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 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
70 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
72 Report and patch from Dmitry Banschikov.
74 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
75 Removed SSL_clear() after SSL_new() which led to protocol negotiation
76 failures. We appear to now support TLS1.1+ with Exim.
78 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
79 lets Exim select keys and certificates based upon TLS SNI from client.
80 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
81 before an outbound SMTP session. New log_selector, +tls_sni.
83 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
84 NULL dereference. Report and patch from Alun Jones.
86 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
87 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
88 Not seeing resolver debug output on NetBSD, but suspect this is a
89 resolver implementation change.
91 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
92 Left warnings. Added "eximon gdb" invocation mode.
94 PP/21 Defaulting "accept_8bitmime" to true, not false.
96 PP/22 Added -bw for inetd wait mode support.
98 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
99 locate the relevant includes and libraries. Made this the default.
101 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
102 Bugzilla 1246, report and most of solution from Tomasz Kusy.
104 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
105 This may cause build issues on older platforms.
107 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
108 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
109 gnutls_require_mac & gnutls_require_protocols (no longer supported).
110 Added SNI support via GnuTLS too.
111 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
113 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
115 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
116 Applied second patch from Janne, implementing suggestion to default
117 multiple-strings-in-record handling to match SPF spec.
119 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
121 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
122 read-only, out of scope).
123 Patch from Wolfgang Breyha, report from Stuart Northfield.
125 PP/29 Fix three issues highlighted by clang analyser static analysis.
126 Only crash-plausible issue would require the Cambridge-specific
127 iplookup router and a misconfiguration.
128 Report from Marcin Mirosław.
130 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
132 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
133 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
134 As part of this, removing so much warning spew let me fix some minor
135 real issues in debug logging.
137 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
138 assignment on my part. Fixed.
140 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
141 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
142 Janne Snabb (who went above and beyond: thank you).
144 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
145 string otherwise requires a connection and a bunch more work and it's
146 relatively easy to get wrong.
152 PP/01 Solaris build fix for Oracle's LDAP libraries.
153 Bugzilla 1109, patch from Stephen Usher.
155 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
157 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
160 TF/02 Fix a couple more cases where we did not log the error message
161 when unlink() failed. See also change 4.74-TF/03.
163 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
164 lock up or crash if it happened to be inside a call to libc when it
165 got a SIGUSR1 from exiwhat.
167 The SIGUSR1 handler appends the current process status to the process
168 log which is later printed by exiwhat. It used to use the general
169 purpose logging code to do this, but several functions it calls are
170 not safe for signals.
172 The new output code in the SIGUSR1 handler is specific to the process
173 log, and simple enough that it's easy to inspect for signal safety.
174 Removing some special cases also simplifies the general logging code.
175 Removing the spurious timestamps from the process log simplifies
178 TF/04 Improved ratelimit ACL condition.
180 The /noupdate option has been deprecated in favour of /readonly which
181 has clearer semantics. The /leaky, /strict, and /readonly update modes
182 are mutually exclusive. The update mode is no longer included in the
183 database key; it just determines when the database is updated. (This
184 means that when you upgrde Exim will forget old rate measurements.)
186 Exim now checks that the per_* options are used with an update mode that
187 makes sense for the current ACL. For example, when Exim is processing a
188 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
189 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
190 must specify per_mail/readonly. If you omit the update mode it defaults to
191 /leaky where that makes sense (as before) or /readonly where required.
193 The /noupdate option is now undocumented but still supported for
194 backwards compatibility. It is equivalent to /readonly except that in
195 ACLs where /readonly is required you may specify /leaky/noupdate or
196 /strict/noupdate which are treated the same as /readonly.
198 A useful new feature is the /count= option. This is a generalization
199 of the per_byte option, so that you can measure the throughput of other
200 aggregate values. For example, the per_byte option is now equivalent
201 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
203 The per_rcpt option has been generalized using the /count= mechanism
204 (though it's more complicated than the per_byte equivalence). When it is
205 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
206 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
207 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
208 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
209 Note that using per_rcpt with a non-readonly update mode in more than
210 one ACL will cause the recipients to be double-counted. (The per_mail
211 and per_byte options don't have this problem.)
213 The handling of very low rates has changed slightly. If the computed rate
214 is less than the event's count (usually one) then this event is the first
215 after a long gap. In this case the rate is set to the same as this event's
216 count, so that the first message of a spam run is counted properly.
218 The major new feature is a mechanism for counting the rate of unique
219 events. The new per_addr option counts the number of different
220 recipients that someone has sent messages to in the last time period. It
221 behaves like per_rcpt if all the recipient addresses are different, but
222 duplicate recipient addresses do not increase the measured rate. Like
223 the /count= option this is a general mechanism, so the per_addr option
224 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
225 example, measure the rate that a client uses different sender addresses
226 with the options per_mail/unique=$sender_address. There are further
227 details in the main documentation.
229 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
231 TF/06 Removed a few PCRE remnants.
233 TF/07 Automatically extract Exim's version number from tags in the git
234 repository when doing development or release builds.
236 PP/02 Raise smtp_cmd_buffer_size to 16kB.
237 Bugzilla 879. Patch from Paul Fisher.
239 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
240 Heavily based on revision 40f9a89a from Simon Arlott's tree.
243 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
245 PP/05 Variable $av_failed, true if the AV scanner deferred.
246 Bugzilla 1078. Patch from John Horne.
248 PP/06 Stop make process more reliably on build failure.
249 Bugzilla 1087. Patch from Heiko Schlittermann.
251 PP/07 Make maildir_use_size_file an _expandable_ boolean.
252 Bugzilla 1089. Patch from Heiko Schlittermann.
254 PP/08 Handle ${run} returning more data than OS pipe buffer size.
255 Bugzilla 1131. Patch from Holger Weiß.
257 PP/09 Handle IPv6 addresses with SPF.
258 Bugzilla 860. Patch from Wolfgang Breyha.
260 PP/10 GnuTLS: support TLS 1.2 & 1.1.
262 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
265 PP/11 match_* no longer expand right-hand-side by default.
266 New compile-time build option, EXPAND_LISTMATCH_RHS.
267 New expansion conditions, "inlist", "inlisti".
269 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
271 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
273 PP/14 fix log_write() format string regression from TF/03.
274 Bugzilla 1152. Patch from Dmitry Isaikin.
280 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
282 PP/02 Harmonised TLS library version reporting; only show if debugging.
283 Layout now matches that introduced for other libraries in 4.74 PP/03.
285 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
287 PP/04 New "dns_use_edns0" global option.
289 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
292 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
293 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
295 TK/01 Updated PolarSSL code to 0.14.2.
296 Bugzilla 1097. Patch from Andreas Metzler.
298 PP/07 Catch divide-by-zero in ${eval:...}.
301 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
304 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
305 format-string attack -- SECURITY: remote arbitrary code execution.
307 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
308 time unintentionally subject to list matching rules, letting the header
309 cause arbitrary Exim lookups (of items which can occur in lists, *not*
310 arbitrary string expansion). This allowed for information disclosure.
312 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
313 INT_MIN/-1 -- value coerced to INT_MAX.
319 NM/01 Workround for PCRE version dependency in version reporting
322 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
323 This fixes portability to compilers other than gcc, notably
324 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
326 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
327 makefiles for portability to HP-UX and POSIX correctness.
329 PP/01 Permit LOOKUP_foo enabling on the make command-line.
330 Also via indented variable definition in the Makefile.
331 (Debugging by Oliver Heesakkers).
333 PP/02 Restore caching of spamd results with expanded spamd_address.
334 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
336 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
337 Improves build reliability. Fix from: Frank Elsner
339 NM/02 Fix wide character breakage in the rfc2047 coding
340 Fixes bug 1064. Patch from Andrey N. Oktyabrski
342 NM/03 Allow underscore in dnslist lookups
343 Fixes bug 1026. Patch from Graeme Fowler
345 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
346 Code patches from Adam Ciarcinski of NetBSD.
348 NM/04 Fixed exiqgrep to cope with mailq missing size issue
351 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
352 is logged, to avoid truncation. Patch from John Horne.
354 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
355 Patch from Jakob Hirsch.
357 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
358 SQL string expansion failure details.
359 Patch from Andrey Oktyabrski.
361 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
362 Patch from Simon Arlott.
364 PP/09 New lookups functionality failed to compile on old gcc which rejects
365 extern declarations in function scope.
366 Patch from Oliver Fleischmann
368 PP/10 Use sig_atomic_t for flags set from signal handlers.
369 Check getgroups() return and improve debugging.
370 Fixed developed for diagnosis in bug 927 (which turned out to be
373 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
374 Patch from Mark Zealey.
376 PP/12 Bugzilla 1056: Improved spamd server selection.
377 Patch from Mark Zealey.
379 PP/13 Bugzilla 1086: Deal with maildir quota file races.
380 Based on patch from Heiko Schlittermann.
382 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
383 Patch from Uwe Doering, sign-off by Michael Haardt.
385 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
386 variable declaration deep within a block. Bug and patch from
389 PP/15 lookups-Makefile IRIX compatibilty coercion.
391 PP/16 Make DISABLE_DKIM build knob functional.
393 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
394 Patch by Simon Arlott
396 TF/03 Fix valgrind.h portability to C89 compilers that do not support
397 variable argument macros. Our copy now differs from upstream.
403 TF/01 Failure to get a lock on a hints database can have serious
404 consequences so log it to the panic log.
406 TF/02 Log LMTP confirmation messages in the same way as SMTP,
407 controlled using the smtp_confirmation log selector.
409 TF/03 Include the error message when we fail to unlink a spool file.
411 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
412 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
413 for maintaining out-of-tree patches for some time.
415 PP/01 Bugzilla 139: Documentation and portability issues.
416 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
417 Handle per-OS dynamic-module compilation flags.
419 PP/02 Let /dev/null have normal permissions.
420 The 4.73 fixes were a little too stringent and complained about the
421 permissions on /dev/null. Exempt it from some checks.
422 Reported by Andreas M. Kirchwitz.
424 PP/03 Report version information for many libraries, including
425 Exim version information for dynamically loaded libraries. Created
426 version.h, now support a version extension string for distributors
427 who patch heavily. Dynamic module ABI change.
429 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
430 privilege escalation vulnerability whereby the Exim run-time user
431 can cause root to append content of the attacker's choosing to
434 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
437 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
438 If dropping privileges for untrusted macros, we disabled normal logging
439 on the basis that it would fail; for the Exim run-time user, this is not
440 the case, and it resulted in successful deliveries going unlogged.
441 Fixed. Reported by Andreas Metzler.
447 PP/01 Date: & Message-Id: revert to normally being appended to a message,
448 only prepend for the Resent-* case. Fixes regression introduced in
449 Exim 4.70 by NM/22 for Bugzilla 607.
451 PP/02 Include check_rfc2047_length in configure.default because we're seeing
452 increasing numbers of administrators be bitten by this.
454 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
456 PP/03 Bugzilla 994: added openssl_options main configuration option.
458 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
460 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
462 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
464 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
465 without a peer certificate, leading to a segfault because of an
466 assumption that peers always have certificates. Be a little more
467 paranoid. Problem reported by Martin Tscholak.
469 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
470 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
471 NB: ClamAV planning to remove STREAM in "middle of 2010".
472 CL also introduces -bmalware, various -d+acl logging additions and
473 more caution in buffer sizes.
475 PP/09 Implemented reverse_ip expansion operator.
477 PP/10 Bugzilla 937: provide a "debug" ACL control.
479 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
481 PP/12 Bugzilla 973: Implement --version.
483 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
485 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
487 PP/15 Bugzilla 816: support multiple condition rules on Routers.
489 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
490 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
491 ignore trailing whitespace.
493 JJ/02 prevent non-panic DKIM error from being sent to paniclog
495 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
498 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
499 Notification from Dr Andrew Aitchison.
501 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
502 ExtendedDetectionInfo response format.
503 Notification from John Horne.
505 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
508 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
509 XSL and documented dependency on system catalogs, with examples of how
512 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
515 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
516 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
517 configuration file which is writeable by the Exim user or group.
519 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
520 of configuration files to cover files specified with the -C option if
521 they are going to be used with root privileges, not just the default
524 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
525 option (effectively making it always true).
527 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
528 files to be used while preserving root privileges.
530 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
531 that rogue child processes cannot use them.
533 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
534 run-time user, instead of root.
536 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
537 Exim run-time user without dropping privileges.
539 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
540 result string, instead of calling string_vformat() twice with the same
543 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
544 for other users. Others should always drop root privileges if they use
545 -C on the command line, even for a whitelisted configure file.
547 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
549 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
555 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
556 $data_path, and $header_path variables; fixed documentation bugs and
559 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
560 exipick to access non-standard spools, including the "frozen" queue
563 NM/01 Bugzilla 965: Support mysql stored procedures.
564 Patch from Alain Williams
566 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
568 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
569 Patch from Andreas Metzler
571 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
572 Patch from Kirill Miazine
574 NM/05 Bugzilla 671: Added umask to procmail example.
576 JJ/03 installed exipick 20100323.0, fixing doc bug
578 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
579 directory. Notification and patch from Dan Rosenberg.
581 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
583 TK/02 Improve log output when DKIM signing operation fails.
585 MH/01 Treat the transport option dkim_domain as a colon separated
586 list, not as a single string, and sign the message with each element,
587 omitting multiple occurences of the same signer.
589 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
590 Bugzilla 985, 986. Patch by Simon Arlott
592 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
593 Patch by Simon Arlott
595 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
596 MBX locking. Notification from Dan Rosenberg.
602 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
604 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
606 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
608 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
610 NM/04 Bugzilla 915: Flag broken perl installation during build.
616 TK/01 Added patch by Johannes Berg that expands the main option
617 "spamd_address" if it starts with a dollar sign.
619 TK/02 Write list of recipients to X-Envelope-Sender header when building
620 the mbox-format spool file for content scanning (suggested by Jakob
623 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
624 (http://www.dcc-servers.net/) support via dccifd. Activated by
625 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
627 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
628 by Mark Daniel Reidel <mr@df.eu>.
630 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
631 When building exim an external PCRE library is now needed -
632 PCRE is a system library on the majority of modern systems.
633 See entry on PCRE_LIBS in EDITME file.
635 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
636 conversation. Added nologin parameter to request.
637 Patch contributed by Kirill Miazine.
639 TF/01 Do not log submission mode rewrites if they do not change the address.
641 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
643 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
644 log files in place. Contributed by Roberto Lima.
646 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
648 TF/03 Bugzilla 615: When checking the local_parts router precondition
649 after a local_part_suffix or local_part_prefix option, Exim now
650 does not use the address's named list lookup cache, since this
651 contains cached lookups for the whole local part.
653 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
654 Robert Millan. Documentation is in experimental-spec.txt.
656 TF/04 Bugzilla 668: Fix parallel build (make -j).
658 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
660 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
661 Patch provided by Jan Srzednicki.
663 TF/05 Leading white space used to be stripped from $spam_report which
664 wrecked the formatting. Now it is preserved.
666 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
667 that they are available at delivery time.
669 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
671 TF/08 TLS error reporting now respects the incoming_interface and
672 incoming_port log selectors.
674 TF/09 Produce a more useful error message if an SMTP transport's hosts
675 setting expands to an empty string.
677 NM/06 Bugzilla 744: EXPN did not work under TLS.
678 Patch provided by Phil Pennock.
680 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
681 Patch provided by Richard Godbee.
683 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
684 acl_smtp_notquit, added index entry.
686 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
687 Patch provided by Eugene Bujak.
689 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
690 accept(). Patch provided by Maxim Dounin.
692 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
693 Patch provided by Phil Pennock.
695 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
697 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
698 Patch provided by Brad "anomie" Jorsch.
700 NM/14 Bugzilla 622: Added timeout setting to transport filter.
701 Patch provided by Dean Brooks.
703 TK/05 Add native DKIM support (does not depend on external libraries).
705 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
706 Patch provided by Graeme Fowler.
708 NM/16 Bugzilla 851: Documentation example syntax fix.
710 NM/17 Changed NOTICE file to remove references to embedded PCRE.
712 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
715 NM/19 Bugzilla 745: TLS version reporting.
716 Patch provided by Phil Pennock.
718 NM/20 Bugzilla 167: bool: condition support.
719 Patch provided by Phil Pennock.
721 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
722 clients. Patch provided by Phil Pennock.
724 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
725 Patch provided by Brad "anomie" Jorsch.
727 NM/23 Bugzilla 687: Fix misparses in eximstats.
728 Patch provided by Heiko Schlittermann.
730 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
731 Patch provided by Heiko Schlittermann.
733 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
734 plus update to original patch.
736 NM/26 Bugzilla 799: Documentation correction for ratelimit.
738 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
739 Patch provided by David Brownlee.
741 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
743 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
745 NM/30 Bugzilla 888: TLS documentation bugfixes.
747 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
749 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
750 Unlike the original bugzilla I have changed all shell scripts in src tree.
752 NM/33 Bugzilla 898: Transport filter timeout fix.
753 Patch by Todd Rinaldo.
755 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
756 Patch by Serge Demonchaux.
758 NM/35 Bugzilla 39: Base64 decode bug fixes.
759 Patch by Jakob Hirsch.
761 NM/36 Bugzilla 909: Correct connect() call in dcc code.
763 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
765 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
767 NM/39 Bugzilla 911: Fixed MakeLinks build script.
773 TK/01 Add preliminary DKIM support. Currently requires a forked version of
774 ALT-N's libdkim that I have put here:
775 http://duncanthrax.net/exim-experimental/
777 Note to Michael Haardt: I had to rename some vars in sieve.c. They
778 were called 'true' and it seems that C99 defines that as a reserved
779 keyword to be used with 'bool' variable types. That means you could
780 not include C99-style headers which use bools without triggering
781 build errors in sieve.c.
783 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
784 as mailq or other aliases. Changed the --help handling significantly
785 to do whats expected. exim_usage() emits usage/help information.
787 SC/01 Added the -bylocaldomain option to eximstats.
789 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
791 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
793 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
799 PH/01 Another patch from the Sieve maintainer.
801 PH/02 When an IPv6 address is converted to a string for single-key lookup
802 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
803 dots are used instead of colons so that keys in lsearch files need not
804 contain colons. This was done some time before quoting was made available
805 in lsearch files. However, iplsearch files do require colons in IPv6 keys
806 (notated using the quote facility) so as to distinguish them from IPv4
807 keys. This meant that lookups for IP addresses in host lists did not work
808 for iplsearch lookups.
810 This has been fixed by arranging for IPv6 addresses to be expressed with
811 colons if the lookup type is iplsearch. This is not incompatible, because
812 previously such lookups could never work.
814 The situation is now rather anomolous, since one *can* have colons in
815 ordinary lsearch keys. However, making the change in all cases is
816 incompatible and would probably break a number of configurations.
818 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
821 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
822 conversion specification without a maximum field width, thereby enabling
823 a rogue spamd server to cause a buffer overflow. While nobody in their
824 right mind would setup Exim to query an untrusted spamd server, an
825 attacker that gains access to a server running spamd could potentially
826 exploit this vulnerability to run arbitrary code as the Exim user.
828 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
829 $primary_hostname instead of what libspf2 thinks the hosts name is.
831 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
832 a directory entry by the name of the lookup key. Previously, if a
833 symlink pointed to a non-existing file or a file in a directory that
834 Exim lacked permissions to read, a lookup for a key matching that
835 symlink would fail. Now it is enough that a matching directory entry
836 exists, symlink or not. (Bugzilla 503.)
838 PH/03 The body_linecount and body_zerocount variables are now exported in the
841 PH/04 Added the $dnslist_matched variable.
843 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
844 This means they are set thereafter only if the connection becomes
847 PH/06 Added the client_condition to authenticators so that some can be skipped
848 by clients under certain conditions.
850 PH/07 The error message for a badly-placed control=no_multiline_responses left
851 "_responses" off the end of the name.
853 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
855 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
856 (without spaces) instead of just copying the configuration text.
858 PH/10 Added the /noupdate option to the ratelimit ACL condition.
860 PH/11 Added $max_received_linelength.
862 PH/12 Added +ignore_defer and +include_defer to host lists.
864 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
865 way in which PCRE > 7.0 is built.
867 PH/14 Implemented queue_only_load_latch.
869 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
870 MAIL command. The effect was to mangle the value on 64-bit systems.
872 PH/16 Another patch from the Sieve maintainer.
874 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
876 PH/18 If a system quota error occurred while trying to create the file for
877 a maildir delivery, the message "Mailbox is full" was not appended to the
878 bounce if the delivery eventually timed out. Change 4.67/27 below applied
879 only to a quota excession during the actual writing of the file.
881 PH/19 It seems that peer DN values may contain newlines (and other non-printing
882 characters?) which causes problems in log lines. The DN values are now
883 passed through string_printing() before being added to log lines.
885 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
886 and InterBase are left for another time.)
888 PH/21 Added message_body_newlines option.
890 PH/22 Guard against possible overflow in moan_check_errorcopy().
892 PH/23 POSIX allows open() to be a macro; guard against that.
894 PH/24 If the recipient of an error message contained an @ in the local part
895 (suitably quoted, of course), incorrect values were put in $domain and
896 $local_part during the evaluation of errors_copy.
902 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
903 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
906 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
907 issue a MAIL command.
909 PH/02 In an ACL statement such as
911 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
913 if a client was not listed at all, or was listed with a value other than
914 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
915 the condition was not true (as it should be), so access was not denied.
916 The bug was that the ! inversion was incorrectly passed on to the second
917 item. This has been fixed.
919 PH/03 Added additional dnslists conditions == and =& which are different from
920 = and & when the dns lookup returns more than one IP address.
922 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
923 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
925 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
926 FSYNC, which compiles an option called disable_fsync that allows for
927 bypassing fsync(). The documentation is heavily laced with warnings.
929 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
931 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
932 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
933 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
934 including adding "make clean"; (3) Added -fPIC when compiling the test
935 dynamically loaded module, to get rid of a warning.
937 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
938 message fails, move_frozen_messages = true and ignore_bounce_errors_after
939 = 0s. The bug is otherwise harmless.
941 PH/07 There was a bug in the dovecot authenticator such that the value of
942 $auth1 could be overwritten, and so not correctly preserved, after a
943 successful authentication. This usually meant that the value preserved by
944 the server_setid option was incorrect.
946 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
948 PH/09 Installed PCRE release 7.0.
950 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
951 run for batched SMTP input. It is now run at the start of every message
952 in the batch. While fixing this I discovered that the process information
953 (output by running exiwhat) was not always getting set for -bs and -bS
954 input. This is fixed, and it now also says "batched" for BSMTP.
956 PH/11 Added control=no_pipelining.
958 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
959 patch, slightly modified), and move the expansion of helo_data till after
960 the connection is made in the smtp transport (so it can use these
963 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
965 PH/14 Added log_selector = +pid.
967 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
969 PH/16 Add ${if forany and ${if forall.
971 PH/17 Added dsn_from option to vary the From: line in DSNs.
973 PH/18 Flush SMTP output before performing a callout, unless control =
974 no_callout_flush is set.
976 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
977 was true (the default) a successful delivery failed to delete the retry
978 item, thus causing premature timeout of the address. The bug is now
981 PH/20 Added hosts_avoid_pipelining to the smtp transport.
983 PH/21 Long custom messages for fakedefer and fakereject are now split up
984 into multiline reponses in the same way that messages for "deny" and
985 other ACL rejections are.
987 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
988 with slight modification.
990 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
991 draft, changing the syntax and factoring some duplicate code".
993 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
994 for deliveries of the second and subsequent messages over the same SMTP
997 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
998 ${reduce, with only minor "tidies".
1000 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1002 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1003 expansion side effects.
1005 PH/27 When a message times out after an over-quota error from an Exim-imposed
1006 quota, the bounce message says "mailbox is full". This message was not
1007 being given when it was a system quota that was exceeded. It now should
1010 MH/03 Made $recipients available in local_scan(). local_scan() already has
1011 better access to the recipient list through recipients_list[], but
1012 $recipients can be useful in postmaster-provided expansion strings.
1014 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
1015 in the case of a MAIL command with additional options following the
1016 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
1017 were accidentally chopped off.
1019 PH/29 SMTP synchronization checks are implemented when a command is read -
1020 there is a check that no more input is waiting when there shouldn't be
1021 any. However, for some commands, a delay in an ACL can mean that it is
1022 some time before the response is written. In this time, more input might
1023 arrive, invalidly. So now there are extra checks after an ACL has run for
1024 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
1025 pipelining has not been advertised.
1027 PH/30 MH's patch to allow iscntrl() characters to be list separators.
1029 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
1030 returned in the SMTP response when smtp_return_error_details was false.
1031 This has been fixed.
1033 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
1034 instead of the C I/O that was originally supplied, because problems were
1035 reported on Solaris.
1037 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
1038 Exim which did not show up earlier: it was assuming that a call to
1039 SSL_CTX_set_info_callback() might give an error value. In fact, there is
1040 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
1041 was a macro that became an assignment, so it seemed to work. This has
1042 changed to a proper function call with a void return, hence the compile
1043 error. Exim's code has been fixed.
1045 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
1048 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
1050 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
1052 PH/37 If a message is not accepted after it has had an id assigned (e.g.
1053 because it turns out to be too big or there is a timeout) there is no
1054 "Completed" line in the log. When some messages of this type were
1055 selected by exigrep, they were listed as "not completed". Others were
1056 picked up by some special patterns. I have improved the selection
1057 criteria to be more general.
1059 PH/38 The host_find_failed option in the manualroute router can now be set
1060 to "ignore", to completely ignore a host whose IP address cannot be
1061 found. If all hosts are ignored, the behaviour is controlled by the new
1062 host_all_ignored option.
1064 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
1065 homing or because of multiple MX records with /mx) generated more than
1066 one IP address, and the following item turned out to be the local host,
1067 all the secondary addresses of the first item were incorrectly removed
1068 from the list, along with the local host and any following hosts (which
1069 is what is supposed to happen).
1071 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
1072 whoever called Exim into the -H file. In the case of the daemon it was
1073 behaving confusingly. When first started, it used values for whoever
1074 started the daemon, but after a SIGHUP it used the Exim user (because it
1075 calls itself on a restart). I have changed the code so that it now always
1078 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
1079 message are rejected with the same error (e.g. no authentication or bad
1080 sender address), and a DATA command is nevertheless sent (as can happen
1081 with PIPELINING or a stupid MUA), the error message that was given to the
1082 RCPT commands is included in the rejection of the DATA command. This is
1083 intended to be helpful for MUAs that show only the final error to their
1086 PH/42 Another patch from the Sieve maintainer.
1088 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
1089 Eximstats - Fixed some broken HTML links and added missing column headers
1091 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
1092 columns for Rejects, Temp Rejects, Ham, and Spam rows.
1094 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
1096 PH/43 Yet another patch from the Sieve maintainer.
1098 PH/44 I found a way to check for a TCP/IP connection going away before sending
1099 the response to the final '.' that terminates a message, but only in the
1100 case where the client has not sent further data following the '.'
1101 (unfortunately, this is allowed). However, in many cases there won't be
1102 any further data because there won't be any more messages to send. A call
1103 to select() can be used: if it shows that the input is "ready", there is
1104 either input waiting, or the socket has been closed. An attempt to read
1105 the next input character can distinguish the two cases. Previously, Exim
1106 would have sent an OK response which the client would never have see.
1107 This could lead to message repetition. This fix should cure that, at
1108 least in a lot of common cases.
1110 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
1111 advertised in response to EHLO.
1117 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
1118 fixed by 4.65/MH/01 (is this a record?) are fixed:
1120 (i) An empty string was always treated as zero by the numeric comparison
1121 operators. This behaviour has been restored.
1123 (ii) It is documented that the numeric comparison operators always treat
1124 their arguments as decimal numbers. This was broken in that numbers
1125 starting with 0 were being interpreted as octal.
1127 While fixing these problems I realized that there was another issue that
1128 hadn't been noticed. Values of message_size_limit (both the global option
1129 and the transport option) were treated as octal if they started with 0.
1130 The documentation was vague. These values are now always treated as
1131 decimal, and I will make that clear in the documentation.
1137 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
1138 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
1141 MH/01 Don't check that the operands of numeric comparison operators are
1142 integers when their expansion is in "skipping" mode (fixes bug
1143 introduced by 4.64-PH/07).
1145 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
1146 child addresses, Exim now panics and dies. Previously, because the count
1147 is held in a short int, deliveries were likely to be lost. As such a
1148 large number of recipients for a single message is ridiculous
1149 (performance will be very, very poor), I have chosen to impose a limit
1150 rather than extend the field.
1156 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
1157 leftover -K file (the existence of which was triggered by #402).
1158 While we were at it, introduced process PID as part of the -K
1159 filename. This should rule out race conditions when creating
1162 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
1163 processing considerably. Previous code took too long for large mails,
1164 triggering a timeout which in turn triggers #401.
1166 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
1167 in the DK code in transports.c. sendfile() is not really portable,
1168 hence the _LINUX specificness.
1170 TF/01 In the add_headers option to the mail command in an Exim filter,
1171 there was a bug that Exim would claim a syntax error in any
1172 header after the first one which had an odd number of characters
1175 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
1176 callout verification, Exim cached a "reject" for the entire domain. This
1177 is correct for most verifications, but it is not correct for a recipient
1178 verification with use_sender or use_postmaster set, because in that case
1179 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
1180 case of MAIL FROM:<> rejection from other early rejections (e.g.
1181 rejection of HELO). When verifying a recipient using a non-null MAIL
1182 address, the cache is ignored if it shows MAIL FROM:<> rejection.
1183 Whatever the result of the callout, the value of the domain cache is
1184 left unchanged (for any other kind of callout, getting as far as trying
1185 RCPT means that the domain itself is ok).
1187 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
1190 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
1191 manifest itself as EPIPE rather than ECONNECT. When tidying away a
1192 session, the daemon ignores ECONNECT errors and logs others; it now
1193 ignores EPIPE as well.
1195 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
1196 (quoted-printable decoding).
1198 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
1199 later the small subsequent patch to fix an introduced bug.
1201 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
1203 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
1205 PH/08 An error is now given if message_size_limit is specified negative.
1207 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
1208 to be given (somewhat) arbitrary names.
1210 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
1213 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
1214 miscellaneous code fixes
1216 PH/10 Added the log_reject_target ACL modifier to specify where to log
1219 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
1220 hostname. This is wrong, because it relates to the incoming message (and
1221 probably the interface on which it is arriving) and not to the outgoing
1222 callout (which could be using a different interface). This has been
1223 changed to use the value of the helo_data option from the smtp transport
1224 instead - this is what is used when a message is actually being sent. If
1225 there is no remote transport (possible with a router that sets up host
1226 addresses), $smtp_active_hostname is used.
1228 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
1229 tweaks were necessary in order to get it to work (see also 21 below):
1230 (a) The code assumed that strncpy() returns a negative number on buffer
1231 overflow, which isn't the case. Replaced with Exim's string_format()
1233 (b) There were several signed/unsigned issues. I just did the minimum
1234 hacking in of casts. There is scope for a larger refactoring.
1235 (c) The code used strcasecmp() which is not a standard C function.
1236 Replaced with Exim's strcmpic() function.
1237 (d) The code set only $1; it now sets $auth1 as well.
1238 (e) A simple test gave the error "authentication client didn't specify
1239 service in request". It would seem that Dovecot has changed its
1240 interface. Fortunately there's a specification; I followed it and
1241 changed what the client sends and it appears to be working now.
1243 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
1246 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
1247 address A is aliased to B and C, where B exists and C does not. Without
1248 -v the output is "A verified" because verification stops after a
1249 successful redirection if more than one address is generated. However,
1250 with -v the child addresses are also verified. Exim was outputting "A
1251 failed to verify" and then showing the successful verification for C,
1252 with its parentage. It now outputs "B failed to verify", showing B's
1253 parentage before showing the successful verification of C.
1255 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
1256 look up a TXT record in a specific list after matching in a combined
1259 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
1260 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
1261 they consult the DNS. I had assumed they would set it the way they
1262 wanted; and indeed my experiments on Linux seem to show that in some
1263 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
1264 To be on the safe side, however, I have now made the interface to
1265 host_find_byname() similar to host_find_bydns(), with an argument
1266 containing the DNS resolver options. The host_find_byname() function now
1267 sets these options at its start, just as host_find_bydns() does. The smtp
1268 transport options dns_qualify_single and dns_search_parents are passed to
1269 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
1270 of host_find_byname() use the default settings of RES_DEFNAMES
1271 (qualify_single) but not RES_DNSRCH (search_parents).
1273 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
1274 spool_read_header() do less string testing, by means of a preliminary
1275 switch on the second character of optional "-foo" lines. (This is
1276 overdue, caused by the large number of possibilities that now exist.
1277 Originally there were few.) While I was there, I also converted the
1278 str(n)cmp tests so they don't re-test the leading "-" and the first
1279 character, in the hope this might squeeze out yet more improvement.
1281 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
1282 flag allowing group syntax was set by the header_syntax check but not
1283 turned off, possible causing trouble later; (2) The flag was not being
1284 set at all for the header_verify test, causing "group"-style headers to
1285 be rejected. I have now set it in this case, and also caused header_
1286 verify to ignore an empty address taken from a group. While doing this, I
1287 came across some other cases where the code for allowing group syntax
1288 while scanning a header line wasn't quite right (mostly, not resetting
1289 the flag correctly in the right place). These bugs could have caused
1290 trouble for malformed header lines. I hope it is now all correct.
1292 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
1293 with the "reply" argument non-NULL. The code, however (which originally
1294 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
1295 but it didn't always do it. This confused somebody who was copying the
1296 code for some other use. I have removed all the tests.
1298 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
1299 feature that was used to support insecure browsers during the U.S. crypto
1300 embargo. It requires special client support, and Exim is probably the
1301 only MTA that supported it -- and would never use it because real RSA is
1302 always available. This code has been removed, because it had the bad
1303 effect of slowing Exim down by computing (never used) parameters for the
1304 RSA_EXPORT functionality.
1306 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
1307 authenticator to fail if there's a tab character in the incoming data
1308 (there should never be unless someone is messing about, as it's supposed
1309 to be base64-encoded). Also added, on Timo's advice, the "secured" option
1310 if the connection is using TLS or if the remote IP is the same as the
1311 local IP, and the "valid-client-cert option" if a client certificate has
1314 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1315 authenticators. This can be used for authorization after authentication
1316 succeeds. (In the case of plaintext, it servers for both authentication
1319 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1320 if any retry times were supplied.
1322 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1323 connection, where there is no client IP address to check. In this
1324 situation, the verify now always succeeds.
1326 PH/25 Applied John Jetmore's -Mset patch.
1328 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1330 PH/27 In a string expansion for a processed (not raw) header when multiple
1331 headers of the same name were present, leading whitespace was being
1332 removed from all of them, but trailing whitespace was being removed only
1333 from the last one. Now trailing whitespace is removed from each header
1334 before concatenation. Completely empty headers in a concatenation (as
1335 before) are ignored.
1337 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1338 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1340 PH/29 [Removed. This was a change that I later backed out, and forgot to
1341 correct the ChangeLog entry (that I had efficiently created) before
1342 committing the later change.]
1344 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1345 address errors (4xx response to RCPT) over the same connection as other
1346 messages routed to the same hosts. Such deliveries are always "forced",
1347 so retry times are not inspected. This resulted in far too many retries
1348 for the affected addresses. The effect occurred only when there were more
1349 hosts than the hosts_max_try setting in the smtp transport when it had
1350 the 4xx errors. Those hosts that it had tried were not added to the list
1351 of hosts for which the message was waiting, so if all were tried, there
1352 was no problem. Two fixes have been applied:
1354 (i) If there are any address or message errors in an SMTP delivery, none
1355 of the hosts (tried or untried) are now added to the list of hosts
1356 for which the message is waiting, so the message should not be a
1357 candidate for sending over the same connection that was used for a
1358 successful delivery of some other message. This seems entirely
1359 reasonable: after all the message is NOT "waiting for some host".
1360 This is so "obvious" that I'm not sure why it wasn't done
1361 previously. Hope I haven't missed anything, but it can't do any
1362 harm, as the worst effect is to miss an optimization.
1364 (ii) If, despite (i), such a delivery is accidentally attempted, the
1365 routing retry time is respected, so at least it doesn't keep
1366 hammering the server.
1368 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1369 in ${readsocket because some servers need this prod.
1371 PH/32 Added some extra debug output when updating a wait-xxx database.
1373 PH/33 The hint "could be header name not terminated by colon", which has been
1374 given for certain expansion errors for a long time, was not being given
1375 for the ${if def:h_colon_omitted{... case.
1377 PH/34 The spec says: "With one important exception, whenever a domain list is
1378 being scanned, $domain contains the subject domain." There was at least
1379 one case where this was not true.
1381 PH/35 The error "getsockname() failed: connection reset by peer" was being
1382 written to the panic log as well as the main log, but it isn't really
1383 panic-worthy as it just means the connection died rather early on. I have
1384 removed the panic log writing for the ECONNRESET error when getsockname()
1387 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1388 runs only) independently of the message's sender address. This meant
1389 that, if the 4xx error was in fact related to the sender, a different
1390 message to the same recipient with a different sender could confuse
1391 things. In particualar, this can happen when sending to a greylisting
1392 server, but other circumstances could also provoke similar problems.
1393 I have changed the default so that the retry time for these errors is now
1394 based a combination of the sender and recipient addresses. This change
1395 can be overridden by setting address_retry_include_sender=false in the
1398 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1399 remote server are returned as part of bounce messages. This was not
1400 happening for LMTP over a pipe (the lmtp transport), but now it is the
1401 same for both kinds of LMTP.
1403 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1404 in header lines that were in fact CNAMEs. This is no longer the case.
1406 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1407 and queue runs started by the daemon processed all messages. This has
1408 been fixed so that -R and -S can now usefully be given with -q<time>.
1410 PH/40 Import PCRE release 6.7 (fixes some bugs).
1412 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1414 PH/42 Give an error if -q is specified more than once.
1416 PH/43 Renamed the variables $interface_address and $interface_port as
1417 $received_ip_address and $received_port, to make it clear that these
1418 values apply to message reception, and not to the outgoing interface when
1419 a message is delivered. (The old names remain recognized, of course.)
1421 PH/44 There was no timeout on the connect() call when using a Unix domain
1422 socket in the ${readsocket expansion. There now is.
1424 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1425 be meaningful with "accept".
1427 SC/01 Eximstats V1.43
1428 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1430 SC/02 Eximstats V1.44
1431 Use a glob alias rather than an array ref in the generated
1432 parser. This improves both readability and performance.
1434 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1435 Collect SpamAssassin and rejection statistics.
1436 Don't display local sender or destination tables unless
1437 there is data to show.
1438 Added average volumes into the top table text output.
1440 SC/04 Eximstats V1.46
1441 Collect data on the number of addresses (recipients)
1442 as well as the number of messages.
1444 SC/05 Eximstats V1.47
1445 Added 'Message too big' to the list of mail rejection
1446 reasons (thanks to Marco Gaiarin).
1448 SC/06 Eximstats V1.48
1449 Mainlog lines which have GMT offsets and are too short to
1450 have a flag are now skipped.
1452 SC/07 Eximstats V1.49 (Alain Williams)
1453 Added the -emptyok flag.
1455 SC/08 Eximstats V1.50
1456 Fixes for obtaining the IP address from reject messages.
1458 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1459 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1460 whitesspace changes from 4.64-PH/27
1462 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1465 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1466 are found, allow negative numbers in numeric criteria)
1468 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1470 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1471 to match changes made in 4.64-PH/43
1473 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1475 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1476 "domain" is actually an IP address, and force a failure. This locks out
1477 those revolvers/nameservers that support "A-for-A" lookups, in
1478 contravention of the specifications.
1480 PH/48 When a host name was looked up from an IP address, and the subsequent
1481 forward lookup of the name timed out, the host name was left in
1482 $sender_host_name, contrary to the specification.
1484 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1485 restricted to single-key lookups, Exim was not diagnosing an error if
1486 * or *@ was used with a query-style lookup.
1488 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1490 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1491 long ago, but noone interested enough thought of it. Let's just say that
1492 the "1.1" means that there are some new functions that weren't there at
1493 some point in the past.
1495 PH/51 Error processing for expansion failure of helo_data from an smtp
1496 transport during callout processing was broken.
1498 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1499 tested/used via the -bh/-bhc/-bs options.
1501 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1502 bug, fixed in subsequent PCRE releases).
1504 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1505 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1507 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1513 SC/01 Use a glob alias rather than an array ref in eximstats generated
1514 parser. This improves both readability and performance.
1516 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1517 Don't display local sender or destination tables in eximstats unless
1518 there is data to show.
1519 Added average volumes into the eximstats top table text output.
1521 SC/03 Collect data on the number of addresses (recipients) as well
1522 as the number of messages in eximstats.
1524 TF/01 Correct an error in the documentation for the redirect router. Exim
1525 does (usually) call initgroups() when daemonizing.
1527 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1528 with consistent privilege compared to when running as a daemon.
1530 TF/03 Note in the spec that $authenticated_id is not set for local
1531 submissions from trusted users.
1533 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1534 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1536 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1537 by adding some example configuration directives to the default
1538 configuration file. A little bit of work is required to uncomment the
1539 directives and define how usernames and passwords are checked, but
1540 there is now a framework to start from.
1542 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1543 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1544 without this. I don't know how relevant this is to other LDAP libraries.
1546 PH/02 Add the verb name to the "unknown ACL verb" error.
1548 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1550 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1552 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1553 directory not expanded when it should be if an expanded home directory
1554 was set for the address (which is overridden by the transport).
1556 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1559 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1560 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1561 because it is too late at that time, and has no effect.
1563 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1564 security issue with \' (bugzilla #107). I could not use the
1565 PQescapeStringConn() function, because it needs a PGconn value as one of
1568 PH/08 When testing addresses using -bt, indicate those final addresses that
1569 are duplicates that would not cause an additional delivery. At least one
1570 person was confused, thinking that -bt output corresponded to deliveries.
1571 (Suppressing duplicates isn't a good idea as you lose the information
1572 about possibly different redirections that led to the duplicates.)
1574 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1575 systems where poll() doesn't work, in particular OS X.
1577 PH/10 Added more information to debugging output for retry time not reached.
1579 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1580 operations in malware.c.
1582 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1585 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1586 syslog_duplication set false, log lines that would normally be written
1587 both the the main log and to the reject log were not written to syslog at
1590 PH/14 In the default configuration, change the use of "message" in ACL warn
1591 statements to "add_header".
1593 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1594 not followed by a command (e.g. "seen endif").
1596 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1597 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1600 PH/17 Added extra conditions to the default value of delay_warning_condition
1604 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1605 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1606 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1609 The Auto-Submitted: and various List- headers are standardised, whereas I
1610 don't think Precedence: ever was.
1612 PH/18 Refactored debugging code in route_finduser() to show more information,
1613 in particular, the error code if getpwnam() issues one.
1615 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1616 This is apparently needed in addition to the PH/07 change above to avoid
1617 any possible encoding problems.
1619 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1620 but not after initializing Perl.
1622 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1623 output them only if debugging. By default they are written stderr,
1624 apparently, which is not desirable.
1626 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1629 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1632 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1634 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1635 authenticated or an ident call has been made. Suppress the default
1636 values for $authenticated_id and $authenticated_sender (but permit -oMai
1637 and -oMas) when testing with -bh.
1639 PH/24 Re-jigged the order of the tests in the default configuration so that the
1640 tests for valid domains and recipients precede the DNS black list and CSA
1641 tests, on the grounds that those ones are more expensive.
1643 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1644 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1645 This bug exists in every version of Exim that I still have, right back to
1648 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1649 However, an attempt to turn on case-sensitivity in a regex key by
1650 including (?-i) didn't work because the subject string was already
1651 lowercased, and the effects were non-intuitive. It turns out that a
1652 one-line patch can be used to allow (?-i) to work as expected.
1658 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1659 other effects) broke the use of negated acl sub-conditions.
1661 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1664 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1665 "Deny" causes Exim to reject the incoming connection with a 554 error.
1666 Unfortunately, if there is a major crisis, such as a disk failure,
1667 tcp-wrappers gives "deny", whereas what one would like would be some
1668 kind of temporary error. A kludge has been added to help with this.
1669 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1670 554 error is used if errno is still zero or contains ENOENT (which occurs
1671 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1674 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1676 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1677 errors. Otherwise a message that provokes a temporary error (when other
1678 messages do not) can cause a whole host to time out.
1680 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1681 addresses were routed directly to files or pipes from a redirect router.
1682 File deliveries just didn't batch; pipe deliveries might have suffered
1685 PH/06 A failure to get a lock for a hints database would erroneously always say
1686 "Failed to get write lock", even when it was really a read lock.
1688 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1689 of 0600. This has been changed to use the value of the lockfile_mode
1690 option (which defaults to 0600).
1692 PH/08 Applied small patch from the Sieve maintainer.
1694 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1695 folder from quota calculations, a direct delivery into this folder messed
1696 up the contents of the maildirsize file. This was because the regex was
1697 used only to exclude .Trash (or whatever) when the size of the mailbox
1698 was calculated. There was no check that a delivery was happening into an
1699 excluded directory. This bug has been fixed by ignoring all quota
1700 processing for deliveries into excluded directories.
1702 PH/10 Added the maildirfolder_create_regex option to appendfile.
1708 PH/01 The code for finding all the local interface addresses on a FreeBSD
1709 system running IPv6 was broken. This may well have applied to all BSD
1710 systems, as well as to others that have similar system calls. The broken
1711 code found IPv4 interfaces correctly, but gave incorrect values for the
1712 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1713 that it would not match correctly against @[] and not recognize the IPv6
1716 PH/02 The ipliteral router was not recognizing addresses of the form user@
1717 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1719 PH/03 Added disable_ipv6.
1721 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1722 decoded form, because it is most often used to construct To: headers
1723 lines in autoreplies, and the decoded form may well be syntactically
1724 invalid. However, $reply_address has leading white space removed, and all
1725 newlines turned into spaces so that the autoreply transport does not
1728 PH/05 If group was specified without a user on a router, and no group or user
1729 was specified on a transport, the group from the router was ignored.
1731 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1732 for visible compile-time settings that can be used to change these
1733 numbers, for those that want even more. Backwards compatibility with old
1734 spool files has been maintained. However, going back to a previous Exim
1735 release will lost any variables that are in spool files.
1737 PH/07 Two small changes when running in the test harness: increase delay when
1738 passing a TCP/IP connection to a new process, in case the original
1739 process has to generate a bounce, and remove special handling of
1740 127.0.0.2 (sic), which is no longer necessary.
1742 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1743 be the same on different OS.
1745 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1748 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1749 whether --show-vars was specified or not
1751 JJ/02 exipick: Added support for new ACL variable spool format introduced
1754 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1755 syntactically invalid From: or Reply-to: line, and a filter used this to
1756 generate an autoreply, and therefore failed to obtain an address for the
1757 autoreply, Exim could try to deliver to a non-existent relative file
1758 name, causing unrelated and misleading errors. What now happens is that
1759 it logs this as a hard delivery error, but does not attempt to create a
1762 PH/11 The exinext utility has a -C option for testing purposes, but although
1763 the given file was scanned by exinext itself; it wasn't being passed on
1764 when Exim was called.
1766 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1767 an end-of-file indication when reading a command response.
1769 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1770 compiled. In many other places in Exim, IPv6 addresses are always
1771 recognized, so I have changed this. It also means that IPv4 domain
1772 literals of the form [IPV4:n.n.n.n] are now always recognized.
1774 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1775 used if the router is not running as root, for example, when verifying at
1776 ACL time, or when using -bh. The debugging output from this situation was
1777 non-existent - all you got was a failure to exec. I have made two
1780 (a) Failures to set uid/gid, the current directory, or a process leader
1781 in a subprocess such as that created by queryprogram now generate
1782 suitable debugging ouput when -d is set.
1784 (b) The queryprogram router detects when it is not running as root,
1785 outputs suitable debugging information if -d is set, and then runs
1786 the subprocess without attempting to change uid/gid.
1788 PH/15 Minor change to Makefile for building test_host (undocumented testing
1791 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1792 additional section of a DNS packet that returns MX or SRV records.
1793 Instead, it always explicitly searches for A/AAAA records. This avoids
1794 major problems that occur when a DNS server includes only records of one
1795 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1796 fixed another bug: if SRV records were looked up and the corresponding
1797 address records were *not* found in the additional section, the port
1798 values from the SRV records were lost.
1800 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1801 using the correct key (the original address) when searching the retry
1802 rules in order to find which one to use for generating the retry hint.
1804 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1805 adding the default one. Similarly, if it contains a Reply-To: header, the
1806 errors_reply_to option, if set, is not used.
1808 PH/19 When calculating a retry time, Exim used to measure the "time since
1809 failure" by looking at the "first failed" field in the retry record. Now
1810 it does not use this if it is later than than the arrival time of the
1811 message. Instead it uses the arrival time. This makes for better
1812 behaviour in cases where some deliveries succeed, thus re-setting the
1813 "first failed" field. An example is a quota failure for a huge message
1814 when small messages continue to be delivered. Without this change, the
1815 "time since failure" will always be short, possible causing more frequent
1816 delivery attempts for the huge message than are intended.
1817 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1819 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1820 $1, $2, $3) because the numerical variables can be reset during some
1821 expansion items (e.g. "match"), thereby losing the authentication data.
1823 PH/21 Make -bV show the size of off_t variables so that the test suite can
1824 decide whether to run tests for quotas > 2G.
1826 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1827 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1828 filecount value is greater than 2G or if a quota value is greater than 2G
1829 on a system where the size of off_t is not greater than 4, a panic error
1832 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1833 never match. The debug and -bh output now contains an explicit error
1834 message indicating a malformed IPv4 address or mask.
1836 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1837 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1838 PH/23 above applies.
1840 PH/25 Do not write to syslog when running in the test harness. The only
1841 occasion when this arises is a failure to open the main or panic logs
1842 (for which there is an explicit test).
1844 PH/26 Added the /no_tell option to "control=freeze".
1846 PH/27 If a host name lookup failed very early in a connection, for example, if
1847 the IP address matched host_lookup and the reverse lookup yielded a name
1848 that did not have a forward lookup, an error message of the form "no IP
1849 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1850 could be logged. Now it outputs the IP address instead of "NULL".
1852 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1853 allows the sender and the authenticated sender to be set when
1854 submitting a message from within Exim. Since child_open_exim() is
1855 documented for local_scan(), the new function should be too.
1857 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1858 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1859 results in an empty string is now treated as unset.
1861 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1863 PH/31 Added sender_verify_fail logging option.
1865 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1866 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1867 tidied the source and removed it altogether.
1869 PH/33 When a queue run was abandoned because the load average was too high, a
1870 log line was always written; now it is written only if the queue_run log
1871 selector is set. In addition, the log line for abandonment now contains
1872 information about the queue run such as the pid. This is always present
1873 in "start" and "stop" lines but was omitted from the "abandon" line.
1875 PH/34 Omit spaces between a header name and the colon in the error message that
1876 is given when verify = headers_syntax fails (if there are lots of them,
1877 the message gets confusing).
1879 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1880 names, as there are now some PTR records that contain slashes. This check
1881 is only to protect against broken name servers that fall over on strange
1882 characters, so the fact that it applies to all lookups doesn't matter.
1884 PH/36 Now that the new test suite is complete, we can remove some of the
1885 special code in Exim that was needed for the old test suite. For example,
1886 sorting DNS records because real resolvers return them in an arbitrary
1887 order. The new test suite's fake resolver always returns records in the
1890 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1891 bounces) except when queue_only is set, to avoid logging races between
1892 the different processes.
1894 PH/38 Panic-die if .include specifies a non-absolute path.
1896 PH/39 A tweak to the "H" retry rule from its user.
1898 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1899 a label. They prevented compilation on older perls.
1901 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1902 a warning to be raised on newish perls.
1904 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1905 on queue. Changes to match documented behaviour of showing count of
1906 messages matching specified criteria.
1908 PH/40 Changed the default ident timeout from 30s to 5s.
1910 PH/41 Added support for the use of login_cap features, on those BSD systems
1911 that have them, for controlling the resources used by pipe deliveries.
1913 PH/42 The content-scanning code uses fopen() to create files in which to put
1914 message data. Previously it was not paying any attention to the mode of
1915 the files. Exim runs with umask(0) because the rest of the code creates
1916 files with open(), and sets the required mode explicitly. Thus, these
1917 files were ending up world-writeable. This was not a big issue, because,
1918 being within the spool directory, they were not world-accessible. I have
1919 created a function called modefopen, which takes an additional mode
1920 argument. It sets umask(777), creates the file, chmods it to the required
1921 mode, then resets the umask. All the relevant calls to fopen() in the
1922 content scanning code have been changed to use this function.
1924 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1925 to 24 hours. This avoids potential overflow problems when processing G
1926 and H retry rules. I suspect nobody ever tinkers with this value.
1928 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1930 PH/45 When the plaintext authenticator is running as a client, the server's
1931 challenges are checked to ensure they are valid base64 strings. By
1932 default, the authentication attempt is cancelled if an invalid string is
1933 received. Setting client_ignore_invalid_base64 true ignores these errors.
1934 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1935 they are received. Thus, the responses can be made to depend on the
1936 challenges. If an invalid string is ignored, an empty string is placed in
1939 PH/46 Messages that are created by the autoreply transport now contains a
1940 References: header, in accordance with RFCs 2822 and 3834.
1942 PH/47 Added authenticated_sender_force to the smtp transport.
1944 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1946 PH/49 Installed latest patch from the Sieve maintainer.
1948 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1949 was also set, the appendfile transport was unnecessarily scanning a
1950 directory of message files (e.g. for maildir delivery) to find the count
1951 of files (along with the size), even though it did not need this
1952 information. It now does the scan only if it needs to find either the
1953 size of the count of files.
1955 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1957 PH/52 Two bugs concerned with error handling when the smtp transport is
1960 (i) Exim was not creating retry information for temporary errors given
1961 for individual recipients after the DATA command when the smtp transport
1962 was used in LMTP mode. This meant that they could be retried too
1963 frequently, and not timed out correctly.
1965 (ii) Exim was setting the flag that allows error details to be returned
1966 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1967 recipients that were returned after the DATA command.
1969 PH/53 This is related to PH/52, but is more general: for any failing address,
1970 when detailed error information was permitted to be returned to the
1971 sender, but the error was temporary, then after the final timeout, only
1972 "retry timeout exceeded" was returned. Now it returns the full error as
1973 well as "retry timeout exceeded".
1975 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1976 do this, and (what is worse) MTAs that accept it.
1978 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1979 will now be deprecated.
1981 PH/56 New os.c-cygwin from the Cygwin maintainer.
1983 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1984 formats (previously only available in exim formats via -bpr, -bpru,
1985 and -bpra. Now also available in native and exiqgrep formats)
1987 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1988 with very large, slow to parse queues
1990 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1992 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1994 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1995 responses to authentication challenges, though it was showing the
1996 challenges; (ii) I've removed the CR characters from the debug output for
1999 PH/58 Allow for the insertion of a newline as well as a space when a string
2000 is turned into more than one encoded-word during RFC 2047 encoding. The
2001 Sieve code now uses this.
2003 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2004 data_4xx, lost_connection, tls_required.
2006 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2007 message was being sent as an SMTP response.
2009 PH/61 Add -l and -k options to exicyclog.
2011 PH/62 When verifying, if an address was redirected to one new address, so that
2012 verification continued, and the new address failed or deferred after
2013 having set something in $address_data, the value of $address_data was not
2014 passed back to the ACL. This was different to the case when no
2015 redirection occurred. The value is now passed back in both cases.
2017 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
2018 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
2019 use login_cap.h, so on its own it isn't the distinguishing feature. The
2020 new name refers directly to the setclassresources() function.
2022 PH/65 Added configuration files for NetBSD3.
2024 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
2026 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
2027 is preferred over IPv4.
2029 PH/68 The bounce_return_message and bounce_return_body options were not being
2030 honoured for bounces generated during the reception of non-SMTP messages.
2031 In particular, this applied to messages rejected by the ACL. This bug has
2032 been fixed. However, if bounce_return_message is true and bounce_return_
2033 body is false, the headers that are returned for a non-SMTP message
2034 include only those that have been read before the error was detected.
2035 (In the case of an ACL rejection, they have all been read.)
2037 PH/69 The HTML version of the specification is now built in a directory called
2038 spec_html instead of spec.html, because the latter looks like a path with
2039 a MIME-type, and this confuses some software.
2041 PH/70 Catch two compiler warnings in sieve.c.
2043 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
2044 function verify_get_ident() calls ip_connect() to connect a socket, but
2045 if the "connect()" function timed out, ip_connect() used to close the
2046 socket. However, verify_get_ident() also closes the socket later, and in
2047 between Exim writes to the log, which may get opened at this point. When
2048 the socket was closed in ip_connect(), the log could get the same file
2049 descriptor number as the socket. This naturally causes chaos. The fix is
2050 not to close the socket in ip_connect(); the socket should be closed by
2051 the function that creates it. There was only one place in the code where
2052 this was missing, in the iplookup router, which I don't think anybody now
2053 uses, but I've fixed it anyway.
2055 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
2056 well as to direct DNS lookups. Otherwise the handling of names in host
2057 lists is inconsistent and therefore confusing.
2063 PH/01 Two changes to the default runtime configuration:
2065 (1) Move the checks for relay_from_hosts and authenticated clients from
2066 after to before the (commented out) DNS black list checks.
2068 (2) Add control=submission to the relay_from_hosts and authenticated
2069 clients checks, on the grounds that messages accepted by these
2070 statements are most likely to be submissions.
2072 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
2074 (1) Generate an error if the third argument for the ${prvs expansion is
2077 (2) Treat a missing third argument of ${prvscheck as if it were an empty
2080 (3) Reset the variables that are obtained from the first argument of
2081 ${prvscheck and used in the second argument before leaving the code,
2082 because their memory is reclaimed, so using them afterwards may do
2085 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
2086 one (it's much easier than Tom thought :-).
2088 (5) Because of (4), we can now allow for the use of $prvscheck_result
2089 inside the third argument.
2091 PH/03 For some reason, the default setting of PATH when running a command from
2092 a pipe transport was just "/usr/bin". I have changed it to
2095 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
2096 anything to be listed in the output from -bV.
2098 PH/05 When a filter generated an autoreply, the entire To: header line was
2099 quoted in the delivery log line, like this:
2101 => >A.N.Other <ano@some.domain> <original@ddress> ...
2103 This has been changed so that it extracts the operative address. There
2104 may be more than one such address. If so, they are comma-separated, like
2107 => >ano@some.domain,ona@other.domain <original@ddress> ...
2109 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
2110 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
2111 not being looked up in the rDNS to get a host name, Exim was showing the
2112 IP address twice in Received: lines, even though the IP addresses were
2113 identical. For example:
2115 Received: from [1.2.3.4] (helo=[1.2.3.4])
2117 However, if the real host name was known, it was omitting the HELO data
2118 if it matched the actual IP address. This has been tidied up so that it
2119 doesn't show the same IP address twice.
2121 PH/07 When both +timestamp and +memory debugging was on, the value given by
2122 $tod_xxx expansions could be wrong, because the tod_stamp() function was
2123 called by the debug printing, thereby overwriting the timestamp buffer.
2124 Debugging no longer uses the tod_stamp() function when +timestamp is set.
2126 PH/08 When the original message was included in an autoreply transport, it
2127 always said "this is a copy of the message, including all the headers",
2128 even if body_only or headers_only was set. It now gives an appropriate
2131 PH/09 Applied a patch from the Sieve maintainer which:
2133 o fixes some comments
2134 o adds the (disabled) notify extension core
2135 o adds some debug output for the result of if/elsif tests
2136 o points to the current vacation draft in the documentation
2137 and documents the missing references header update
2141 o fixes a bug in processing the envelope test (when testing
2142 multiple envelope elements, the last element determinted the
2145 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
2146 Electronic Mail") by including:
2148 Auto-submitted: auto-generated
2150 in the messages that it generates (bounce messages and others, such as
2151 warnings). In the case of bounce messages for non-SMTP mesages, there was
2152 also a typo: it was using "Auto_submitted" (underscore instead of
2153 hyphen). Since every message generated by Exim is necessarily in response
2154 to another message, thes have all been changed to:
2156 Auto-Submitted: auto-replied
2158 in accordance with these statements in the RFC:
2160 The auto-replied keyword:
2162 - SHOULD be used on messages sent in direct response to another
2163 message by an automatic process,
2165 - MUST NOT be used on manually-generated messages,
2167 - MAY be used on Delivery Status Notifications (DSNs) and Message
2168 Disposition Notifications (MDNs),
2170 - MUST NOT be used on messages generated by automatic or periodic
2171 processes, except for messages which are automatic responses to
2174 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
2175 to the default Received: header definition.
2177 PH/12 Added log selector acl_warn_skipped (default on).
2179 PH/13 After a successful wildlsearch lookup, discard the values of numeric
2180 variables because (a) they are in the wrong storage pool and (b) even if
2181 they were copied, it wouldn't work properly because of the caching.
2183 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
2184 checking when decoding. Apparently there are clients that generate
2185 overlong encoded strings. Why am I not surprised?
2187 PH/15 If the first argument of "${if match_address" was not empty, but did not
2188 contain an "@" character, Exim crashed. Now it writes a panic log message
2189 and treats the condition as false.
2191 PH/16 In autoreply, treat an empty string for "once" the same as unset.
2193 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
2194 extension "envelope-auth". The code is finished and in agreement with
2195 other implementations, but there is no documentation so far and in fact,
2196 nobody wrote the draft yet. This extension is currently #undef'ed, thus
2197 not changing the active code.
2199 Print executed "if" and "elsif" statements when debugging is used. This
2200 helps a great deal to understand what a filter does.
2202 Document more things not specified clearly in RFC3028. I had all this
2203 sorted out, when out of a sudden new issues came to my mind. Oops."
2205 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
2208 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
2209 canonical form (as documented). However, after a host name lookup from
2210 the IP address, check_host() was doing a simple string comparison with
2211 addresses acquired from the DNS when checking that the found name did
2212 have the original IP as one of its addresses. Since any found IPv6
2213 addresses are likely to be in abbreviated form, the comparison could
2214 fail. Luckily, there already exists a function for doing the comparison
2215 by converting both addresses to binary, so now that is used instead of
2216 the text comparison.
2218 PH/20 There was another similar case to PH/19, when a complete host name was
2219 given in a host list; looking up its IP address could give an abbreviated
2220 form, whereas the current host's name might or might not be abbreviated.
2221 The same fix has been applied.
2227 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
2228 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
2231 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
2232 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
2234 PH/03 Typo: missing ".o" in src/pcre/Makefile.
2236 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
2237 header line, restrict the check to what is listed in RFCs 2369 and 2929.
2238 Also, for "Auto-Submitted", treat anything other than "no" as
2239 non-personal, in accordance with RFC 3834. (Previously it treated
2240 anything starting "auto-" as non-personal.)
2242 TF/01 The control=submission/name=... option had a problem with syntax
2243 errors if the name included a slash character. The /name= option
2244 now slurps the rest of the string, so it can include any characters
2245 but it must come last in the list of options (after /sender_retain
2248 PH/05 Some modifications to the interface to the fake nameserver for the new
2256 TK/01 Added the "success_on_redirect" address verification option. See
2257 NewStuff for rationale and an example.
2259 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
2261 PH/02 Patch to exigrep to allow it to work on syslog lines.
2263 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
2264 fread() to skip over the body file's header line, because in Cygwin the
2265 header line is locked and is inaccessible.
2267 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
2268 co-exist for some time) to make it clear that it is the Exim ID that is
2269 referenced, not the Message-ID: header line.
2271 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
2272 string_format() function, because snprintf() does not exist on all
2275 PH/06 The use of forbid_filter_existstest now also locks out the use of the
2276 ${stat: expansion item.
2278 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
2279 protocol synchronization error", to keep the pedants happy.
2281 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
2282 well as for IRIX systems, when gcc is being used. See the host.c source
2285 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
2287 PH/10 Named domain lists were not working if used in a queue_smtp_domains
2290 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
2291 transport and to the smtp transport in LMTP mode.
2293 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
2295 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
2296 run a filter in a subprocess. This could lead to confusion in subsequent
2297 lookups in the parent process. There should also be a search_tidyup() at
2298 the end of the subprocess.
2300 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
2301 only if the host matched helo_try_verify_hosts, which caused the
2302 verification to occur when the EHLO/HELO command was issued. The ACL just
2303 tested the remembered result. Now, if a previous verification attempt has
2304 not happened, "verify = helo" does it there and then.
2306 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
2308 TK/03 Fix log output including CR from clamd.
2310 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
2311 exist provoked a memory error which could cause a segfault.
2313 PH/15 Installed PCRE 6.2
2315 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2317 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2318 of the problem. Specifically, suggested +O2 rather than +O1 for the
2321 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2323 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2324 record, the port was not correctly logged when the outgoing_port log
2325 selector was set (it logged the transort's default port).
2327 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2328 fallback_hosts, and "hosts" in the smtp transport.
2330 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2331 host errors such as "Connection refused".
2333 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2334 authentication with radiusclient 0.4.9:
2336 - Error returned from rc_read_config was caught wrongly
2337 - Username/password not passed on to radius server due to wrong length.
2339 The presumption is that some radiusclient API changes for 4.51/PH/17
2340 were not taken care of correctly. The code is still untested by me (my
2341 Linux distribution still has 0.3.2 of radiusclient), but it was
2342 contributed by a Radius user.
2344 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2345 expanding the "port" option of the smtp transport.
2347 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2348 while reading a MIME header. Thanks to Tom Hughes for a patch.
2350 PH/24 Include config.h inside local_scan.h so that configuration settings are
2353 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2354 that in an ACL for RCPT (for example), you can examine exactly what was
2357 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2358 commands, but it was not correctly comparing the address with the actual
2359 client host address. Thus, it would show the EHLO address in Received:
2360 header lines when this was not necessary.
2362 PH/27 Added the % operator to ${eval:}.
2364 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2365 it should be ignoring failures (because with -C, for example, it has lost
2366 privilege). It wasn't ignoring creation failures other than "already
2369 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2372 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2373 because an input line was too long, either on its own, or by virtue of
2374 too many continuations, the temporary file was not being removed, and the
2375 return code was incorrect.
2377 PH/31 Missing "BOOL" in function definition in filtertest.c.
2379 PH/32 Applied Sieve patches from the maintainer.
2381 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2383 PH/33 Added "verify = not_blind".
2385 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2386 Local/Makefile (with some defaults set). These are used in built scripts
2387 such as exicyclog, but they have never been used in the exim_install
2388 script (though there are many overriding facilities there). I have
2389 arranged that the exim_install script now takes note of these two
2392 PH/35 Installed configuration files for Dragonfly.
2394 PH/36 When a locally submitted message by a trusted user did not contain a
2395 From: header, and the sender address was obtained from -f or from an SMTP
2396 MAIL command, and the trusted user did not use -F to supply a sender
2397 name, $originator_name was incorrectly used when constructing a From:
2398 header. Furthermore, $originator_name was used for submission mode
2399 messages from external hosts without From: headers in a similar way,
2400 which is clearly wrong.
2402 PH/37 Added control=suppress_local_fixups.
2404 PH/38 When log_selector = +received_sender was set, and the addition of the
2405 sender made the log line's construction buffer exactly full, or one byte
2406 less than full, an overflow happened when the terminating "\n" was
2409 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2410 when the result of a list match is failure because a DNS lookup failed.
2412 PH/40 RM_COMMAND is now used in the building process.
2414 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2415 the "build-* directories that it finds.
2417 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2418 address in a domain literal was a prefix of an interface address.
2420 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2421 when verifying a sender address, unless rewrite_headers is false.
2423 PH/44 Wrote a long comment about why errors_to addresses are verified as
2424 recipients, not senders.
2426 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2427 the ratelimit ACL was added.
2429 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2431 PH/46 Added extra information about PostgreSQL errors to the error string.
2433 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2434 avoiding the need to install special zones in a real server. This is
2435 backwards compatible; if it can't find the fake resolver, it drops back.
2436 Thus, both old and new test suites can be run.
2438 TF/02 Added util/ratelimit.pl
2440 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2441 clock is set back in time.
2443 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2444 Candler <B.Candler@pobox.com>.
2446 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2447 OK for addresses that are the result of redirections.
2449 PH/48 A number of further additions for the benefit of the new test suite,
2450 including a fake gethostbyname() that interfaces to the fake DNS resolver
2453 TF/06 The fix for widen_domains has also been applied to qualify_single and
2454 search_parents which are the other dnslookup options that can cause
2457 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2460 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2462 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2463 service" if it encountered an MX record with an empty target hostname.
2464 The message is now "an MX or SRV record indicated no SMTP service".
2466 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2467 if the DNS of the sending site is misconfigured. This is quite a
2468 common situation. This change restores the behaviour of treating a
2469 helo verification defer as a failure.
2471 PH/51 If self=fail was set on a router, the bounce message did not include the
2472 actual error message.
2478 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2480 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2481 command itself ended in error, the underlying message about the transport
2482 filter timeout was being overwritten with the pipe command error. Now the
2483 underlying error message should be appended to the second error message.
2485 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2487 PH/02 Reduce the amount of output that "make" produces by default. Full output
2488 can still be requested.
2490 PH/03 The warning log line about a condition test deferring for a "warn" verb
2491 was being output only once per connection, rather than after each
2492 occurrence (because it was using the same function as for successful
2493 "warn" verbs). This seems wrong, so I have changed it.
2495 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2496 it should not have, which might have caused a crash in the right
2497 circumstances, but probably never did.
2499 PH/04 Installed a modified version of Tony Finch's patch to make submission
2500 mode fix the return path as well as the Sender: header line, and to
2501 add a /name= option so that you can make the user's friendly name appear
2504 TF/03 Added the control = fakedefer ACL modifier.
2506 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2507 Mark Lowes for thorough testing.
2509 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2511 TK/03 Merged latest SRS patch from Miles Wilton.
2513 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2514 with the definition in sysexits.h (which is #included earlier).
2515 Fortunately, Exim does not actually use EX_OK. The code used to try to
2516 preserve the sysexits.h value, by assumimg that macro definitions were
2517 scanned for macro replacements. I have been disabused of this notion,
2518 so now the code just undefines EX_OK before #including unistd.h.
2520 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2521 in the smtp transport. When a block could not be written in a single
2522 write() function, the timeout was being re-applied to each part-write.
2523 This seems wrong - if the receiver was accepting one byte at a time it
2524 would take for ever. The timeout is now adjusted when this happens. It
2525 doesn't have to be particularly precise.
2527 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2528 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2530 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2531 without a domain if the check to <postmaster@domain> fails.
2533 SC/01 Eximstats: added -xls and the ability to specify output files
2534 (patch written by Frank Heydlauf).
2536 SC/02 Eximstats: use FileHandles for outputing results.
2538 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2540 SC/04 Eximstats: fixed display of large numbers with -nvr option
2542 SC/05 Eximstats: fixed merging of reports with empty tables.
2544 SC/06 Eximstats: added the -include_original_destination flag
2546 SC/07 Eximstats: removed tabs and trailing whitespace.
2548 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2550 TK/06 MBOX spool code: Add real "From " MBOX separator line
2551 so the .eml file is really in mbox format (even though
2552 most programs do not really care). Patch from Alex Miller.
2554 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2555 The latter is generated from $received_to and is only set if the
2556 message has one envelope recipient. SA can use these headers,
2557 obviously out-of-the-box. Patch from Alex Miller.
2559 PH/08 The ${def test on a variable was returning false if the variable's
2560 value was "0", contrary to what the specification has always said!
2561 The result should be true unless the variable is empty.
2563 PH/09 The syntax error of a character other than { following "${if
2564 def:variable_name" (after optional whitespace) was not being diagnosed.
2565 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2566 accidental colon was present, for example, could give incorrect results.
2568 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2569 result is used (not including appendfile, where other changes are about
2572 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2573 This involved changing a lot of size variables from int to off_t. It
2574 should work with maildirs and everything.
2576 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2577 spamd dying while we are connected to it.
2579 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2582 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2583 function for BDB 4.3.
2585 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2587 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2588 the sources for PCRE have been re-arranged and more files are now
2591 PH/15 The code I had for printing potentially long long variables in PH/11
2592 above was not the best (it lost precision). The length of off_t variables
2593 is now inspected at build time, and an appropriate printing format (%ld
2594 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2595 to be "long long int" or "long int". This is needed for the internal
2596 formatting function string_vformat().
2598 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2599 the configuration file to be ":syslog", then the script "guesses" where
2600 the logs files are, rather than using the compiled in default. In our
2601 case the guess is not the same as the compiled default, so the script
2602 suddenly stopped working when I started to use syslog. The patch checks
2603 to see if log_file_path is "". If so, it attempts to read it from exim
2604 with no configuration file to get the compiled in version, before it
2605 falls back to the previous guessing code."
2607 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2608 implementing BATV in an Exim configuration. See NewStuff for the gory
2611 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2612 Makefile that are specific to HP-UX.
2614 PH/18 If the "use_postmaster" option was set for a recipient callout together
2615 with the "random" option, the postmaster address was used as the MAIL
2616 FROM address for the random test, but not for the subsequent recipient
2617 test. It is now used for both.
2619 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2620 patch removes a few documentation additions to RFC 3028, because the
2621 latest draft now contains them. It adds the new en;ascii-case comparator
2622 and a new error check for 8bit text in MIME parts. Comparator and
2623 require names are now matched exactly. I enabled the subaddress
2624 extension, but it is not well tested yet (read: it works for me)."
2626 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2627 rework some of the code of TK/09 above to avoid the hardwired use of
2628 "%lld" and "long long". Replaced the call to snprintf() with a call to
2631 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2632 records point to non-existent hosts", "retry timeout exceeded", and
2633 "retry time not reached for any host after a long failure period".
2635 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2636 experimental DomainKeys support:
2638 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2639 (2) On an error such as an illegally used "control", the wrong name for
2640 the control was given.
2642 These problems did NOT occur unless DomainKeys support was compiled.
2644 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2646 PH/24 Added ${if match_ip condition.
2648 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2649 databases so that it will be absolutely obvious if a crash occurs in the
2650 DB library. This is a regular occurrence (often caused by mis-matched
2653 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2654 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2655 picked up on a user's system that detects such things. There doesn't seem
2656 to be a gcc warning option for this - only an attribute that has to be
2657 put on the function's prototype. It seems that in Fedora Core 4 they have
2658 set this on a number of new functions. No doubt there will be more in due
2661 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2662 specify a transport. However, if an address that was verified by such a
2663 router was the subject of a callout, Exim crashed because it tried to
2664 read the rcpt_include_affixes from the non-existent transport. Now it
2665 just assumes that the setting of that option is false. This bug was
2666 introduced by 4.51/PH/31.
2668 PH/28 Changed -d+all to exclude +memory, because that information is very
2669 rarely of interest, but it makes the output a lot bigger. People tend to
2670 do -d+all out of habit.
2672 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2673 code in os-type was giving problems when libc.so lives in lib64, like on
2676 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2677 aren't the modern standard, and it seems that some systems' include files
2678 don't always have them. Exim was already checking for some of the newer
2679 ones like T_AAAA, and defining it itself. I've added checks for all the
2680 record types that Exim uses.
2682 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2683 not automatically generating a new one, as it is supposed to. This
2684 prevented TLS from working. If the file did exist, but contained invalid
2685 data, a new version was generated, as expected. It was only the case of a
2686 non-existent file that was broken.
2688 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2689 with a change in libdomainkeys > 0.64.
2691 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2692 from DNS. If the selector record carries the flag, it now has
2693 precedence over the domain-wide flag.
2695 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2697 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2698 the use of a port name that isn't defined in /etc/services) occurred, the
2699 message was deferred as in a normal delivery, and thus remained on the
2700 spool, instead of being failed because of the mua_wrapper setting. This
2701 is now fixed, and I tidied up some of the mua_wrapper messages at the
2704 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2705 the messages in a hash of arrays rather than using individual hashes.
2706 This is a bit cleaner and results in dramatic memory savings, albeit
2707 at a slight CPU cost.
2709 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2710 as requested by Marc Sherman.
2712 SC/10 Eximstats: added histograms for user specified patterns as requested
2715 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2717 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2718 fopen() in the content-scanning modules that did not already have it.
2724 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2725 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2727 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2729 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2731 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2732 to test Sieve filters that use "vacation".
2734 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2735 that changes the way the GnuTLS parameters are stored in the cache file.
2736 The new format can be generated externally. For backward compatibility,
2737 if the data in the cache doesn't make sense, Exim assumes it has read an
2738 old-format file, and it generates new data and writes a new file. This
2739 means that you can't go back to an older release without removing the
2742 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2743 work if there are any delivery delays because "one_time" forces the
2744 parent to be marked "delivered", so its unseen clone is never tried
2745 again. For this reason, Exim now forbids the simultaneous setting of
2748 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2749 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2750 entry if you want to know the details. The fix, however, neglected to
2751 consider the case when local delivery batching is involved. The test for
2752 "previously delivered" was not happening when checking to see if an
2753 address could be batched with a previous (undelivered) one; under
2754 certain circumstances this could lead to multiple deliveries to the same
2757 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2758 in its include files, and this causes problems building Exim.
2760 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2761 header_syntax) but Exim was just ignoring anything given after a slash.
2762 In particular, this caused confusion with an attempt to use "verify =
2763 reverse_host_lookup/defer_ok". An error is now given when options are
2764 supplied for verify items that do not have them. (Maybe reverse_host_
2765 lookup should have a defer_ok option, but that's a different point.)
2767 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2768 defined by RFC 821) to 2048, because there were problems with some AUTH
2769 commands, and RFC 1869 says the size should be increased for extended
2770 SMTP commands that take arguments.
2772 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2775 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2776 "unknown" error; now it says that the functionality isn't in the binary.
2778 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2779 an address' error message when a string expansion fails (syntax or
2780 whatever). Otherwise the password may appear in the log. Following change
2781 PH/42 below, there is no longer a chance of it appearing in a bounce
2784 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2786 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2787 of its addresses was ever tried. (Bugzilla bug #2.)
2789 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2790 the result incorrectly in the debug output. (It correctly added a newline
2791 to what was transported.)
2793 TF/01 Added $received_time.
2795 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2796 commented out examples of how to interface to a virus scanner and to
2797 SpamAssassin. Also added commented examples of av_scanner and
2798 spamd_address settings.
2800 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2801 and controls are allowed in which ACLs. There were a couple of minor
2802 errors. Some of the entries in the conditions table (which is a table of
2803 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2804 negation of where the condition IS allowed.
2806 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2808 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2809 header file does not have a version number, so I've had to invent a new
2810 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2811 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2812 radiusclient), but it was contributed by a Radius user.
2814 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2815 files or directories, for OpenSSL.
2817 PH/19 When an Exim process that is running as root has to create an Exim log
2818 file, it does so in a subprocess that runs as exim:exim so as to get the
2819 ownership right at creation (otherwise, other Exim processes might see
2820 the file with the wrong ownership). There was no test for failure of this
2821 fork() call, which would lead to the process getting stuck as it waited
2822 for a non-existent subprocess. Forks do occasionally fail when resources
2823 run out. I reviewed all the other calls to fork(); they all seem to check
2826 PH/20 When checking for unexpected SMTP input at connect time (before writing
2827 the banner), Exim was not dealing correctly with a non-positive return
2828 from the read() function. If the client had disconnected by this time,
2829 the result was a log entry for a synchronization error with an empty
2830 string after "input=" when read() returned zero. If read() returned -1
2831 (an event I could not check), uninitialized data bytes were printed.
2832 There were reports of junk text (parts of files, etc) appearing after
2835 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2837 PH/22 Added support for macro redefinition, and (re)definition in between
2838 driver and ACL definitions.
2840 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2841 forgetting to use the resulting value; it was using the unexpanded value.
2843 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2844 hadn't been configured. The fix is from Juergen Kreileder, who
2845 understands it better than I do:
2847 "Here's what I see happening with three configured cyrus_sasl
2848 authenticators configured (plain, login, cram-md5):
2850 On startup auth_cyrus_sasl_init() gets called for each of these.
2851 This means three calls to sasl_listmech() without a specified mech_list.
2852 => SASL tests which mechs of all available mechs actually work
2853 => three warnings about OTP not working
2854 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2856 With the patch, sasl_listmech() also gets called three times. But now
2857 SASL's mech_list option is set to the server_mech specified in the the
2858 authenticator. Or in other words, the answer from sasl_listmech()
2859 gets limited to just the mech you're testing for (which is different
2861 => the return list contains just 'plain' or 'login', 'cram-md5' or
2862 nothing depending on the value of ob->server_mech.
2864 I've just tested the patch: Authentication still works fine,
2865 unavailable mechs specified in the exim configuration are still
2866 caught, and the auth.log warnings about OTP are gone."
2868 PH/25 When debugging is enabled, the contents of the command line are added
2869 to the debugging output, even when log_selector=+arguments is not
2872 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2873 answer is "GNU", and only if the return is "GNU/something" is the answer
2876 PH/27 $acl_verify_message is now set immediately after the failure of a
2877 verification in an ACL, and so is available in subsequent modifiers. In
2878 particular, the message can be preserved by coding like this:
2880 warn !verify = sender
2881 set acl_m0 = $acl_verify_message
2883 Previously, $acl_verify_message was set only while expanding "message"
2884 and "log_message" when a very denied access.
2886 PH/28 Modified OS/os.c-Linux with
2888 -#ifndef OS_LOAD_AVERAGE
2889 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2891 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2892 nomenclature these days.)
2894 PH/29 Installed patch from the Sieve maintainer that adds the options
2895 sieve_useraddress and sieve_subaddress to the redirect router.
2897 PH/30 In these circumstances:
2898 . Two addresses routed to the same list of hosts;
2899 . First host does not offer TLS;
2900 . First host accepts first address;
2901 . First host gives temporary error to second address;
2902 . Second host offers TLS and a TLS session is established;
2903 . Second host accepts second address.
2904 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2905 and peerdn, if requested) that were in fact used only for the second
2908 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2909 attention to any local part prefix or suffix that was matched by the
2910 router that accepted the address. It now behaves in the same way as it
2911 does for delivery: the affixes are removed from the local part unless
2912 rcpt_include_affixes is set on the transport.
2914 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2915 timeout during the DATA phase of an incoming message.
2917 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2918 applied a patch sanctioned by the Sieve maintainer.
2920 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2921 the uid or gid is negative. A case of a negative gid caused this to be
2922 noticed. The fix allows for either to be negative.
2924 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2925 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2928 PH/36 The change PH/12 above was broken. Fixed it.
2930 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2931 the grounds that routing the same address twice would always produce the
2932 same answer. This might have been true once, but it is certainly no
2933 longer true now. Routing a child address may depend on the previous
2934 routing that produced that child. Some complicated redirection strategies
2935 went wrong when messages had multiple recipients, and made Exim's
2936 behaviour dependent on the order in which the addresses were given.
2938 I have moved the duplicate checking until after the routing is complete.
2939 Exim scans the addresses that are assigned to local and remote
2940 transports, and removes any duplicates. This means that more work will be
2941 done, as duplicates will always all be routed, but duplicates are
2942 presumably rare, so I don't expect this is of any significance.
2944 For deliveries to pipes, files, and autoreplies, the duplicate checking
2945 still happens during the routing process, since they are not going to be
2948 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2949 It corrects a timeout issue with spamd. This is Ian's comment: "The
2950 background is that sometimes spamd either never reads data from a
2951 connection it has accepted, or it never writes response data. The exiscan
2952 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2953 blindly assumes that writes won't block so it may never time out."
2955 PH/39 Allow G after quota size as well as K and M.
2957 PH/40 The value set for $authenticated_id in an authenticator may not contain
2958 binary zeroes or newlines because the value is written to log lines and
2959 to spool files. There was no check on this. Now the value is run through
2960 the string_printing() function so that such characters are converted to
2961 printable escape sequences.
2963 PH/41 $message_linecount is a new variable that contains the total number of
2964 lines in the message. Compare $body_linecount, which is the count for the
2967 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2968 bounce and delay warning messages, except in certain special cases, which
2971 (a) An SMTP error message from a remote host;
2972 (b) A message specified in a :fail: redirection;
2973 (c) A message specified in a "fail" command in a system filter;
2974 (d) A message specified in a FAIL return from the queryprogram router;
2975 (e) A message specified by the cannot_route_message router option.
2977 In these cases only, Exim does include the error details in bounce and
2978 warning messages. There are also a few cases where bland messages such
2979 as "unrouteable address" or "local delivery error" are given.
2981 PH/43 $value is now also set for the "else" part of a ${run expansion.
2983 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2984 being worked on, but at least Exim now implements the latest version to
2987 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2988 process to complete was treated as a delivery failure, a timeout while
2989 writing the message to the pipe was logged, but erroneously treated as a
2990 successful delivery. Such timeouts include transport filter timeouts. For
2991 consistency with the overall process timeout, these timeouts are now
2992 treated as errors, giving rise to delivery failures by default. However,
2993 there is now a new Boolean option for the pipe transport called
2994 timeout_defer, which, if set TRUE, converts the failures into defers for
2995 both kinds of timeout. A transport filter timeout is now identified in
2998 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2999 systems where "make" and "gmake" are different, calling "gmake" at top
3000 level broke things. I've arranged for the value of $(MAKE) to be passed
3001 from the Makefile to this script so that it can call the same version of
3005 A note about Exim versions 4.44 and 4.50
3006 ----------------------------------------
3008 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3009 changes of various kinds. As a consequence, a big documentation update was
3010 needed. This delayed the release for rather longer than seemed good, especially
3011 in the light of a couple of (minor) security issues. Therefore, the changes
3012 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
3013 release. So 4.44 and 4.50 are in effect two different branches that both start
3016 I have left the 4.50 change log unchanged; it contains all the changes since
3017 4.43. The change log for 4.44 is below; many of its items are identical to
3018 those for 4.50. This seems to be the most sensible way to preserve the
3019 historical information.
3025 1. Minor wording change to the doc/README.SIEVE file.
3027 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
3028 computation of the current number of files was incorrect.
3030 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3031 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3034 4. Give more explanation in the error message when the command for a transport
3035 filter fails to execute.
3037 5. There are several places where Exim runs a non-Exim command in a
3038 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3039 was being done only for the command run by the queryprogram router. It is
3040 now done for all such subprocesses. The other cases are: ${run, transport
3041 filters, and the commands run by the lmtp and pipe transports.
3043 6. Added CONFIGURE_GROUP build-time option.
3045 7. Some older OS have a limit of 256 on the maximum number of file
3046 descriptors. Exim was using setrlimit() to set 1000 as a large value
3047 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3048 systems. I've change it so that if it can't get 1000, it tries for 256.
3050 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3051 was an oversight, and furthermore, ever since the addition of extra
3052 controls (e.g. 4.43/32), the checks on when to allow different forms of
3053 "control" were broken. There should now be diagnostics for all cases when a
3054 control that does not make sense is encountered.
3056 9. Added the /retain_sender option to "control=submission".
3058 10. $recipients is now available in the predata ACL (oversight).
3060 11. Tidy the search cache before the fork to do a delivery from a message
3061 received from the command line. Otherwise the child will trigger a lookup
3062 failure and thereby defer the delivery if it tries to use (for example) a
3063 cached ldap connection that the parent has called unbind on.
3065 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3066 of $address_data from the recipient verification was clobbered by the
3067 sender verification.
3069 13. The value of address_data from a sender verification is now available in
3070 $sender_address_data in subsequent conditions in the ACL statement.
3072 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
3074 15. Added a new option "connect=<time>" to callout options, to set a different
3077 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3078 was its contents. (It was OK if the option was not defined at all.)
3080 17. A "Completed" log line is now written for messages that are removed from
3081 the spool by the -Mrm option.
3083 18. New variables $sender_verify_failure and $recipient_verify_failure contain
3084 information about exactly what failed.
3086 19. Added -dd to debug only the daemon process.
3088 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
3089 handles timeouts, both on the server side and network timeouts. Renamed the
3090 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
3092 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
3093 It is now set to "smtps".
3095 22. $host_address is now set to the target address during the checking of
3096 ignore_target_hosts.
3098 23. When checking ignore_target_hosts for an ipliteral router, no host name was
3099 being passed; this would have caused $sender_host_name to have been used if
3100 matching the list had actually called for a host name (not very likely,
3101 since this list is usually IP addresses). A host name is now passed as
3104 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3105 code that specifies a non-restarting handler (typically sigaction() in
3106 modern systems) in an attempt to fix a rare and obscure crash bug.
3108 25. Narrowed the window for a race in the daemon that could cause it to ignore
3109 SIGCHLD signals. This is not a major problem, because they are used only to
3110 wake it up if nothing else does.
3112 26. A malformed maildirsize file could cause Exim to calculate negative values
3113 for the mailbox size or file count. Odd effects could occur as a result.
3114 The maildirsize information is now recalculated if the size or filecount
3117 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3118 support for a long time. Removed HAVE_SYS_VFS_H.
3120 28. Installed the latest version of exipick from John Jetmore.
3122 29. In an address list, if the pattern was not a regular expression, an empty
3123 subject address (from a bounce message) matched only if the pattern was an
3124 empty string. Non-empty patterns were not even tested. This was the wrong
3125 because it is perfectly reasonable to use an empty address as part of a
3126 database query. An empty address is now tested by patterns that are
3127 lookups. However, all the other forms of pattern expect the subject to
3128 contain a local part and a domain, and therefore, for them, an empty
3129 address still always fails if the pattern is not itself empty.
3131 30. Exim went into a mad DNS loop when attempting to do a callout where the
3132 host was specified on an smtp transport, and looking it up yielded more
3133 than one IP address.
3135 31. Re-factored the code for checking spool and log partition space into a
3136 function that finds that data and another that does the check. The former
3137 is then used to implement four new variables: $spool_space, $log_space,
3138 $spool_inodes, and $log_inodes.
3140 32. The RFC2047 encoding function was originally intended for short strings
3141 such as real names; it was not keeping to the 75-character limit for
3142 encoded words that the RFC imposes. It now respects the limit, and
3143 generates multiple encoded words if necessary. To be on the safe side, I
3144 have increased the buffer size for the ${rfc2047: expansion operator from
3147 33. It is now permitted to omit both strings after an "if" condition; if the
3148 condition is true, the result is "true". As before, when the second string
3149 is omitted, a false condition yields an empty string. This makes it less
3150 cumbersome to write custom ACL and router conditions.
3152 34. Failure to deliver a bounce message always caused it to be frozen, even if
3153 there was an errors_to setting on the router. The errors_to setting is now
3156 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3157 canonical form (fully expanded) before being placed in
3158 $sender_host_address.
3160 36. The table in the code that translates DNS record types into text (T_A to
3161 "A" for instance) was missing entries for NS and CNAME. It is just possible
3162 that this could have caused confusion if both these types were looked up
3163 for the same domain, because the text type is used as part of Exim's
3164 per-process caching. But the chance of anyone hitting this buglet seems
3167 37. The dnsdb lookup has been extended in a number of ways.
3169 (1) There is a new type, "zns", which walks up the domain tree until it
3170 finds some nameserver records. It should be used with care.
3172 (2) There is a new type, "mxh", which is like "mx" except that it returns
3173 just the host names, not the priorities.
3175 (3) It is now possible to give a list of domains (or IP addresses) to be
3176 looked up. The behaviour when one of the lookups defers can be
3177 controlled by a keyword.
3179 (4) It is now possible to specify the separator character for use when
3180 multiple records are returned.
3182 38. The dnslists ACL condition has been extended: it is now possible to supply
3183 a list of IP addresses and/or domains to be looked up in a particular DNS
3186 39. Added log_selector=+queue_time_overall.
3188 40. When running the queue in the test harness, wait just a tad after forking a
3189 delivery process, to get repeatability of debugging output.
3191 41. Include certificate and key file names in error message when GnuTLS fails
3192 to set them up, because the GnuTLS error message doesn't include the name
3193 of the failing file when there is a problem reading it.
3195 42. Allow both -bf and -bF in the same test run.
3197 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
3199 44. The "Exiscan patch" is now merged into the mainline Exim source.
3201 45. Sometimes the final signoff response after QUIT could fail to get
3202 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3203 < 0 before doing a fflush(). This bug looks as though it goes back to the
3204 introduction of TLS in release 3.20, but "sometimes" must have been rare
3205 because the tests only now provoked it.
3207 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
3208 (this can affect the format of dates).
3210 47. exim_tidydb, when checking for the continued existence of a message for
3211 which it has found a message-specific retry record, was not finding
3212 messages that were in split spool directories. Consequently, it was
3213 deleting retry records that should have stayed in existence.
3215 48. Steve fixed some bugs in eximstats.
3217 49. The SPA authentication driver was not abandoning authentication and moving
3218 on to the next authenticator when an expansion was forced to fail,
3219 contradicting the general specification for all authenticators. Instead it
3220 was generating a temporary error. It now behaves as specified.
3222 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
3223 (the order specifies the preference for clients). The order is now AES256,
3224 AES128, 3DES, ARCFOUR128.
3226 51. Small patch to Sieve code - explicitly set From: when generating an
3229 52. Exim crashed if a remote delivery caused a very long error message to be
3230 recorded - for instance if somebody sent an entire SpamAssassin report back
3231 as a large number of 550 error lines. This bug was coincidentally fixed by
3232 increasing the size of one of Exim's internal buffers (big_buffer) that
3233 happened as part of the Exiscan merge. However, to be on the safe side, I
3234 have made the code more robust (and fixed the comments that describe what
3237 53. Now that there can be additional text after "Completed" in log lines (if
3238 the queue_time_overall log selector is set), a one-byte patch to exigrep
3239 was needed to allow it to recognize "Completed" as not the last thing in
3242 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
3243 patch that reportedly fixes this has been added. I am not expert enough to
3244 create a test for it. This is what the patch creator wrote:
3246 "I found a little strange behaviour of ldap code when working with
3247 Windows 2003 AD Domain, where users was placed in more than one
3248 Organization Units. When I tried to give exim partial DN, the exit code
3249 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
3250 But simultaneously result of request was absolutely normal ldap result,
3251 so I produce this patch..."
3253 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
3254 REFERENCE, so I have modified the code to exclude the patch when that macro
3257 55. Some experimental protocols are using DNS PTR records for new purposes. The
3258 keys for these records are domain names, not reversed IP addresses. The
3259 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3260 leaves it alone. Component reversal etc. now happens only for IP addresses.
3263 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
3265 57. Double the size of the debug message buffer (to 2048) so that more of very
3266 long debug lines gets shown.
3268 58. The exicyclog utility now does better if the number of log files to keep
3269 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
3271 59. Two changes related to the smtp_active_hostname option:
3273 (1) $smtp_active_hostname is now available as a variable.
3274 (2) The default for smtp_banner uses $smtp_active_hostname instead
3275 of $primary_hostname.
3277 60. The host_aton() function is supposed to be passed a string that is known
3278 to be a valid IP address. However, in the case of IPv6 addresses, it was
3279 not checking this. This is a hostage to fortune. Exim now panics and dies
3280 if the condition is not met. A case was found where this could be provoked
3281 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3282 components; fortuitously, this particular loophole had already been fixed
3283 by change 4.50/55 above.
3285 If there are any other similar loopholes, the new check in host_aton()
3286 itself should stop them being exploited. The report I received stated that
3287 data on the command line could provoke the exploit when Exim was running as
3288 exim, but did not say which command line option was involved. All I could
3289 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3290 running as the user.
3293 61. There was a buffer overflow vulnerability in the SPA authentication code
3294 (which came originally from the Samba project). I have added a test to the
3295 spa_base64_to_bits() function which I hope fixes it.
3298 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
3299 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
3301 63. The daemon start-up calls getloadavg() while still root for those OS that
3302 need the first call to be done as root, but it missed one case: when
3303 deliver_queue_load_max is set with deliver_drop_privilege. This is
3304 necessary for the benefit of the queue runner, because there is no re-exec
3305 when deliver_drop_privilege is set.
3307 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
3308 This has been fixed.
3310 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3311 was in use, was not putting the data itself into the right store pool;
3312 consequently, it could be overwritten for a subsequent message in the same
3313 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3316 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3318 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3319 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3320 the function were treating the return as a boolean value, which happened to
3321 work because 0=false and not-0=true, but is not correct code.
3323 68. The host_aton() function was not handling scoped IPv6 addresses (those
3324 with, for example, "%eth0" on the end) correctly.
3326 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3327 negated items (that is, ~something) in unsigned ints. Some compilers
3328 apparently mutter when there is no cast.
3330 70. If an address verification called from an ACL failed, and did not produce a
3331 user-specific message (i.e. there was only a "system" message), nothing was
3332 put in $acl_verify_message. In this situation, it now puts the system
3335 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3336 session; change 4.31/43 added the unwanted input to the log line - except
3337 that it did not do this in the start of session case. It now does.
3339 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3340 This is wrong and can cause the other end to generate a synchronization
3341 error if it is another Exim or anything else that does the synchronization
3342 check. A QUIT command is no longer sent after a timeout.
3344 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3345 during host lookups.
3347 74. The defer_ok option of callout verification was not working if it was used
3348 when verifying addresses in header lines, that is, for this case:
3350 verify = header_sender/callout=defer_ok
3352 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3353 those file descriptors could be used for SMTP connections. If anything
3354 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3355 could be sent to the SMTP client, causing chaos. The daemon now opens
3356 stdin, stdout, and stderr to /dev/null when it puts itself into the
3359 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3360 log by default. The user can override this with suitable Perl magic.
3362 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3363 the log message when discard triggers, was not working for the DATA ACL or
3364 for the non-SMTP ACL.
3366 78. Error message wording change in sieve.c.
3368 79. If smtp_accept_max_per_host was set, the number of connections could be
3369 restricted to fewer than expected, because the daemon was trying to set up
3370 a new connection before checking whether the processes handling previous
3371 connections had finished. The check for completed processes is now done
3372 earlier. On busy systems, this bug wouldn't be noticed because something
3373 else would have woken the daemon, and it would have reaped the completed
3376 80. If a message was submitted locally by a user whose login name contained one
3377 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3378 It caused a spool format error. I have fixed the spool reading code. A
3379 related problem was that the "from" clause in the Received: line became
3380 illegal because of the space(s). It is now covered by ${quote_local_part.
3382 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3385 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3387 83. Patch from Sieve maintainer to fix unterminated string problem in
3388 "vacation" handling.
3390 84. Some minor changes to the Linux configuration files to help with other
3391 OS variants using glibc.
3393 85. One more patch for Sieve to update vacation handling to latest spec.
3396 ----------------------------------------------------
3397 See the note above about the 4.44 and 4.50 releases.
3398 ----------------------------------------------------
3404 1. Change 4.43/35 introduced a bug that caused file counts to be
3405 incorrectly computed when quota_filecount was set in an appendfile
3408 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3409 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3412 3. Give more explanation in the error message when the command for a transport
3413 filter fails to execute.
3415 4. There are several places where Exim runs a non-Exim command in a
3416 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3417 was being done only for the command run by the queryprogram router. It is
3418 now done for all such subprocesses. The other cases are: ${run, transport
3419 filters, and the commands run by the lmtp and pipe transports.
3421 5. Some older OS have a limit of 256 on the maximum number of file
3422 descriptors. Exim was using setrlimit() to set 1000 as a large value
3423 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3424 systems. I've change it so that if it can't get 1000, it tries for 256.
3426 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3427 was an oversight, and furthermore, ever since the addition of extra
3428 controls (e.g. 4.43/32), the checks on when to allow different forms of
3429 "control" were broken. There should now be diagnostics for all cases when a
3430 control that does not make sense is encountered.
3432 7. $recipients is now available in the predata ACL (oversight).
3434 8. Tidy the search cache before the fork to do a delivery from a message
3435 received from the command line. Otherwise the child will trigger a lookup
3436 failure and thereby defer the delivery if it tries to use (for example) a
3437 cached ldap connection that the parent has called unbind on.
3439 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3440 of $address_data from the recipient verification was clobbered by the
3441 sender verification.
3443 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3444 was its contents. (It was OK if the option was not defined at all.)
3446 11. A "Completed" log line is now written for messages that are removed from
3447 the spool by the -Mrm option.
3449 12. $host_address is now set to the target address during the checking of
3450 ignore_target_hosts.
3452 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3453 being passed; this would have caused $sender_host_name to have been used if
3454 matching the list had actually called for a host name (not very likely,
3455 since this list is usually IP addresses). A host name is now passed as
3458 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3459 code that specifies a non-restarting handler (typically sigaction() in
3460 modern systems) in an attempt to fix a rare and obscure crash bug.
3462 15. Narrowed the window for a race in the daemon that could cause it to ignore
3463 SIGCHLD signals. This is not a major problem, because they are used only to
3464 wake it up if nothing else does.
3466 16. A malformed maildirsize file could cause Exim to calculate negative values
3467 for the mailbox size or file count. Odd effects could occur as a result.
3468 The maildirsize information is now recalculated if the size or filecount
3471 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3472 support for a long time. Removed HAVE_SYS_VFS_H.
3474 18. Updated exipick to current release from John Jetmore.
3476 19. Allow an empty sender to be matched against a lookup in an address list.
3477 Previously the only cases considered were a regular expression, or an
3480 20. Exim went into a mad DNS lookup loop when doing a callout where the
3481 host was specified on the transport, if the DNS lookup yielded more than
3484 21. The RFC2047 encoding function was originally intended for short strings
3485 such as real names; it was not keeping to the 75-character limit for
3486 encoded words that the RFC imposes. It now respects the limit, and
3487 generates multiple encoded words if necessary. To be on the safe side, I
3488 have increased the buffer size for the ${rfc2047: expansion operator from
3491 22. Failure to deliver a bounce message always caused it to be frozen, even if
3492 there was an errors_to setting on the router. The errors_to setting is now
3495 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3496 canonical form (fully expanded) before being placed in
3497 $sender_host_address.
3499 24. Updated eximstats to version 1.33
3501 25. Include certificate and key file names in error message when GnuTLS fails
3502 to set them up, because the GnuTLS error message doesn't include the name
3503 of the failing file when there is a problem reading it.
3505 26. Expand error message when OpenSSL has problems setting up cert/key files.
3508 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3509 (this can affect the format of dates).
3511 28. exim_tidydb, when checking for the continued existence of a message for
3512 which it has found a message-specific retry record, was not finding
3513 messages that were in split spool directories. Consequently, it was
3514 deleting retry records that should have stayed in existence.
3516 29. eximstats updated to version 1.35
3517 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3518 1.35 - bugfix such that pie charts by volume are generated correctly
3520 30. The SPA authentication driver was not abandoning authentication and moving
3521 on to the next authenticator when an expansion was forced to fail,
3522 contradicting the general specification for all authenticators. Instead it
3523 was generating a temporary error. It now behaves as specified.
3525 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3526 (the order specifies the preference for clients). The order is now AES256,
3527 AES128, 3DES, ARCFOUR128.
3529 31. Small patch to Sieve code - explicitly set From: when generating an
3532 32. Exim crashed if a remote delivery caused a very long error message to be
3533 recorded - for instance if somebody sent an entire SpamAssassin report back
3534 as a large number of 550 error lines. This bug was coincidentally fixed by
3535 increasing the size of one of Exim's internal buffers (big_buffer) that
3536 happened as part of the Exiscan merge. However, to be on the safe side, I
3537 have made the code more robust (and fixed the comments that describe what
3540 33. Some experimental protocols are using DNS PTR records for new purposes. The
3541 keys for these records are domain names, not reversed IP addresses. The
3542 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3543 leaves it alone. Component reversal etc. now happens only for IP addresses.
3546 34. The host_aton() function is supposed to be passed a string that is known
3547 to be a valid IP address. However, in the case of IPv6 addresses, it was
3548 not checking this. This is a hostage to fortune. Exim now panics and dies
3549 if the condition is not met. A case was found where this could be provoked
3550 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3551 components; fortuitously, this particular loophole had already been fixed
3552 by change 4.50/55 or 4.44/33 above.
3554 If there are any other similar loopholes, the new check in host_aton()
3555 itself should stop them being exploited. The report I received stated that
3556 data on the command line could provoke the exploit when Exim was running as
3557 exim, but did not say which command line option was involved. All I could
3558 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3559 running as the user.
3562 35. There was a buffer overflow vulnerability in the SPA authentication code
3563 (which came originally from the Samba project). I have added a test to the
3564 spa_base64_to_bits() function which I hope fixes it.
3567 36. The daemon start-up calls getloadavg() while still root for those OS that
3568 need the first call to be done as root, but it missed one case: when
3569 deliver_queue_load_max is set with deliver_drop_privilege. This is
3570 necessary for the benefit of the queue runner, because there is no re-exec
3571 when deliver_drop_privilege is set.
3573 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3574 was in use, was not putting the data itself into the right store pool;
3575 consequently, it could be overwritten for a subsequent message in the same
3576 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3579 38. Sometimes the final signoff response after QUIT could fail to get
3580 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3581 < 0 before doing a fflush(). This bug looks as though it goes back to the
3582 introduction of TLS in release 3.20, but "sometimes" must have been rare
3583 because the tests only now provoked it.
3589 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3590 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3591 Now it returns an integer. A number of calls were still expecting a T/F
3592 return. Fortuitously, in all cases, the tests worked in OK situations,
3593 which is the norm. However, things would have gone wrong on any write
3594 failures on the smtp file descriptor. This function is used when sending
3595 messages over SMTP and also when doing verify callouts.
3597 2. When Exim is called to do synchronous delivery of a locally submitted
3598 message (the -odf or -odi options), it no longer closes stderr before doing
3601 3. Implemented the mua_wrapper option.
3603 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3605 5. Implemented the functions header_remove(), header_testname(),
3606 header_add_at_position(), and receive_remove_recipient(), and exported them
3609 6. If an ACL "warn" statement specified the addition of headers, Exim already
3610 inserted X-ACL-Warn: at the start if there was no header name. However, it
3611 was not making this test for the second and subsequent header lines if
3612 there were newlines in the string. This meant that an invalid header could
3613 be inserted if Exim was badly configured.
3615 7. Allow an ACL "warn" statement to add header lines at the start or after all
3616 the Received: headers, as well as at the end.
3618 8. Added the rcpt_4xx retry error code.
3620 9. Added postmaster_mailfrom=xxx to callout verification option.
3622 10. Added mailfrom=xxxx to the callout verification option, for verify=
3625 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3626 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3628 12. Inserted some casts to stop certain compilers warning when using pointer
3629 differences as field lengths or precisions in printf-type calls (mostly
3630 affecting debugging statements).
3632 13. Added optional readline() support for -be (dynamically loaded).
3634 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3635 same clock tick as a message's arrival, so that its received time was the
3636 same as the "first fail" time on the retry record, and that message
3637 remained on the queue past the ultimate address timeout, every queue runner
3638 would try a delivery (because it was past the ultimate address timeout) but
3639 after another failure, the ultimate address timeout, which should have then
3640 bounced the address, did not kick in. This was a "< instead of <=" error;
3641 in most cases the first failure would have been in the next clock tick
3642 after the received time, and all would be well.
3644 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3645 being recognized when the domain list was tested by the match_domain
3646 condition in an expansion string.
3648 16. Added the ${str2b64: operator.
3650 17. Exim was always calling setrlimit() to set a large limit for the number of
3651 processes, without checking whether the existing limit was already
3652 adequate. (It did check for the limit on file descriptors.) Furthermore,
3653 errors from getrlimit() and setrlimit() were being ignored. Now they are
3654 logged to the main and panic logs, but Exim does carry on, to try to do its
3655 job under whatever limits there are.
3657 18. Imported PCRE 5.0.
3659 19. Trivial typo in log message " temporarily refused connection" (the leading
3662 20. If the log selector return_path_on_delivery was set and an address was
3663 redirected to /dev/null, the delivery process crashed because it assumed
3664 that a return path would always be set for a "successful" delivery. In this
3665 case, the whole delivery is bypassed as an optimization, and therefore no
3668 21. Internal re-arrangement: the function for sending a challenge and reading
3669 a response while authentication was assuming a zero-terminated challenge
3670 string. It's now changed to take a pointer and a length, to allow for
3671 binary data in such strings.
3673 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3675 23. Exim was not respecting finduser_retries when seeking the login of the
3676 uid under which it was called; it was always trying 10 times. (The default
3677 setting of finduser_retries is zero.) Also, it was sleeping after the final
3678 failure, which is pointless.
3680 24. Implemented tls_on_connect_ports.
3682 25. Implemented acl_smtp_predata.
3684 26. If the domain in control=submission is set empty, Exim assumes that the
3685 authenticated id is a complete email address when it generates From: or
3686 Sender: header lines.
3688 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3689 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3690 chown and chgrp in /bin and hostname in /usr/bin.
3692 28. Exim was keeping the "process log" file open after each use, just as it
3693 does for the main log. This opens the possibility of it remaining open for
3694 long periods when the USR1 signal hits a daemon. Occasional processlog
3695 errors were reported, that could have been caused by this. Anyway, it seems
3696 much more sensible not to leave this file open at all, so that is what now
3699 29. The long-running daemon process does not normally write to the log once it
3700 has entered its main loop, and it closes the log before doing so. This is
3701 so that log files can straightforwardly be renamed and moved. However,
3702 there are a couple of unusual error situations where the daemon does write
3703 log entries, and I had neglected to close the log afterwards.
3705 30. The text of an SMTP error response that was received during a remote
3706 delivery was being truncated at 512 bytes. This is too short for some of
3707 the long messages that one sometimes sees. I've increased the limit to
3710 31. It is now possible to make retry rules that apply only when a message has a
3711 specific sender, in particular, an empty sender.
3713 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3714 it possible to be selective about when SMTP synchronization is enforced.
3716 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3718 32. Implemented hosts_connection_nolog.
3720 33. Added an ACL for QUIT.
3722 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3725 35. Added mailbox_size and mailbox_filecount to appendfile.
3727 36. Added control = no_multiline_responses to ACLs.
3729 37. There was a bug in the logic of the code that waits for the clock to tick
3730 in the case where the clock went backwards by a substantial amount such
3731 that the microsecond fraction of "now" was more than the microsecond
3732 fraction of "then" (but the whole seconds number was less).
3734 38. Added support for the libradius Radius client library this is found on
3735 FreeBSD (previously only the radiusclient library was supported).
3741 1. When certain lookups returned multiple values in the form name=value, the
3742 quoting of the values was not always being done properly. Specifically:
3743 (a) If the value started with a double quote, but contained no whitespace,
3745 (b) If the value contained whitespace other than a space character (i.e.
3746 tabs or newlines or carriage returns) it was not quoted.
3747 This fix has been applied to the mysql and pgsql lookups by writing a
3748 separate quoting function and calling it from the lookup code. The fix
3749 should probably also be applied to nisplus, ibase and oracle lookups, but
3750 since I cannot test any of those, I have not disturbed their existing code.
3752 2. A hit in the callout cache for a specific address caused a log line with no
3753 reason for rejecting RCPT. Now it says "Previous (cached) callout
3754 verification failure".
3756 3. There was an off-by-one bug in the queryprogram router. An over-long
3757 return line was truncated at 256 instead of 255 characters, thereby
3758 overflowing its buffer with the terminating zero. As well as fixing this, I
3759 have increased the buffer size to 1024 (and made a note to document this).
3761 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3762 when Exim is waiting for an SMTP response from a remote server, Exim
3763 restarts its select() call on the socket, thereby resetting its timeout.
3764 This is not a problem when such interrupts are rare. Somebody set up a cron
3765 job to run exiwhat every 2 minutes, which is less than the normal select()
3766 timeout (5 or 10 minutes). This meant that the select() timeout never
3767 kicked in because it was always reset. I have fixed this by comparing the
3768 time when an interrupt arrives with the time at the start of the first call
3769 to select(). If more time than the timeout has elapsed, the interrupt is
3770 treated as a timeout.
3772 5. Some internal re-factoring in preparation for the addition of Sieve
3773 extensions (by MH). In particular, the "personal" test is moved to a
3774 separate function, and given an option for scanning Cc: and Bcc: (which is
3775 not set for Exim filters).
3777 6. When Exim created an email address using the login of the caller as the
3778 local part (e.g. when creating a From: or Sender: header line), it was not
3779 quoting the local part when it contained special characters such as @.
3781 7. Installed new OpenBSD configuration files.
3783 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3784 try to make them clearer.
3786 9. Callout options, other than the timeout value, were being ignored when
3787 verifying sender addresses in header lines. For example, when using
3789 verify = header_sender/callout=no_cache
3791 the cache was (incorrectly) being used.
3793 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3794 only the Cygwin environment.
3796 11. When return_path_on_delivery was set as a log selector, if different remote
3797 addresses in the same message used different return paths and parallel
3798 remote delivery occurred, the wrong values would sometimes be logged.
3799 (Whenever a remote delivery process finished, the return path value from
3800 the most recently started remote delivery process was logged.)
3802 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3803 lines when AUTH and/or TLS are in use. This is the "received protocol"
3804 field. Exim used to use "asmtp" for authenticated SMTP, without any
3805 indication (in the protocol name) for TLS use. Now it follows the RFC and
3806 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3807 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3808 names appear in log lines as well as in Received: header lines.
3810 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3811 extensions, and comparison tests, and to fix some bugs.
3813 14. Changes to the "personal" filter test:
3815 (1) The test was buggy in that it was just doing the equivalent of
3816 "contains" tests on header lines. For example, if a user's address was
3817 anne@some.where, the "personal" test would incorrectly be true for
3819 To: susanne@some.where
3821 This test is now done by extracting each address from the header in turn,
3822 and checking the entire address. Other tests that are part of "personal"
3823 are now done using regular expressions (for example, to check local parts
3824 of addresses in From: header lines).
3826 (2) The list of non-personal local parts in From: addresses has been
3827 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3828 taken from the Sieve specification recommendations.
3830 (3) If the message contains any header line starting with "List-" it is
3831 treated as non-personal.
3833 (4) The test for "circular" in the Subject: header line has been removed
3834 because it now seems ill-conceived.
3836 15. Minor typos in src/EDITME comments corrected.
3838 16. Installed latest exipick from John Jetmore.
3840 17. If headers_add on a router specified a text string that was too long for
3841 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3842 of string_sprintf() is now avoided.
3844 18. $message_body_size was not set (it was always zero) when running the DATA
3845 ACL and the local_scan() function.
3847 19. For the "mail" command in an Exim filter, no default was being set for
3848 the once_repeat time, causing a random time value to be used if "once" was
3849 specified. (If the value happened to be <= 0, no repeat happened.) The
3850 default is now 0s, meaning "never repeat". The "vacation" command was OK
3851 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3852 (I found it when inspecting the code).
3854 20. There is now an overall timeout for performing a callout verification. It
3855 defaults to 4 times the callout timeout, which applies to individual SMTP
3856 commands during the callout. The overall timeout applies when there is more
3857 than one host that can be tried. The timeout is checked before trying the
3858 next host. This prevents very long delays if there are a large number of
3859 hosts and all are timing out (e.g. when the network connections are timing
3860 out). The value of the overall timeout can be changed by specifying an
3861 additional sub-option for "callout", called "maxwait". For example:
3863 verify = sender/callout=5s,maxwait=20s
3865 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3866 to the end before writing, but this should make it even safer).
3868 22. Exim was forgetting that it had advertised PIPELINING for the second and
3869 subsequent messages on an SMTP connection. It was also not resetting its
3870 memory on STARTTLS and an internal HELO.
3872 23. When Exim logs an SMTP synchronization error within a session, it now
3873 records whether PIPELINING has been advertised or not.
3875 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3876 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3877 rather than long int.
3879 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3881 26. Added the never_mail option to autoreply.
3887 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3888 crash if the getsockname() call failed; this can happen if a connection is
3889 closed very soon after it is established. The problem was simply in the
3890 order in which certain operations were done, causing Exim to try to write
3891 to the SMTP stream before it had set up the file descriptor. The bug has
3892 been fixed by making things happen in the correct order.
3898 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3899 before the connection was closed, thus losing the rejection response.
3901 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3902 some early Solaris releases, but causes trouble in current releases where
3903 socklen_t is defined.
3905 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3908 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3911 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3912 that releasing the top of it at the end releases what was used for sub-
3913 expansions (unless the block got too big). However, discard this block if
3914 the first thing is a variable or header, so that we can use its block when
3915 it is dynamic (useful for very large $message_headers, for example).
3917 6. Lookups now cache *every* query, not just the most recent. A new, separate
3918 store pool is used for this. It can be recovered when all lookup caches are
3919 flushed. Lookups now release memory at the end of their result strings.
3920 This has involved some general refactoring of the lookup sources.
3922 7. Some code has been added to the store_xxx() functions to reduce the amount
3923 of flapping under certain conditions.
3925 8. log_incoming_interface used to affect only the <= reception log lines. Now
3926 it causes the local interface and port to be added to several more SMTP log
3927 lines, for example "SMTP connection from", and rejection lines.
3929 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3931 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3933 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3934 could be overwritten at the end of the current message (or the start of a
3935 new message if it was set in a HELO ACL). The value is now preserved for
3936 the duration of the SMTP connection.
3938 12. If a transport had a headers_rewrite setting, and a matching header line
3939 contained an unqualified address, that address was qualified, even if it
3940 did not match any rewriting rules. The underlying bug was that the values
3941 of the flags that permit the existence of unqualified sender and recipient
3942 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3943 non-local messages, and by -bnq for local messages) were not being
3944 preserved with the message after it was received.
3946 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3947 "next input=" as part of the text comprising the host identity instead of
3948 the correct text. The code was using the same buffer for two different
3949 strings. However, depending on which order the printing function evaluated
3950 its arguments, the bug did not always show up. Under Linux, for example, my
3951 test suite worked just fine.
3953 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3954 doesn't work with some older versions of Perl. It has been changed to "my",
3955 which in any case is probably the better facility to use.
3957 15. A really picky compiler found some instances of statements for creating
3958 error messages that either had too many or two few arguments for the format
3961 16. The size of the buffer for calls to the DNS resolver has been increased
3962 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3963 for addresses that have a lot of PTR records. This alleviates a problem; it
3964 does not fully solve it.
3966 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3967 buffer now truncates the list and logs the incident, which is the same
3968 action as happens when Exim is looking up a host name and its aliases.
3969 Previously in this situation something unpredictable would happen;
3970 sometimes it was "internal error: store_reset failed".
3972 18. If a server dropped the connection unexpectedly when an Exim client was
3973 using GnuTLS and trying to read a response, the client delivery process
3974 crashed while trying to generate an error log message.
3976 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3977 string, for example:
3979 warn message = H1: something\nH2: something
3981 the text was added as a single header line from Exim's point of view
3982 though it ended up OK in the delivered message. However, searching for the
3983 second and subsequent header lines using $h_h2: did not work. This has been
3984 fixed. Similarly, if a system filter added multiple headers in this way,
3985 the routers could not see them.
3987 20. Expanded the error message when iplsearch is called with an invalid key to
3988 suggest using net-iplsearch in a host list.
3990 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3991 ACLs are no longer actually imposed (and a message to that effect is
3994 22. If a "gecos" field in a passwd entry contained escaped characters, in
3995 particular, if it contained a \" sequence, Exim got it wrong when building
3996 a From: or a Sender: header from that name. A second bug also caused
3997 incorrect handling when an unquoted " was present following a character
3998 that needed quoting.
4000 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4001 was not being matched caselessly.
4003 24. Arranged for all hyphens in the exim.8 source to be escaped with
4006 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4007 properly. Recipient callout cache records were still being keyed to include
4008 the sender, even when use_sender was set false. This led to far more
4009 callouts that were necessary. The sender is no longer included in the key
4010 when use_sender is false.
4012 26. Added "control = submission" modifier to ACLs.
4014 27. Added the ${base62d: operator to decode base 62 numbers.
4016 28. dnsdb lookups can now access SRV records.
4018 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
4019 the configuration file.
4021 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
4022 (-v) mode. This makes the output for a verbose queue run more intelligible.
4024 31. Added a use_postmaster feature to recipient callouts.
4026 32. Added the $body_zerocount variable, containing the number of binary zero
4027 bytes in the message body.
4029 33. The time of last modification of the "new" subdirectory is now used as the
4030 "mailbox time last read" when there is a quota error for a maildir
4033 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
4035 35. Added +ignore_unknown as a special item in host lists.
4037 36. Code for decoding IPv6 addresses in host lists is now included, even if
4038 IPv6 support is not being compiled. This fixes a bug in which an IPv6
4039 address was recognized as an IP address, but was then not correctly decoded
4040 into binary, causing unexpected and incorrect effects when compared with
4047 1. Very minor rewording of debugging text in manualroute to say "list of
4048 hosts" instead of "hostlist".
4050 2. If verify=header_syntax was set, and a header line with an unqualified
4051 address (no domain) and a large number of spaces between the end of the
4052 name and the colon was received, the reception process suffered a buffer
4053 overflow, and (when I tested it) crashed. This was caused by some obsolete
4054 code that should have been removed. The fix is to remove it!
4056 3. When running in the test harness, delay a bit after writing a bounce
4057 message to get a bit more predictability in the log output.
4059 4. Added a call to search_tidyup() just before forking a reception process. In
4060 theory, someone could use a lookup in the expansion of smtp_accept_max_
4061 per_host which, without the tidyup, could leave open a database connection.
4063 5. Added the variables $recipient_data and $sender_data which get set from a
4064 lookup success in an ACL "recipients" or "senders" condition, or a router
4065 "senders" option, similar to $domain_data and $local_part_data.
4067 6. Moved the writing of debug_print from before to after the "senders" test
4070 7. Change 4.31/66 (moving the time when the Received: is generated) caused
4071 problems for message scanning, either using a data ACL, or using
4072 local_scan() because the Received: header was not generated till after they
4073 were called (in order to set the time as the time of reception completion).
4074 I have revised the way this works. The header is now generated after the
4075 body is received, but before the ACL or local_scan() are called. After they
4076 are run, the timestamp in the header is updated.
4082 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
4083 before starting a queue runner without re-exec. This happened only when
4084 deliver_drop_privilege was set or when the Exim user was set to root. The
4085 effect of the bug was that timeouts during subsequent deliveries caused
4086 crashes instead of being properly handled. The handler is now left at its
4087 default (and expected) setting.
4089 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
4090 message, again when deliver_drop_privilege is set or Exim is run as root.
4091 The bug described in (1) was not present in this case, but the tidying up
4092 of the other signals was missing. I have made the two cases consistent.
4094 3. The ignore_target_hosts setting on a manualroute router was being ignored
4095 for hosts that were looked up using the /MX notation.
4097 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
4100 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
4101 operated on the sender address. After changing the $sender_address to <>
4102 for the sender address verify, Exim was re-instated it as the original
4103 (before rewriting) address, but remembering that it had rewritten it, so it
4104 wasn't rewriting it again. This bug also had the effect of breaking the
4105 sender address verification caching when the sender address was rewritten.
4107 6. The ignore_target_hosts option was being ignored by the ipliteral router.
4108 This has been changed so that if the ip literal address matches
4109 ignore_target_hosts, the router declines.
4111 7. Added expansion conditions match_domain, match_address, and match_local_
4112 part (NOT match_host).
4114 8. The placeholder for the Received: header didn't have a length field set.
4116 9. Added code to Exim itself and to exim_lock to test for a specific race
4117 condition that could lead to file corruption when using MBX delivery. The
4118 issue is with the lockfile that is created in /tmp. If this file is removed
4119 after a process has opened it but before that process has acquired a lock,
4120 there is the potential for a second process to recreate the file and also
4121 acquire a lock. This could lead to two Exim processes writing to the file
4122 at the same time. The added code performs the same test as UW imapd; it
4123 checks after acquiring the lock that its file descriptor still refers to
4124 the same named file.
4126 10. The buffer for building added header lines was of fixed size, 8192 bytes.
4127 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
4130 11. Added the smtp_active_hostname option. If used, this will typically be made
4131 to depend on the incoming interface address. Because $interface_address is
4132 not set up until the daemon has forked a reception process, error responses
4133 that can happen earlier (such as "too many connections") no longer contain
4136 12. If an expansion in a condition on a "warn" statement fails because a lookup
4137 defers, the "warn" statement is abandoned, and the next ACL statement is
4138 processed. Previously this caused the whole ACL to be aborted.
4140 13. Added the iplsearch lookup type.
4142 14. Added ident_timeout as a log selector.
4144 15. Added tls_certificate_verified as a log selector.
4146 16. Added a global option tls_require_ciphers (compare the smtp transport
4147 option of the same name). This controls incoming TLS connections.
4149 17. I finally figured out how to make tls_require_ciphers do a similar thing
4150 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
4151 before starting the TLS session.
4153 18. Tabs are now shown as \t in -bP output.
4155 19. If the log selector return_path_on_delivery was set, Exim crashed when
4156 bouncing a message because it had too many Received: header lines.
4158 20. If two routers both had headers_remove settings, and the first one included
4159 a superfluous trailing colon, the final name in the first list and the
4160 first name in the second list were incorrectly joined into one item (with a
4161 colon in the middle).
4167 1. Added -C and -D options to the exinext utility, mainly to make it easier
4168 to include in the automated testing, but these could be helpful when
4169 multiple configurations are in use.
4171 2. The exinext utility was not formatting the output nicely when there was
4172 an alternate port involved in the retry record key, nor when there was a
4173 message id as well (for retries that were specific to a specific message
4174 and a specific host). It was also confused by IPv6 addresses, because of
4175 the additional colons they contain. I have fixed the IPv4 problem, and
4176 patched it up to do a reasonable job for IPv6.
4178 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
4179 delivery, the log line now contains "pipelined" if PIPELINING was used.
4181 4. An SMTP transport process used to panic and die if the bind() call to set
4182 an explicit outgoing interface failed. This has been changed; it is now
4183 treated in the same way as a connect() failure.
4185 5. A reference to $sender_host_name in the part of a conditional expansion
4186 that was being skipped was still causing a DNS lookup. This no longer
4189 6. The def: expansion condition was not recognizing references to header lines
4190 that used bh_ and bheader_.
4192 7. Added the _cache feature to named lists.
4194 8. The code for checking quota_filecount in the appendfile transport was
4195 allowing one more file than it should have been.
4197 9. For compatibility with Sendmail, the command line option
4205 and sets the incoming protocol and host name (for trusted callers). The
4206 host name and its colon can be omitted when only the protocol is to be set.
4207 Note the Exim already has two private options, -pd and -ps, that refer to
4208 embedded Perl. It is therefore impossible to set a protocol value of "d" or
4209 "s", but I don't think that's a major issue.
4211 10. A number of refactoring changes to the code, none of which should affect
4214 (a) The number of logging options was getting close to filling up the
4215 32-bit word that was used as a bit map. I have split them into two classes:
4216 those that are passed in the argument to log_write(), and those that are
4217 only ever tested independently outside of that function. These are now in
4218 separate 32-bit words, so there is plenty of room for expansion again.
4219 There is no change in the user interface or the logging behaviour.
4221 (b) When building, for example, log lines, the code previously used a
4222 macro that called string_cat() twice, in order to add two strings. This is
4223 not really sufficiently general. Furthermore, there was one instance where
4224 it was actually wrong because one of the argument was used twice, and in
4225 one call a function was used. (As it happened, calling the function twice
4226 did not affect the overall behaviour.) The macro has been replaced by a
4227 function that can join an arbitrary number of extra strings onto a growing
4230 (c) The code for expansion conditions now uses a table and a binary chop
4231 instead of a serial search (which was left over from when there were very
4232 few conditions). Also, it now recognizes conditions like "pam" even when
4233 the relevant support is not compiled in: a suitably worded error message is
4234 given if an attempt is made to use such a condition.
4236 11. Added ${time_interval:xxxxx}.
4238 12. A bug was causing one of the ddress fields not to be passed back correctly
4239 from remote delivery subprocesses. The field in question was not being
4240 subsequently used, so this caused to problems in practice.
4242 13. Added new log selectors queue_time and deliver_time.
4244 14. Might have fixed a bug in maildirsizefile handling that threw up
4245 "unexpected character" debug warnings, and recalculated the data
4246 unnecessarily. In any case, I expanded the warning message to give more
4249 15. Added the message "Restricted characters in address" to the statements in
4250 the default ACL that block characters like @ and % in local parts.
4252 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
4253 Three changes have been made:
4255 (a) There was a serious bug; a negative response to MAIL caused the whole
4256 recipient domain to be cached as invalid, thereby blocking all messages
4257 to all local parts at the same domain, from all senders. This bug has
4258 been fixed. The domain is no longer cached after a negative response to
4259 MAIL if the sender used is not empty.
4261 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
4264 (c) A new callout option, "use_sender" has been added for people who want
4265 the modified behaviour.
4271 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
4274 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
4275 indeed breaks things for older releases.
4277 3. Added additional logging to the case where there is a problem reading data
4278 from a filter that is running in a subprocess using a pipe, in order to
4279 try to track down a specific problem.
4281 4. Testing facility fudge: when running in the test harness and attempting
4282 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
4283 getting "No route to host". Convert this to a timeout.
4285 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
4288 6. Some OS don't have socklen_t but use size_t instead. This affects the
4289 fifth argument of getsockopt() amongst other things. This is now
4290 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
4291 can be set for individual OS. I have set it for SunOS5, OSF1, and
4292 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
4293 some earlier ones do not.
4295 7. Change 4.30/15 was not doing the test caselessly.
4297 8. The standard form for an IPv6 address literal was being rejected by address
4298 parsing in, for example, MAIL and RCPT commands. An example of this kind of
4299 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4300 this, as well as the form without the "IPv6" on the front (but only when
4301 address literals are enabled, of course).
4303 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
4305 10. Exim crashed if a message with an empty sender address specified by -f
4306 encountered a router with an errors_to setting. This could be provoked only
4307 by a command such as
4311 where an empty string was supplied; "<>" did not hit this bug.
4313 11. Installed PCRE release 4.5.
4315 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4316 remained set. It is now erased.
4318 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4319 times from message ids (which are base 36 rather than the normal 62).
4321 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4322 were being counted as actual protocol errors, and logged if the log
4323 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4324 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4325 and DATA following a set of rejected RCPTs do not count as protocol errors.
4326 In other words, Exim assumes they were pipelined, though this may not
4327 actually be the case. Of course, in all cases the client gets an
4328 appropriate error code.
4330 15. If a lookup fails in an ACL condition, a message about the failure may
4331 be available; it is used if testing the ACL cannot continue, because most
4332 such messages specify what the cause of the deferral is. However, some
4333 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4334 that caused an old message to be retained and used if a later statement
4335 caused a defer, replacing the real cause of the deferral.
4337 16. If an IP address had so many PTR records that the DNS lookup buffer
4338 was not large enough to hold them, Exim could crash while trying to process
4339 the truncated data. It now detects and logs this case.
4341 17. Further to 4.21/58, another change has been made: if (and only if) the
4342 first line of a message (the first header line) ends with CRLF, a bare LF
4343 in a subsequent header line has a space inserted after it, so as not to
4344 terminate the header.
4346 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4347 unnecessarily, used atoi() instead of strtol(), and didn't check the
4348 termination when getting file sizes from file names by regex.
4350 19. Completely re-implemented the support for maildirsize files, in the light
4351 of a number of problems with the previous contributed implementation
4352 (4.30/29). In particular:
4354 . If the quota is zero, the maildirsize file is maintained, but no quota is
4357 . If the maildir directory does not exist, it is created before any attempt
4358 to write a maildirsize file.
4360 . The quota value in the file is just a cache; if the quota is changed in
4361 the transport, the new value overrides.
4363 . A regular expression is available for excluding directories from the
4366 20. The autoreply transport checks the characters in options that define the
4367 message's headers; it allows continued headers, but it was checking with
4368 isspace() after an embedded newline instead of explicitly looking for a
4371 21. If all the "regular" hosts to which an address was routed had passed their
4372 expiry times, and had not reached their retry times, the address was
4373 bounced, even if fallback hosts were defined. Now Exim should go on to try
4376 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4377 equivalent code in the SMTP transport. Some hosts send humungous responses
4378 to HELO/EHLO, more than 1024 it seems.
4380 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4381 gives compiler warnings in some environments. I've now done it "properly",
4384 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4385 (because of problems with the built-in one) was declared to return uschar *
4386 instead of char *, causing compiler failure.
4388 25. Fixed a file descriptor leak when processing alias/forward files.
4390 26. Fixed a minor format string issue in dbfn.c.
4392 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4394 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4395 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4398 29. When a sender address is verified, it is cached, to save repeating the test
4399 when there is more than one recipient in a message. However, when the
4400 verification involves a callout, it is possible for different callout
4401 options to be set for different recipients. It is too complicated to keep
4402 track of this in the cache, so now Exim always runs a verification when a
4403 callout is required, relying on the callout cache for the optimization.
4404 The overhead is duplication of the address routing, but this should not be
4407 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4408 to be verified with callout=postmaster, and the main callout worked but the
4409 postmaster check failed, the verification correctly failed. However, if a
4410 subsequent RCPT command asked for sender verification *without* the
4411 postmaster check, incorrect caching caused this verification also to fail,
4414 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4415 it was not caching the DNS options (qualify_single, search_parents) that
4416 were used when the lookup failed. A subsequent lookup with different
4417 options therefore always gave the same answer, though there were cases
4418 where it should not have. (Example: a "domains = !$mx_any" option on a
4419 dnslookup router: the "domains" option is always processed without any
4420 widening, but the router might have qualify_single set.) Now Exim uses the
4421 cached value only when the same options are set.
4423 32. Added John Jetmore's "exipick" utility to the distribution.
4425 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4426 than a timeout (e.g. a certificate is required, and is not provided), an
4427 Exim server now closes the connection immediately. Previously it waited for
4428 the client to close - but if the client is SSL, it seems that they each
4429 wait for each other, leading to a delay before one of them times out.
4431 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4432 maintained 0.8.x compatibility because I don't think many are using it, and
4433 it is clearly obsolete.
4435 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4438 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4439 client certificate was expired. A simple patch fixes this, though I don't
4440 understand the full logic of why the verify callback is called multiple
4443 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4444 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4445 which causes problems with some clients (such as the Certicom SSL Plus
4446 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4447 disables the coutermeasure allowing Eudora to connect."
4449 38. Exim was not checking that a write() to a log file succeeded. This could
4450 lead to Bad Things if a log got too big, in particular if it hit a file
4451 size limit. Exim now panics and dies if it cannot write to a log file, just
4452 as it does if it cannot open a log file.
4454 39. Modified OS/Makefile-Linux so that it now contains
4456 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4458 The two -D definitions ensure that Exim is compiled with large file
4459 support, which makes it possible to handle log files that are bigger than
4462 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4463 instance) a domain was checked against a named list that involved a lookup,
4464 causing $domain_data to be set, then another domain was checked against the
4465 same list, then the first domain was re-checked, the value of $domain_data
4466 after the final check could be wrong. In particular, if the second check
4467 failed, it could be set empty. This bug probably also applied to
4470 41. The strip_trailing_dot option was not being applied to the address given
4471 with the -f command-line option.
4473 42. The code for reading a message's header from the spool was incrementing
4474 $received_count, but never initializing it. This meant that the value was
4475 incorrect (doubled) while delivering a message in the same process in which
4476 it was received. In the most common configuration of Exim, this never
4477 happens - a fresh exec is done - but it can happen when
4478 deliver_drop_privilege is set.
4480 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4481 it now includes up to 150 characters of the unexpected data in the log
4484 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4485 and building data strings. The size of both of these buffers was 10 000
4486 bytes - far larger than anybody would *ever* want, thought I. Needless to
4487 say, somebody hit the limit. I have increased the maximum line length to
4488 20 000 and the maximum data length of concatenated lines to 100 000. I have
4489 also fixed two bugs, because there was no checking on these buffers. Tsk,
4490 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4491 buffer is too small.
4493 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4494 lsearch lookups. Now it does.
4496 46. When parsing a route_list item in a manualroute router, a fixed-length
4497 buffer was used for the list of hosts. I made this 1024 bytes long,
4498 thinking that nobody would ever have a list of hosts that long. Wrong.
4499 Somebody had a whole pile of complicated expansion conditions, and the
4500 string was silently truncated, leading to an expansion error. It turns out
4501 that it is easier to change to an unlimited length (owing to other changes
4502 that have happened since this code was originally written) than to build
4503 structure for giving a limitation error. The length of the item that
4504 expands into the list of hosts is now unlimited.
4506 47. The lsearch lookup could not handle data where the length of text line was
4507 more than 4095 characters. Such lines were truncated, leading to shortened
4508 data being returned. It should now handle lines of any length.
4510 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4511 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4514 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4515 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4516 no longer happens in comment lines. A list of replacements is now placed
4517 at the head of all of the source files, except those whose only change is
4518 to replace PERL_COMMAND in the very first #! line.
4520 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4521 messages on the queue, with a bottom-up merge sort, using code contributed
4522 by Michael Haardt. This should make operations like -bp somewhat faster on
4523 large queues. It won't affect queue runners, except when queue_run_in_order
4526 51. Installed eximstats 1.31 in the distribution.
4528 52. Added support for SRV lookups to the dnslookup router.
4530 53. If an ACL referred to $message_body or $message_body_end, the value was not
4531 reset for any messages that followed in the same SMTP session.
4533 54. The store-handling optimization for building very long strings was not
4534 differentiating between the different store pools. I don't think this
4535 actually made any difference in practice, but I've tidied it.
4537 55. While running the routers to verify a sender address, $sender_address
4538 was still set to the sender address. This is wrong, because when routing to
4539 send a bounce to the sender, it would be empty. Therefore, I have changed
4540 it so that, while verifying a sender address, $sender_address is set to <>.
4541 (There is no change to what happens when verifying a recipient address.)
4543 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4544 target A or AAAA records (if not already returned) without resetting the
4545 qualify_single or search_parents options of the DNS resolver. These are
4546 inappropriate in this case because the targets of MX and SRV records must
4547 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4548 target that, when qualified, matched something in the local domain. These
4549 two options are now turned off when doing these lookups.
4551 57. It seems that at least some releases of Reiserfs (which does not have the
4552 concept of a fixed number of inodes) returns zero and not -1 for the
4553 number of available inodes. This interacted badly with check_spool_inodes,
4554 which assumed that -1 was the "no such thing" setting. What I have done is
4555 to check that the total number of inodes is greater than zero before doing
4556 the test of how many are available.
4558 58. When a "warn" ACL statement has a log_message modifier, the message is
4559 remembered, and not repeated. This is to avoid a lot of repetition when a
4560 message has many recipients that cause the same warning to be written.
4561 Howewer, Exim was preserving the list of already written lines for an
4562 entire SMTP session, which doesn't seem right. The memory is now reset if a
4563 new message is started.
4565 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4566 debugging output unless log_rewrite was also set.
4568 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4569 of (int)(handle) when we know that handle contains (void *)(-1).
4571 61. The Exim daemon panic-logs an error return when it closes the incoming
4572 connection. However "connection reset by peer" seems to be common, and
4573 isn't really an error worthy of noting specially, so that particular error
4576 62. When Exim is trying to find all the local interfaces, it used to panic and
4577 die if the ioctl to get the interface flags failed. However, it seems that
4578 on at least one OS (Solaris 9) it is possible to have an interface that is
4579 included in the list of interfaces, but for which you get a failure error
4580 for this call. This happens when the interface is not "plumbed" into a
4581 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4582 failure of the "get flags" call assumes that the interface is down.
4584 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4585 makes life easier for people who are doing arithmetic on fields extracted
4586 from dates, where you often get leading zeros that should not be
4587 interpreted as octal.
4589 64. Added qualify_domain to the redirect router, to override the global
4592 65. If a pathologically long header line contained very many addresses (the
4593 report of this problem mentioned 10 000) and each of them was rewritten,
4594 Exim could use up a very large amount of memory. (It kept on making new
4595 copies of the header line as it rewrote, and never released the old ones.)
4596 At the expense of a bit more processing, the header rewriting function has
4597 been changed so that it no longer eats memory in this way.
4599 66. The generation of the Received: header has been moved from the time that a
4600 message starts to be received, to the time that it finishes. The timestamp
4601 in the Received: header should now be very close to that of the <= log
4602 line. There are two side-effects of this change:
4604 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4605 logged header lines no longer include the local Received: line, because
4606 it has not yet been created. The same applies to a copy of the message
4607 that is returned to a non-SMTP sender when a message is rejected.
4609 (b) When a filter file is tested using -bf, no additional Received: header
4610 is added to the test message. After some thought, I decided that this
4613 This change does not affect the value of $received_for. It is still set
4614 after address rewriting, but before local_scan() is called.
4616 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4618 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4619 gave an unhelpful panic error message, and a defer error. I have managed to
4620 change this behaviour so that it now rejects any supplied certificate,
4621 which seems right, as the list of acceptable certificates is empty.
4623 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4624 gave an unhelpful defer error. I have not managed to make this reject any
4625 supplied certificates, but the error message it gives is "no certificate
4626 supplied", which is not helpful.
4628 70. exigrep's output now also includes lines that are not associated with any
4629 message, but which match the given pattern. Implemented by a patch from
4630 Martin Sluka, which also tidied up the Perl a bit.
4632 71. Recipient callout verification, like sender verification, was using <> in
4633 the MAIL FROM command. This isn't really the right thing, since the actual
4634 sender may affect whether the remote host accepts the recipient or not. I
4635 have changed it to use the actual sender in the callout; this means that
4636 the cache record is now keyed on a recipient/sender pair, not just the
4637 recipient address. There doesn't seem to be a real danger of callout loops,
4638 since a callout by the remote host to check the sender would use <>.
4639 [SEE ABOVE: changed after hitting problems.]
4641 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4642 temporary errors. However, in the case of such a code being given after
4643 the end of a data transmission (i.e. after ".") Exim was failing to write
4644 a retry record for the message. (Yes, there was some broken host that was
4645 actually sending 8xx at this point.)
4647 73. An unknown lookup type in a host list could cause Exim to panic-die when
4648 the list was checked. (An example that provoked this was putting <; in the
4649 middle of a list instead of at the start.) If this happened during a DATA
4650 ACL check, a -D file could be left lying around. This kind of configuration
4651 error no longer causes Exim to die; instead it causes a defer errror. The
4652 incident is still logged to the main and panic logs.
4654 74. Buglet left over from Exim 3 conversion. The message "too many messages
4655 in one connection" was written to the rejectlog but not the mainlog, except
4656 when address rewriting (yes!) was being logged.
4658 75. Added write_rejectlog option.
4660 76. When a system filter was run not as root (that is, when system_filter_user
4661 was set), the values of the $n variables were not being returned to the
4662 main process; thus, they were not subsequently available in the $sn
4665 77. Added +return_path_on_delivery log selector.
4667 78. A connection timeout was being treated differently from recipients deferred
4668 when testing hosts_max_try with a message that was older than the host's
4669 retry timeout. (The host should not be counted, thus allowing all hosts to
4670 be tried at least once before bouncing.) This may have been the cause of an
4671 occasionally reported bug whereby a message would remain on the queue
4672 longer than the retry timeout, but would be bounced if a delivery was
4673 forced. I say "may" because I never totally pinned down the problem;
4674 setting up timeout/retry tests is difficult. See also the next item.
4676 79. The ultimate address timeout was not being applied to errors that involved
4677 a combination of host plus message (for example, a timeout on a MAIL
4678 command). When an address resolved to a number of possible hosts, and they
4679 were not all tried for each delivery (e.g. because of hosts_max_try), a
4680 message could remain on the queue longer than the retry timeout.
4682 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4685 81. Fixed an obscure SMTP outgoing bug which required at least the following
4686 conditions: (a) there was another message waiting for the same server;
4687 (b) the server returned 5xx to all RCPT commands in the first message so
4688 that the message was not completed; (c) the server dropped the connection
4689 or gave a negative response to the RSET that Exim sends to abort the
4690 transaction. The observed case was a dropped connection after DATA that had
4691 been sent in pipelining mode. That is, the server had advertised PIPELINING
4692 but was not implementing it correctly. The effect of the bug was incorrect
4693 behaviour, such as trying another host, and this could lead to a crash.
4699 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4700 and daemon.c were passed as pointers to ints; they should have been
4701 pointers to socklen_t variables (which are typically unsigned ints).
4703 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4706 3. Fixed a really odd bug that affected only the testing scheme; patching a
4707 certain fixed string in the binary changed the value of another string that
4708 happened to be identical to the end of the original first string.
4710 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4711 name", it returns that address as the IP address. On some operating
4712 systems (e.g. Solaris), it also passes back the IP address string as the
4713 "host name". However, on others (e.g. Linux), it passes back an empty
4714 string. Exim wasn't checking for this, and was changing the host name to an
4715 empty string, assuming it had been canonicized.
4717 5. Although rare, it is permitted to have more than one PTR record for a given
4718 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4719 all the names associated with an address, because they do in Solaris.
4720 However, it seems that they do not in Linux for data that comes from the
4721 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4722 I found this out when I moved to a new Linux workstation and tried to run
4723 the Exim test suite.
4725 To get round this problem I have changed the code so that it now does its
4726 own call to the DNS to look up PTR records when searching for a host name.
4727 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4728 addresses that are only in /etc/hosts are still found.
4730 This behaviour is, however, controlled by an option called host_lookup_
4731 order, which defaults to "bydns:byaddr". If people want to use the other
4732 order, or indeed, just use one or the other means of lookup, they can
4733 specify it in this variable.
4735 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4736 some operating systems, this comes back from gethostbyaddr() as an empty
4737 string, and this is what Exim used to test for. However, it seems that in
4738 other systems, "." is yielded. Exim now tests for this case too.
4740 7. The values of check_spool_space and check_log_space are now held internally
4741 as a number of kilobytes instead of an absolute number of bytes. If a
4742 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4743 kilobyte. This means that much larger values can be stored.
4745 8. Exim monitor: an attempt to get the action menu when not actually pointing
4746 at a message produces an empty menu entitled "No message selected". This
4747 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4748 no entries in it ("Shell widget menu has zero width and/or height"). So I
4749 have added a single, blank menu entry in this case.
4751 9. Added ${quote_local_part.
4753 10. MIME decoding is now applied to the contents of Subject: header lines when
4756 11. Now that a reference to $sender_host_address automatically causes a reverse
4757 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4758 host lookup before query-style lookups in lists that might use this
4759 variable. This has therefore been abolished, and the "net-" prefix is no
4760 longer necessary for query-style lookups.
4762 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4763 have been a typo for LFLAGS, so it has been changed.
4765 13. The install script calls Exim with "-C /dev/null" in order to find the
4766 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4767 to be output. Howeve, since Exim outputs its version number before the
4768 error, it didn't break the script. It just looked ugly. I fixed this by
4769 always allowing "-C /dev/null" if the caller is root.
4771 14. Ignore overlarge ACL variable number when reading spool file - insurance
4772 against a later release with more variables having written the file.
4774 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4775 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4776 this, as well as the form without the "IPv6" on the front.
4778 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4779 OS/Makefile-Darwin file.
4781 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4782 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4784 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4785 variable. However, some OS use ints for inodes. I've added cast to long int
4786 to get rid of the compiler warning.
4788 19. I had forgotten to lock out "/../" in configuration file names when
4789 ALT_CONFIG_PREFIX was set.
4791 20. Routers used for verification do not need to specify transports. However,
4792 if such a router generated a host list, and callout was configured, Exim
4793 crashed, because it could not find a port number from the (non-existent)
4794 transport. It now assumes port 25 in this circumstance.
4796 21. Added the -t option to exigrep.
4798 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4799 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4800 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4801 them is set, LOOKUP_LSEARCH is forced.
4803 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4805 24. Added sender and host information to the "rejected by local_scan()" log
4806 line; previously there was no indication of these.
4808 25. Added .include_if_exists.
4810 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4811 Linux. It turns out that not all file systems support this. Apparently some
4812 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4813 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4814 not supported on the file descriptor, is now ignored when Exim is trying to
4815 sync a directory. This applies only to Linux.
4817 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4819 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4820 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4823 29. Added support for maildirsize files from supplied patch (modified a bit).
4825 30. The use of :fail: followed by an empty string could lead Exim to respond to
4826 sender verification failures with (e.g.):
4828 550 Verification failed for <xxx>
4829 550 Sender verify failed
4831 where the first response line was missing the '-' that indicates it is not
4832 the final line of the response.
4834 31. The loop for finding the name of the user that called Exim had a hardwired
4835 limit of 10; it now uses the value of finduser_retries, which is used for
4836 all other user lookups.
4838 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4841 33. Exim was neglecting to zero errno before one call of strtol() when
4842 expanding a string and expecting an integer value. On some systems this
4843 resulted in spurious "integer overflow" errors. Also, it was casting the
4844 result into an int without checking.
4846 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4847 did not work. The code looks as if it has *never* worked, though it appears
4848 to have been documented since at least releast 1.62. I have made it work.
4850 35. The "timeout_DNS" error in retry rules, also documented since at least
4851 1.62, also never worked. As it isn't clear exactly what this means, and
4852 clearly it isn't a major issue, I have abolished the feature by treating it
4853 as "timeout", and writing a warning to the main and panic logs.
4855 36. The display of retry rules for -brt wasn't always showing the error code
4858 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4859 timeout_connect_A, timeout_connect_MX.
4861 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4862 to the empty sender.
4864 39. The daemon was not analysing the content of -oX till after it had closed
4865 stderr and disconnected from the controlling terminal. This meant that any
4866 syntax errors were only noted on the panic log, and the return code from
4867 the command was 0. By re-arranging the code a little, I've made the
4868 decoding happen first, so such errors now appear on stderr, and the return
4869 code is 1. However, the actual setting up of the sockets still happens in
4870 the disconnected process, so errors there are still only recorded on the
4873 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4874 connections (as happens on some IP stacks) was logged at start up time as
4875 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4876 it from "IPv6 and IPv4", which means that two separate sockets are being
4879 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4880 says whether AF_INET or AF_INET6 was passed as an argument.
4882 42. Exiwhat output was messed up when time zones were included in log
4885 43. Exiwhat now gives more information about the daemon's listening ports,
4886 and whether -tls-on-connect was used.
4888 44. The "port" option of the smtp transport is now expanded.
4890 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4891 silently ignored. Now an error message is written to the main and panic
4894 46. There's a new ACL modifier called "logwrite" which writes to a log file
4895 as soon as it is encountered.
4897 47. Added $local_user_uid and $local_user_gid at routing time.
4899 48. Exim crashed when trying to verify a sender address that was being
4902 49. Exim was recognizing only a space character after ".include". It now also
4903 recognizes a tab character.
4905 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4906 extracting the relevant information from the specification. The man page no
4907 longer contains scrambled data for the -d option, and I've added a section
4908 at the front about calling Exim under different names.
4910 51. Added "extra_headers" argument to the "mail" command in filter files.
4912 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4915 53. Installed eximstats 1.29.
4917 54. Added transport_filter_timeout as a generic transport option.
4919 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4920 Cc: header lines. This was required by RFC 822, but it not required by RFC
4923 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4924 incoming messages that did not have them. Now it does so only if the
4925 message originates locally, that is, if there is no associated remote host
4926 address. When Resent- header lines are present, this applies to the Resent-
4927 lines rather than the non-Resent- lines.
4929 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4930 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4932 58. Messages for configuration errors now include the name of the main
4933 configuration file - useful now that there may be more than one file in a
4934 list (.included file names were always shown).
4936 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4937 for those rare installations that do not start the daemon as root or run it
4938 setuid root. I've cut out the call to initgroups() if the daemon is not
4941 60. The Exim user and group can now be bound into the binary as text strings
4942 that are looked up at the start of Exim's processing.
4944 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4946 62. Added $mailstore_basename variable.
4948 63. Installed patch to sieve.c from Michael Haardt.
4950 64. When Exim failed to open the panic log after failing to open the main log,
4951 the original message it was trying to log was written to stderr and debug
4952 output, but if they were not available (the usual case in production), it
4953 was lost. Now it is written to syslog before the two lines that record the
4954 failures to open the logs.
4956 65. Users' Exim filters run in subprocesses under the user's uid. It is
4957 possible for a "deliver" command or an alias in a "personal" command to
4958 provoke an address rewrite. If logging of address rewriting is configured,
4959 this fails because the process is not running as root or exim. There may be
4960 a better way of dealing with this, but for the moment (because 4.30 needs
4961 to be released), I have disabled address rewrite logging when running a
4962 filter in a non-root, non-exim process.
4968 1. The buildconfig auxiliary program wasn't quoting the value set for
4969 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4970 not defined. This bug was masked in 4.22 by the effect that was fixed in
4973 2. Some messages that were rejected after a message id was allocated were
4974 shown as "incomplete" by exigrep. It no longer does this for messages that
4975 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4977 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4978 have allow_domain_literals set, the ID did not get logged in the <= line.
4979 Domain literals are now always recognized in Message-ID: header lines.
4981 4. The first argument for a ${extract expansion item is the key name or field
4982 number. Leading and trailing spaces in this item were not being ignored,
4983 causing some misleading effects.
4985 5. When deliver_drop_privilege was set, single queue runner processes started
4986 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4987 same command in the process it spins off) were not dropping privilege.
4989 6. When the daemon running as "exim" started a queue runner, it always
4990 re-executed Exim in the spun-off process. This is a waste of effort when
4991 deliver_drop_privilege is set. The new process now just calls the
4992 queue-runner function directly.
4998 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5001 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5002 ignored. Though the use of -oP was forcing the writing of a pid file, it
5003 was always written to the default place.
5005 3. If the message "no IP address found for host xxxx" is generated during
5006 incoming verification, it is now followed by identification of the incoming
5007 connection (so you can more easily find what provoked it).
5009 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
5011 5. Added some features to "harden" Exim a bit more against certain attacks:
5013 (a) There is now a build-time option called FIXED_NEVER_USERS that can
5014 be put in Local/Makefile. This is like the never_users runtime option,
5015 but it cannot be overridden. The default setting is "root".
5017 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
5018 prefix string with which any file named in a -C command line option
5021 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
5022 is retained for -C and -D only if the caller of Exim is root. Without
5023 it, the exim user may also use -C and -D and retain privilege.
5025 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
5026 command line option is disabled.
5028 6. Macro names set by the -D option must start with an upper case letter, just
5029 like macro names defined in the configuration file.
5031 7. Added "dereference=" facility to LDAP.
5033 8. Two instances of the typo "uknown" in the source files are fixed.
5035 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
5036 the Configure-Makefile script screwed up while processing it.
5038 10. Incorporated PCRE 4.4.
5040 11. The SMTP synchronization check was not operating right at the start of an
5041 SMTP session. For example, it could not catch a HELO sent before the client
5042 waited for the greeting. There is now a check for outstanding input at the
5043 point when the greeting is written. Because of the duplex, asynchronous
5044 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
5045 way, but not yet received, when the check is performed.
5047 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
5048 on TCP/IP sockets, because this apparently causes some broken clients to
5051 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
5052 unchanged) from the Cygwin maintainer.
5054 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
5055 was supported instead of testing for mbx. Effectively a typo.
5057 15. The spa authenticator server code was not checking that the input it
5058 received was valid base64.
5060 16. The debug output line for the "set" modifier in ACLs was not showing the
5061 name of the variable that was being set.
5063 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
5065 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
5066 lookup on its own. Something else was needed to trigger the lookup. For
5067 example, a match in host_lookup or the need for a host name in a host list.
5068 Now, if $sender_host_name is referenced and the host name has not yet been
5069 looked up, a lookup is performed. If the lookup fails, the variable remains
5070 empty, and $host_lookup_failed is set to "1".
5072 19. Added "eqi" as a case-independent comparison operator.
5074 20. The saslauthd authentication condition could segfault if neither service
5075 nor realm was specified.
5077 21. If an overflowing value such as "2048M" was set for message_size_limit, the
5078 error message that was logged was misleading, and incoming SMTP
5079 connections were dropped. The message is now more accurate, and temporary
5080 errors are given to SMTP connections.
5082 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
5083 (except RSET) with a 421 error, until QUIT is received. However, it was
5084 failing to send a response to QUIT.
5086 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
5087 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
5088 ACL is now run after the helo_try_verify_hosts code.
5090 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
5091 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
5092 case-independent, so other case variants are also recognized). Apparently
5093 some systems use these upper case variants.
5095 25. If more than two messages were waiting for the same host, and a transport
5096 filter was specified for the transport, Exim sent two messages over the
5097 same TCP/IP connection, and then failed with "socket operation on non-
5098 socket" when it tried to send the third.
5100 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
5102 27. The extern definition of crypt16() in expand.c was not being excluded when
5103 the OS had its own crypt16() function.
5105 28. Added bounce_return_body as a new option, and bounce_return_size_limit
5106 as a preferred synonym for return_size_limit, both as an option and as an
5109 29. Added LIBS=-liconv to OS/Makefile-OSF1.
5111 30. Changed the default configuration ACL to relax the local part checking rule
5112 for addresses that are not in any local domains. For these addresses,
5113 slashes and pipe symbols are allowed within local parts, but the sequence
5114 /../ is explicitly forbidden.
5116 31. SPA server authentication was not clearing the challenge buffer before
5119 32. log_message in a "warn" ACL statement was writing to the reject log as
5120 well as to the main log, which contradicts the documentation and doesn't
5121 seem right (because no rejection is happening). So I have stopped it.
5123 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
5124 However, I am unable to do any testing of this.
5126 34. Fixed an infelicity in the appendfile transport. When checking directories
5127 for a mailbox, to see if any needed to be created, it was accidentally
5128 using path names with one or more superfluous leading slashes; tracing
5129 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
5131 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
5132 contents are added to the log line that is written for every discarded
5133 recipient. (Previously a log_message setting was ignored.)
5135 36. The ${quote: operator now quotes the string if it is empty.
5137 37. The install script runs exim in order to find its version number. If for
5138 some reason other than non-existence or emptiness, which it checks, it
5139 could not run './exim', it was installing it with an empty version number,
5140 i.e. as "exim-". This error state is now caught, and the installation is
5143 38. An argument was missing from the function that creates an error message
5144 when Exim fails to connect to the socket for saslauthd authentication.
5145 This could cause Exim to crash, or give a corrupted message.
5147 39. Added isip, isip4, and isip6 to ${if conditions.
5149 40. The ACL variables $acl_xx are now saved with the message, and can be
5150 accessed later in routers, transports, and filters.
5152 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
5153 strings in the file are not string-expanded.
5155 42. If a MAIL command specified a SIZE value that was too large to fit into an
5156 int variable, the check against message_size_limit failed. Such values are
5157 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
5158 day this will have to be increased, but I don't think I want to be around
5159 when emails are that large.
5166 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
5167 iconv() is not standard in FreeBSD.
5169 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
5170 IPv6 enabled. The observed symptom was a segmentation fault on return from
5171 the function os_common_find_running_interfaces() in src/os.c.
5173 3. In the check_special_case() function in daemon.c I had used "errno" as an
5174 argument name, which causes warnings on some systems. This was basically a
5175 typo, since it was named "eno" in the comments!
5177 4. The code that waits for the clock to tick (at a resolution of some fraction
5178 of a second) so as to ensure message-id uniqueness was always waiting for
5179 at least one whole tick, when it could have waited for less. [This is
5180 almost certainly not relevant at current processor speeds, where it is
5181 unlikely to ever wait at all. But we try to future-proof.]
5183 5. The function that sleeps for a time interval that includes fractions of a
5184 second contained a race. It did not block SIGALRM between setting the
5185 timer, and suspending (a couple of lines later). If the interval was short
5186 and the sigsuspend() was delayed until after it had expired, the suspension
5187 never ended. On busy systems this could lead to processes getting stuck for
5190 6. Some uncommon configurations may cause a lookup to happen in a queue runner
5191 process, before it forks any delivery processes. The open lookup caching
5192 mechanism meant that the open file or database connection was passed into
5193 the delivery process. The problem was that delivery processes always tidy
5194 up cached lookup data. This could cause a problem for the next delivery
5195 process started by the queue runner, because the external queue runner
5196 process does not know about the closure. So the next delivery process
5197 still has data in the lookup cache. In the case of a file lookup, there was
5198 no problem because closing a file descriptor in a subprocess doesn't affect
5199 the parent. However, if the lookup was caching a connection to a database,
5200 the connection was closed, and the second delivery process was likely to
5201 see errors such as "PGSQL: query failed: server closed the connection
5202 unexpectedly". The problem has been fixed by closing all cached lookups
5203 in a queue runner before running a delivery process.
5205 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
5206 seem to have the "const" qualifier which it has on other OS. I've
5209 8. Change 4.21/2 was too strict. It is only if there are two authenticators
5210 *of the same type* (client or server) with the same public name that an
5211 error should be diagnosed.
5213 9. When Exim looked up a host name for an IP address, but failed to find the
5214 original IP address when looking up the host name (a safety check), it
5215 output the message "<ip address> does not match any IP for NULL", which was
5216 confusing, to say the least. The bug was that the host name should have
5217 appeared instead of "NULL".
5219 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
5220 user that is built into the binary, and the -C or -D options is used, root
5221 privilege is dropped before the configuration file is read. In addition,
5222 logging is switched to stderr instead of the normal log files. If the
5223 configuration then re-defines the Exim user, the unprivileged environment
5224 is probably not what is expected, so Exim logs a panic warning message (but
5227 However, if deliver_drop_privilege is set, the unprivileged state may well
5228 be exactly what is intended, so the warning has been cut out in that case,
5229 and Exim is allowed to try to write to its normal log files.
5235 1. smtp_return_error_details was not giving details for temporary sender
5236 or receiver verification errors.
5238 2. Diagnose a configuration error if two authenticators have the same public
5241 3. Exim used not to create the message log file for a message until the first
5242 delivery attempt. This could be confusing when incoming messages were held
5243 for policy or load reasons. The message log file is now created at the time
5244 the message is received, and an initial "Received" line is written to it.
5246 4. The automatically generated man page for command line options had a minor
5247 bug that caused no ill effects; however, a more serious problem was that
5248 the procedure for building the man page automatically didn't always
5249 operate. Consequently, release 4.20 contains an out-of-date version. This
5250 shouldn't happen again.
5252 5. When building Exim with embedded Perl support, the script that builds the
5253 Makefile was calling 'perl' to find its compile-time parameters, ignoring
5254 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
5256 6. The freeze_tell option was not being used for messages that were frozen on
5257 arrival, either by an ACL or by local_scan().
5259 7. Added the smtp_incomplete_transaction log selector.
5261 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
5262 was accepting AUTH without a new EHLO.
5264 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
5265 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
5268 10. Logging of TCP/IP connections (when configured) now happens in the main
5269 daemon process instead of the child process, so that the TCP/IP connection
5270 count is more accurate (but it can never be perfect).
5272 11. The use of "drop" in a nested ACL was not being handled correctly in the
5273 outer ACL. Now, if condition failure induced by the nested "drop" causes
5274 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
5275 or "require"), the connection is dropped.
5277 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
5278 that yield "discard" can now be used with an "accept" or a "discard" verb,
5279 but an error is generated for any others (because I can't see a useful way
5280 to define what should happen).
5282 13. When an ACL is read dynamically from a file (or anywhere else), the lines
5283 are now processed in the same way as lines in the Exim configuration file.
5284 In particular, continuation lines are supported.
5286 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
5288 15. Added -ti meaning -t -i.
5290 16. Check for letters, digits, hyphens, and dots in the names of dnslist
5291 domains, and warn by logging if others are found.
5293 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
5294 returned from GIFCONF when Exim is trying to find the list of interfaces on
5295 a host. The code in os.c has been modified to copy each ifreq to an aligned
5296 structure in all cases.
5298 Also, in some cases, the returned ifreq's were being copied to a 'struct
5299 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
5300 means the last couple of bytes of an IPv6 address could be chopped if the
5301 ifreq contained only a normal sockaddr (14 bytes storage).
5303 18. Named domain lists were not supported in the hosts_treat_as_local option.
5304 An entry such as +xxxx was not recognized, and was treated as a literal
5307 19. Ensure that header lines added by a DATA ACL are included in the reject log
5308 if the ACL subsequently rejects the message.
5310 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
5311 MD5 (which is deprecated).
5313 21. When testing a filter file using -bf, Exim was writing a message when it
5314 took the sender from a "From " line in the message, but it was not doing so
5315 when it took $return_path from a Return-Path: header line. It now does.
5317 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5318 with a valid header line field name (a series of printing characters
5319 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5321 23. Changed "disc" in the source to "disk" to conform to the documentation and
5322 the book and for uniformity.
5324 24. Ignore Sendmail's -Ooption=value command line item.
5326 25. When execve() failed while trying to run a command in a pipe transport,
5327 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5328 could be confused with a return value of 69 from the command itself. This
5329 has been changed to 127, the value the shell returns if it is asked to run
5330 a non-existent command. The wording for the related log line suggests a
5331 non-existent command as the problem.
5333 26. If received_header_text expands to an empty string, do not add a Received:
5334 header line to the message. (Well, it adds a token one on the spool, but
5335 marks it "old" so that it doesn't get used or transmitted.)
5337 27. Installed eximstats 1.28 (addition of -nt option).
5339 28. There was no check for failure on the call to getsockname() in the daemon
5340 code. This can fail if there is a shortage of resources on the system, with
5341 ENOMEM, for example. A temporary error is now given on failure.
5343 29. Contrary to the C standard, it seems that in some environments, the
5344 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5345 program. Exim now does this explicitly; it affects the formatting of
5346 timestamps using strftime().
5348 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5349 complaints. I've now initialized all the variables, to avoid this.
5351 32. Header lines added by a system filter were not being "seen" during
5352 transport-time rewrites.
5354 33. The info_callback() function passed to OpenSSL is set up with type void
5355 (*)(SSL *, int, int), as described somewhere. However, when calling the
5356 function (actually a macro) that sets it up, the type void(*)() is
5357 expected. I've put in a cast to prevent warnings from picky compilers.
5359 34. If a DNS black list lookup found a CNAME record, but there were no A
5360 records associated with the domain it pointed at, Exim crashed.
5362 35. If a DNS black list lookup returned more than one A record, Exim ignored
5363 all but the first. It now scans all returned addresses if a particular IP
5364 value is being sought. In this situation, the contents of the
5365 $dnslist_value variable are a list of all the addresses, separated by a
5368 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5369 to accept a host name and all its aliases if the forward lookup for any of
5370 them yielded the IP address of the incoming connection. Now it accepts only
5371 those names whose forward lookup yields the correct IP address. Any other
5372 names are discarded. This closes a loophole whereby a rogue DNS
5373 administrator could create reverse DNS records to break through a
5374 wildcarded host restriction in an ACL.
5376 37. If a user filter or a system filter that ran in a subprocess used any of
5377 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5378 the wrong values were passed to the pipe command ($thisaddress had the
5379 value of $0, $0 had the value of $1, etc). This bug was introduced by
5380 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5382 38. Improved the line breaking for long SMTP error messages from ACLs.
5383 Previously, if there was no break point between 40 and 75 characters, Exim
5384 left the rest of the message alone. Two changes have been made: (a) I've
5385 reduced the minimum length to 35 characters; (b) if it can't find a break
5386 point between 35 and 75 characters, it looks ahead and uses the first one
5387 that it finds. This may give the occasional overlong line, but at least the
5388 remaining text gets split now.
5390 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5391 file descriptors might be low, and that setting 1000 would always raise it.
5392 It turns out that in some environments, the limit is already over 1000 and
5393 that lowering it causes trouble. So now Exim takes care not to decrease it.
5395 40. When delivering a message, the value of $return_path is set to $sender_
5396 address at the start of routing (routers may change the value). By an
5397 oversight, this default was not being set up when an address was tested by
5398 -bt or -bv, which affected the outcome if any router or filter referred to
5401 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5402 log only when "message" or "log_message" are set. However, if one of the
5403 conditions was an address verification, or a call to a nested ACL, the
5404 messages generated by the underlying test were being passed through. This
5405 no longer happens. The underlying message is available in $acl_verify_
5406 message for both "message" and "log_message" expansions, so it can be
5407 passed through if needed.
5409 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5410 new expansion $bh_ to give the encoded byte string without charset
5411 translation. Translation happens only if iconv() is available; HAVE_ICONV
5412 indicates this at build time. HEADERS_CHARSET gives the charset to
5413 translate to; headers_charset can change it in the configuration, and
5414 "headers charset" can change it in an individual filter file.
5416 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5417 that creates RFC 2047 encoded "words" labels them as that charset instead
5418 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5419 expansion operator; (ii) when Exim creates a From: line for a local
5420 message; (iii) when a header line is rewritten to include a "phrase" part.
5422 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5423 buggy, causing it to skip the first lines of messages whose message ID
5424 ended in 'D'. This would not have bitten before Exim release 4.14, because
5425 message IDs were unlikely to end in 'D' before then. The effect was to have
5426 incorrect size information for certain domains.
5428 45. #include "config.h" was missing at the start of the crypt16.c module. This
5429 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5432 46. If there was a timeout during a "random" callout check, Exim treated it as
5433 a failure of the random address, and carried on sending RSET and the real
5434 address. If the delay was just some slowness somewhere, the response to the
5435 original RCPT would be taken as a response to RSET and so on, causing
5436 mayhem of various kinds.
5438 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5439 when I implemented it. It didn't allow for the fact that some option values
5440 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5441 the right test for positive values.
5443 48. Domain names in DNS records are case-independent. Exim always looks them up
5444 in lower case. Some resolvers return domain names in exactly the case they
5445 appear in the zone file, that is, they may contain uppercase letters. Not
5446 all resolvers do this - some return always lower case. Exim was treating a
5447 change of case by a resolver as a change of domain, similar to a widening
5448 of a domain abbreviation. This triggered its re-routing code and so it was
5449 trying to route what was effectively the same domain again. This normally
5450 caused routing to fail (because the router wouldn't handle the domain
5451 twice). Now Exim checks for this case specially, and just changes the
5452 casing of the domain that it ultimately uses when it transmits the message
5455 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5458 50. If a filter generated a file delivery with a non-absolute name (possible if
5459 no home directory exists for the router), the forbid_file option was not
5462 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5463 the existing equality matching.
5465 52. Exim was using ints instead of ino_t variables in some places where it was
5466 dealing with inode numbers.
5468 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5469 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5470 called TMPDIR, and if it finds it is different, it changes its value.
5472 54. The smtp_printf() function is now made available to local_scan() so
5473 additional output lines can be written before returning. There is also an
5474 smtp_fflush() function to enable the detection of a dropped connection.
5475 The variables smtp_input and smtp_batched_input are exported to
5478 55. Changed the default runtime configuration: the message "Unknown user"
5479 has been removed from the ACL, and instead placed on the localuser router,
5480 using the cannot_route_message feature. This means that any verification
5481 failures that generate their own messages won't get overridden. Similarly,
5482 the "Unrouteable address" message that was in the ACL for unverifiable
5483 relay addresses has also been removed.
5485 56. Added hosts_avoid_esmtp to the smtp transport.
5487 57. The exicyclog script was not checking for the esoteric option
5488 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5489 will work only if exicyclog is run under the appropriate euid.
5491 58. Following a discussion on the list, the rules by which Exim recognises line
5492 endings on incoming messages have been changed. The -dropcr and drop_cr
5493 options are now no-ops, retained only for backwards compatibility. The
5494 following line terminators are recognized: LF CRLF CR. However, special
5495 processing applies to CR:
5497 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5498 nor a local message in the state where . is a terminator.
5500 (ii) If a bare CR is encountered in a header line, an extra space is added
5501 after the line terminator so as not to end the header. The reasoning
5502 behind this is that bare CRs in header lines are most likely either
5503 to be mistakes, or people trying to play silly games.
5505 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5506 was being incorrectly given as 18 bytes larger than it should have been.
5507 This is a VOB (very old bug).
5509 60. This may never have affected anything current, but just in case it has:
5510 When the local host is found other than at the start of a list of hosts,
5511 the local host, those with the same MX, and any that follow, are discarded.
5512 When the list in question was part of a longer list of hosts, the following
5513 hosts (not currently being processed) were also being discarded. This no
5514 longer happens. I'm not sure if this situation could ever has previously
5517 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5520 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5521 header. This is something that is recommended in a new Internet Draft, and
5522 is something that is documented as being done by Sendmail. There are two
5523 possible values. For messages generated by the autoreply transport, Exim
5526 Auto-Submitted: auto-replied
5528 whereas for all other generated messages (e.g. bounces) it adds
5530 Auto-Submitted: auto-generated
5532 63. The "personal" condition in filters now includes a test for the
5533 Auto-Submitted: header. If it contains the string "auto-" the message it
5534 not considered personal.
5536 64. Added rcpt_include_affixes as a generic transport option.
5538 65. Added queue_only_override (default true).
5540 66. Added the syslog_duplication option.
5542 67. If what should have been the first header line of a message consisted of
5543 a space followed by a colon, Exim was mis-interpreting it as a header line.
5544 It isn't of course - it is syntactically invalid and should therefore be
5545 treated as the start of the message body. The misbehaviour could have
5546 caused a number of strange effects, including loss of data in subsequent
5547 header lines, and spool format errors.
5549 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5550 client host had authenticated. This control can now be exercised by an ACL
5551 for more flexibility.
5553 69. By default, callouts do not happen when testing with -bh. There is now a
5554 variant, -bhc, which does actually run the callout code, including
5555 consulting and updating the callout cache.
5557 70. Added support for saslauthd authentication, courtesy of Alexander
5560 71. If statvfs() failed on the spool or log directories while checking their
5561 size for availability, Exim confusingly gave the error "space shortage".
5562 Furthermore, in debugging mode it crashed with a floating point exception.
5563 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5564 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5565 serious problem, Exim now writes to the main and panic logs when this
5566 happens, with details of the failure. It then refuses to accept the
5567 incoming message, giving the message "spool directory problem" or "log
5568 directory problem" with a 421 code for SMTP messages.
5570 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5571 0, 1, and 2 exist, because some OS complain for execs without them (see
5572 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5573 descriptors. However, the code omitted to check that the open succeeded,
5574 causing mysterious errors if for some reason the permissions on /dev/null
5575 got screwed. Now Exim writes a message to the main and panic logs, and
5576 bombs out if it can't open /dev/null.
5578 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5579 interact so that it is all more flexible. It is supposed to remain
5580 backwards compatible. Also added extra_local_interfaces.
5582 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5583 to bomb out with an assertion failure - to the client this appears as a
5584 connection drop. This problem occurs in the part of the code that was taken
5585 from the Samba project. Fortunately, the assertion is in a very simple
5586 function, so I have fixed this by reproducing the function inline in the
5587 one place where it is called, and arranging for authentication to fail
5588 instead of killing the process with assert().
5590 75. The SPA client code was not working when the server requested OEM rather
5591 than Unicode encoding.
5593 76. Added code to make require_files with a specific uid setting more usable in
5594 the case where statting the file as root fails - usually a non-root-mounted
5595 NFS file system. When this happens and the failure is EACCES, Exim now
5596 forks a subprocess and does the per-uid checking as the relevant uid.
5598 77. Added process_log_path.
5600 78. If log_file_path was not explicitly set, a setting of check_log_space or
5601 check_log_inodes was ignored.
5603 79. If a space check for the spool or log partitions fails, the incident is now
5604 logged. Of course, in the latter case the data may get lost...
5606 80. Added the %p formatting code to string_format() so that it can be used to
5607 print addresses in debug_print(). Adjusted all the address printing in the
5608 debugging in store.c to use %p rather than %d.
5610 81. There was a concern that a line of code in smtp_in.c could overflow a
5611 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5612 initially expressed, the concern was not well-founded, because trailing
5613 spaces are removed early. However, if the trailing spaces were followed by
5614 a NULL, they did not get removed, so the overflow was possible. Two fixes
5617 (a) I re-wrote the offending code in a cleaner fashion.
5618 (b) If an incoming SMTP command contains a NULL character, it is rejected
5621 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5622 runs initgroups(), so that if the Exim user is in any additional groups,
5623 they will be used during message reception.
5629 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob