5370 conditions was an address verification, or a call to a nested ACL, the
5371 messages generated by the underlying test were being passed through. This
5372 no longer happens. The underlying message is available in $acl_verify_
5373 message for both "message" and "log_message" expansions, so it can be
5374 passed through if needed.
5376 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5377 new expansion $bh_ to give the encoded byte string without charset
5378 translation. Translation happens only if iconv() is available; HAVE_ICONV
5379 indicates this at build time. HEADERS_CHARSET gives the charset to
5380 translate to; headers_charset can change it in the configuration, and
5381 "headers charset" can change it in an individual filter file.
5383 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5384 that creates RFC 2047 encoded "words" labels them as that charset instead
5385 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5386 expansion operator; (ii) when Exim creates a From: line for a local
5387 message; (iii) when a header line is rewritten to include a "phrase" part.
5389 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5390 buggy, causing it to skip the first lines of messages whose message ID
5391 ended in 'D'. This would not have bitten before Exim release 4.14, because
5392 message IDs were unlikely to end in 'D' before then. The effect was to have
5393 incorrect size information for certain domains.
5395 45. #include "config.h" was missing at the start of the crypt16.c module. This
5396 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5399 46. If there was a timeout during a "random" callout check, Exim treated it as
5400 a failure of the random address, and carried on sending RSET and the real
5401 address. If the delay was just some slowness somewhere, the response to the
5402 original RCPT would be taken as a response to RSET and so on, causing
5403 mayhem of various kinds.
5405 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5406 when I implemented it. It didn't allow for the fact that some option values
5407 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5408 the right test for positive values.
5410 48. Domain names in DNS records are case-independent. Exim always looks them up
5411 in lower case. Some resolvers return domain names in exactly the case they
5412 appear in the zone file, that is, they may contain uppercase letters. Not
5413 all resolvers do this - some return always lower case. Exim was treating a
5414 change of case by a resolver as a change of domain, similar to a widening
5415 of a domain abbreviation. This triggered its re-routing code and so it was
5416 trying to route what was effectively the same domain again. This normally
5417 caused routing to fail (because the router wouldn't handle the domain
5418 twice). Now Exim checks for this case specially, and just changes the
5419 casing of the domain that it ultimately uses when it transmits the message
5422 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5425 50. If a filter generated a file delivery with a non-absolute name (possible if
5426 no home directory exists for the router), the forbid_file option was not
5429 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5430 the existing equality matching.
5432 52. Exim was using ints instead of ino_t variables in some places where it was
5433 dealing with inode numbers.
5435 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5436 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5437 called TMPDIR, and if it finds it is different, it changes its value.
5439 54. The smtp_printf() function is now made available to local_scan() so
5440 additional output lines can be written before returning. There is also an
5441 smtp_fflush() function to enable the detection of a dropped connection.
5442 The variables smtp_input and smtp_batched_input are exported to
5445 55. Changed the default runtime configuration: the message "Unknown user"
5446 has been removed from the ACL, and instead placed on the localuser router,
5447 using the cannot_route_message feature. This means that any verification
5448 failures that generate their own messages won't get overridden. Similarly,
5449 the "Unrouteable address" message that was in the ACL for unverifiable
5450 relay addresses has also been removed.
5452 56. Added hosts_avoid_esmtp to the smtp transport.
5454 57. The exicyclog script was not checking for the esoteric option
5455 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5456 will work only if exicyclog is run under the appropriate euid.
5458 58. Following a discussion on the list, the rules by which Exim recognises line
5459 endings on incoming messages have been changed. The -dropcr and drop_cr
5460 options are now no-ops, retained only for backwards compatibility. The
5461 following line terminators are recognized: LF CRLF CR. However, special
5462 processing applies to CR:
5464 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5465 nor a local message in the state where . is a terminator.
5467 (ii) If a bare CR is encountered in a header line, an extra space is added
5468 after the line terminator so as not to end the header. The reasoning
5469 behind this is that bare CRs in header lines are most likely either
5470 to be mistakes, or people trying to play silly games.
5472 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5473 was being incorrectly given as 18 bytes larger than it should have been.
5474 This is a VOB (very old bug).
5476 60. This may never have affected anything current, but just in case it has:
5477 When the local host is found other than at the start of a list of hosts,
5478 the local host, those with the same MX, and any that follow, are discarded.
5479 When the list in question was part of a longer list of hosts, the following
5480 hosts (not currently being processed) were also being discarded. This no
5481 longer happens. I'm not sure if this situation could ever has previously
5484 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5487 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5488 header. This is something that is recommended in a new Internet Draft, and
5489 is something that is documented as being done by Sendmail. There are two
5490 possible values. For messages generated by the autoreply transport, Exim
5493 Auto-Submitted: auto-replied
5495 whereas for all other generated messages (e.g. bounces) it adds
5497 Auto-Submitted: auto-generated
5499 63. The "personal" condition in filters now includes a test for the
5500 Auto-Submitted: header. If it contains the string "auto-" the message it
5501 not considered personal.
5503 64. Added rcpt_include_affixes as a generic transport option.
5505 65. Added queue_only_override (default true).
5507 66. Added the syslog_duplication option.
5509 67. If what should have been the first header line of a message consisted of
5510 a space followed by a colon, Exim was mis-interpreting it as a header line.
5511 It isn't of course - it is syntactically invalid and should therefore be
5512 treated as the start of the message body. The misbehaviour could have
5513 caused a number of strange effects, including loss of data in subsequent
5514 header lines, and spool format errors.
5516 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5517 client host had authenticated. This control can now be exercised by an ACL
5518 for more flexibility.
5520 69. By default, callouts do not happen when testing with -bh. There is now a
5521 variant, -bhc, which does actually run the callout code, including
5522 consulting and updating the callout cache.
5524 70. Added support for saslauthd authentication, courtesy of Alexander
5527 71. If statvfs() failed on the spool or log directories while checking their
5528 size for availability, Exim confusingly gave the error "space shortage".
5529 Furthermore, in debugging mode it crashed with a floating point exception.
5530 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5531 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5532 serious problem, Exim now writes to the main and panic logs when this
5533 happens, with details of the failure. It then refuses to accept the
5534 incoming message, giving the message "spool directory problem" or "log
5535 directory problem" with a 421 code for SMTP messages.
5537 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5538 0, 1, and 2 exist, because some OS complain for execs without them (see
5539 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5540 descriptors. However, the code omitted to check that the open succeeded,
5541 causing mysterious errors if for some reason the permissions on /dev/null
5542 got screwed. Now Exim writes a message to the main and panic logs, and
5543 bombs out if it can't open /dev/null.
5545 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5546 interact so that it is all more flexible. It is supposed to remain
5547 backwards compatible. Also added extra_local_interfaces.
5549 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5550 to bomb out with an assertion failure - to the client this appears as a
5551 connection drop. This problem occurs in the part of the code that was taken
5552 from the Samba project. Fortunately, the assertion is in a very simple
5553 function, so I have fixed this by reproducing the function inline in the
5554 one place where it is called, and arranging for authentication to fail
5555 instead of killing the process with assert().
5557 75. The SPA client code was not working when the server requested OEM rather
5558 than Unicode encoding.
5560 76. Added code to make require_files with a specific uid setting more usable in
5561 the case where statting the file as root fails - usually a non-root-mounted
5562 NFS file system. When this happens and the failure is EACCES, Exim now
5563 forks a subprocess and does the per-uid checking as the relevant uid.
5565 77. Added process_log_path.
5567 78. If log_file_path was not explicitly set, a setting of check_log_space or
5568 check_log_inodes was ignored.
5570 79. If a space check for the spool or log partitions fails, the incident is now
5571 logged. Of course, in the latter case the data may get lost...
5573 80. Added the %p formatting code to string_format() so that it can be used to
5574 print addresses in debug_print(). Adjusted all the address printing in the
5575 debugging in store.c to use %p rather than %d.
5577 81. There was a concern that a line of code in smtp_in.c could overflow a
5578 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5579 initially expressed, the concern was not well-founded, because trailing
5580 spaces are removed early. However, if the trailing spaces were followed by
5581 a NULL, they did not get removed, so the overflow was possible. Two fixes
5584 (a) I re-wrote the offending code in a cleaner fashion.
5585 (b) If an incoming SMTP command contains a NULL character, it is rejected
5588 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5589 runs initgroups(), so that if the Exim user is in any additional groups,
5590 they will be used during message reception.
5596 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob