1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.632 2010/06/12 15:21:25 jetmore Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 NM/01 Workround for PCRE version dependency in version reporting
12 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
13 This fixes portability to compilers other than gcc, notably
14 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
16 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
17 makefiles for portability to HP-UX and POSIX correctness.
19 PP/01 Permit LOOKUP_foo enabling on the make command-line.
20 Also via indented variable definition in the Makefile.
21 (Debugging by Oliver Heesakkers).
23 PP/02 Restore caching of spamd results with expanded spamd_address.
24 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
26 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
27 Improves build reliability. Fix from: Frank Elsner
29 NM/02 Fix wide character breakage in the rfc2047 coding
30 Fixes bug 1064. Patch from Andrey N. Oktyabrski
32 NM/03 Allow underscore in dnslist lookups
33 Fixes bug 1026. Patch from Graeme Fowler
35 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
36 Code patches from Adam Ciarcinski of NetBSD.
38 NM/04 Fixed exiqgrep to cope with mailq missing size issue
41 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
42 is logged, to avoid truncation. Patch from John Horne.
44 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
45 Patch from Jakob Hirsch.
47 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
48 SQL string expansion failure details.
49 Patch from Andrey Oktyabrski.
51 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
52 Patch from Simon Arlott.
54 PP/09 New lookups functionality failed to compile on old gcc which rejects
55 extern declarations in function scope.
56 Patch from Oliver Fleischmann
58 PP/10 Use sig_atomic_t for flags set from signal handlers.
59 Check getgroups() return and improve debugging.
60 Fixed developed for diagnosis in bug 927 (which turned out to be
63 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
64 Patch from Mark Zealey.
66 PP/12 Bugzilla 1056: Improved spamd server selection.
67 Patch from Mark Zealey.
69 PP/13 Bugzilla 1086: Deal with maildir quota file races.
70 Based on patch from Heiko Schlittermann.
72 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
73 Patch from Uwe Doering, sign-off by Michael Haardt.
75 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
76 variable declaration deep within a block. Bug and patch from
79 PP/15 lookups-Makefile IRIX compatibilty coercion.
81 PP/16 Make DISABLE_DKIM build knob functional.
83 NM/06 Bugzilla 698: child_open_uid: restore default SIGPIPE handler
89 TF/01 Failure to get a lock on a hints database can have serious
90 consequences so log it to the panic log.
92 TF/02 Log LMTP confirmation messages in the same way as SMTP,
93 controlled using the smtp_confirmation log selector.
95 TF/03 Include the error message when we fail to unlink a spool file.
97 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
98 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
99 for maintaining out-of-tree patches for some time.
101 PP/01 Bugzilla 139: Documentation and portability issues.
102 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
103 Handle per-OS dynamic-module compilation flags.
105 PP/02 Let /dev/null have normal permissions.
106 The 4.73 fixes were a little too stringent and complained about the
107 permissions on /dev/null. Exempt it from some checks.
108 Reported by Andreas M. Kirchwitz.
110 PP/03 Report version information for many libraries, including
111 Exim version information for dynamically loaded libraries. Created
112 version.h, now support a version extension string for distributors
113 who patch heavily. Dynamic module ABI change.
115 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
116 privilege escalation vulnerability whereby the Exim run-time user
117 can cause root to append content of the attacker's choosing to
120 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
123 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
124 If dropping privileges for untrusted macros, we disabled normal logging
125 on the basis that it would fail; for the Exim run-time user, this is not
126 the case, and it resulted in successful deliveries going unlogged.
127 Fixed. Reported by Andreas Metzler.
133 PP/01 Date: & Message-Id: revert to normally being appended to a message,
134 only prepend for the Resent-* case. Fixes regression introduced in
135 Exim 4.70 by NM/22 for Bugzilla 607.
137 PP/02 Include check_rfc2047_length in configure.default because we're seeing
138 increasing numbers of administrators be bitten by this.
140 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
142 PP/03 Bugzilla 994: added openssl_options main configuration option.
144 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
146 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
148 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
150 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
151 without a peer certificate, leading to a segfault because of an
152 assumption that peers always have certificates. Be a little more
153 paranoid. Problem reported by Martin Tscholak.
155 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
156 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
157 NB: ClamAV planning to remove STREAM in "middle of 2010".
158 CL also introduces -bmalware, various -d+acl logging additions and
159 more caution in buffer sizes.
161 PP/09 Implemented reverse_ip expansion operator.
163 PP/10 Bugzilla 937: provide a "debug" ACL control.
165 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
167 PP/12 Bugzilla 973: Implement --version.
169 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
171 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
173 PP/15 Bugzilla 816: support multiple condition rules on Routers.
175 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
176 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
177 ignore trailing whitespace.
179 JJ/02 prevent non-panic DKIM error from being sent to paniclog
181 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
184 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
185 Notification from Dr Andrew Aitchison.
187 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
188 ExtendedDetectionInfo response format.
189 Notification from John Horne.
191 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
194 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
195 XSL and documented dependency on system catalogs, with examples of how
198 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
201 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
202 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
203 configuration file which is writeable by the Exim user or group.
205 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
206 of configuration files to cover files specified with the -C option if
207 they are going to be used with root privileges, not just the default
210 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
211 option (effectively making it always true).
213 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
214 files to be used while preserving root privileges.
216 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
217 that rogue child processes cannot use them.
219 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
220 run-time user, instead of root.
222 PP/28 Add WHITELIST_D_MACROS option to let some macros be overriden by the
223 Exim run-time user without dropping privileges.
225 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
226 result string, instead of calling string_vformat() twice with the same
229 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
230 for other users. Others should always drop root privileges if they use
231 -C on the command line, even for a whitelisted configure file.
233 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
235 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
241 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
242 $data_path, and $header_path variables; fixed documentation bugs and
245 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
246 exipick to access non-standard spools, including the "frozen" queue
249 NM/01 Bugzilla 965: Support mysql stored procedures.
250 Patch from Alain Williams
252 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
254 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
255 Patch from Andreas Metzler
257 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
258 Patch from Kirill Miazine
260 NM/05 Bugzilla 671: Added umask to procmail example.
262 JJ/03 installed exipick 20100323.0, fixing doc bug
264 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
265 directory. Notification and patch from Dan Rosenberg.
267 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
269 TK/02 Improve log output when DKIM signing operation fails.
271 MH/01 Treat the transport option dkim_domain as a colon separated
272 list, not as a single string, and sign the message with each element,
273 omitting multiple occurences of the same signer.
275 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
276 Bugzilla 985, 986. Patch by Simon Arlott
278 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
279 Patch by Simon Arlott
281 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
282 MBX locking. Notification from Dan Rosenberg.
288 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
290 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
292 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
294 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
296 NM/04 Bugzilla 915: Flag broken perl installation during build.
302 TK/01 Added patch by Johannes Berg that expands the main option
303 "spamd_address" if it starts with a dollar sign.
305 TK/02 Write list of recipients to X-Envelope-Sender header when building
306 the mbox-format spool file for content scanning (suggested by Jakob
309 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
310 (http://www.dcc-servers.net/) support via dccifd. Activated by
311 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
313 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
314 by Mark Daniel Reidel <mr@df.eu>.
316 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
317 When building exim an external PCRE library is now needed -
318 PCRE is a system library on the majority of modern systems.
319 See entry on PCRE_LIBS in EDITME file.
321 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
322 conversation. Added nologin parameter to request.
323 Patch contributed by Kirill Miazine.
325 TF/01 Do not log submission mode rewrites if they do not change the address.
327 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
329 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
330 log files in place. Contributed by Roberto Lima.
332 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
334 TF/03 Bugzilla 615: When checking the local_parts router precondition
335 after a local_part_suffix or local_part_prefix option, Exim now
336 does not use the address's named list lookup cache, since this
337 contains cached lookups for the whole local part.
339 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
340 Robert Millan. Documentation is in experimental-spec.txt.
342 TF/04 Bugzilla 668: Fix parallel build (make -j).
344 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
346 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
347 Patch provided by Jan Srzednicki.
349 TF/05 Leading white space used to be stripped from $spam_report which
350 wrecked the formatting. Now it is preserved.
352 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
353 that they are available at delivery time.
355 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
357 TF/08 TLS error reporting now respects the incoming_interface and
358 incoming_port log selectors.
360 TF/09 Produce a more useful error message if an SMTP transport's hosts
361 setting expands to an empty string.
363 NM/06 Bugzilla 744: EXPN did not work under TLS.
364 Patch provided by Phil Pennock.
366 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
367 Patch provided by Richard Godbee.
369 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
370 acl_smtp_notquit, added index entry.
372 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
373 Patch provided by Eugene Bujak.
375 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
376 accept(). Patch provided by Maxim Dounin.
378 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
379 Patch provided by Phil Pennock.
381 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
383 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
384 Patch provided by Brad "anomie" Jorsch.
386 NM/14 Bugzilla 622: Added timeout setting to transport filter.
387 Patch provided by Dean Brooks.
389 TK/05 Add native DKIM support (does not depend on external libraries).
391 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
392 Patch provided by Graeme Fowler.
394 NM/16 Bugzilla 851: Documentation example syntax fix.
396 NM/17 Changed NOTICE file to remove references to embedded PCRE.
398 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
401 NM/19 Bugzilla 745: TLS version reporting.
402 Patch provided by Phil Pennock.
404 NM/20 Bugzilla 167: bool: condition support.
405 Patch provided by Phil Pennock.
407 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
408 clients. Patch provided by Phil Pennock.
410 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
411 Patch provided by Brad "anomie" Jorsch.
413 NM/23 Bugzilla 687: Fix misparses in eximstats.
414 Patch provided by Heiko Schlittermann.
416 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
417 Patch provided by Heiko Schlittermann.
419 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
420 plus update to original patch.
422 NM/26 Bugzilla 799: Documentation correction for ratelimit.
424 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
425 Patch provided by David Brownlee.
427 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
429 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
431 NM/30 Bugzilla 888: TLS documentation bugfixes.
433 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
435 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
436 Unlike the original bugzilla I have changed all shell scripts in src tree.
438 NM/33 Bugzilla 898: Transport filter timeout fix.
439 Patch by Todd Rinaldo.
441 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches.
442 Patch by Serge Demonchaux.
444 NM/35 Bugzilla 39: Base64 decode bug fixes.
445 Patch by Jakob Hirsch.
447 NM/36 Bugzilla 909: Correct connect() call in dcc code.
449 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
451 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
453 NM/39 Bugzilla 911: Fixed MakeLinks build script.
459 TK/01 Add preliminary DKIM support. Currently requires a forked version of
460 ALT-N's libdkim that I have put here:
461 http://duncanthrax.net/exim-experimental/
463 Note to Michael Haardt: I had to rename some vars in sieve.c. They
464 were called 'true' and it seems that C99 defines that as a reserved
465 keyword to be used with 'bool' variable types. That means you could
466 not include C99-style headers which use bools without triggering
467 build errors in sieve.c.
469 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
470 as mailq or other aliases. Changed the --help handling significantly
471 to do whats expected. exim_usage() emits usage/help information.
473 SC/01 Added the -bylocaldomain option to eximstats.
475 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
477 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
479 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
485 PH/01 Another patch from the Sieve maintainer.
487 PH/02 When an IPv6 address is converted to a string for single-key lookup
488 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
489 dots are used instead of colons so that keys in lsearch files need not
490 contain colons. This was done some time before quoting was made available
491 in lsearch files. However, iplsearch files do require colons in IPv6 keys
492 (notated using the quote facility) so as to distinguish them from IPv4
493 keys. This meant that lookups for IP addresses in host lists did not work
494 for iplsearch lookups.
496 This has been fixed by arranging for IPv6 addresses to be expressed with
497 colons if the lookup type is iplsearch. This is not incompatible, because
498 previously such lookups could never work.
500 The situation is now rather anomolous, since one *can* have colons in
501 ordinary lsearch keys. However, making the change in all cases is
502 incompatible and would probably break a number of configurations.
504 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
507 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
508 conversion specification without a maximum field width, thereby enabling
509 a rogue spamd server to cause a buffer overflow. While nobody in their
510 right mind would setup Exim to query an untrusted spamd server, an
511 attacker that gains access to a server running spamd could potentially
512 exploit this vulnerability to run arbitrary code as the Exim user.
514 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
515 $primary_hostname instead of what libspf2 thinks the hosts name is.
517 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
518 a directory entry by the name of the lookup key. Previously, if a
519 symlink pointed to a non-existing file or a file in a directory that
520 Exim lacked permissions to read, a lookup for a key matching that
521 symlink would fail. Now it is enough that a matching directory entry
522 exists, symlink or not. (Bugzilla 503.)
524 PH/03 The body_linecount and body_zerocount variables are now exported in the
527 PH/04 Added the $dnslist_matched variable.
529 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
530 This means they are set thereafter only if the connection becomes
533 PH/06 Added the client_condition to authenticators so that some can be skipped
534 by clients under certain conditions.
536 PH/07 The error message for a badly-placed control=no_multiline_responses left
537 "_responses" off the end of the name.
539 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
541 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
542 (without spaces) instead of just copying the configuration text.
544 PH/10 Added the /noupdate option to the ratelimit ACL condition.
546 PH/11 Added $max_received_linelength.
548 PH/12 Added +ignore_defer and +include_defer to host lists.
550 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
551 way in which PCRE > 7.0 is built.
553 PH/14 Implemented queue_only_load_latch.
555 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
556 MAIL command. The effect was to mangle the value on 64-bit systems.
558 PH/16 Another patch from the Sieve maintainer.
560 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
562 PH/18 If a system quota error occurred while trying to create the file for
563 a maildir delivery, the message "Mailbox is full" was not appended to the
564 bounce if the delivery eventually timed out. Change 4.67/27 below applied
565 only to a quota excession during the actual writing of the file.
567 PH/19 It seems that peer DN values may contain newlines (and other non-printing
568 characters?) which causes problems in log lines. The DN values are now
569 passed through string_printing() before being added to log lines.
571 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
572 and InterBase are left for another time.)
574 PH/21 Added message_body_newlines option.
576 PH/22 Guard against possible overflow in moan_check_errorcopy().
578 PH/23 POSIX allows open() to be a macro; guard against that.
580 PH/24 If the recipient of an error message contained an @ in the local part
581 (suitably quoted, of course), incorrect values were put in $domain and
582 $local_part during the evaluation of errors_copy.
588 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
589 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
592 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
593 issue a MAIL command.
595 PH/02 In an ACL statement such as
597 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
599 if a client was not listed at all, or was listed with a value other than
600 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
601 the condition was not true (as it should be), so access was not denied.
602 The bug was that the ! inversion was incorrectly passed on to the second
603 item. This has been fixed.
605 PH/03 Added additional dnslists conditions == and =& which are different from
606 = and & when the dns lookup returns more than one IP address.
608 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
609 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
611 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
612 FSYNC, which compiles an option called disable_fsync that allows for
613 bypassing fsync(). The documentation is heavily laced with warnings.
615 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
617 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
618 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
619 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
620 including adding "make clean"; (3) Added -fPIC when compiling the test
621 dynamically loaded module, to get rid of a warning.
623 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
624 message fails, move_frozen_messages = true and ignore_bounce_errors_after
625 = 0s. The bug is otherwise harmless.
627 PH/07 There was a bug in the dovecot authenticator such that the value of
628 $auth1 could be overwritten, and so not correctly preserved, after a
629 successful authentication. This usually meant that the value preserved by
630 the server_setid option was incorrect.
632 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
634 PH/09 Installed PCRE release 7.0.
636 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
637 run for batched SMTP input. It is now run at the start of every message
638 in the batch. While fixing this I discovered that the process information
639 (output by running exiwhat) was not always getting set for -bs and -bS
640 input. This is fixed, and it now also says "batched" for BSMTP.
642 PH/11 Added control=no_pipelining.
644 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
645 patch, slightly modified), and move the expansion of helo_data till after
646 the connection is made in the smtp transport (so it can use these
649 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
651 PH/14 Added log_selector = +pid.
653 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
655 PH/16 Add ${if forany and ${if forall.
657 PH/17 Added dsn_from option to vary the From: line in DSNs.
659 PH/18 Flush SMTP output before performing a callout, unless control =
660 no_callout_flush is set.
662 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
663 was true (the default) a successful delivery failed to delete the retry
664 item, thus causing premature timeout of the address. The bug is now
667 PH/20 Added hosts_avoid_pipelining to the smtp transport.
669 PH/21 Long custom messages for fakedefer and fakereject are now split up
670 into multiline reponses in the same way that messages for "deny" and
671 other ACL rejections are.
673 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
674 with slight modification.
676 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
677 draft, changing the syntax and factoring some duplicate code".
679 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
680 for deliveries of the second and subsequent messages over the same SMTP
683 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
684 ${reduce, with only minor "tidies".
686 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
688 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
689 expansion side effects.
691 PH/27 When a message times out after an over-quota error from an Exim-imposed
692 quota, the bounce message says "mailbox is full". This message was not
693 being given when it was a system quota that was exceeded. It now should
696 MH/03 Made $recipients available in local_scan(). local_scan() already has
697 better access to the recipient list through recipients_list[], but
698 $recipients can be useful in postmaster-provided expansion strings.
700 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
701 in the case of a MAIL command with additional options following the
702 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
703 were accidentally chopped off.
705 PH/29 SMTP synchronization checks are implemented when a command is read -
706 there is a check that no more input is waiting when there shouldn't be
707 any. However, for some commands, a delay in an ACL can mean that it is
708 some time before the response is written. In this time, more input might
709 arrive, invalidly. So now there are extra checks after an ACL has run for
710 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
711 pipelining has not been advertised.
713 PH/30 MH's patch to allow iscntrl() characters to be list separators.
715 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
716 returned in the SMTP response when smtp_return_error_details was false.
719 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
720 instead of the C I/O that was originally supplied, because problems were
723 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
724 Exim which did not show up earlier: it was assuming that a call to
725 SSL_CTX_set_info_callback() might give an error value. In fact, there is
726 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
727 was a macro that became an assignment, so it seemed to work. This has
728 changed to a proper function call with a void return, hence the compile
729 error. Exim's code has been fixed.
731 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
734 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
736 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
738 PH/37 If a message is not accepted after it has had an id assigned (e.g.
739 because it turns out to be too big or there is a timeout) there is no
740 "Completed" line in the log. When some messages of this type were
741 selected by exigrep, they were listed as "not completed". Others were
742 picked up by some special patterns. I have improved the selection
743 criteria to be more general.
745 PH/38 The host_find_failed option in the manualroute router can now be set
746 to "ignore", to completely ignore a host whose IP address cannot be
747 found. If all hosts are ignored, the behaviour is controlled by the new
748 host_all_ignored option.
750 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
751 homing or because of multiple MX records with /mx) generated more than
752 one IP address, and the following item turned out to be the local host,
753 all the secondary addresses of the first item were incorrectly removed
754 from the list, along with the local host and any following hosts (which
755 is what is supposed to happen).
757 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
758 whoever called Exim into the -H file. In the case of the daemon it was
759 behaving confusingly. When first started, it used values for whoever
760 started the daemon, but after a SIGHUP it used the Exim user (because it
761 calls itself on a restart). I have changed the code so that it now always
764 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
765 message are rejected with the same error (e.g. no authentication or bad
766 sender address), and a DATA command is nevertheless sent (as can happen
767 with PIPELINING or a stupid MUA), the error message that was given to the
768 RCPT commands is included in the rejection of the DATA command. This is
769 intended to be helpful for MUAs that show only the final error to their
772 PH/42 Another patch from the Sieve maintainer.
774 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
775 Eximstats - Fixed some broken HTML links and added missing column headers
777 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
778 columns for Rejects, Temp Rejects, Ham, and Spam rows.
780 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
782 PH/43 Yet another patch from the Sieve maintainer.
784 PH/44 I found a way to check for a TCP/IP connection going away before sending
785 the response to the final '.' that terminates a message, but only in the
786 case where the client has not sent further data following the '.'
787 (unfortunately, this is allowed). However, in many cases there won't be
788 any further data because there won't be any more messages to send. A call
789 to select() can be used: if it shows that the input is "ready", there is
790 either input waiting, or the socket has been closed. An attempt to read
791 the next input character can distinguish the two cases. Previously, Exim
792 would have sent an OK response which the client would never have see.
793 This could lead to message repetition. This fix should cure that, at
794 least in a lot of common cases.
796 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
797 advertised in response to EHLO.
803 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
804 fixed by 4.65/MH/01 (is this a record?) are fixed:
806 (i) An empty string was always treated as zero by the numeric comparison
807 operators. This behaviour has been restored.
809 (ii) It is documented that the numeric comparison operators always treat
810 their arguments as decimal numbers. This was broken in that numbers
811 starting with 0 were being interpreted as octal.
813 While fixing these problems I realized that there was another issue that
814 hadn't been noticed. Values of message_size_limit (both the global option
815 and the transport option) were treated as octal if they started with 0.
816 The documentation was vague. These values are now always treated as
817 decimal, and I will make that clear in the documentation.
823 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
824 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
827 MH/01 Don't check that the operands of numeric comparison operators are
828 integers when their expansion is in "skipping" mode (fixes bug
829 introduced by 4.64-PH/07).
831 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
832 child addresses, Exim now panics and dies. Previously, because the count
833 is held in a short int, deliveries were likely to be lost. As such a
834 large number of recipients for a single message is ridiculous
835 (performance will be very, very poor), I have chosen to impose a limit
836 rather than extend the field.
842 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
843 leftover -K file (the existence of which was triggered by #402).
844 While we were at it, introduced process PID as part of the -K
845 filename. This should rule out race conditions when creating
848 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
849 processing considerably. Previous code took too long for large mails,
850 triggering a timeout which in turn triggers #401.
852 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
853 in the DK code in transports.c. sendfile() is not really portable,
854 hence the _LINUX specificness.
856 TF/01 In the add_headers option to the mail command in an Exim filter,
857 there was a bug that Exim would claim a syntax error in any
858 header after the first one which had an odd number of characters
861 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
862 callout verification, Exim cached a "reject" for the entire domain. This
863 is correct for most verifications, but it is not correct for a recipient
864 verification with use_sender or use_postmaster set, because in that case
865 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
866 case of MAIL FROM:<> rejection from other early rejections (e.g.
867 rejection of HELO). When verifying a recipient using a non-null MAIL
868 address, the cache is ignored if it shows MAIL FROM:<> rejection.
869 Whatever the result of the callout, the value of the domain cache is
870 left unchanged (for any other kind of callout, getting as far as trying
871 RCPT means that the domain itself is ok).
873 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
876 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
877 manifest itself as EPIPE rather than ECONNECT. When tidying away a
878 session, the daemon ignores ECONNECT errors and logs others; it now
879 ignores EPIPE as well.
881 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
882 (quoted-printable decoding).
884 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
885 later the small subsequent patch to fix an introduced bug.
887 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
889 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
891 PH/08 An error is now given if message_size_limit is specified negative.
893 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
894 to be given (somewhat) arbitrary names.
896 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
899 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
900 miscellaneous code fixes
902 PH/10 Added the log_reject_target ACL modifier to specify where to log
905 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
906 hostname. This is wrong, because it relates to the incoming message (and
907 probably the interface on which it is arriving) and not to the outgoing
908 callout (which could be using a different interface). This has been
909 changed to use the value of the helo_data option from the smtp transport
910 instead - this is what is used when a message is actually being sent. If
911 there is no remote transport (possible with a router that sets up host
912 addresses), $smtp_active_hostname is used.
914 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
915 tweaks were necessary in order to get it to work (see also 21 below):
916 (a) The code assumed that strncpy() returns a negative number on buffer
917 overflow, which isn't the case. Replaced with Exim's string_format()
919 (b) There were several signed/unsigned issues. I just did the minimum
920 hacking in of casts. There is scope for a larger refactoring.
921 (c) The code used strcasecmp() which is not a standard C function.
922 Replaced with Exim's strcmpic() function.
923 (d) The code set only $1; it now sets $auth1 as well.
924 (e) A simple test gave the error "authentication client didn't specify
925 service in request". It would seem that Dovecot has changed its
926 interface. Fortunately there's a specification; I followed it and
927 changed what the client sends and it appears to be working now.
929 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
932 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
933 address A is aliased to B and C, where B exists and C does not. Without
934 -v the output is "A verified" because verification stops after a
935 successful redirection if more than one address is generated. However,
936 with -v the child addresses are also verified. Exim was outputting "A
937 failed to verify" and then showing the successful verification for C,
938 with its parentage. It now outputs "B failed to verify", showing B's
939 parentage before showing the successful verification of C.
941 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
942 look up a TXT record in a specific list after matching in a combined
945 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
946 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
947 they consult the DNS. I had assumed they would set it the way they
948 wanted; and indeed my experiments on Linux seem to show that in some
949 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
950 To be on the safe side, however, I have now made the interface to
951 host_find_byname() similar to host_find_bydns(), with an argument
952 containing the DNS resolver options. The host_find_byname() function now
953 sets these options at its start, just as host_find_bydns() does. The smtp
954 transport options dns_qualify_single and dns_search_parents are passed to
955 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
956 of host_find_byname() use the default settings of RES_DEFNAMES
957 (qualify_single) but not RES_DNSRCH (search_parents).
959 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
960 spool_read_header() do less string testing, by means of a preliminary
961 switch on the second character of optional "-foo" lines. (This is
962 overdue, caused by the large number of possibilities that now exist.
963 Originally there were few.) While I was there, I also converted the
964 str(n)cmp tests so they don't re-test the leading "-" and the first
965 character, in the hope this might squeeze out yet more improvement.
967 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
968 flag allowing group syntax was set by the header_syntax check but not
969 turned off, possible causing trouble later; (2) The flag was not being
970 set at all for the header_verify test, causing "group"-style headers to
971 be rejected. I have now set it in this case, and also caused header_
972 verify to ignore an empty address taken from a group. While doing this, I
973 came across some other cases where the code for allowing group syntax
974 while scanning a header line wasn't quite right (mostly, not resetting
975 the flag correctly in the right place). These bugs could have caused
976 trouble for malformed header lines. I hope it is now all correct.
978 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
979 with the "reply" argument non-NULL. The code, however (which originally
980 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
981 but it didn't always do it. This confused somebody who was copying the
982 code for some other use. I have removed all the tests.
984 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
985 feature that was used to support insecure browsers during the U.S. crypto
986 embargo. It requires special client support, and Exim is probably the
987 only MTA that supported it -- and would never use it because real RSA is
988 always available. This code has been removed, because it had the bad
989 effect of slowing Exim down by computing (never used) parameters for the
990 RSA_EXPORT functionality.
992 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
993 authenticator to fail if there's a tab character in the incoming data
994 (there should never be unless someone is messing about, as it's supposed
995 to be base64-encoded). Also added, on Timo's advice, the "secured" option
996 if the connection is using TLS or if the remote IP is the same as the
997 local IP, and the "valid-client-cert option" if a client certificate has
1000 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
1001 authenticators. This can be used for authorization after authentication
1002 succeeds. (In the case of plaintext, it servers for both authentication
1005 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
1006 if any retry times were supplied.
1008 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
1009 connection, where there is no client IP address to check. In this
1010 situation, the verify now always succeeds.
1012 PH/25 Applied John Jetmore's -Mset patch.
1014 PH/26 Added -bem to be like -Mset, but loading a message from a file.
1016 PH/27 In a string expansion for a processed (not raw) header when multiple
1017 headers of the same name were present, leading whitespace was being
1018 removed from all of them, but trailing whitespace was being removed only
1019 from the last one. Now trailing whitespace is removed from each header
1020 before concatenation. Completely empty headers in a concatenation (as
1021 before) are ignored.
1023 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
1024 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
1026 PH/29 [Removed. This was a change that I later backed out, and forgot to
1027 correct the ChangeLog entry (that I had efficiently created) before
1028 committing the later change.]
1030 PH/30 Exim was sometimes attempting to deliver messages that had suffered
1031 address errors (4xx response to RCPT) over the same connection as other
1032 messages routed to the same hosts. Such deliveries are always "forced",
1033 so retry times are not inspected. This resulted in far too many retries
1034 for the affected addresses. The effect occurred only when there were more
1035 hosts than the hosts_max_try setting in the smtp transport when it had
1036 the 4xx errors. Those hosts that it had tried were not added to the list
1037 of hosts for which the message was waiting, so if all were tried, there
1038 was no problem. Two fixes have been applied:
1040 (i) If there are any address or message errors in an SMTP delivery, none
1041 of the hosts (tried or untried) are now added to the list of hosts
1042 for which the message is waiting, so the message should not be a
1043 candidate for sending over the same connection that was used for a
1044 successful delivery of some other message. This seems entirely
1045 reasonable: after all the message is NOT "waiting for some host".
1046 This is so "obvious" that I'm not sure why it wasn't done
1047 previously. Hope I haven't missed anything, but it can't do any
1048 harm, as the worst effect is to miss an optimization.
1050 (ii) If, despite (i), such a delivery is accidentally attempted, the
1051 routing retry time is respected, so at least it doesn't keep
1052 hammering the server.
1054 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
1055 in ${readsocket because some servers need this prod.
1057 PH/32 Added some extra debug output when updating a wait-xxx database.
1059 PH/33 The hint "could be header name not terminated by colon", which has been
1060 given for certain expansion errors for a long time, was not being given
1061 for the ${if def:h_colon_omitted{... case.
1063 PH/34 The spec says: "With one important exception, whenever a domain list is
1064 being scanned, $domain contains the subject domain." There was at least
1065 one case where this was not true.
1067 PH/35 The error "getsockname() failed: connection reset by peer" was being
1068 written to the panic log as well as the main log, but it isn't really
1069 panic-worthy as it just means the connection died rather early on. I have
1070 removed the panic log writing for the ECONNRESET error when getsockname()
1073 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
1074 runs only) independently of the message's sender address. This meant
1075 that, if the 4xx error was in fact related to the sender, a different
1076 message to the same recipient with a different sender could confuse
1077 things. In particualar, this can happen when sending to a greylisting
1078 server, but other circumstances could also provoke similar problems.
1079 I have changed the default so that the retry time for these errors is now
1080 based a combination of the sender and recipient addresses. This change
1081 can be overridden by setting address_retry_include_sender=false in the
1084 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
1085 remote server are returned as part of bounce messages. This was not
1086 happening for LMTP over a pipe (the lmtp transport), but now it is the
1087 same for both kinds of LMTP.
1089 PH/38 Despite being documented as not happening, Exim was rewriting addresses
1090 in header lines that were in fact CNAMEs. This is no longer the case.
1092 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
1093 and queue runs started by the daemon processed all messages. This has
1094 been fixed so that -R and -S can now usefully be given with -q<time>.
1096 PH/40 Import PCRE release 6.7 (fixes some bugs).
1098 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
1100 PH/42 Give an error if -q is specified more than once.
1102 PH/43 Renamed the variables $interface_address and $interface_port as
1103 $received_ip_address and $received_port, to make it clear that these
1104 values apply to message reception, and not to the outgoing interface when
1105 a message is delivered. (The old names remain recognized, of course.)
1107 PH/44 There was no timeout on the connect() call when using a Unix domain
1108 socket in the ${readsocket expansion. There now is.
1110 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
1111 be meaningful with "accept".
1113 SC/01 Eximstats V1.43
1114 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
1116 SC/02 Eximstats V1.44
1117 Use a glob alias rather than an array ref in the generated
1118 parser. This improves both readability and performance.
1120 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
1121 Collect SpamAssassin and rejection statistics.
1122 Don't display local sender or destination tables unless
1123 there is data to show.
1124 Added average volumes into the top table text output.
1126 SC/04 Eximstats V1.46
1127 Collect data on the number of addresses (recipients)
1128 as well as the number of messages.
1130 SC/05 Eximstats V1.47
1131 Added 'Message too big' to the list of mail rejection
1132 reasons (thanks to Marco Gaiarin).
1134 SC/06 Eximstats V1.48
1135 Mainlog lines which have GMT offsets and are too short to
1136 have a flag are now skipped.
1138 SC/07 Eximstats V1.49 (Alain Williams)
1139 Added the -emptyok flag.
1141 SC/08 Eximstats V1.50
1142 Fixes for obtaining the IP address from reject messages.
1144 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
1145 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
1146 whitesspace changes from 4.64-PH/27
1148 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
1151 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
1152 are found, allow negative numbers in numeric criteria)
1154 JJ/06 exipick.20061117.2, added new $message_body_missing variable
1156 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
1157 to match changes made in 4.64-PH/43
1159 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
1161 PH/47 Put in an explicit test for a DNS lookup of an address record where the
1162 "domain" is actually an IP address, and force a failure. This locks out
1163 those revolvers/nameservers that support "A-for-A" lookups, in
1164 contravention of the specifications.
1166 PH/48 When a host name was looked up from an IP address, and the subsequent
1167 forward lookup of the name timed out, the host name was left in
1168 $sender_host_name, contrary to the specification.
1170 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
1171 restricted to single-key lookups, Exim was not diagnosing an error if
1172 * or *@ was used with a query-style lookup.
1174 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
1176 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
1177 long ago, but noone interested enough thought of it. Let's just say that
1178 the "1.1" means that there are some new functions that weren't there at
1179 some point in the past.
1181 PH/51 Error processing for expansion failure of helo_data from an smtp
1182 transport during callout processing was broken.
1184 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
1185 tested/used via the -bh/-bhc/-bs options.
1187 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
1188 bug, fixed in subsequent PCRE releases).
1190 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
1191 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
1193 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
1199 SC/01 Use a glob alias rather than an array ref in eximstats generated
1200 parser. This improves both readability and performance.
1202 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
1203 Don't display local sender or destination tables in eximstats unless
1204 there is data to show.
1205 Added average volumes into the eximstats top table text output.
1207 SC/03 Collect data on the number of addresses (recipients) as well
1208 as the number of messages in eximstats.
1210 TF/01 Correct an error in the documentation for the redirect router. Exim
1211 does (usually) call initgroups() when daemonizing.
1213 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
1214 with consistent privilege compared to when running as a daemon.
1216 TF/03 Note in the spec that $authenticated_id is not set for local
1217 submissions from trusted users.
1219 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
1220 Thanks to Dean Brooks <dean@iglou.com> for the patch.
1222 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
1223 by adding some example configuration directives to the default
1224 configuration file. A little bit of work is required to uncomment the
1225 directives and define how usernames and passwords are checked, but
1226 there is now a framework to start from.
1228 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
1229 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
1230 without this. I don't know how relevant this is to other LDAP libraries.
1232 PH/02 Add the verb name to the "unknown ACL verb" error.
1234 PH/03 Magnus Holmgren's patch for filter_prepend_home.
1236 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
1238 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
1239 directory not expanded when it should be if an expanded home directory
1240 was set for the address (which is overridden by the transport).
1242 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
1245 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
1246 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
1247 because it is too late at that time, and has no effect.
1249 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
1250 security issue with \' (bugzilla #107). I could not use the
1251 PQescapeStringConn() function, because it needs a PGconn value as one of
1254 PH/08 When testing addresses using -bt, indicate those final addresses that
1255 are duplicates that would not cause an additional delivery. At least one
1256 person was confused, thinking that -bt output corresponded to deliveries.
1257 (Suppressing duplicates isn't a good idea as you lose the information
1258 about possibly different redirections that led to the duplicates.)
1260 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
1261 systems where poll() doesn't work, in particular OS X.
1263 PH/10 Added more information to debugging output for retry time not reached.
1265 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
1266 operations in malware.c.
1268 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
1271 PH/13 If write_rejectlog was set false when logging was sent to syslog with
1272 syslog_duplication set false, log lines that would normally be written
1273 both the the main log and to the reject log were not written to syslog at
1276 PH/14 In the default configuration, change the use of "message" in ACL warn
1277 statements to "add_header".
1279 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
1280 not followed by a command (e.g. "seen endif").
1282 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
1283 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
1286 PH/17 Added extra conditions to the default value of delay_warning_condition
1290 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
1291 { match{$h_precedence:}{(?i)bulk|list|junk} } \
1292 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1295 The Auto-Submitted: and various List- headers are standardised, whereas I
1296 don't think Precedence: ever was.
1298 PH/18 Refactored debugging code in route_finduser() to show more information,
1299 in particular, the error code if getpwnam() issues one.
1301 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1302 This is apparently needed in addition to the PH/07 change above to avoid
1303 any possible encoding problems.
1305 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1306 but not after initializing Perl.
1308 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1309 output them only if debugging. By default they are written stderr,
1310 apparently, which is not desirable.
1312 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1315 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1318 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1320 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1321 authenticated or an ident call has been made. Suppress the default
1322 values for $authenticated_id and $authenticated_sender (but permit -oMai
1323 and -oMas) when testing with -bh.
1325 PH/24 Re-jigged the order of the tests in the default configuration so that the
1326 tests for valid domains and recipients precede the DNS black list and CSA
1327 tests, on the grounds that those ones are more expensive.
1329 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1330 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1331 This bug exists in every version of Exim that I still have, right back to
1334 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1335 However, an attempt to turn on case-sensitivity in a regex key by
1336 including (?-i) didn't work because the subject string was already
1337 lowercased, and the effects were non-intuitive. It turns out that a
1338 one-line patch can be used to allow (?-i) to work as expected.
1344 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1345 other effects) broke the use of negated acl sub-conditions.
1347 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1350 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1351 "Deny" causes Exim to reject the incoming connection with a 554 error.
1352 Unfortunately, if there is a major crisis, such as a disk failure,
1353 tcp-wrappers gives "deny", whereas what one would like would be some
1354 kind of temporary error. A kludge has been added to help with this.
1355 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1356 554 error is used if errno is still zero or contains ENOENT (which occurs
1357 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1360 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1362 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1363 errors. Otherwise a message that provokes a temporary error (when other
1364 messages do not) can cause a whole host to time out.
1366 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1367 addresses were routed directly to files or pipes from a redirect router.
1368 File deliveries just didn't batch; pipe deliveries might have suffered
1371 PH/06 A failure to get a lock for a hints database would erroneously always say
1372 "Failed to get write lock", even when it was really a read lock.
1374 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1375 of 0600. This has been changed to use the value of the lockfile_mode
1376 option (which defaults to 0600).
1378 PH/08 Applied small patch from the Sieve maintainer.
1380 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1381 folder from quota calculations, a direct delivery into this folder messed
1382 up the contents of the maildirsize file. This was because the regex was
1383 used only to exclude .Trash (or whatever) when the size of the mailbox
1384 was calculated. There was no check that a delivery was happening into an
1385 excluded directory. This bug has been fixed by ignoring all quota
1386 processing for deliveries into excluded directories.
1388 PH/10 Added the maildirfolder_create_regex option to appendfile.
1394 PH/01 The code for finding all the local interface addresses on a FreeBSD
1395 system running IPv6 was broken. This may well have applied to all BSD
1396 systems, as well as to others that have similar system calls. The broken
1397 code found IPv4 interfaces correctly, but gave incorrect values for the
1398 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1399 that it would not match correctly against @[] and not recognize the IPv6
1402 PH/02 The ipliteral router was not recognizing addresses of the form user@
1403 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1405 PH/03 Added disable_ipv6.
1407 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1408 decoded form, because it is most often used to construct To: headers
1409 lines in autoreplies, and the decoded form may well be syntactically
1410 invalid. However, $reply_address has leading white space removed, and all
1411 newlines turned into spaces so that the autoreply transport does not
1414 PH/05 If group was specified without a user on a router, and no group or user
1415 was specified on a transport, the group from the router was ignored.
1417 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1418 for visible compile-time settings that can be used to change these
1419 numbers, for those that want even more. Backwards compatibility with old
1420 spool files has been maintained. However, going back to a previous Exim
1421 release will lost any variables that are in spool files.
1423 PH/07 Two small changes when running in the test harness: increase delay when
1424 passing a TCP/IP connection to a new process, in case the original
1425 process has to generate a bounce, and remove special handling of
1426 127.0.0.2 (sic), which is no longer necessary.
1428 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1429 be the same on different OS.
1431 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1434 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1435 whether --show-vars was specified or not
1437 JJ/02 exipick: Added support for new ACL variable spool format introduced
1440 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1441 syntactically invalid From: or Reply-to: line, and a filter used this to
1442 generate an autoreply, and therefore failed to obtain an address for the
1443 autoreply, Exim could try to deliver to a non-existent relative file
1444 name, causing unrelated and misleading errors. What now happens is that
1445 it logs this as a hard delivery error, but does not attempt to create a
1448 PH/11 The exinext utility has a -C option for testing purposes, but although
1449 the given file was scanned by exinext itself; it wasn't being passed on
1450 when Exim was called.
1452 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1453 an end-of-file indication when reading a command response.
1455 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1456 compiled. In many other places in Exim, IPv6 addresses are always
1457 recognized, so I have changed this. It also means that IPv4 domain
1458 literals of the form [IPV4:n.n.n.n] are now always recognized.
1460 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1461 used if the router is not running as root, for example, when verifying at
1462 ACL time, or when using -bh. The debugging output from this situation was
1463 non-existent - all you got was a failure to exec. I have made two
1466 (a) Failures to set uid/gid, the current directory, or a process leader
1467 in a subprocess such as that created by queryprogram now generate
1468 suitable debugging ouput when -d is set.
1470 (b) The queryprogram router detects when it is not running as root,
1471 outputs suitable debugging information if -d is set, and then runs
1472 the subprocess without attempting to change uid/gid.
1474 PH/15 Minor change to Makefile for building test_host (undocumented testing
1477 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1478 additional section of a DNS packet that returns MX or SRV records.
1479 Instead, it always explicitly searches for A/AAAA records. This avoids
1480 major problems that occur when a DNS server includes only records of one
1481 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1482 fixed another bug: if SRV records were looked up and the corresponding
1483 address records were *not* found in the additional section, the port
1484 values from the SRV records were lost.
1486 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1487 using the correct key (the original address) when searching the retry
1488 rules in order to find which one to use for generating the retry hint.
1490 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1491 adding the default one. Similarly, if it contains a Reply-To: header, the
1492 errors_reply_to option, if set, is not used.
1494 PH/19 When calculating a retry time, Exim used to measure the "time since
1495 failure" by looking at the "first failed" field in the retry record. Now
1496 it does not use this if it is later than than the arrival time of the
1497 message. Instead it uses the arrival time. This makes for better
1498 behaviour in cases where some deliveries succeed, thus re-setting the
1499 "first failed" field. An example is a quota failure for a huge message
1500 when small messages continue to be delivered. Without this change, the
1501 "time since failure" will always be short, possible causing more frequent
1502 delivery attempts for the huge message than are intended.
1503 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1505 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1506 $1, $2, $3) because the numerical variables can be reset during some
1507 expansion items (e.g. "match"), thereby losing the authentication data.
1509 PH/21 Make -bV show the size of off_t variables so that the test suite can
1510 decide whether to run tests for quotas > 2G.
1512 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1513 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1514 filecount value is greater than 2G or if a quota value is greater than 2G
1515 on a system where the size of off_t is not greater than 4, a panic error
1518 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1519 never match. The debug and -bh output now contains an explicit error
1520 message indicating a malformed IPv4 address or mask.
1522 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1523 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1524 PH/23 above applies.
1526 PH/25 Do not write to syslog when running in the test harness. The only
1527 occasion when this arises is a failure to open the main or panic logs
1528 (for which there is an explicit test).
1530 PH/26 Added the /no_tell option to "control=freeze".
1532 PH/27 If a host name lookup failed very early in a connection, for example, if
1533 the IP address matched host_lookup and the reverse lookup yielded a name
1534 that did not have a forward lookup, an error message of the form "no IP
1535 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1536 could be logged. Now it outputs the IP address instead of "NULL".
1538 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1539 allows the sender and the authenticated sender to be set when
1540 submitting a message from within Exim. Since child_open_exim() is
1541 documented for local_scan(), the new function should be too.
1543 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1544 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1545 results in an empty string is now treated as unset.
1547 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1549 PH/31 Added sender_verify_fail logging option.
1551 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1552 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1553 tidied the source and removed it altogether.
1555 PH/33 When a queue run was abandoned because the load average was too high, a
1556 log line was always written; now it is written only if the queue_run log
1557 selector is set. In addition, the log line for abandonment now contains
1558 information about the queue run such as the pid. This is always present
1559 in "start" and "stop" lines but was omitted from the "abandon" line.
1561 PH/34 Omit spaces between a header name and the colon in the error message that
1562 is given when verify = headers_syntax fails (if there are lots of them,
1563 the message gets confusing).
1565 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1566 names, as there are now some PTR records that contain slashes. This check
1567 is only to protect against broken name servers that fall over on strange
1568 characters, so the fact that it applies to all lookups doesn't matter.
1570 PH/36 Now that the new test suite is complete, we can remove some of the
1571 special code in Exim that was needed for the old test suite. For example,
1572 sorting DNS records because real resolvers return them in an arbitrary
1573 order. The new test suite's fake resolver always returns records in the
1576 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1577 bounces) except when queue_only is set, to avoid logging races between
1578 the different processes.
1580 PH/38 Panic-die if .include specifies a non-absolute path.
1582 PH/39 A tweak to the "H" retry rule from its user.
1584 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1585 a label. They prevented compilation on older perls.
1587 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1588 a warning to be raised on newish perls.
1590 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1591 on queue. Changes to match documented behaviour of showing count of
1592 messages matching specified criteria.
1594 PH/40 Changed the default ident timeout from 30s to 5s.
1596 PH/41 Added support for the use of login_cap features, on those BSD systems
1597 that have them, for controlling the resources used by pipe deliveries.
1599 PH/42 The content-scanning code uses fopen() to create files in which to put
1600 message data. Previously it was not paying any attention to the mode of
1601 the files. Exim runs with umask(0) because the rest of the code creates
1602 files with open(), and sets the required mode explicitly. Thus, these
1603 files were ending up world-writeable. This was not a big issue, because,
1604 being within the spool directory, they were not world-accessible. I have
1605 created a function called modefopen, which takes an additional mode
1606 argument. It sets umask(777), creates the file, chmods it to the required
1607 mode, then resets the umask. All the relevant calls to fopen() in the
1608 content scanning code have been changed to use this function.
1610 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1611 to 24 hours. This avoids potential overflow problems when processing G
1612 and H retry rules. I suspect nobody ever tinkers with this value.
1614 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1616 PH/45 When the plaintext authenticator is running as a client, the server's
1617 challenges are checked to ensure they are valid base64 strings. By
1618 default, the authentication attempt is cancelled if an invalid string is
1619 received. Setting client_ignore_invalid_base64 true ignores these errors.
1620 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1621 they are received. Thus, the responses can be made to depend on the
1622 challenges. If an invalid string is ignored, an empty string is placed in
1625 PH/46 Messages that are created by the autoreply transport now contains a
1626 References: header, in accordance with RFCs 2822 and 3834.
1628 PH/47 Added authenticated_sender_force to the smtp transport.
1630 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1632 PH/49 Installed latest patch from the Sieve maintainer.
1634 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1635 was also set, the appendfile transport was unnecessarily scanning a
1636 directory of message files (e.g. for maildir delivery) to find the count
1637 of files (along with the size), even though it did not need this
1638 information. It now does the scan only if it needs to find either the
1639 size of the count of files.
1641 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1643 PH/52 Two bugs concerned with error handling when the smtp transport is
1646 (i) Exim was not creating retry information for temporary errors given
1647 for individual recipients after the DATA command when the smtp transport
1648 was used in LMTP mode. This meant that they could be retried too
1649 frequently, and not timed out correctly.
1651 (ii) Exim was setting the flag that allows error details to be returned
1652 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1653 recipients that were returned after the DATA command.
1655 PH/53 This is related to PH/52, but is more general: for any failing address,
1656 when detailed error information was permitted to be returned to the
1657 sender, but the error was temporary, then after the final timeout, only
1658 "retry timeout exceeded" was returned. Now it returns the full error as
1659 well as "retry timeout exceeded".
1661 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1662 do this, and (what is worse) MTAs that accept it.
1664 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1665 will now be deprecated.
1667 PH/56 New os.c-cygwin from the Cygwin maintainer.
1669 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1670 formats (previously only available in exim formats via -bpr, -bpru,
1671 and -bpra. Now also available in native and exiqgrep formats)
1673 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1674 with very large, slow to parse queues
1676 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1678 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1680 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1681 responses to authentication challenges, though it was showing the
1682 challenges; (ii) I've removed the CR characters from the debug output for
1685 PH/58 Allow for the insertion of a newline as well as a space when a string
1686 is turned into more than one encoded-word during RFC 2047 encoding. The
1687 Sieve code now uses this.
1689 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1690 data_4xx, lost_connection, tls_required.
1692 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1693 message was being sent as an SMTP response.
1695 PH/61 Add -l and -k options to exicyclog.
1697 PH/62 When verifying, if an address was redirected to one new address, so that
1698 verification continued, and the new address failed or deferred after
1699 having set something in $address_data, the value of $address_data was not
1700 passed back to the ACL. This was different to the case when no
1701 redirection occurred. The value is now passed back in both cases.
1703 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1704 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1705 use login_cap.h, so on its own it isn't the distinguishing feature. The
1706 new name refers directly to the setclassresources() function.
1708 PH/65 Added configuration files for NetBSD3.
1710 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1712 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1713 is preferred over IPv4.
1715 PH/68 The bounce_return_message and bounce_return_body options were not being
1716 honoured for bounces generated during the reception of non-SMTP messages.
1717 In particular, this applied to messages rejected by the ACL. This bug has
1718 been fixed. However, if bounce_return_message is true and bounce_return_
1719 body is false, the headers that are returned for a non-SMTP message
1720 include only those that have been read before the error was detected.
1721 (In the case of an ACL rejection, they have all been read.)
1723 PH/69 The HTML version of the specification is now built in a directory called
1724 spec_html instead of spec.html, because the latter looks like a path with
1725 a MIME-type, and this confuses some software.
1727 PH/70 Catch two compiler warnings in sieve.c.
1729 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1730 function verify_get_ident() calls ip_connect() to connect a socket, but
1731 if the "connect()" function timed out, ip_connect() used to close the
1732 socket. However, verify_get_ident() also closes the socket later, and in
1733 between Exim writes to the log, which may get opened at this point. When
1734 the socket was closed in ip_connect(), the log could get the same file
1735 descriptor number as the socket. This naturally causes chaos. The fix is
1736 not to close the socket in ip_connect(); the socket should be closed by
1737 the function that creates it. There was only one place in the code where
1738 this was missing, in the iplookup router, which I don't think anybody now
1739 uses, but I've fixed it anyway.
1741 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1742 well as to direct DNS lookups. Otherwise the handling of names in host
1743 lists is inconsistent and therefore confusing.
1749 PH/01 Two changes to the default runtime configuration:
1751 (1) Move the checks for relay_from_hosts and authenticated clients from
1752 after to before the (commented out) DNS black list checks.
1754 (2) Add control=submission to the relay_from_hosts and authenticated
1755 clients checks, on the grounds that messages accepted by these
1756 statements are most likely to be submissions.
1758 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1760 (1) Generate an error if the third argument for the ${prvs expansion is
1763 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1766 (3) Reset the variables that are obtained from the first argument of
1767 ${prvscheck and used in the second argument before leaving the code,
1768 because their memory is reclaimed, so using them afterwards may do
1771 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1772 one (it's much easier than Tom thought :-).
1774 (5) Because of (4), we can now allow for the use of $prvscheck_result
1775 inside the third argument.
1777 PH/03 For some reason, the default setting of PATH when running a command from
1778 a pipe transport was just "/usr/bin". I have changed it to
1781 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1782 anything to be listed in the output from -bV.
1784 PH/05 When a filter generated an autoreply, the entire To: header line was
1785 quoted in the delivery log line, like this:
1787 => >A.N.Other <ano@some.domain> <original@ddress> ...
1789 This has been changed so that it extracts the operative address. There
1790 may be more than one such address. If so, they are comma-separated, like
1793 => >ano@some.domain,ona@other.domain <original@ddress> ...
1795 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1796 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1797 not being looked up in the rDNS to get a host name, Exim was showing the
1798 IP address twice in Received: lines, even though the IP addresses were
1799 identical. For example:
1801 Received: from [1.2.3.4] (helo=[1.2.3.4])
1803 However, if the real host name was known, it was omitting the HELO data
1804 if it matched the actual IP address. This has been tidied up so that it
1805 doesn't show the same IP address twice.
1807 PH/07 When both +timestamp and +memory debugging was on, the value given by
1808 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1809 called by the debug printing, thereby overwriting the timestamp buffer.
1810 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1812 PH/08 When the original message was included in an autoreply transport, it
1813 always said "this is a copy of the message, including all the headers",
1814 even if body_only or headers_only was set. It now gives an appropriate
1817 PH/09 Applied a patch from the Sieve maintainer which:
1819 o fixes some comments
1820 o adds the (disabled) notify extension core
1821 o adds some debug output for the result of if/elsif tests
1822 o points to the current vacation draft in the documentation
1823 and documents the missing references header update
1827 o fixes a bug in processing the envelope test (when testing
1828 multiple envelope elements, the last element determinted the
1831 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1832 Electronic Mail") by including:
1834 Auto-submitted: auto-generated
1836 in the messages that it generates (bounce messages and others, such as
1837 warnings). In the case of bounce messages for non-SMTP mesages, there was
1838 also a typo: it was using "Auto_submitted" (underscore instead of
1839 hyphen). Since every message generated by Exim is necessarily in response
1840 to another message, thes have all been changed to:
1842 Auto-Submitted: auto-replied
1844 in accordance with these statements in the RFC:
1846 The auto-replied keyword:
1848 - SHOULD be used on messages sent in direct response to another
1849 message by an automatic process,
1851 - MUST NOT be used on manually-generated messages,
1853 - MAY be used on Delivery Status Notifications (DSNs) and Message
1854 Disposition Notifications (MDNs),
1856 - MUST NOT be used on messages generated by automatic or periodic
1857 processes, except for messages which are automatic responses to
1860 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1861 to the default Received: header definition.
1863 PH/12 Added log selector acl_warn_skipped (default on).
1865 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1866 variables because (a) they are in the wrong storage pool and (b) even if
1867 they were copied, it wouldn't work properly because of the caching.
1869 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1870 checking when decoding. Apparently there are clients that generate
1871 overlong encoded strings. Why am I not surprised?
1873 PH/15 If the first argument of "${if match_address" was not empty, but did not
1874 contain an "@" character, Exim crashed. Now it writes a panic log message
1875 and treats the condition as false.
1877 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1879 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1880 extension "envelope-auth". The code is finished and in agreement with
1881 other implementations, but there is no documentation so far and in fact,
1882 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1883 not changing the active code.
1885 Print executed "if" and "elsif" statements when debugging is used. This
1886 helps a great deal to understand what a filter does.
1888 Document more things not specified clearly in RFC3028. I had all this
1889 sorted out, when out of a sudden new issues came to my mind. Oops."
1891 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1894 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1895 canonical form (as documented). However, after a host name lookup from
1896 the IP address, check_host() was doing a simple string comparison with
1897 addresses acquired from the DNS when checking that the found name did
1898 have the original IP as one of its addresses. Since any found IPv6
1899 addresses are likely to be in abbreviated form, the comparison could
1900 fail. Luckily, there already exists a function for doing the comparison
1901 by converting both addresses to binary, so now that is used instead of
1902 the text comparison.
1904 PH/20 There was another similar case to PH/19, when a complete host name was
1905 given in a host list; looking up its IP address could give an abbreviated
1906 form, whereas the current host's name might or might not be abbreviated.
1907 The same fix has been applied.
1913 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1914 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1917 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1918 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1920 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1922 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1923 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1924 Also, for "Auto-Submitted", treat anything other than "no" as
1925 non-personal, in accordance with RFC 3834. (Previously it treated
1926 anything starting "auto-" as non-personal.)
1928 TF/01 The control=submission/name=... option had a problem with syntax
1929 errors if the name included a slash character. The /name= option
1930 now slurps the rest of the string, so it can include any characters
1931 but it must come last in the list of options (after /sender_retain
1934 PH/05 Some modifications to the interface to the fake nameserver for the new
1942 TK/01 Added the "success_on_redirect" address verification option. See
1943 NewStuff for rationale and an example.
1945 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1947 PH/02 Patch to exigrep to allow it to work on syslog lines.
1949 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1950 fread() to skip over the body file's header line, because in Cygwin the
1951 header line is locked and is inaccessible.
1953 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1954 co-exist for some time) to make it clear that it is the Exim ID that is
1955 referenced, not the Message-ID: header line.
1957 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1958 string_format() function, because snprintf() does not exist on all
1961 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1962 ${stat: expansion item.
1964 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1965 protocol synchronization error", to keep the pedants happy.
1967 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1968 well as for IRIX systems, when gcc is being used. See the host.c source
1971 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1973 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1976 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1977 transport and to the smtp transport in LMTP mode.
1979 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1981 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1982 run a filter in a subprocess. This could lead to confusion in subsequent
1983 lookups in the parent process. There should also be a search_tidyup() at
1984 the end of the subprocess.
1986 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1987 only if the host matched helo_try_verify_hosts, which caused the
1988 verification to occur when the EHLO/HELO command was issued. The ACL just
1989 tested the remembered result. Now, if a previous verification attempt has
1990 not happened, "verify = helo" does it there and then.
1992 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1994 TK/03 Fix log output including CR from clamd.
1996 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1997 exist provoked a memory error which could cause a segfault.
1999 PH/15 Installed PCRE 6.2
2001 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
2003 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
2004 of the problem. Specifically, suggested +O2 rather than +O1 for the
2007 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
2009 PH/20 If a delivery was routed to a non-standard port by means of an SRV
2010 record, the port was not correctly logged when the outgoing_port log
2011 selector was set (it logged the transort's default port).
2013 PH/21 Added support for host-specific ports to manualroute, queryprogram,
2014 fallback_hosts, and "hosts" in the smtp transport.
2016 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
2017 host errors such as "Connection refused".
2019 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
2020 authentication with radiusclient 0.4.9:
2022 - Error returned from rc_read_config was caught wrongly
2023 - Username/password not passed on to radius server due to wrong length.
2025 The presumption is that some radiusclient API changes for 4.51/PH/17
2026 were not taken care of correctly. The code is still untested by me (my
2027 Linux distribution still has 0.3.2 of radiusclient), but it was
2028 contributed by a Radius user.
2030 PH/24 When doing a callout, the value of $domain wasn't set correctly when
2031 expanding the "port" option of the smtp transport.
2033 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
2034 while reading a MIME header. Thanks to Tom Hughes for a patch.
2036 PH/24 Include config.h inside local_scan.h so that configuration settings are
2039 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
2040 that in an ACL for RCPT (for example), you can examine exactly what was
2043 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
2044 commands, but it was not correctly comparing the address with the actual
2045 client host address. Thus, it would show the EHLO address in Received:
2046 header lines when this was not necessary.
2048 PH/27 Added the % operator to ${eval:}.
2050 PH/28 Exim tries to create and chdir to its spool directory when it starts;
2051 it should be ignoring failures (because with -C, for example, it has lost
2052 privilege). It wasn't ignoring creation failures other than "already
2055 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
2058 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
2059 because an input line was too long, either on its own, or by virtue of
2060 too many continuations, the temporary file was not being removed, and the
2061 return code was incorrect.
2063 PH/31 Missing "BOOL" in function definition in filtertest.c.
2065 PH/32 Applied Sieve patches from the maintainer.
2067 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
2069 PH/33 Added "verify = not_blind".
2071 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
2072 Local/Makefile (with some defaults set). These are used in built scripts
2073 such as exicyclog, but they have never been used in the exim_install
2074 script (though there are many overriding facilities there). I have
2075 arranged that the exim_install script now takes note of these two
2078 PH/35 Installed configuration files for Dragonfly.
2080 PH/36 When a locally submitted message by a trusted user did not contain a
2081 From: header, and the sender address was obtained from -f or from an SMTP
2082 MAIL command, and the trusted user did not use -F to supply a sender
2083 name, $originator_name was incorrectly used when constructing a From:
2084 header. Furthermore, $originator_name was used for submission mode
2085 messages from external hosts without From: headers in a similar way,
2086 which is clearly wrong.
2088 PH/37 Added control=suppress_local_fixups.
2090 PH/38 When log_selector = +received_sender was set, and the addition of the
2091 sender made the log line's construction buffer exactly full, or one byte
2092 less than full, an overflow happened when the terminating "\n" was
2095 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
2096 when the result of a list match is failure because a DNS lookup failed.
2098 PH/40 RM_COMMAND is now used in the building process.
2100 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
2101 the "build-* directories that it finds.
2103 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
2104 address in a domain literal was a prefix of an interface address.
2106 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
2107 when verifying a sender address, unless rewrite_headers is false.
2109 PH/44 Wrote a long comment about why errors_to addresses are verified as
2110 recipients, not senders.
2112 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
2113 the ratelimit ACL was added.
2115 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
2117 PH/46 Added extra information about PostgreSQL errors to the error string.
2119 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
2120 avoiding the need to install special zones in a real server. This is
2121 backwards compatible; if it can't find the fake resolver, it drops back.
2122 Thus, both old and new test suites can be run.
2124 TF/02 Added util/ratelimit.pl
2126 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
2127 clock is set back in time.
2129 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
2130 Candler <B.Candler@pobox.com>.
2132 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
2133 OK for addresses that are the result of redirections.
2135 PH/48 A number of further additions for the benefit of the new test suite,
2136 including a fake gethostbyname() that interfaces to the fake DNS resolver
2139 TF/06 The fix for widen_domains has also been applied to qualify_single and
2140 search_parents which are the other dnslookup options that can cause
2143 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
2146 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
2148 TF/07 Exim produced the error message "an SRV record indicated no SMTP
2149 service" if it encountered an MX record with an empty target hostname.
2150 The message is now "an MX or SRV record indicated no SMTP service".
2152 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
2153 if the DNS of the sending site is misconfigured. This is quite a
2154 common situation. This change restores the behaviour of treating a
2155 helo verification defer as a failure.
2157 PH/51 If self=fail was set on a router, the bounce message did not include the
2158 actual error message.
2164 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
2166 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
2167 command itself ended in error, the underlying message about the transport
2168 filter timeout was being overwritten with the pipe command error. Now the
2169 underlying error message should be appended to the second error message.
2171 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
2173 PH/02 Reduce the amount of output that "make" produces by default. Full output
2174 can still be requested.
2176 PH/03 The warning log line about a condition test deferring for a "warn" verb
2177 was being output only once per connection, rather than after each
2178 occurrence (because it was using the same function as for successful
2179 "warn" verbs). This seems wrong, so I have changed it.
2181 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
2182 it should not have, which might have caused a crash in the right
2183 circumstances, but probably never did.
2185 PH/04 Installed a modified version of Tony Finch's patch to make submission
2186 mode fix the return path as well as the Sender: header line, and to
2187 add a /name= option so that you can make the user's friendly name appear
2190 TF/03 Added the control = fakedefer ACL modifier.
2192 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
2193 Mark Lowes for thorough testing.
2195 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
2197 TK/03 Merged latest SRS patch from Miles Wilton.
2199 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
2200 with the definition in sysexits.h (which is #included earlier).
2201 Fortunately, Exim does not actually use EX_OK. The code used to try to
2202 preserve the sysexits.h value, by assumimg that macro definitions were
2203 scanned for macro replacements. I have been disabused of this notion,
2204 so now the code just undefines EX_OK before #including unistd.h.
2206 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
2207 in the smtp transport. When a block could not be written in a single
2208 write() function, the timeout was being re-applied to each part-write.
2209 This seems wrong - if the receiver was accepting one byte at a time it
2210 would take for ever. The timeout is now adjusted when this happens. It
2211 doesn't have to be particularly precise.
2213 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
2214 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
2216 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
2217 without a domain if the check to <postmaster@domain> fails.
2219 SC/01 Eximstats: added -xls and the ability to specify output files
2220 (patch written by Frank Heydlauf).
2222 SC/02 Eximstats: use FileHandles for outputing results.
2224 SC/03 Eximstats: allow any combination of xls, txt, and html output.
2226 SC/04 Eximstats: fixed display of large numbers with -nvr option
2228 SC/05 Eximstats: fixed merging of reports with empty tables.
2230 SC/06 Eximstats: added the -include_original_destination flag
2232 SC/07 Eximstats: removed tabs and trailing whitespace.
2234 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
2236 TK/06 MBOX spool code: Add real "From " MBOX separator line
2237 so the .eml file is really in mbox format (even though
2238 most programs do not really care). Patch from Alex Miller.
2240 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
2241 The latter is generated from $received_to and is only set if the
2242 message has one envelope recipient. SA can use these headers,
2243 obviously out-of-the-box. Patch from Alex Miller.
2245 PH/08 The ${def test on a variable was returning false if the variable's
2246 value was "0", contrary to what the specification has always said!
2247 The result should be true unless the variable is empty.
2249 PH/09 The syntax error of a character other than { following "${if
2250 def:variable_name" (after optional whitespace) was not being diagnosed.
2251 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
2252 accidental colon was present, for example, could give incorrect results.
2254 PH/10 Tidied the code in a number of places where the st_size field of a stat()
2255 result is used (not including appendfile, where other changes are about
2258 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
2259 This involved changing a lot of size variables from int to off_t. It
2260 should work with maildirs and everything.
2262 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
2263 spamd dying while we are connected to it.
2265 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
2268 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
2269 function for BDB 4.3.
2271 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
2273 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
2274 the sources for PCRE have been re-arranged and more files are now
2277 PH/15 The code I had for printing potentially long long variables in PH/11
2278 above was not the best (it lost precision). The length of off_t variables
2279 is now inspected at build time, and an appropriate printing format (%ld
2280 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
2281 to be "long long int" or "long int". This is needed for the internal
2282 formatting function string_vformat().
2284 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
2285 the configuration file to be ":syslog", then the script "guesses" where
2286 the logs files are, rather than using the compiled in default. In our
2287 case the guess is not the same as the compiled default, so the script
2288 suddenly stopped working when I started to use syslog. The patch checks
2289 to see if log_file_path is "". If so, it attempts to read it from exim
2290 with no configuration file to get the compiled in version, before it
2291 falls back to the previous guessing code."
2293 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2294 implementing BATV in an Exim configuration. See NewStuff for the gory
2297 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2298 Makefile that are specific to HP-UX.
2300 PH/18 If the "use_postmaster" option was set for a recipient callout together
2301 with the "random" option, the postmaster address was used as the MAIL
2302 FROM address for the random test, but not for the subsequent recipient
2303 test. It is now used for both.
2305 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2306 patch removes a few documentation additions to RFC 3028, because the
2307 latest draft now contains them. It adds the new en;ascii-case comparator
2308 and a new error check for 8bit text in MIME parts. Comparator and
2309 require names are now matched exactly. I enabled the subaddress
2310 extension, but it is not well tested yet (read: it works for me)."
2312 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2313 rework some of the code of TK/09 above to avoid the hardwired use of
2314 "%lld" and "long long". Replaced the call to snprintf() with a call to
2317 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2318 records point to non-existent hosts", "retry timeout exceeded", and
2319 "retry time not reached for any host after a long failure period".
2321 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2322 experimental DomainKeys support:
2324 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2325 (2) On an error such as an illegally used "control", the wrong name for
2326 the control was given.
2328 These problems did NOT occur unless DomainKeys support was compiled.
2330 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2332 PH/24 Added ${if match_ip condition.
2334 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2335 databases so that it will be absolutely obvious if a crash occurs in the
2336 DB library. This is a regular occurrence (often caused by mis-matched
2339 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2340 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2341 picked up on a user's system that detects such things. There doesn't seem
2342 to be a gcc warning option for this - only an attribute that has to be
2343 put on the function's prototype. It seems that in Fedora Core 4 they have
2344 set this on a number of new functions. No doubt there will be more in due
2347 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2348 specify a transport. However, if an address that was verified by such a
2349 router was the subject of a callout, Exim crashed because it tried to
2350 read the rcpt_include_affixes from the non-existent transport. Now it
2351 just assumes that the setting of that option is false. This bug was
2352 introduced by 4.51/PH/31.
2354 PH/28 Changed -d+all to exclude +memory, because that information is very
2355 rarely of interest, but it makes the output a lot bigger. People tend to
2356 do -d+all out of habit.
2358 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2359 code in os-type was giving problems when libc.so lives in lib64, like on
2362 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2363 aren't the modern standard, and it seems that some systems' include files
2364 don't always have them. Exim was already checking for some of the newer
2365 ones like T_AAAA, and defining it itself. I've added checks for all the
2366 record types that Exim uses.
2368 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2369 not automatically generating a new one, as it is supposed to. This
2370 prevented TLS from working. If the file did exist, but contained invalid
2371 data, a new version was generated, as expected. It was only the case of a
2372 non-existent file that was broken.
2374 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2375 with a change in libdomainkeys > 0.64.
2377 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2378 from DNS. If the selector record carries the flag, it now has
2379 precedence over the domain-wide flag.
2381 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2383 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2384 the use of a port name that isn't defined in /etc/services) occurred, the
2385 message was deferred as in a normal delivery, and thus remained on the
2386 spool, instead of being failed because of the mua_wrapper setting. This
2387 is now fixed, and I tidied up some of the mua_wrapper messages at the
2390 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2391 the messages in a hash of arrays rather than using individual hashes.
2392 This is a bit cleaner and results in dramatic memory savings, albeit
2393 at a slight CPU cost.
2395 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2396 as requested by Marc Sherman.
2398 SC/10 Eximstats: added histograms for user specified patterns as requested
2401 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2403 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2404 fopen() in the content-scanning modules that did not already have it.
2410 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2411 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2413 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2415 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2417 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2418 to test Sieve filters that use "vacation".
2420 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2421 that changes the way the GnuTLS parameters are stored in the cache file.
2422 The new format can be generated externally. For backward compatibility,
2423 if the data in the cache doesn't make sense, Exim assumes it has read an
2424 old-format file, and it generates new data and writes a new file. This
2425 means that you can't go back to an older release without removing the
2428 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2429 work if there are any delivery delays because "one_time" forces the
2430 parent to be marked "delivered", so its unseen clone is never tried
2431 again. For this reason, Exim now forbids the simultaneous setting of
2434 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2435 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2436 entry if you want to know the details. The fix, however, neglected to
2437 consider the case when local delivery batching is involved. The test for
2438 "previously delivered" was not happening when checking to see if an
2439 address could be batched with a previous (undelivered) one; under
2440 certain circumstances this could lead to multiple deliveries to the same
2443 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2444 in its include files, and this causes problems building Exim.
2446 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2447 header_syntax) but Exim was just ignoring anything given after a slash.
2448 In particular, this caused confusion with an attempt to use "verify =
2449 reverse_host_lookup/defer_ok". An error is now given when options are
2450 supplied for verify items that do not have them. (Maybe reverse_host_
2451 lookup should have a defer_ok option, but that's a different point.)
2453 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2454 defined by RFC 821) to 2048, because there were problems with some AUTH
2455 commands, and RFC 1869 says the size should be increased for extended
2456 SMTP commands that take arguments.
2458 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2461 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2462 "unknown" error; now it says that the functionality isn't in the binary.
2464 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2465 an address' error message when a string expansion fails (syntax or
2466 whatever). Otherwise the password may appear in the log. Following change
2467 PH/42 below, there is no longer a chance of it appearing in a bounce
2470 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2472 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2473 of its addresses was ever tried. (Bugzilla bug #2.)
2475 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2476 the result incorrectly in the debug output. (It correctly added a newline
2477 to what was transported.)
2479 TF/01 Added $received_time.
2481 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2482 commented out examples of how to interface to a virus scanner and to
2483 SpamAssassin. Also added commented examples of av_scanner and
2484 spamd_address settings.
2486 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2487 and controls are allowed in which ACLs. There were a couple of minor
2488 errors. Some of the entries in the conditions table (which is a table of
2489 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2490 negation of where the condition IS allowed.
2492 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2494 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2495 header file does not have a version number, so I've had to invent a new
2496 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2497 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2498 radiusclient), but it was contributed by a Radius user.
2500 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2501 files or directories, for OpenSSL.
2503 PH/19 When an Exim process that is running as root has to create an Exim log
2504 file, it does so in a subprocess that runs as exim:exim so as to get the
2505 ownership right at creation (otherwise, other Exim processes might see
2506 the file with the wrong ownership). There was no test for failure of this
2507 fork() call, which would lead to the process getting stuck as it waited
2508 for a non-existent subprocess. Forks do occasionally fail when resources
2509 run out. I reviewed all the other calls to fork(); they all seem to check
2512 PH/20 When checking for unexpected SMTP input at connect time (before writing
2513 the banner), Exim was not dealing correctly with a non-positive return
2514 from the read() function. If the client had disconnected by this time,
2515 the result was a log entry for a synchronization error with an empty
2516 string after "input=" when read() returned zero. If read() returned -1
2517 (an event I could not check), uninitialized data bytes were printed.
2518 There were reports of junk text (parts of files, etc) appearing after
2521 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2523 PH/22 Added support for macro redefinition, and (re)definition in between
2524 driver and ACL definitions.
2526 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2527 forgetting to use the resulting value; it was using the unexpanded value.
2529 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2530 hadn't been configured. The fix is from Juergen Kreileder, who
2531 understands it better than I do:
2533 "Here's what I see happening with three configured cyrus_sasl
2534 authenticators configured (plain, login, cram-md5):
2536 On startup auth_cyrus_sasl_init() gets called for each of these.
2537 This means three calls to sasl_listmech() without a specified mech_list.
2538 => SASL tests which mechs of all available mechs actually work
2539 => three warnings about OTP not working
2540 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2542 With the patch, sasl_listmech() also gets called three times. But now
2543 SASL's mech_list option is set to the server_mech specified in the the
2544 authenticator. Or in other words, the answer from sasl_listmech()
2545 gets limited to just the mech you're testing for (which is different
2547 => the return list contains just 'plain' or 'login', 'cram-md5' or
2548 nothing depending on the value of ob->server_mech.
2550 I've just tested the patch: Authentication still works fine,
2551 unavailable mechs specified in the exim configuration are still
2552 caught, and the auth.log warnings about OTP are gone."
2554 PH/25 When debugging is enabled, the contents of the command line are added
2555 to the debugging output, even when log_selector=+arguments is not
2558 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2559 answer is "GNU", and only if the return is "GNU/something" is the answer
2562 PH/27 $acl_verify_message is now set immediately after the failure of a
2563 verification in an ACL, and so is available in subsequent modifiers. In
2564 particular, the message can be preserved by coding like this:
2566 warn !verify = sender
2567 set acl_m0 = $acl_verify_message
2569 Previously, $acl_verify_message was set only while expanding "message"
2570 and "log_message" when a very denied access.
2572 PH/28 Modified OS/os.c-Linux with
2574 -#ifndef OS_LOAD_AVERAGE
2575 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2577 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2578 nomenclature these days.)
2580 PH/29 Installed patch from the Sieve maintainer that adds the options
2581 sieve_useraddress and sieve_subaddress to the redirect router.
2583 PH/30 In these circumstances:
2584 . Two addresses routed to the same list of hosts;
2585 . First host does not offer TLS;
2586 . First host accepts first address;
2587 . First host gives temporary error to second address;
2588 . Second host offers TLS and a TLS session is established;
2589 . Second host accepts second address.
2590 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2591 and peerdn, if requested) that were in fact used only for the second
2594 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2595 attention to any local part prefix or suffix that was matched by the
2596 router that accepted the address. It now behaves in the same way as it
2597 does for delivery: the affixes are removed from the local part unless
2598 rcpt_include_affixes is set on the transport.
2600 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2601 timeout during the DATA phase of an incoming message.
2603 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2604 applied a patch sanctioned by the Sieve maintainer.
2606 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2607 the uid or gid is negative. A case of a negative gid caused this to be
2608 noticed. The fix allows for either to be negative.
2610 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2611 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2614 PH/36 The change PH/12 above was broken. Fixed it.
2616 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2617 the grounds that routing the same address twice would always produce the
2618 same answer. This might have been true once, but it is certainly no
2619 longer true now. Routing a child address may depend on the previous
2620 routing that produced that child. Some complicated redirection strategies
2621 went wrong when messages had multiple recipients, and made Exim's
2622 behaviour dependent on the order in which the addresses were given.
2624 I have moved the duplicate checking until after the routing is complete.
2625 Exim scans the addresses that are assigned to local and remote
2626 transports, and removes any duplicates. This means that more work will be
2627 done, as duplicates will always all be routed, but duplicates are
2628 presumably rare, so I don't expect this is of any significance.
2630 For deliveries to pipes, files, and autoreplies, the duplicate checking
2631 still happens during the routing process, since they are not going to be
2634 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2635 It corrects a timeout issue with spamd. This is Ian's comment: "The
2636 background is that sometimes spamd either never reads data from a
2637 connection it has accepted, or it never writes response data. The exiscan
2638 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2639 blindly assumes that writes won't block so it may never time out."
2641 PH/39 Allow G after quota size as well as K and M.
2643 PH/40 The value set for $authenticated_id in an authenticator may not contain
2644 binary zeroes or newlines because the value is written to log lines and
2645 to spool files. There was no check on this. Now the value is run through
2646 the string_printing() function so that such characters are converted to
2647 printable escape sequences.
2649 PH/41 $message_linecount is a new variable that contains the total number of
2650 lines in the message. Compare $body_linecount, which is the count for the
2653 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2654 bounce and delay warning messages, except in certain special cases, which
2657 (a) An SMTP error message from a remote host;
2658 (b) A message specified in a :fail: redirection;
2659 (c) A message specified in a "fail" command in a system filter;
2660 (d) A message specified in a FAIL return from the queryprogram router;
2661 (e) A message specified by the cannot_route_message router option.
2663 In these cases only, Exim does include the error details in bounce and
2664 warning messages. There are also a few cases where bland messages such
2665 as "unrouteable address" or "local delivery error" are given.
2667 PH/43 $value is now also set for the "else" part of a ${run expansion.
2669 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2670 being worked on, but at least Exim now implements the latest version to
2673 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2674 process to complete was treated as a delivery failure, a timeout while
2675 writing the message to the pipe was logged, but erroneously treated as a
2676 successful delivery. Such timeouts include transport filter timeouts. For
2677 consistency with the overall process timeout, these timeouts are now
2678 treated as errors, giving rise to delivery failures by default. However,
2679 there is now a new Boolean option for the pipe transport called
2680 timeout_defer, which, if set TRUE, converts the failures into defers for
2681 both kinds of timeout. A transport filter timeout is now identified in
2684 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2685 systems where "make" and "gmake" are different, calling "gmake" at top
2686 level broke things. I've arranged for the value of $(MAKE) to be passed
2687 from the Makefile to this script so that it can call the same version of
2691 A note about Exim versions 4.44 and 4.50
2692 ----------------------------------------
2694 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2695 changes of various kinds. As a consequence, a big documentation update was
2696 needed. This delayed the release for rather longer than seemed good, especially
2697 in the light of a couple of (minor) security issues. Therefore, the changes
2698 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2699 release. So 4.44 and 4.50 are in effect two different branches that both start
2702 I have left the 4.50 change log unchanged; it contains all the changes since
2703 4.43. The change log for 4.44 is below; many of its items are identical to
2704 those for 4.50. This seems to be the most sensible way to preserve the
2705 historical information.
2711 1. Minor wording change to the doc/README.SIEVE file.
2713 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2714 computation of the current number of files was incorrect.
2716 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2717 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2720 4. Give more explanation in the error message when the command for a transport
2721 filter fails to execute.
2723 5. There are several places where Exim runs a non-Exim command in a
2724 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2725 was being done only for the command run by the queryprogram router. It is
2726 now done for all such subprocesses. The other cases are: ${run, transport
2727 filters, and the commands run by the lmtp and pipe transports.
2729 6. Added CONFIGURE_GROUP build-time option.
2731 7. Some older OS have a limit of 256 on the maximum number of file
2732 descriptors. Exim was using setrlimit() to set 1000 as a large value
2733 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2734 systems. I've change it so that if it can't get 1000, it tries for 256.
2736 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2737 was an oversight, and furthermore, ever since the addition of extra
2738 controls (e.g. 4.43/32), the checks on when to allow different forms of
2739 "control" were broken. There should now be diagnostics for all cases when a
2740 control that does not make sense is encountered.
2742 9. Added the /retain_sender option to "control=submission".
2744 10. $recipients is now available in the predata ACL (oversight).
2746 11. Tidy the search cache before the fork to do a delivery from a message
2747 received from the command line. Otherwise the child will trigger a lookup
2748 failure and thereby defer the delivery if it tries to use (for example) a
2749 cached ldap connection that the parent has called unbind on.
2751 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2752 of $address_data from the recipient verification was clobbered by the
2753 sender verification.
2755 13. The value of address_data from a sender verification is now available in
2756 $sender_address_data in subsequent conditions in the ACL statement.
2758 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2760 15. Added a new option "connect=<time>" to callout options, to set a different
2763 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2764 was its contents. (It was OK if the option was not defined at all.)
2766 17. A "Completed" log line is now written for messages that are removed from
2767 the spool by the -Mrm option.
2769 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2770 information about exactly what failed.
2772 19. Added -dd to debug only the daemon process.
2774 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2775 handles timeouts, both on the server side and network timeouts. Renamed the
2776 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2778 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2779 It is now set to "smtps".
2781 22. $host_address is now set to the target address during the checking of
2782 ignore_target_hosts.
2784 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2785 being passed; this would have caused $sender_host_name to have been used if
2786 matching the list had actually called for a host name (not very likely,
2787 since this list is usually IP addresses). A host name is now passed as
2790 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2791 code that specifies a non-restarting handler (typically sigaction() in
2792 modern systems) in an attempt to fix a rare and obscure crash bug.
2794 25. Narrowed the window for a race in the daemon that could cause it to ignore
2795 SIGCHLD signals. This is not a major problem, because they are used only to
2796 wake it up if nothing else does.
2798 26. A malformed maildirsize file could cause Exim to calculate negative values
2799 for the mailbox size or file count. Odd effects could occur as a result.
2800 The maildirsize information is now recalculated if the size or filecount
2803 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2804 support for a long time. Removed HAVE_SYS_VFS_H.
2806 28. Installed the latest version of exipick from John Jetmore.
2808 29. In an address list, if the pattern was not a regular expression, an empty
2809 subject address (from a bounce message) matched only if the pattern was an
2810 empty string. Non-empty patterns were not even tested. This was the wrong
2811 because it is perfectly reasonable to use an empty address as part of a
2812 database query. An empty address is now tested by patterns that are
2813 lookups. However, all the other forms of pattern expect the subject to
2814 contain a local part and a domain, and therefore, for them, an empty
2815 address still always fails if the pattern is not itself empty.
2817 30. Exim went into a mad DNS loop when attempting to do a callout where the
2818 host was specified on an smtp transport, and looking it up yielded more
2819 than one IP address.
2821 31. Re-factored the code for checking spool and log partition space into a
2822 function that finds that data and another that does the check. The former
2823 is then used to implement four new variables: $spool_space, $log_space,
2824 $spool_inodes, and $log_inodes.
2826 32. The RFC2047 encoding function was originally intended for short strings
2827 such as real names; it was not keeping to the 75-character limit for
2828 encoded words that the RFC imposes. It now respects the limit, and
2829 generates multiple encoded words if necessary. To be on the safe side, I
2830 have increased the buffer size for the ${rfc2047: expansion operator from
2833 33. It is now permitted to omit both strings after an "if" condition; if the
2834 condition is true, the result is "true". As before, when the second string
2835 is omitted, a false condition yields an empty string. This makes it less
2836 cumbersome to write custom ACL and router conditions.
2838 34. Failure to deliver a bounce message always caused it to be frozen, even if
2839 there was an errors_to setting on the router. The errors_to setting is now
2842 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2843 canonical form (fully expanded) before being placed in
2844 $sender_host_address.
2846 36. The table in the code that translates DNS record types into text (T_A to
2847 "A" for instance) was missing entries for NS and CNAME. It is just possible
2848 that this could have caused confusion if both these types were looked up
2849 for the same domain, because the text type is used as part of Exim's
2850 per-process caching. But the chance of anyone hitting this buglet seems
2853 37. The dnsdb lookup has been extended in a number of ways.
2855 (1) There is a new type, "zns", which walks up the domain tree until it
2856 finds some nameserver records. It should be used with care.
2858 (2) There is a new type, "mxh", which is like "mx" except that it returns
2859 just the host names, not the priorities.
2861 (3) It is now possible to give a list of domains (or IP addresses) to be
2862 looked up. The behaviour when one of the lookups defers can be
2863 controlled by a keyword.
2865 (4) It is now possible to specify the separator character for use when
2866 multiple records are returned.
2868 38. The dnslists ACL condition has been extended: it is now possible to supply
2869 a list of IP addresses and/or domains to be looked up in a particular DNS
2872 39. Added log_selector=+queue_time_overall.
2874 40. When running the queue in the test harness, wait just a tad after forking a
2875 delivery process, to get repeatability of debugging output.
2877 41. Include certificate and key file names in error message when GnuTLS fails
2878 to set them up, because the GnuTLS error message doesn't include the name
2879 of the failing file when there is a problem reading it.
2881 42. Allow both -bf and -bF in the same test run.
2883 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2885 44. The "Exiscan patch" is now merged into the mainline Exim source.
2887 45. Sometimes the final signoff response after QUIT could fail to get
2888 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2889 < 0 before doing a fflush(). This bug looks as though it goes back to the
2890 introduction of TLS in release 3.20, but "sometimes" must have been rare
2891 because the tests only now provoked it.
2893 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2894 (this can affect the format of dates).
2896 47. exim_tidydb, when checking for the continued existence of a message for
2897 which it has found a message-specific retry record, was not finding
2898 messages that were in split spool directories. Consequently, it was
2899 deleting retry records that should have stayed in existence.
2901 48. Steve fixed some bugs in eximstats.
2903 49. The SPA authentication driver was not abandoning authentication and moving
2904 on to the next authenticator when an expansion was forced to fail,
2905 contradicting the general specification for all authenticators. Instead it
2906 was generating a temporary error. It now behaves as specified.
2908 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2909 (the order specifies the preference for clients). The order is now AES256,
2910 AES128, 3DES, ARCFOUR128.
2912 51. Small patch to Sieve code - explicitly set From: when generating an
2915 52. Exim crashed if a remote delivery caused a very long error message to be
2916 recorded - for instance if somebody sent an entire SpamAssassin report back
2917 as a large number of 550 error lines. This bug was coincidentally fixed by
2918 increasing the size of one of Exim's internal buffers (big_buffer) that
2919 happened as part of the Exiscan merge. However, to be on the safe side, I
2920 have made the code more robust (and fixed the comments that describe what
2923 53. Now that there can be additional text after "Completed" in log lines (if
2924 the queue_time_overall log selector is set), a one-byte patch to exigrep
2925 was needed to allow it to recognize "Completed" as not the last thing in
2928 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2929 patch that reportedly fixes this has been added. I am not expert enough to
2930 create a test for it. This is what the patch creator wrote:
2932 "I found a little strange behaviour of ldap code when working with
2933 Windows 2003 AD Domain, where users was placed in more than one
2934 Organization Units. When I tried to give exim partial DN, the exit code
2935 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2936 But simultaneously result of request was absolutely normal ldap result,
2937 so I produce this patch..."
2939 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2940 REFERENCE, so I have modified the code to exclude the patch when that macro
2943 55. Some experimental protocols are using DNS PTR records for new purposes. The
2944 keys for these records are domain names, not reversed IP addresses. The
2945 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2946 leaves it alone. Component reversal etc. now happens only for IP addresses.
2949 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2951 57. Double the size of the debug message buffer (to 2048) so that more of very
2952 long debug lines gets shown.
2954 58. The exicyclog utility now does better if the number of log files to keep
2955 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2957 59. Two changes related to the smtp_active_hostname option:
2959 (1) $smtp_active_hostname is now available as a variable.
2960 (2) The default for smtp_banner uses $smtp_active_hostname instead
2961 of $primary_hostname.
2963 60. The host_aton() function is supposed to be passed a string that is known
2964 to be a valid IP address. However, in the case of IPv6 addresses, it was
2965 not checking this. This is a hostage to fortune. Exim now panics and dies
2966 if the condition is not met. A case was found where this could be provoked
2967 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2968 components; fortuitously, this particular loophole had already been fixed
2969 by change 4.50/55 above.
2971 If there are any other similar loopholes, the new check in host_aton()
2972 itself should stop them being exploited. The report I received stated that
2973 data on the command line could provoke the exploit when Exim was running as
2974 exim, but did not say which command line option was involved. All I could
2975 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2976 running as the user.
2979 61. There was a buffer overflow vulnerability in the SPA authentication code
2980 (which came originally from the Samba project). I have added a test to the
2981 spa_base64_to_bits() function which I hope fixes it.
2984 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2985 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2987 63. The daemon start-up calls getloadavg() while still root for those OS that
2988 need the first call to be done as root, but it missed one case: when
2989 deliver_queue_load_max is set with deliver_drop_privilege. This is
2990 necessary for the benefit of the queue runner, because there is no re-exec
2991 when deliver_drop_privilege is set.
2993 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2994 This has been fixed.
2996 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2997 was in use, was not putting the data itself into the right store pool;
2998 consequently, it could be overwritten for a subsequent message in the same
2999 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3002 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
3004 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
3005 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
3006 the function were treating the return as a boolean value, which happened to
3007 work because 0=false and not-0=true, but is not correct code.
3009 68. The host_aton() function was not handling scoped IPv6 addresses (those
3010 with, for example, "%eth0" on the end) correctly.
3012 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
3013 negated items (that is, ~something) in unsigned ints. Some compilers
3014 apparently mutter when there is no cast.
3016 70. If an address verification called from an ACL failed, and did not produce a
3017 user-specific message (i.e. there was only a "system" message), nothing was
3018 put in $acl_verify_message. In this situation, it now puts the system
3021 71. Change 4.23/11 added synchronization checking at the start of an SMTP
3022 session; change 4.31/43 added the unwanted input to the log line - except
3023 that it did not do this in the start of session case. It now does.
3025 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
3026 This is wrong and can cause the other end to generate a synchronization
3027 error if it is another Exim or anything else that does the synchronization
3028 check. A QUIT command is no longer sent after a timeout.
3030 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
3031 during host lookups.
3033 74. The defer_ok option of callout verification was not working if it was used
3034 when verifying addresses in header lines, that is, for this case:
3036 verify = header_sender/callout=defer_ok
3038 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
3039 those file descriptors could be used for SMTP connections. If anything
3040 wrote to stderr (the example that came up was "warn" in embedded Perl), it
3041 could be sent to the SMTP client, causing chaos. The daemon now opens
3042 stdin, stdout, and stderr to /dev/null when it puts itself into the
3045 76. Arrange for output from Perl's "warn" command to be written to Exim's main
3046 log by default. The user can override this with suitable Perl magic.
3048 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
3049 the log message when discard triggers, was not working for the DATA ACL or
3050 for the non-SMTP ACL.
3052 78. Error message wording change in sieve.c.
3054 79. If smtp_accept_max_per_host was set, the number of connections could be
3055 restricted to fewer than expected, because the daemon was trying to set up
3056 a new connection before checking whether the processes handling previous
3057 connections had finished. The check for completed processes is now done
3058 earlier. On busy systems, this bug wouldn't be noticed because something
3059 else would have woken the daemon, and it would have reaped the completed
3062 80. If a message was submitted locally by a user whose login name contained one
3063 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
3064 It caused a spool format error. I have fixed the spool reading code. A
3065 related problem was that the "from" clause in the Received: line became
3066 illegal because of the space(s). It is now covered by ${quote_local_part.
3068 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
3071 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
3073 83. Patch from Sieve maintainer to fix unterminated string problem in
3074 "vacation" handling.
3076 84. Some minor changes to the Linux configuration files to help with other
3077 OS variants using glibc.
3079 85. One more patch for Sieve to update vacation handling to latest spec.
3082 ----------------------------------------------------
3083 See the note above about the 4.44 and 4.50 releases.
3084 ----------------------------------------------------
3090 1. Change 4.43/35 introduced a bug that caused file counts to be
3091 incorrectly computed when quota_filecount was set in an appendfile
3094 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
3095 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
3098 3. Give more explanation in the error message when the command for a transport
3099 filter fails to execute.
3101 4. There are several places where Exim runs a non-Exim command in a
3102 subprocess. The SIGUSR1 signal should be disabled for these processes. This
3103 was being done only for the command run by the queryprogram router. It is
3104 now done for all such subprocesses. The other cases are: ${run, transport
3105 filters, and the commands run by the lmtp and pipe transports.
3107 5. Some older OS have a limit of 256 on the maximum number of file
3108 descriptors. Exim was using setrlimit() to set 1000 as a large value
3109 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
3110 systems. I've change it so that if it can't get 1000, it tries for 256.
3112 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
3113 was an oversight, and furthermore, ever since the addition of extra
3114 controls (e.g. 4.43/32), the checks on when to allow different forms of
3115 "control" were broken. There should now be diagnostics for all cases when a
3116 control that does not make sense is encountered.
3118 7. $recipients is now available in the predata ACL (oversight).
3120 8. Tidy the search cache before the fork to do a delivery from a message
3121 received from the command line. Otherwise the child will trigger a lookup
3122 failure and thereby defer the delivery if it tries to use (for example) a
3123 cached ldap connection that the parent has called unbind on.
3125 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
3126 of $address_data from the recipient verification was clobbered by the
3127 sender verification.
3129 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
3130 was its contents. (It was OK if the option was not defined at all.)
3132 11. A "Completed" log line is now written for messages that are removed from
3133 the spool by the -Mrm option.
3135 12. $host_address is now set to the target address during the checking of
3136 ignore_target_hosts.
3138 13. When checking ignore_target_hosts for an ipliteral router, no host name was
3139 being passed; this would have caused $sender_host_name to have been used if
3140 matching the list had actually called for a host name (not very likely,
3141 since this list is usually IP addresses). A host name is now passed as
3144 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
3145 code that specifies a non-restarting handler (typically sigaction() in
3146 modern systems) in an attempt to fix a rare and obscure crash bug.
3148 15. Narrowed the window for a race in the daemon that could cause it to ignore
3149 SIGCHLD signals. This is not a major problem, because they are used only to
3150 wake it up if nothing else does.
3152 16. A malformed maildirsize file could cause Exim to calculate negative values
3153 for the mailbox size or file count. Odd effects could occur as a result.
3154 The maildirsize information is now recalculated if the size or filecount
3157 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
3158 support for a long time. Removed HAVE_SYS_VFS_H.
3160 18. Updated exipick to current release from John Jetmore.
3162 19. Allow an empty sender to be matched against a lookup in an address list.
3163 Previously the only cases considered were a regular expression, or an
3166 20. Exim went into a mad DNS lookup loop when doing a callout where the
3167 host was specified on the transport, if the DNS lookup yielded more than
3170 21. The RFC2047 encoding function was originally intended for short strings
3171 such as real names; it was not keeping to the 75-character limit for
3172 encoded words that the RFC imposes. It now respects the limit, and
3173 generates multiple encoded words if necessary. To be on the safe side, I
3174 have increased the buffer size for the ${rfc2047: expansion operator from
3177 22. Failure to deliver a bounce message always caused it to be frozen, even if
3178 there was an errors_to setting on the router. The errors_to setting is now
3181 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
3182 canonical form (fully expanded) before being placed in
3183 $sender_host_address.
3185 24. Updated eximstats to version 1.33
3187 25. Include certificate and key file names in error message when GnuTLS fails
3188 to set them up, because the GnuTLS error message doesn't include the name
3189 of the failing file when there is a problem reading it.
3191 26. Expand error message when OpenSSL has problems setting up cert/key files.
3194 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
3195 (this can affect the format of dates).
3197 28. exim_tidydb, when checking for the continued existence of a message for
3198 which it has found a message-specific retry record, was not finding
3199 messages that were in split spool directories. Consequently, it was
3200 deleting retry records that should have stayed in existence.
3202 29. eximstats updated to version 1.35
3203 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
3204 1.35 - bugfix such that pie charts by volume are generated correctly
3206 30. The SPA authentication driver was not abandoning authentication and moving
3207 on to the next authenticator when an expansion was forced to fail,
3208 contradicting the general specification for all authenticators. Instead it
3209 was generating a temporary error. It now behaves as specified.
3211 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
3212 (the order specifies the preference for clients). The order is now AES256,
3213 AES128, 3DES, ARCFOUR128.
3215 31. Small patch to Sieve code - explicitly set From: when generating an
3218 32. Exim crashed if a remote delivery caused a very long error message to be
3219 recorded - for instance if somebody sent an entire SpamAssassin report back
3220 as a large number of 550 error lines. This bug was coincidentally fixed by
3221 increasing the size of one of Exim's internal buffers (big_buffer) that
3222 happened as part of the Exiscan merge. However, to be on the safe side, I
3223 have made the code more robust (and fixed the comments that describe what
3226 33. Some experimental protocols are using DNS PTR records for new purposes. The
3227 keys for these records are domain names, not reversed IP addresses. The
3228 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
3229 leaves it alone. Component reversal etc. now happens only for IP addresses.
3232 34. The host_aton() function is supposed to be passed a string that is known
3233 to be a valid IP address. However, in the case of IPv6 addresses, it was
3234 not checking this. This is a hostage to fortune. Exim now panics and dies
3235 if the condition is not met. A case was found where this could be provoked
3236 from a dnsdb PTR lookup with an IPv6 address that had more than 8
3237 components; fortuitously, this particular loophole had already been fixed
3238 by change 4.50/55 or 4.44/33 above.
3240 If there are any other similar loopholes, the new check in host_aton()
3241 itself should stop them being exploited. The report I received stated that
3242 data on the command line could provoke the exploit when Exim was running as
3243 exim, but did not say which command line option was involved. All I could
3244 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
3245 running as the user.
3248 35. There was a buffer overflow vulnerability in the SPA authentication code
3249 (which came originally from the Samba project). I have added a test to the
3250 spa_base64_to_bits() function which I hope fixes it.
3253 36. The daemon start-up calls getloadavg() while still root for those OS that
3254 need the first call to be done as root, but it missed one case: when
3255 deliver_queue_load_max is set with deliver_drop_privilege. This is
3256 necessary for the benefit of the queue runner, because there is no re-exec
3257 when deliver_drop_privilege is set.
3259 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
3260 was in use, was not putting the data itself into the right store pool;
3261 consequently, it could be overwritten for a subsequent message in the same
3262 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
3265 38. Sometimes the final signoff response after QUIT could fail to get
3266 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
3267 < 0 before doing a fflush(). This bug looks as though it goes back to the
3268 introduction of TLS in release 3.20, but "sometimes" must have been rare
3269 because the tests only now provoked it.
3275 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
3276 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
3277 Now it returns an integer. A number of calls were still expecting a T/F
3278 return. Fortuitously, in all cases, the tests worked in OK situations,
3279 which is the norm. However, things would have gone wrong on any write
3280 failures on the smtp file descriptor. This function is used when sending
3281 messages over SMTP and also when doing verify callouts.
3283 2. When Exim is called to do synchronous delivery of a locally submitted
3284 message (the -odf or -odi options), it no longer closes stderr before doing
3287 3. Implemented the mua_wrapper option.
3289 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
3291 5. Implemented the functions header_remove(), header_testname(),
3292 header_add_at_position(), and receive_remove_recipient(), and exported them
3295 6. If an ACL "warn" statement specified the addition of headers, Exim already
3296 inserted X-ACL-Warn: at the start if there was no header name. However, it
3297 was not making this test for the second and subsequent header lines if
3298 there were newlines in the string. This meant that an invalid header could
3299 be inserted if Exim was badly configured.
3301 7. Allow an ACL "warn" statement to add header lines at the start or after all
3302 the Received: headers, as well as at the end.
3304 8. Added the rcpt_4xx retry error code.
3306 9. Added postmaster_mailfrom=xxx to callout verification option.
3308 10. Added mailfrom=xxxx to the callout verification option, for verify=
3311 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3312 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3314 12. Inserted some casts to stop certain compilers warning when using pointer
3315 differences as field lengths or precisions in printf-type calls (mostly
3316 affecting debugging statements).
3318 13. Added optional readline() support for -be (dynamically loaded).
3320 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3321 same clock tick as a message's arrival, so that its received time was the
3322 same as the "first fail" time on the retry record, and that message
3323 remained on the queue past the ultimate address timeout, every queue runner
3324 would try a delivery (because it was past the ultimate address timeout) but
3325 after another failure, the ultimate address timeout, which should have then
3326 bounced the address, did not kick in. This was a "< instead of <=" error;
3327 in most cases the first failure would have been in the next clock tick
3328 after the received time, and all would be well.
3330 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3331 being recognized when the domain list was tested by the match_domain
3332 condition in an expansion string.
3334 16. Added the ${str2b64: operator.
3336 17. Exim was always calling setrlimit() to set a large limit for the number of
3337 processes, without checking whether the existing limit was already
3338 adequate. (It did check for the limit on file descriptors.) Furthermore,
3339 errors from getrlimit() and setrlimit() were being ignored. Now they are
3340 logged to the main and panic logs, but Exim does carry on, to try to do its
3341 job under whatever limits there are.
3343 18. Imported PCRE 5.0.
3345 19. Trivial typo in log message " temporarily refused connection" (the leading
3348 20. If the log selector return_path_on_delivery was set and an address was
3349 redirected to /dev/null, the delivery process crashed because it assumed
3350 that a return path would always be set for a "successful" delivery. In this
3351 case, the whole delivery is bypassed as an optimization, and therefore no
3354 21. Internal re-arrangement: the function for sending a challenge and reading
3355 a response while authentication was assuming a zero-terminated challenge
3356 string. It's now changed to take a pointer and a length, to allow for
3357 binary data in such strings.
3359 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3361 23. Exim was not respecting finduser_retries when seeking the login of the
3362 uid under which it was called; it was always trying 10 times. (The default
3363 setting of finduser_retries is zero.) Also, it was sleeping after the final
3364 failure, which is pointless.
3366 24. Implemented tls_on_connect_ports.
3368 25. Implemented acl_smtp_predata.
3370 26. If the domain in control=submission is set empty, Exim assumes that the
3371 authenticated id is a complete email address when it generates From: or
3372 Sender: header lines.
3374 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3375 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3376 chown and chgrp in /bin and hostname in /usr/bin.
3378 28. Exim was keeping the "process log" file open after each use, just as it
3379 does for the main log. This opens the possibility of it remaining open for
3380 long periods when the USR1 signal hits a daemon. Occasional processlog
3381 errors were reported, that could have been caused by this. Anyway, it seems
3382 much more sensible not to leave this file open at all, so that is what now
3385 29. The long-running daemon process does not normally write to the log once it
3386 has entered its main loop, and it closes the log before doing so. This is
3387 so that log files can straightforwardly be renamed and moved. However,
3388 there are a couple of unusual error situations where the daemon does write
3389 log entries, and I had neglected to close the log afterwards.
3391 30. The text of an SMTP error response that was received during a remote
3392 delivery was being truncated at 512 bytes. This is too short for some of
3393 the long messages that one sometimes sees. I've increased the limit to
3396 31. It is now possible to make retry rules that apply only when a message has a
3397 specific sender, in particular, an empty sender.
3399 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3400 it possible to be selective about when SMTP synchronization is enforced.
3402 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3404 32. Implemented hosts_connection_nolog.
3406 33. Added an ACL for QUIT.
3408 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3411 35. Added mailbox_size and mailbox_filecount to appendfile.
3413 36. Added control = no_multiline_responses to ACLs.
3415 37. There was a bug in the logic of the code that waits for the clock to tick
3416 in the case where the clock went backwards by a substantial amount such
3417 that the microsecond fraction of "now" was more than the microsecond
3418 fraction of "then" (but the whole seconds number was less).
3420 38. Added support for the libradius Radius client library this is found on
3421 FreeBSD (previously only the radiusclient library was supported).
3427 1. When certain lookups returned multiple values in the form name=value, the
3428 quoting of the values was not always being done properly. Specifically:
3429 (a) If the value started with a double quote, but contained no whitespace,
3431 (b) If the value contained whitespace other than a space character (i.e.
3432 tabs or newlines or carriage returns) it was not quoted.
3433 This fix has been applied to the mysql and pgsql lookups by writing a
3434 separate quoting function and calling it from the lookup code. The fix
3435 should probably also be applied to nisplus, ibase and oracle lookups, but
3436 since I cannot test any of those, I have not disturbed their existing code.
3438 2. A hit in the callout cache for a specific address caused a log line with no
3439 reason for rejecting RCPT. Now it says "Previous (cached) callout
3440 verification failure".
3442 3. There was an off-by-one bug in the queryprogram router. An over-long
3443 return line was truncated at 256 instead of 255 characters, thereby
3444 overflowing its buffer with the terminating zero. As well as fixing this, I
3445 have increased the buffer size to 1024 (and made a note to document this).
3447 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3448 when Exim is waiting for an SMTP response from a remote server, Exim
3449 restarts its select() call on the socket, thereby resetting its timeout.
3450 This is not a problem when such interrupts are rare. Somebody set up a cron
3451 job to run exiwhat every 2 minutes, which is less than the normal select()
3452 timeout (5 or 10 minutes). This meant that the select() timeout never
3453 kicked in because it was always reset. I have fixed this by comparing the
3454 time when an interrupt arrives with the time at the start of the first call
3455 to select(). If more time than the timeout has elapsed, the interrupt is
3456 treated as a timeout.
3458 5. Some internal re-factoring in preparation for the addition of Sieve
3459 extensions (by MH). In particular, the "personal" test is moved to a
3460 separate function, and given an option for scanning Cc: and Bcc: (which is
3461 not set for Exim filters).
3463 6. When Exim created an email address using the login of the caller as the
3464 local part (e.g. when creating a From: or Sender: header line), it was not
3465 quoting the local part when it contained special characters such as @.
3467 7. Installed new OpenBSD configuration files.
3469 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3470 try to make them clearer.
3472 9. Callout options, other than the timeout value, were being ignored when
3473 verifying sender addresses in header lines. For example, when using
3475 verify = header_sender/callout=no_cache
3477 the cache was (incorrectly) being used.
3479 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3480 only the Cygwin environment.
3482 11. When return_path_on_delivery was set as a log selector, if different remote
3483 addresses in the same message used different return paths and parallel
3484 remote delivery occurred, the wrong values would sometimes be logged.
3485 (Whenever a remote delivery process finished, the return path value from
3486 the most recently started remote delivery process was logged.)
3488 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3489 lines when AUTH and/or TLS are in use. This is the "received protocol"
3490 field. Exim used to use "asmtp" for authenticated SMTP, without any
3491 indication (in the protocol name) for TLS use. Now it follows the RFC and
3492 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3493 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3494 names appear in log lines as well as in Received: header lines.
3496 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3497 extensions, and comparison tests, and to fix some bugs.
3499 14. Changes to the "personal" filter test:
3501 (1) The test was buggy in that it was just doing the equivalent of
3502 "contains" tests on header lines. For example, if a user's address was
3503 anne@some.where, the "personal" test would incorrectly be true for
3505 To: susanne@some.where
3507 This test is now done by extracting each address from the header in turn,
3508 and checking the entire address. Other tests that are part of "personal"
3509 are now done using regular expressions (for example, to check local parts
3510 of addresses in From: header lines).
3512 (2) The list of non-personal local parts in From: addresses has been
3513 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3514 taken from the Sieve specification recommendations.
3516 (3) If the message contains any header line starting with "List-" it is
3517 treated as non-personal.
3519 (4) The test for "circular" in the Subject: header line has been removed
3520 because it now seems ill-conceived.
3522 15. Minor typos in src/EDITME comments corrected.
3524 16. Installed latest exipick from John Jetmore.
3526 17. If headers_add on a router specified a text string that was too long for
3527 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3528 of string_sprintf() is now avoided.
3530 18. $message_body_size was not set (it was always zero) when running the DATA
3531 ACL and the local_scan() function.
3533 19. For the "mail" command in an Exim filter, no default was being set for
3534 the once_repeat time, causing a random time value to be used if "once" was
3535 specified. (If the value happened to be <= 0, no repeat happened.) The
3536 default is now 0s, meaning "never repeat". The "vacation" command was OK
3537 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3538 (I found it when inspecting the code).
3540 20. There is now an overall timeout for performing a callout verification. It
3541 defaults to 4 times the callout timeout, which applies to individual SMTP
3542 commands during the callout. The overall timeout applies when there is more
3543 than one host that can be tried. The timeout is checked before trying the
3544 next host. This prevents very long delays if there are a large number of
3545 hosts and all are timing out (e.g. when the network connections are timing
3546 out). The value of the overall timeout can be changed by specifying an
3547 additional sub-option for "callout", called "maxwait". For example:
3549 verify = sender/callout=5s,maxwait=20s
3551 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3552 to the end before writing, but this should make it even safer).
3554 22. Exim was forgetting that it had advertised PIPELINING for the second and
3555 subsequent messages on an SMTP connection. It was also not resetting its
3556 memory on STARTTLS and an internal HELO.
3558 23. When Exim logs an SMTP synchronization error within a session, it now
3559 records whether PIPELINING has been advertised or not.
3561 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3562 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3563 rather than long int.
3565 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3567 26. Added the never_mail option to autoreply.
3573 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3574 crash if the getsockname() call failed; this can happen if a connection is
3575 closed very soon after it is established. The problem was simply in the
3576 order in which certain operations were done, causing Exim to try to write
3577 to the SMTP stream before it had set up the file descriptor. The bug has
3578 been fixed by making things happen in the correct order.
3584 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3585 before the connection was closed, thus losing the rejection response.
3587 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3588 some early Solaris releases, but causes trouble in current releases where
3589 socklen_t is defined.
3591 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3594 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3597 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3598 that releasing the top of it at the end releases what was used for sub-
3599 expansions (unless the block got too big). However, discard this block if
3600 the first thing is a variable or header, so that we can use its block when
3601 it is dynamic (useful for very large $message_headers, for example).
3603 6. Lookups now cache *every* query, not just the most recent. A new, separate
3604 store pool is used for this. It can be recovered when all lookup caches are
3605 flushed. Lookups now release memory at the end of their result strings.
3606 This has involved some general refactoring of the lookup sources.
3608 7. Some code has been added to the store_xxx() functions to reduce the amount
3609 of flapping under certain conditions.
3611 8. log_incoming_interface used to affect only the <= reception log lines. Now
3612 it causes the local interface and port to be added to several more SMTP log
3613 lines, for example "SMTP connection from", and rejection lines.
3615 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3617 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3619 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3620 could be overwritten at the end of the current message (or the start of a
3621 new message if it was set in a HELO ACL). The value is now preserved for
3622 the duration of the SMTP connection.
3624 12. If a transport had a headers_rewrite setting, and a matching header line
3625 contained an unqualified address, that address was qualified, even if it
3626 did not match any rewriting rules. The underlying bug was that the values
3627 of the flags that permit the existence of unqualified sender and recipient
3628 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3629 non-local messages, and by -bnq for local messages) were not being
3630 preserved with the message after it was received.
3632 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3633 "next input=" as part of the text comprising the host identity instead of
3634 the correct text. The code was using the same buffer for two different
3635 strings. However, depending on which order the printing function evaluated
3636 its arguments, the bug did not always show up. Under Linux, for example, my
3637 test suite worked just fine.
3639 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3640 doesn't work with some older versions of Perl. It has been changed to "my",
3641 which in any case is probably the better facility to use.
3643 15. A really picky compiler found some instances of statements for creating
3644 error messages that either had too many or two few arguments for the format
3647 16. The size of the buffer for calls to the DNS resolver has been increased
3648 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3649 for addresses that have a lot of PTR records. This alleviates a problem; it
3650 does not fully solve it.
3652 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3653 buffer now truncates the list and logs the incident, which is the same
3654 action as happens when Exim is looking up a host name and its aliases.
3655 Previously in this situation something unpredictable would happen;
3656 sometimes it was "internal error: store_reset failed".
3658 18. If a server dropped the connection unexpectedly when an Exim client was
3659 using GnuTLS and trying to read a response, the client delivery process
3660 crashed while trying to generate an error log message.
3662 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3663 string, for example:
3665 warn message = H1: something\nH2: something
3667 the text was added as a single header line from Exim's point of view
3668 though it ended up OK in the delivered message. However, searching for the
3669 second and subsequent header lines using $h_h2: did not work. This has been
3670 fixed. Similarly, if a system filter added multiple headers in this way,
3671 the routers could not see them.
3673 20. Expanded the error message when iplsearch is called with an invalid key to
3674 suggest using net-iplsearch in a host list.
3676 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3677 ACLs are no longer actually imposed (and a message to that effect is
3680 22. If a "gecos" field in a passwd entry contained escaped characters, in
3681 particular, if it contained a \" sequence, Exim got it wrong when building
3682 a From: or a Sender: header from that name. A second bug also caused
3683 incorrect handling when an unquoted " was present following a character
3684 that needed quoting.
3686 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3687 was not being matched caselessly.
3689 24. Arranged for all hyphens in the exim.8 source to be escaped with
3692 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3693 properly. Recipient callout cache records were still being keyed to include
3694 the sender, even when use_sender was set false. This led to far more
3695 callouts that were necessary. The sender is no longer included in the key
3696 when use_sender is false.
3698 26. Added "control = submission" modifier to ACLs.
3700 27. Added the ${base62d: operator to decode base 62 numbers.
3702 28. dnsdb lookups can now access SRV records.
3704 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3705 the configuration file.
3707 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3708 (-v) mode. This makes the output for a verbose queue run more intelligible.
3710 31. Added a use_postmaster feature to recipient callouts.
3712 32. Added the $body_zerocount variable, containing the number of binary zero
3713 bytes in the message body.
3715 33. The time of last modification of the "new" subdirectory is now used as the
3716 "mailbox time last read" when there is a quota error for a maildir
3719 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3721 35. Added +ignore_unknown as a special item in host lists.
3723 36. Code for decoding IPv6 addresses in host lists is now included, even if
3724 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3725 address was recognized as an IP address, but was then not correctly decoded
3726 into binary, causing unexpected and incorrect effects when compared with
3733 1. Very minor rewording of debugging text in manualroute to say "list of
3734 hosts" instead of "hostlist".
3736 2. If verify=header_syntax was set, and a header line with an unqualified
3737 address (no domain) and a large number of spaces between the end of the
3738 name and the colon was received, the reception process suffered a buffer
3739 overflow, and (when I tested it) crashed. This was caused by some obsolete
3740 code that should have been removed. The fix is to remove it!
3742 3. When running in the test harness, delay a bit after writing a bounce
3743 message to get a bit more predictability in the log output.
3745 4. Added a call to search_tidyup() just before forking a reception process. In
3746 theory, someone could use a lookup in the expansion of smtp_accept_max_
3747 per_host which, without the tidyup, could leave open a database connection.
3749 5. Added the variables $recipient_data and $sender_data which get set from a
3750 lookup success in an ACL "recipients" or "senders" condition, or a router
3751 "senders" option, similar to $domain_data and $local_part_data.
3753 6. Moved the writing of debug_print from before to after the "senders" test
3756 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3757 problems for message scanning, either using a data ACL, or using
3758 local_scan() because the Received: header was not generated till after they
3759 were called (in order to set the time as the time of reception completion).
3760 I have revised the way this works. The header is now generated after the
3761 body is received, but before the ACL or local_scan() are called. After they
3762 are run, the timestamp in the header is updated.
3768 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3769 before starting a queue runner without re-exec. This happened only when
3770 deliver_drop_privilege was set or when the Exim user was set to root. The
3771 effect of the bug was that timeouts during subsequent deliveries caused
3772 crashes instead of being properly handled. The handler is now left at its
3773 default (and expected) setting.
3775 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3776 message, again when deliver_drop_privilege is set or Exim is run as root.
3777 The bug described in (1) was not present in this case, but the tidying up
3778 of the other signals was missing. I have made the two cases consistent.
3780 3. The ignore_target_hosts setting on a manualroute router was being ignored
3781 for hosts that were looked up using the /MX notation.
3783 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3786 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3787 operated on the sender address. After changing the $sender_address to <>
3788 for the sender address verify, Exim was re-instated it as the original
3789 (before rewriting) address, but remembering that it had rewritten it, so it
3790 wasn't rewriting it again. This bug also had the effect of breaking the
3791 sender address verification caching when the sender address was rewritten.
3793 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3794 This has been changed so that if the ip literal address matches
3795 ignore_target_hosts, the router declines.
3797 7. Added expansion conditions match_domain, match_address, and match_local_
3798 part (NOT match_host).
3800 8. The placeholder for the Received: header didn't have a length field set.
3802 9. Added code to Exim itself and to exim_lock to test for a specific race
3803 condition that could lead to file corruption when using MBX delivery. The
3804 issue is with the lockfile that is created in /tmp. If this file is removed
3805 after a process has opened it but before that process has acquired a lock,
3806 there is the potential for a second process to recreate the file and also
3807 acquire a lock. This could lead to two Exim processes writing to the file
3808 at the same time. The added code performs the same test as UW imapd; it
3809 checks after acquiring the lock that its file descriptor still refers to
3810 the same named file.
3812 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3813 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3816 11. Added the smtp_active_hostname option. If used, this will typically be made
3817 to depend on the incoming interface address. Because $interface_address is
3818 not set up until the daemon has forked a reception process, error responses
3819 that can happen earlier (such as "too many connections") no longer contain
3822 12. If an expansion in a condition on a "warn" statement fails because a lookup
3823 defers, the "warn" statement is abandoned, and the next ACL statement is
3824 processed. Previously this caused the whole ACL to be aborted.
3826 13. Added the iplsearch lookup type.
3828 14. Added ident_timeout as a log selector.
3830 15. Added tls_certificate_verified as a log selector.
3832 16. Added a global option tls_require_ciphers (compare the smtp transport
3833 option of the same name). This controls incoming TLS connections.
3835 17. I finally figured out how to make tls_require_ciphers do a similar thing
3836 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3837 before starting the TLS session.
3839 18. Tabs are now shown as \t in -bP output.
3841 19. If the log selector return_path_on_delivery was set, Exim crashed when
3842 bouncing a message because it had too many Received: header lines.
3844 20. If two routers both had headers_remove settings, and the first one included
3845 a superfluous trailing colon, the final name in the first list and the
3846 first name in the second list were incorrectly joined into one item (with a
3847 colon in the middle).
3853 1. Added -C and -D options to the exinext utility, mainly to make it easier
3854 to include in the automated testing, but these could be helpful when
3855 multiple configurations are in use.
3857 2. The exinext utility was not formatting the output nicely when there was
3858 an alternate port involved in the retry record key, nor when there was a
3859 message id as well (for retries that were specific to a specific message
3860 and a specific host). It was also confused by IPv6 addresses, because of
3861 the additional colons they contain. I have fixed the IPv4 problem, and
3862 patched it up to do a reasonable job for IPv6.
3864 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3865 delivery, the log line now contains "pipelined" if PIPELINING was used.
3867 4. An SMTP transport process used to panic and die if the bind() call to set
3868 an explicit outgoing interface failed. This has been changed; it is now
3869 treated in the same way as a connect() failure.
3871 5. A reference to $sender_host_name in the part of a conditional expansion
3872 that was being skipped was still causing a DNS lookup. This no longer
3875 6. The def: expansion condition was not recognizing references to header lines
3876 that used bh_ and bheader_.
3878 7. Added the _cache feature to named lists.
3880 8. The code for checking quota_filecount in the appendfile transport was
3881 allowing one more file than it should have been.
3883 9. For compatibility with Sendmail, the command line option
3891 and sets the incoming protocol and host name (for trusted callers). The
3892 host name and its colon can be omitted when only the protocol is to be set.
3893 Note the Exim already has two private options, -pd and -ps, that refer to
3894 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3895 "s", but I don't think that's a major issue.
3897 10. A number of refactoring changes to the code, none of which should affect
3900 (a) The number of logging options was getting close to filling up the
3901 32-bit word that was used as a bit map. I have split them into two classes:
3902 those that are passed in the argument to log_write(), and those that are
3903 only ever tested independently outside of that function. These are now in
3904 separate 32-bit words, so there is plenty of room for expansion again.
3905 There is no change in the user interface or the logging behaviour.
3907 (b) When building, for example, log lines, the code previously used a
3908 macro that called string_cat() twice, in order to add two strings. This is
3909 not really sufficiently general. Furthermore, there was one instance where
3910 it was actually wrong because one of the argument was used twice, and in
3911 one call a function was used. (As it happened, calling the function twice
3912 did not affect the overall behaviour.) The macro has been replaced by a
3913 function that can join an arbitrary number of extra strings onto a growing
3916 (c) The code for expansion conditions now uses a table and a binary chop
3917 instead of a serial search (which was left over from when there were very
3918 few conditions). Also, it now recognizes conditions like "pam" even when
3919 the relevant support is not compiled in: a suitably worded error message is
3920 given if an attempt is made to use such a condition.
3922 11. Added ${time_interval:xxxxx}.
3924 12. A bug was causing one of the ddress fields not to be passed back correctly
3925 from remote delivery subprocesses. The field in question was not being
3926 subsequently used, so this caused to problems in practice.
3928 13. Added new log selectors queue_time and deliver_time.
3930 14. Might have fixed a bug in maildirsizefile handling that threw up
3931 "unexpected character" debug warnings, and recalculated the data
3932 unnecessarily. In any case, I expanded the warning message to give more
3935 15. Added the message "Restricted characters in address" to the statements in
3936 the default ACL that block characters like @ and % in local parts.
3938 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3939 Three changes have been made:
3941 (a) There was a serious bug; a negative response to MAIL caused the whole
3942 recipient domain to be cached as invalid, thereby blocking all messages
3943 to all local parts at the same domain, from all senders. This bug has
3944 been fixed. The domain is no longer cached after a negative response to
3945 MAIL if the sender used is not empty.
3947 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3950 (c) A new callout option, "use_sender" has been added for people who want
3951 the modified behaviour.
3957 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3960 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3961 indeed breaks things for older releases.
3963 3. Added additional logging to the case where there is a problem reading data
3964 from a filter that is running in a subprocess using a pipe, in order to
3965 try to track down a specific problem.
3967 4. Testing facility fudge: when running in the test harness and attempting
3968 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3969 getting "No route to host". Convert this to a timeout.
3971 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3974 6. Some OS don't have socklen_t but use size_t instead. This affects the
3975 fifth argument of getsockopt() amongst other things. This is now
3976 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3977 can be set for individual OS. I have set it for SunOS5, OSF1, and
3978 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3979 some earlier ones do not.
3981 7. Change 4.30/15 was not doing the test caselessly.
3983 8. The standard form for an IPv6 address literal was being rejected by address
3984 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3985 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3986 this, as well as the form without the "IPv6" on the front (but only when
3987 address literals are enabled, of course).
3989 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3991 10. Exim crashed if a message with an empty sender address specified by -f
3992 encountered a router with an errors_to setting. This could be provoked only
3993 by a command such as
3997 where an empty string was supplied; "<>" did not hit this bug.
3999 11. Installed PCRE release 4.5.
4001 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
4002 remained set. It is now erased.
4004 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
4005 times from message ids (which are base 36 rather than the normal 62).
4007 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
4008 were being counted as actual protocol errors, and logged if the log
4009 selector +smtp_protocol_error was set. One cannot be perfect in this test,
4010 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
4011 and DATA following a set of rejected RCPTs do not count as protocol errors.
4012 In other words, Exim assumes they were pipelined, though this may not
4013 actually be the case. Of course, in all cases the client gets an
4014 appropriate error code.
4016 15. If a lookup fails in an ACL condition, a message about the failure may
4017 be available; it is used if testing the ACL cannot continue, because most
4018 such messages specify what the cause of the deferral is. However, some
4019 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
4020 that caused an old message to be retained and used if a later statement
4021 caused a defer, replacing the real cause of the deferral.
4023 16. If an IP address had so many PTR records that the DNS lookup buffer
4024 was not large enough to hold them, Exim could crash while trying to process
4025 the truncated data. It now detects and logs this case.
4027 17. Further to 4.21/58, another change has been made: if (and only if) the
4028 first line of a message (the first header line) ends with CRLF, a bare LF
4029 in a subsequent header line has a space inserted after it, so as not to
4030 terminate the header.
4032 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
4033 unnecessarily, used atoi() instead of strtol(), and didn't check the
4034 termination when getting file sizes from file names by regex.
4036 19. Completely re-implemented the support for maildirsize files, in the light
4037 of a number of problems with the previous contributed implementation
4038 (4.30/29). In particular:
4040 . If the quota is zero, the maildirsize file is maintained, but no quota is
4043 . If the maildir directory does not exist, it is created before any attempt
4044 to write a maildirsize file.
4046 . The quota value in the file is just a cache; if the quota is changed in
4047 the transport, the new value overrides.
4049 . A regular expression is available for excluding directories from the
4052 20. The autoreply transport checks the characters in options that define the
4053 message's headers; it allows continued headers, but it was checking with
4054 isspace() after an embedded newline instead of explicitly looking for a
4057 21. If all the "regular" hosts to which an address was routed had passed their
4058 expiry times, and had not reached their retry times, the address was
4059 bounced, even if fallback hosts were defined. Now Exim should go on to try
4062 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
4063 equivalent code in the SMTP transport. Some hosts send humungous responses
4064 to HELO/EHLO, more than 1024 it seems.
4066 23. Refactoring: code in filter.c used (void *) for "any old type" but this
4067 gives compiler warnings in some environments. I've now done it "properly",
4070 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
4071 (because of problems with the built-in one) was declared to return uschar *
4072 instead of char *, causing compiler failure.
4074 25. Fixed a file descriptor leak when processing alias/forward files.
4076 26. Fixed a minor format string issue in dbfn.c.
4078 27. Typo in exim.c: ("dmbnz" for "dbmnz").
4080 28. If a filter file refered to $h_xxx or $message_headers, and the headers
4081 contained RFC 2047 "words", Exim's memory could, under certain conditions,
4084 29. When a sender address is verified, it is cached, to save repeating the test
4085 when there is more than one recipient in a message. However, when the
4086 verification involves a callout, it is possible for different callout
4087 options to be set for different recipients. It is too complicated to keep
4088 track of this in the cache, so now Exim always runs a verification when a
4089 callout is required, relying on the callout cache for the optimization.
4090 The overhead is duplication of the address routing, but this should not be
4093 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
4094 to be verified with callout=postmaster, and the main callout worked but the
4095 postmaster check failed, the verification correctly failed. However, if a
4096 subsequent RCPT command asked for sender verification *without* the
4097 postmaster check, incorrect caching caused this verification also to fail,
4100 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
4101 it was not caching the DNS options (qualify_single, search_parents) that
4102 were used when the lookup failed. A subsequent lookup with different
4103 options therefore always gave the same answer, though there were cases
4104 where it should not have. (Example: a "domains = !$mx_any" option on a
4105 dnslookup router: the "domains" option is always processed without any
4106 widening, but the router might have qualify_single set.) Now Exim uses the
4107 cached value only when the same options are set.
4109 32. Added John Jetmore's "exipick" utility to the distribution.
4111 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
4112 than a timeout (e.g. a certificate is required, and is not provided), an
4113 Exim server now closes the connection immediately. Previously it waited for
4114 the client to close - but if the client is SSL, it seems that they each
4115 wait for each other, leading to a delay before one of them times out.
4117 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
4118 maintained 0.8.x compatibility because I don't think many are using it, and
4119 it is clearly obsolete.
4121 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
4124 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
4125 client certificate was expired. A simple patch fixes this, though I don't
4126 understand the full logic of why the verify callback is called multiple
4129 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
4130 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
4131 which causes problems with some clients (such as the Certicom SSL Plus
4132 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
4133 disables the coutermeasure allowing Eudora to connect."
4135 38. Exim was not checking that a write() to a log file succeeded. This could
4136 lead to Bad Things if a log got too big, in particular if it hit a file
4137 size limit. Exim now panics and dies if it cannot write to a log file, just
4138 as it does if it cannot open a log file.
4140 39. Modified OS/Makefile-Linux so that it now contains
4142 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
4144 The two -D definitions ensure that Exim is compiled with large file
4145 support, which makes it possible to handle log files that are bigger than
4148 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
4149 instance) a domain was checked against a named list that involved a lookup,
4150 causing $domain_data to be set, then another domain was checked against the
4151 same list, then the first domain was re-checked, the value of $domain_data
4152 after the final check could be wrong. In particular, if the second check
4153 failed, it could be set empty. This bug probably also applied to
4156 41. The strip_trailing_dot option was not being applied to the address given
4157 with the -f command-line option.
4159 42. The code for reading a message's header from the spool was incrementing
4160 $received_count, but never initializing it. This meant that the value was
4161 incorrect (doubled) while delivering a message in the same process in which
4162 it was received. In the most common configuration of Exim, this never
4163 happens - a fresh exec is done - but it can happen when
4164 deliver_drop_privilege is set.
4166 43. When Exim logs an SMTP synchronization error - client data sent too soon -
4167 it now includes up to 150 characters of the unexpected data in the log
4170 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
4171 and building data strings. The size of both of these buffers was 10 000
4172 bytes - far larger than anybody would *ever* want, thought I. Needless to
4173 say, somebody hit the limit. I have increased the maximum line length to
4174 20 000 and the maximum data length of concatenated lines to 100 000. I have
4175 also fixed two bugs, because there was no checking on these buffers. Tsk,
4176 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
4177 buffer is too small.
4179 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
4180 lsearch lookups. Now it does.
4182 46. When parsing a route_list item in a manualroute router, a fixed-length
4183 buffer was used for the list of hosts. I made this 1024 bytes long,
4184 thinking that nobody would ever have a list of hosts that long. Wrong.
4185 Somebody had a whole pile of complicated expansion conditions, and the
4186 string was silently truncated, leading to an expansion error. It turns out
4187 that it is easier to change to an unlimited length (owing to other changes
4188 that have happened since this code was originally written) than to build
4189 structure for giving a limitation error. The length of the item that
4190 expands into the list of hosts is now unlimited.
4192 47. The lsearch lookup could not handle data where the length of text line was
4193 more than 4095 characters. Such lines were truncated, leading to shortened
4194 data being returned. It should now handle lines of any length.
4196 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
4197 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
4200 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
4201 replacing strings such as BIN_DIRECTORY in a source file: the replacement
4202 no longer happens in comment lines. A list of replacements is now placed
4203 at the head of all of the source files, except those whose only change is
4204 to replace PERL_COMMAND in the very first #! line.
4206 50. Replaced the slow insertion sort in queue.c, for sorting the list of
4207 messages on the queue, with a bottom-up merge sort, using code contributed
4208 by Michael Haardt. This should make operations like -bp somewhat faster on
4209 large queues. It won't affect queue runners, except when queue_run_in_order
4212 51. Installed eximstats 1.31 in the distribution.
4214 52. Added support for SRV lookups to the dnslookup router.
4216 53. If an ACL referred to $message_body or $message_body_end, the value was not
4217 reset for any messages that followed in the same SMTP session.
4219 54. The store-handling optimization for building very long strings was not
4220 differentiating between the different store pools. I don't think this
4221 actually made any difference in practice, but I've tidied it.
4223 55. While running the routers to verify a sender address, $sender_address
4224 was still set to the sender address. This is wrong, because when routing to
4225 send a bounce to the sender, it would be empty. Therefore, I have changed
4226 it so that, while verifying a sender address, $sender_address is set to <>.
4227 (There is no change to what happens when verifying a recipient address.)
4229 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
4230 target A or AAAA records (if not already returned) without resetting the
4231 qualify_single or search_parents options of the DNS resolver. These are
4232 inappropriate in this case because the targets of MX and SRV records must
4233 be FQDNs. A broken DNS record could cause trouble if it happened to have a
4234 target that, when qualified, matched something in the local domain. These
4235 two options are now turned off when doing these lookups.
4237 57. It seems that at least some releases of Reiserfs (which does not have the
4238 concept of a fixed number of inodes) returns zero and not -1 for the
4239 number of available inodes. This interacted badly with check_spool_inodes,
4240 which assumed that -1 was the "no such thing" setting. What I have done is
4241 to check that the total number of inodes is greater than zero before doing
4242 the test of how many are available.
4244 58. When a "warn" ACL statement has a log_message modifier, the message is
4245 remembered, and not repeated. This is to avoid a lot of repetition when a
4246 message has many recipients that cause the same warning to be written.
4247 Howewer, Exim was preserving the list of already written lines for an
4248 entire SMTP session, which doesn't seem right. The memory is now reset if a
4249 new message is started.
4251 59. The "rewrite" debugging flag was not showing the result of rewriting in the
4252 debugging output unless log_rewrite was also set.
4254 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
4255 of (int)(handle) when we know that handle contains (void *)(-1).
4257 61. The Exim daemon panic-logs an error return when it closes the incoming
4258 connection. However "connection reset by peer" seems to be common, and
4259 isn't really an error worthy of noting specially, so that particular error
4262 62. When Exim is trying to find all the local interfaces, it used to panic and
4263 die if the ioctl to get the interface flags failed. However, it seems that
4264 on at least one OS (Solaris 9) it is possible to have an interface that is
4265 included in the list of interfaces, but for which you get a failure error
4266 for this call. This happens when the interface is not "plumbed" into a
4267 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
4268 failure of the "get flags" call assumes that the interface is down.
4270 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
4271 makes life easier for people who are doing arithmetic on fields extracted
4272 from dates, where you often get leading zeros that should not be
4273 interpreted as octal.
4275 64. Added qualify_domain to the redirect router, to override the global
4278 65. If a pathologically long header line contained very many addresses (the
4279 report of this problem mentioned 10 000) and each of them was rewritten,
4280 Exim could use up a very large amount of memory. (It kept on making new
4281 copies of the header line as it rewrote, and never released the old ones.)
4282 At the expense of a bit more processing, the header rewriting function has
4283 been changed so that it no longer eats memory in this way.
4285 66. The generation of the Received: header has been moved from the time that a
4286 message starts to be received, to the time that it finishes. The timestamp
4287 in the Received: header should now be very close to that of the <= log
4288 line. There are two side-effects of this change:
4290 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
4291 logged header lines no longer include the local Received: line, because
4292 it has not yet been created. The same applies to a copy of the message
4293 that is returned to a non-SMTP sender when a message is rejected.
4295 (b) When a filter file is tested using -bf, no additional Received: header
4296 is added to the test message. After some thought, I decided that this
4299 This change does not affect the value of $received_for. It is still set
4300 after address rewriting, but before local_scan() is called.
4302 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4304 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4305 gave an unhelpful panic error message, and a defer error. I have managed to
4306 change this behaviour so that it now rejects any supplied certificate,
4307 which seems right, as the list of acceptable certificates is empty.
4309 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4310 gave an unhelpful defer error. I have not managed to make this reject any
4311 supplied certificates, but the error message it gives is "no certificate
4312 supplied", which is not helpful.
4314 70. exigrep's output now also includes lines that are not associated with any
4315 message, but which match the given pattern. Implemented by a patch from
4316 Martin Sluka, which also tidied up the Perl a bit.
4318 71. Recipient callout verification, like sender verification, was using <> in
4319 the MAIL FROM command. This isn't really the right thing, since the actual
4320 sender may affect whether the remote host accepts the recipient or not. I
4321 have changed it to use the actual sender in the callout; this means that
4322 the cache record is now keyed on a recipient/sender pair, not just the
4323 recipient address. There doesn't seem to be a real danger of callout loops,
4324 since a callout by the remote host to check the sender would use <>.
4325 [SEE ABOVE: changed after hitting problems.]
4327 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4328 temporary errors. However, in the case of such a code being given after
4329 the end of a data transmission (i.e. after ".") Exim was failing to write
4330 a retry record for the message. (Yes, there was some broken host that was
4331 actually sending 8xx at this point.)
4333 73. An unknown lookup type in a host list could cause Exim to panic-die when
4334 the list was checked. (An example that provoked this was putting <; in the
4335 middle of a list instead of at the start.) If this happened during a DATA
4336 ACL check, a -D file could be left lying around. This kind of configuration
4337 error no longer causes Exim to die; instead it causes a defer errror. The
4338 incident is still logged to the main and panic logs.
4340 74. Buglet left over from Exim 3 conversion. The message "too many messages
4341 in one connection" was written to the rejectlog but not the mainlog, except
4342 when address rewriting (yes!) was being logged.
4344 75. Added write_rejectlog option.
4346 76. When a system filter was run not as root (that is, when system_filter_user
4347 was set), the values of the $n variables were not being returned to the
4348 main process; thus, they were not subsequently available in the $sn
4351 77. Added +return_path_on_delivery log selector.
4353 78. A connection timeout was being treated differently from recipients deferred
4354 when testing hosts_max_try with a message that was older than the host's
4355 retry timeout. (The host should not be counted, thus allowing all hosts to
4356 be tried at least once before bouncing.) This may have been the cause of an
4357 occasionally reported bug whereby a message would remain on the queue
4358 longer than the retry timeout, but would be bounced if a delivery was
4359 forced. I say "may" because I never totally pinned down the problem;
4360 setting up timeout/retry tests is difficult. See also the next item.
4362 79. The ultimate address timeout was not being applied to errors that involved
4363 a combination of host plus message (for example, a timeout on a MAIL
4364 command). When an address resolved to a number of possible hosts, and they
4365 were not all tried for each delivery (e.g. because of hosts_max_try), a
4366 message could remain on the queue longer than the retry timeout.
4368 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4371 81. Fixed an obscure SMTP outgoing bug which required at least the following
4372 conditions: (a) there was another message waiting for the same server;
4373 (b) the server returned 5xx to all RCPT commands in the first message so
4374 that the message was not completed; (c) the server dropped the connection
4375 or gave a negative response to the RSET that Exim sends to abort the
4376 transaction. The observed case was a dropped connection after DATA that had
4377 been sent in pipelining mode. That is, the server had advertised PIPELINING
4378 but was not implementing it correctly. The effect of the bug was incorrect
4379 behaviour, such as trying another host, and this could lead to a crash.
4385 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4386 and daemon.c were passed as pointers to ints; they should have been
4387 pointers to socklen_t variables (which are typically unsigned ints).
4389 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4392 3. Fixed a really odd bug that affected only the testing scheme; patching a
4393 certain fixed string in the binary changed the value of another string that
4394 happened to be identical to the end of the original first string.
4396 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4397 name", it returns that address as the IP address. On some operating
4398 systems (e.g. Solaris), it also passes back the IP address string as the
4399 "host name". However, on others (e.g. Linux), it passes back an empty
4400 string. Exim wasn't checking for this, and was changing the host name to an
4401 empty string, assuming it had been canonicized.
4403 5. Although rare, it is permitted to have more than one PTR record for a given
4404 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4405 all the names associated with an address, because they do in Solaris.
4406 However, it seems that they do not in Linux for data that comes from the
4407 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4408 I found this out when I moved to a new Linux workstation and tried to run
4409 the Exim test suite.
4411 To get round this problem I have changed the code so that it now does its
4412 own call to the DNS to look up PTR records when searching for a host name.
4413 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4414 addresses that are only in /etc/hosts are still found.
4416 This behaviour is, however, controlled by an option called host_lookup_
4417 order, which defaults to "bydns:byaddr". If people want to use the other
4418 order, or indeed, just use one or the other means of lookup, they can
4419 specify it in this variable.
4421 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4422 some operating systems, this comes back from gethostbyaddr() as an empty
4423 string, and this is what Exim used to test for. However, it seems that in
4424 other systems, "." is yielded. Exim now tests for this case too.
4426 7. The values of check_spool_space and check_log_space are now held internally
4427 as a number of kilobytes instead of an absolute number of bytes. If a
4428 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4429 kilobyte. This means that much larger values can be stored.
4431 8. Exim monitor: an attempt to get the action menu when not actually pointing
4432 at a message produces an empty menu entitled "No message selected". This
4433 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4434 no entries in it ("Shell widget menu has zero width and/or height"). So I
4435 have added a single, blank menu entry in this case.
4437 9. Added ${quote_local_part.
4439 10. MIME decoding is now applied to the contents of Subject: header lines when
4442 11. Now that a reference to $sender_host_address automatically causes a reverse
4443 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4444 host lookup before query-style lookups in lists that might use this
4445 variable. This has therefore been abolished, and the "net-" prefix is no
4446 longer necessary for query-style lookups.
4448 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4449 have been a typo for LFLAGS, so it has been changed.
4451 13. The install script calls Exim with "-C /dev/null" in order to find the
4452 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4453 to be output. Howeve, since Exim outputs its version number before the
4454 error, it didn't break the script. It just looked ugly. I fixed this by
4455 always allowing "-C /dev/null" if the caller is root.
4457 14. Ignore overlarge ACL variable number when reading spool file - insurance
4458 against a later release with more variables having written the file.
4460 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4461 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4462 this, as well as the form without the "IPv6" on the front.
4464 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4465 OS/Makefile-Darwin file.
4467 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4468 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4470 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4471 variable. However, some OS use ints for inodes. I've added cast to long int
4472 to get rid of the compiler warning.
4474 19. I had forgotten to lock out "/../" in configuration file names when
4475 ALT_CONFIG_PREFIX was set.
4477 20. Routers used for verification do not need to specify transports. However,
4478 if such a router generated a host list, and callout was configured, Exim
4479 crashed, because it could not find a port number from the (non-existent)
4480 transport. It now assumes port 25 in this circumstance.
4482 21. Added the -t option to exigrep.
4484 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4485 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4486 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4487 them is set, LOOKUP_LSEARCH is forced.
4489 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4491 24. Added sender and host information to the "rejected by local_scan()" log
4492 line; previously there was no indication of these.
4494 25. Added .include_if_exists.
4496 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4497 Linux. It turns out that not all file systems support this. Apparently some
4498 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4499 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4500 not supported on the file descriptor, is now ignored when Exim is trying to
4501 sync a directory. This applies only to Linux.
4503 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4505 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4506 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4509 29. Added support for maildirsize files from supplied patch (modified a bit).
4511 30. The use of :fail: followed by an empty string could lead Exim to respond to
4512 sender verification failures with (e.g.):
4514 550 Verification failed for <xxx>
4515 550 Sender verify failed
4517 where the first response line was missing the '-' that indicates it is not
4518 the final line of the response.
4520 31. The loop for finding the name of the user that called Exim had a hardwired
4521 limit of 10; it now uses the value of finduser_retries, which is used for
4522 all other user lookups.
4524 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4527 33. Exim was neglecting to zero errno before one call of strtol() when
4528 expanding a string and expecting an integer value. On some systems this
4529 resulted in spurious "integer overflow" errors. Also, it was casting the
4530 result into an int without checking.
4532 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4533 did not work. The code looks as if it has *never* worked, though it appears
4534 to have been documented since at least releast 1.62. I have made it work.
4536 35. The "timeout_DNS" error in retry rules, also documented since at least
4537 1.62, also never worked. As it isn't clear exactly what this means, and
4538 clearly it isn't a major issue, I have abolished the feature by treating it
4539 as "timeout", and writing a warning to the main and panic logs.
4541 36. The display of retry rules for -brt wasn't always showing the error code
4544 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4545 timeout_connect_A, timeout_connect_MX.
4547 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4548 to the empty sender.
4550 39. The daemon was not analysing the content of -oX till after it had closed
4551 stderr and disconnected from the controlling terminal. This meant that any
4552 syntax errors were only noted on the panic log, and the return code from
4553 the command was 0. By re-arranging the code a little, I've made the
4554 decoding happen first, so such errors now appear on stderr, and the return
4555 code is 1. However, the actual setting up of the sockets still happens in
4556 the disconnected process, so errors there are still only recorded on the
4559 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4560 connections (as happens on some IP stacks) was logged at start up time as
4561 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4562 it from "IPv6 and IPv4", which means that two separate sockets are being
4565 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4566 says whether AF_INET or AF_INET6 was passed as an argument.
4568 42. Exiwhat output was messed up when time zones were included in log
4571 43. Exiwhat now gives more information about the daemon's listening ports,
4572 and whether -tls-on-connect was used.
4574 44. The "port" option of the smtp transport is now expanded.
4576 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4577 silently ignored. Now an error message is written to the main and panic
4580 46. There's a new ACL modifier called "logwrite" which writes to a log file
4581 as soon as it is encountered.
4583 47. Added $local_user_uid and $local_user_gid at routing time.
4585 48. Exim crashed when trying to verify a sender address that was being
4588 49. Exim was recognizing only a space character after ".include". It now also
4589 recognizes a tab character.
4591 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4592 extracting the relevant information from the specification. The man page no
4593 longer contains scrambled data for the -d option, and I've added a section
4594 at the front about calling Exim under different names.
4596 51. Added "extra_headers" argument to the "mail" command in filter files.
4598 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4601 53. Installed eximstats 1.29.
4603 54. Added transport_filter_timeout as a generic transport option.
4605 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4606 Cc: header lines. This was required by RFC 822, but it not required by RFC
4609 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4610 incoming messages that did not have them. Now it does so only if the
4611 message originates locally, that is, if there is no associated remote host
4612 address. When Resent- header lines are present, this applies to the Resent-
4613 lines rather than the non-Resent- lines.
4615 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4616 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4618 58. Messages for configuration errors now include the name of the main
4619 configuration file - useful now that there may be more than one file in a
4620 list (.included file names were always shown).
4622 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4623 for those rare installations that do not start the daemon as root or run it
4624 setuid root. I've cut out the call to initgroups() if the daemon is not
4627 60. The Exim user and group can now be bound into the binary as text strings
4628 that are looked up at the start of Exim's processing.
4630 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4632 62. Added $mailstore_basename variable.
4634 63. Installed patch to sieve.c from Michael Haardt.
4636 64. When Exim failed to open the panic log after failing to open the main log,
4637 the original message it was trying to log was written to stderr and debug
4638 output, but if they were not available (the usual case in production), it
4639 was lost. Now it is written to syslog before the two lines that record the
4640 failures to open the logs.
4642 65. Users' Exim filters run in subprocesses under the user's uid. It is
4643 possible for a "deliver" command or an alias in a "personal" command to
4644 provoke an address rewrite. If logging of address rewriting is configured,
4645 this fails because the process is not running as root or exim. There may be
4646 a better way of dealing with this, but for the moment (because 4.30 needs
4647 to be released), I have disabled address rewrite logging when running a
4648 filter in a non-root, non-exim process.
4654 1. The buildconfig auxiliary program wasn't quoting the value set for
4655 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4656 not defined. This bug was masked in 4.22 by the effect that was fixed in
4659 2. Some messages that were rejected after a message id was allocated were
4660 shown as "incomplete" by exigrep. It no longer does this for messages that
4661 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4663 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4664 have allow_domain_literals set, the ID did not get logged in the <= line.
4665 Domain literals are now always recognized in Message-ID: header lines.
4667 4. The first argument for a ${extract expansion item is the key name or field
4668 number. Leading and trailing spaces in this item were not being ignored,
4669 causing some misleading effects.
4671 5. When deliver_drop_privilege was set, single queue runner processes started
4672 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4673 same command in the process it spins off) were not dropping privilege.
4675 6. When the daemon running as "exim" started a queue runner, it always
4676 re-executed Exim in the spun-off process. This is a waste of effort when
4677 deliver_drop_privilege is set. The new process now just calls the
4678 queue-runner function directly.
4684 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4687 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4688 ignored. Though the use of -oP was forcing the writing of a pid file, it
4689 was always written to the default place.
4691 3. If the message "no IP address found for host xxxx" is generated during
4692 incoming verification, it is now followed by identification of the incoming
4693 connection (so you can more easily find what provoked it).
4695 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4697 5. Added some features to "harden" Exim a bit more against certain attacks:
4699 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4700 be put in Local/Makefile. This is like the never_users runtime option,
4701 but it cannot be overridden. The default setting is "root".
4703 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4704 prefix string with which any file named in a -C command line option
4707 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4708 is retained for -C and -D only if the caller of Exim is root. Without
4709 it, the exim user may also use -C and -D and retain privilege.
4711 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4712 command line option is disabled.
4714 6. Macro names set by the -D option must start with an upper case letter, just
4715 like macro names defined in the configuration file.
4717 7. Added "dereference=" facility to LDAP.
4719 8. Two instances of the typo "uknown" in the source files are fixed.
4721 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4722 the Configure-Makefile script screwed up while processing it.
4724 10. Incorporated PCRE 4.4.
4726 11. The SMTP synchronization check was not operating right at the start of an
4727 SMTP session. For example, it could not catch a HELO sent before the client
4728 waited for the greeting. There is now a check for outstanding input at the
4729 point when the greeting is written. Because of the duplex, asynchronous
4730 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4731 way, but not yet received, when the check is performed.
4733 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4734 on TCP/IP sockets, because this apparently causes some broken clients to
4737 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4738 unchanged) from the Cygwin maintainer.
4740 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4741 was supported instead of testing for mbx. Effectively a typo.
4743 15. The spa authenticator server code was not checking that the input it
4744 received was valid base64.
4746 16. The debug output line for the "set" modifier in ACLs was not showing the
4747 name of the variable that was being set.
4749 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4751 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4752 lookup on its own. Something else was needed to trigger the lookup. For
4753 example, a match in host_lookup or the need for a host name in a host list.
4754 Now, if $sender_host_name is referenced and the host name has not yet been
4755 looked up, a lookup is performed. If the lookup fails, the variable remains
4756 empty, and $host_lookup_failed is set to "1".
4758 19. Added "eqi" as a case-independent comparison operator.
4760 20. The saslauthd authentication condition could segfault if neither service
4761 nor realm was specified.
4763 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4764 error message that was logged was misleading, and incoming SMTP
4765 connections were dropped. The message is now more accurate, and temporary
4766 errors are given to SMTP connections.
4768 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4769 (except RSET) with a 421 error, until QUIT is received. However, it was
4770 failing to send a response to QUIT.
4772 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4773 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4774 ACL is now run after the helo_try_verify_hosts code.
4776 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4777 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4778 case-independent, so other case variants are also recognized). Apparently
4779 some systems use these upper case variants.
4781 25. If more than two messages were waiting for the same host, and a transport
4782 filter was specified for the transport, Exim sent two messages over the
4783 same TCP/IP connection, and then failed with "socket operation on non-
4784 socket" when it tried to send the third.
4786 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4788 27. The extern definition of crypt16() in expand.c was not being excluded when
4789 the OS had its own crypt16() function.
4791 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4792 as a preferred synonym for return_size_limit, both as an option and as an
4795 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4797 30. Changed the default configuration ACL to relax the local part checking rule
4798 for addresses that are not in any local domains. For these addresses,
4799 slashes and pipe symbols are allowed within local parts, but the sequence
4800 /../ is explicitly forbidden.
4802 31. SPA server authentication was not clearing the challenge buffer before
4805 32. log_message in a "warn" ACL statement was writing to the reject log as
4806 well as to the main log, which contradicts the documentation and doesn't
4807 seem right (because no rejection is happening). So I have stopped it.
4809 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4810 However, I am unable to do any testing of this.
4812 34. Fixed an infelicity in the appendfile transport. When checking directories
4813 for a mailbox, to see if any needed to be created, it was accidentally
4814 using path names with one or more superfluous leading slashes; tracing
4815 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4817 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4818 contents are added to the log line that is written for every discarded
4819 recipient. (Previously a log_message setting was ignored.)
4821 36. The ${quote: operator now quotes the string if it is empty.
4823 37. The install script runs exim in order to find its version number. If for
4824 some reason other than non-existence or emptiness, which it checks, it
4825 could not run './exim', it was installing it with an empty version number,
4826 i.e. as "exim-". This error state is now caught, and the installation is
4829 38. An argument was missing from the function that creates an error message
4830 when Exim fails to connect to the socket for saslauthd authentication.
4831 This could cause Exim to crash, or give a corrupted message.
4833 39. Added isip, isip4, and isip6 to ${if conditions.
4835 40. The ACL variables $acl_xx are now saved with the message, and can be
4836 accessed later in routers, transports, and filters.
4838 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4839 strings in the file are not string-expanded.
4841 42. If a MAIL command specified a SIZE value that was too large to fit into an
4842 int variable, the check against message_size_limit failed. Such values are
4843 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4844 day this will have to be increased, but I don't think I want to be around
4845 when emails are that large.
4852 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4853 iconv() is not standard in FreeBSD.
4855 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4856 IPv6 enabled. The observed symptom was a segmentation fault on return from
4857 the function os_common_find_running_interfaces() in src/os.c.
4859 3. In the check_special_case() function in daemon.c I had used "errno" as an
4860 argument name, which causes warnings on some systems. This was basically a
4861 typo, since it was named "eno" in the comments!
4863 4. The code that waits for the clock to tick (at a resolution of some fraction
4864 of a second) so as to ensure message-id uniqueness was always waiting for
4865 at least one whole tick, when it could have waited for less. [This is
4866 almost certainly not relevant at current processor speeds, where it is
4867 unlikely to ever wait at all. But we try to future-proof.]
4869 5. The function that sleeps for a time interval that includes fractions of a
4870 second contained a race. It did not block SIGALRM between setting the
4871 timer, and suspending (a couple of lines later). If the interval was short
4872 and the sigsuspend() was delayed until after it had expired, the suspension
4873 never ended. On busy systems this could lead to processes getting stuck for
4876 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4877 process, before it forks any delivery processes. The open lookup caching
4878 mechanism meant that the open file or database connection was passed into
4879 the delivery process. The problem was that delivery processes always tidy
4880 up cached lookup data. This could cause a problem for the next delivery
4881 process started by the queue runner, because the external queue runner
4882 process does not know about the closure. So the next delivery process
4883 still has data in the lookup cache. In the case of a file lookup, there was
4884 no problem because closing a file descriptor in a subprocess doesn't affect
4885 the parent. However, if the lookup was caching a connection to a database,
4886 the connection was closed, and the second delivery process was likely to
4887 see errors such as "PGSQL: query failed: server closed the connection
4888 unexpectedly". The problem has been fixed by closing all cached lookups
4889 in a queue runner before running a delivery process.
4891 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4892 seem to have the "const" qualifier which it has on other OS. I've
4895 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4896 *of the same type* (client or server) with the same public name that an
4897 error should be diagnosed.
4899 9. When Exim looked up a host name for an IP address, but failed to find the
4900 original IP address when looking up the host name (a safety check), it
4901 output the message "<ip address> does not match any IP for NULL", which was
4902 confusing, to say the least. The bug was that the host name should have
4903 appeared instead of "NULL".
4905 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4906 user that is built into the binary, and the -C or -D options is used, root
4907 privilege is dropped before the configuration file is read. In addition,
4908 logging is switched to stderr instead of the normal log files. If the
4909 configuration then re-defines the Exim user, the unprivileged environment
4910 is probably not what is expected, so Exim logs a panic warning message (but
4913 However, if deliver_drop_privilege is set, the unprivileged state may well
4914 be exactly what is intended, so the warning has been cut out in that case,
4915 and Exim is allowed to try to write to its normal log files.
4921 1. smtp_return_error_details was not giving details for temporary sender
4922 or receiver verification errors.
4924 2. Diagnose a configuration error if two authenticators have the same public
4927 3. Exim used not to create the message log file for a message until the first
4928 delivery attempt. This could be confusing when incoming messages were held
4929 for policy or load reasons. The message log file is now created at the time
4930 the message is received, and an initial "Received" line is written to it.
4932 4. The automatically generated man page for command line options had a minor
4933 bug that caused no ill effects; however, a more serious problem was that
4934 the procedure for building the man page automatically didn't always
4935 operate. Consequently, release 4.20 contains an out-of-date version. This
4936 shouldn't happen again.
4938 5. When building Exim with embedded Perl support, the script that builds the
4939 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4940 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4942 6. The freeze_tell option was not being used for messages that were frozen on
4943 arrival, either by an ACL or by local_scan().
4945 7. Added the smtp_incomplete_transaction log selector.
4947 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4948 was accepting AUTH without a new EHLO.
4950 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4951 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4954 10. Logging of TCP/IP connections (when configured) now happens in the main
4955 daemon process instead of the child process, so that the TCP/IP connection
4956 count is more accurate (but it can never be perfect).
4958 11. The use of "drop" in a nested ACL was not being handled correctly in the
4959 outer ACL. Now, if condition failure induced by the nested "drop" causes
4960 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4961 or "require"), the connection is dropped.
4963 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4964 that yield "discard" can now be used with an "accept" or a "discard" verb,
4965 but an error is generated for any others (because I can't see a useful way
4966 to define what should happen).
4968 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4969 are now processed in the same way as lines in the Exim configuration file.
4970 In particular, continuation lines are supported.
4972 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4974 15. Added -ti meaning -t -i.
4976 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4977 domains, and warn by logging if others are found.
4979 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4980 returned from GIFCONF when Exim is trying to find the list of interfaces on
4981 a host. The code in os.c has been modified to copy each ifreq to an aligned
4982 structure in all cases.
4984 Also, in some cases, the returned ifreq's were being copied to a 'struct
4985 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4986 means the last couple of bytes of an IPv6 address could be chopped if the
4987 ifreq contained only a normal sockaddr (14 bytes storage).
4989 18. Named domain lists were not supported in the hosts_treat_as_local option.
4990 An entry such as +xxxx was not recognized, and was treated as a literal
4993 19. Ensure that header lines added by a DATA ACL are included in the reject log
4994 if the ACL subsequently rejects the message.
4996 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4997 MD5 (which is deprecated).
4999 21. When testing a filter file using -bf, Exim was writing a message when it
5000 took the sender from a "From " line in the message, but it was not doing so
5001 when it took $return_path from a Return-Path: header line. It now does.
5003 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
5004 with a valid header line field name (a series of printing characters
5005 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
5007 23. Changed "disc" in the source to "disk" to conform to the documentation and
5008 the book and for uniformity.
5010 24. Ignore Sendmail's -Ooption=value command line item.
5012 25. When execve() failed while trying to run a command in a pipe transport,
5013 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
5014 could be confused with a return value of 69 from the command itself. This
5015 has been changed to 127, the value the shell returns if it is asked to run
5016 a non-existent command. The wording for the related log line suggests a
5017 non-existent command as the problem.
5019 26. If received_header_text expands to an empty string, do not add a Received:
5020 header line to the message. (Well, it adds a token one on the spool, but
5021 marks it "old" so that it doesn't get used or transmitted.)
5023 27. Installed eximstats 1.28 (addition of -nt option).
5025 28. There was no check for failure on the call to getsockname() in the daemon
5026 code. This can fail if there is a shortage of resources on the system, with
5027 ENOMEM, for example. A temporary error is now given on failure.
5029 29. Contrary to the C standard, it seems that in some environments, the
5030 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
5031 program. Exim now does this explicitly; it affects the formatting of
5032 timestamps using strftime().
5034 30. If exiqsumm was given junk data, it threw up some uninitialized variable
5035 complaints. I've now initialized all the variables, to avoid this.
5037 32. Header lines added by a system filter were not being "seen" during
5038 transport-time rewrites.
5040 33. The info_callback() function passed to OpenSSL is set up with type void
5041 (*)(SSL *, int, int), as described somewhere. However, when calling the
5042 function (actually a macro) that sets it up, the type void(*)() is
5043 expected. I've put in a cast to prevent warnings from picky compilers.
5045 34. If a DNS black list lookup found a CNAME record, but there were no A
5046 records associated with the domain it pointed at, Exim crashed.
5048 35. If a DNS black list lookup returned more than one A record, Exim ignored
5049 all but the first. It now scans all returned addresses if a particular IP
5050 value is being sought. In this situation, the contents of the
5051 $dnslist_value variable are a list of all the addresses, separated by a
5054 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
5055 to accept a host name and all its aliases if the forward lookup for any of
5056 them yielded the IP address of the incoming connection. Now it accepts only
5057 those names whose forward lookup yields the correct IP address. Any other
5058 names are discarded. This closes a loophole whereby a rogue DNS
5059 administrator could create reverse DNS records to break through a
5060 wildcarded host restriction in an ACL.
5062 37. If a user filter or a system filter that ran in a subprocess used any of
5063 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
5064 the wrong values were passed to the pipe command ($thisaddress had the
5065 value of $0, $0 had the value of $1, etc). This bug was introduced by
5066 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
5068 38. Improved the line breaking for long SMTP error messages from ACLs.
5069 Previously, if there was no break point between 40 and 75 characters, Exim
5070 left the rest of the message alone. Two changes have been made: (a) I've
5071 reduced the minimum length to 35 characters; (b) if it can't find a break
5072 point between 35 and 75 characters, it looks ahead and uses the first one
5073 that it finds. This may give the occasional overlong line, but at least the
5074 remaining text gets split now.
5076 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
5077 file descriptors might be low, and that setting 1000 would always raise it.
5078 It turns out that in some environments, the limit is already over 1000 and
5079 that lowering it causes trouble. So now Exim takes care not to decrease it.
5081 40. When delivering a message, the value of $return_path is set to $sender_
5082 address at the start of routing (routers may change the value). By an
5083 oversight, this default was not being set up when an address was tested by
5084 -bt or -bv, which affected the outcome if any router or filter referred to
5087 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
5088 log only when "message" or "log_message" are set. However, if one of the
5089 conditions was an address verification, or a call to a nested ACL, the
5090 messages generated by the underlying test were being passed through. This
5091 no longer happens. The underlying message is available in $acl_verify_
5092 message for both "message" and "log_message" expansions, so it can be
5093 passed through if needed.
5095 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
5096 new expansion $bh_ to give the encoded byte string without charset
5097 translation. Translation happens only if iconv() is available; HAVE_ICONV
5098 indicates this at build time. HEADERS_CHARSET gives the charset to
5099 translate to; headers_charset can change it in the configuration, and
5100 "headers charset" can change it in an individual filter file.
5102 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
5103 that creates RFC 2047 encoded "words" labels them as that charset instead
5104 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
5105 expansion operator; (ii) when Exim creates a From: line for a local
5106 message; (iii) when a header line is rewritten to include a "phrase" part.
5108 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
5109 buggy, causing it to skip the first lines of messages whose message ID
5110 ended in 'D'. This would not have bitten before Exim release 4.14, because
5111 message IDs were unlikely to end in 'D' before then. The effect was to have
5112 incorrect size information for certain domains.
5114 45. #include "config.h" was missing at the start of the crypt16.c module. This
5115 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
5118 46. If there was a timeout during a "random" callout check, Exim treated it as
5119 a failure of the random address, and carried on sending RSET and the real
5120 address. If the delay was just some slowness somewhere, the response to the
5121 original RCPT would be taken as a response to RSET and so on, causing
5122 mayhem of various kinds.
5124 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
5125 when I implemented it. It didn't allow for the fact that some option values
5126 may legitimatetly be negative (e.g. size_addition), and it didn't even do
5127 the right test for positive values.
5129 48. Domain names in DNS records are case-independent. Exim always looks them up
5130 in lower case. Some resolvers return domain names in exactly the case they
5131 appear in the zone file, that is, they may contain uppercase letters. Not
5132 all resolvers do this - some return always lower case. Exim was treating a
5133 change of case by a resolver as a change of domain, similar to a widening
5134 of a domain abbreviation. This triggered its re-routing code and so it was
5135 trying to route what was effectively the same domain again. This normally
5136 caused routing to fail (because the router wouldn't handle the domain
5137 twice). Now Exim checks for this case specially, and just changes the
5138 casing of the domain that it ultimately uses when it transmits the message
5141 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
5144 50. If a filter generated a file delivery with a non-absolute name (possible if
5145 no home directory exists for the router), the forbid_file option was not
5148 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
5149 the existing equality matching.
5151 52. Exim was using ints instead of ino_t variables in some places where it was
5152 dealing with inode numbers.
5154 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
5155 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
5156 called TMPDIR, and if it finds it is different, it changes its value.
5158 54. The smtp_printf() function is now made available to local_scan() so
5159 additional output lines can be written before returning. There is also an
5160 smtp_fflush() function to enable the detection of a dropped connection.
5161 The variables smtp_input and smtp_batched_input are exported to
5164 55. Changed the default runtime configuration: the message "Unknown user"
5165 has been removed from the ACL, and instead placed on the localuser router,
5166 using the cannot_route_message feature. This means that any verification
5167 failures that generate their own messages won't get overridden. Similarly,
5168 the "Unrouteable address" message that was in the ACL for unverifiable
5169 relay addresses has also been removed.
5171 56. Added hosts_avoid_esmtp to the smtp transport.
5173 57. The exicyclog script was not checking for the esoteric option
5174 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
5175 will work only if exicyclog is run under the appropriate euid.
5177 58. Following a discussion on the list, the rules by which Exim recognises line
5178 endings on incoming messages have been changed. The -dropcr and drop_cr
5179 options are now no-ops, retained only for backwards compatibility. The
5180 following line terminators are recognized: LF CRLF CR. However, special
5181 processing applies to CR:
5183 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
5184 nor a local message in the state where . is a terminator.
5186 (ii) If a bare CR is encountered in a header line, an extra space is added
5187 after the line terminator so as not to end the header. The reasoning
5188 behind this is that bare CRs in header lines are most likely either
5189 to be mistakes, or people trying to play silly games.
5191 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
5192 was being incorrectly given as 18 bytes larger than it should have been.
5193 This is a VOB (very old bug).
5195 60. This may never have affected anything current, but just in case it has:
5196 When the local host is found other than at the start of a list of hosts,
5197 the local host, those with the same MX, and any that follow, are discarded.
5198 When the list in question was part of a longer list of hosts, the following
5199 hosts (not currently being processed) were also being discarded. This no
5200 longer happens. I'm not sure if this situation could ever has previously
5203 61. Added the "/MX" feature to lists of hosts in the manualroute and query
5206 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
5207 header. This is something that is recommended in a new Internet Draft, and
5208 is something that is documented as being done by Sendmail. There are two
5209 possible values. For messages generated by the autoreply transport, Exim
5212 Auto-Submitted: auto-replied
5214 whereas for all other generated messages (e.g. bounces) it adds
5216 Auto-Submitted: auto-generated
5218 63. The "personal" condition in filters now includes a test for the
5219 Auto-Submitted: header. If it contains the string "auto-" the message it
5220 not considered personal.
5222 64. Added rcpt_include_affixes as a generic transport option.
5224 65. Added queue_only_override (default true).
5226 66. Added the syslog_duplication option.
5228 67. If what should have been the first header line of a message consisted of
5229 a space followed by a colon, Exim was mis-interpreting it as a header line.
5230 It isn't of course - it is syntactically invalid and should therefore be
5231 treated as the start of the message body. The misbehaviour could have
5232 caused a number of strange effects, including loss of data in subsequent
5233 header lines, and spool format errors.
5235 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
5236 client host had authenticated. This control can now be exercised by an ACL
5237 for more flexibility.
5239 69. By default, callouts do not happen when testing with -bh. There is now a
5240 variant, -bhc, which does actually run the callout code, including
5241 consulting and updating the callout cache.
5243 70. Added support for saslauthd authentication, courtesy of Alexander
5246 71. If statvfs() failed on the spool or log directories while checking their
5247 size for availability, Exim confusingly gave the error "space shortage".
5248 Furthermore, in debugging mode it crashed with a floating point exception.
5249 These checks are done if check_{spool,log}_{space,inodes} are set, and when
5250 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
5251 serious problem, Exim now writes to the main and panic logs when this
5252 happens, with details of the failure. It then refuses to accept the
5253 incoming message, giving the message "spool directory problem" or "log
5254 directory problem" with a 421 code for SMTP messages.
5256 72. When Exim is about to re-exec itself, it ensures that the file descriptors
5257 0, 1, and 2 exist, because some OS complain for execs without them (see
5258 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
5259 descriptors. However, the code omitted to check that the open succeeded,
5260 causing mysterious errors if for some reason the permissions on /dev/null
5261 got screwed. Now Exim writes a message to the main and panic logs, and
5262 bombs out if it can't open /dev/null.
5264 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
5265 interact so that it is all more flexible. It is supposed to remain
5266 backwards compatible. Also added extra_local_interfaces.
5268 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
5269 to bomb out with an assertion failure - to the client this appears as a
5270 connection drop. This problem occurs in the part of the code that was taken
5271 from the Samba project. Fortunately, the assertion is in a very simple
5272 function, so I have fixed this by reproducing the function inline in the
5273 one place where it is called, and arranging for authentication to fail
5274 instead of killing the process with assert().
5276 75. The SPA client code was not working when the server requested OEM rather
5277 than Unicode encoding.
5279 76. Added code to make require_files with a specific uid setting more usable in
5280 the case where statting the file as root fails - usually a non-root-mounted
5281 NFS file system. When this happens and the failure is EACCES, Exim now
5282 forks a subprocess and does the per-uid checking as the relevant uid.
5284 77. Added process_log_path.
5286 78. If log_file_path was not explicitly set, a setting of check_log_space or
5287 check_log_inodes was ignored.
5289 79. If a space check for the spool or log partitions fails, the incident is now
5290 logged. Of course, in the latter case the data may get lost...
5292 80. Added the %p formatting code to string_format() so that it can be used to
5293 print addresses in debug_print(). Adjusted all the address printing in the
5294 debugging in store.c to use %p rather than %d.
5296 81. There was a concern that a line of code in smtp_in.c could overflow a
5297 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5298 initially expressed, the concern was not well-founded, because trailing
5299 spaces are removed early. However, if the trailing spaces were followed by
5300 a NULL, they did not get removed, so the overflow was possible. Two fixes
5303 (a) I re-wrote the offending code in a cleaner fashion.
5304 (b) If an incoming SMTP command contains a NULL character, it is rejected
5307 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5308 runs initgroups(), so that if the Exim user is in any additional groups,
5309 they will be used during message reception.
5315 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob