1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.593 2009/11/13 15:58:16 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000
54 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
55 Patch provided by Jan Srzednicki
57 TF/05 Leading white space used to be stripped from $spam_report which
58 wrecked the formatting. Now it is preserved.
60 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
61 that they are available at delivery time.
63 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
65 TF/08 TLS error reporting now respects the incoming_interface and
66 incoming_port log selectors.
68 TF/09 Produce a more useful error message if an SMTP transport's hosts
69 setting expands to an empty string.
71 NM/06 Bugzilla 744: EXPN did not work under TLS.
72 Patch provided by Phil Pennock
74 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
75 Patch provided by Richard Godbee
77 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
78 acl_smtp_notquit, added index entry.
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
83 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to accept()
84 Patch provided by Maxim Dounin
86 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
87 Patch provided by Phil Pennock
89 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
91 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
92 Patch provided by Brad "anomie" Jorsch
94 NM/14 Bugzilla 622: Added timeout setting to transport filter.
95 Patch provided by Dean Brooks
97 TK/05 Add native DKIM support (does not depend on external libraries).
99 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
100 Patch provided by Graeme Fowler
102 NM/16 Bugzilla 851: Documentation example syntax fix.
104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
106 NM/18 Bugzilla 894: Fix issue with very long lines including comments in lsearch
108 NM/19 Bugzilla 745: TLS version reporting
109 Patch provided by Phil Pennock
111 NM/20 Bugzilla 167: bool: condition support
112 Patch provided by Phil Pennock
114 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken clients
115 Patch provided by Phil Pennock
117 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date
118 Patch provided by Brad "anomie" Jorsch
120 NM/23 Bugzilla 687: Fix misparses in eximstats
121 Patch provided by Heiko Schlittermann
123 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid
124 Patch provided by Heiko Schlittermann
126 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file
127 plus update to original patch.
129 NM/26 Bugzilla 799: Documentation correction for ratelimit
131 NM/27 Bugzilla 802: Improvements to local interface IP addr detection
132 Patch provided by David Brownlee
134 NM/28 Bugzilla 807: Improvements to LMTP delivery logging
136 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes
138 NM/30 Bugzilla 888: TLS documentation bugfixes
140 NM/31 Bugzilla 896: Dovecot buffer overrun fix
142 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
143 Unlike the original bugzilla I have changed all shell scripts in src tree
145 NM/33 Bugzilla 898: Transport filter timeout fix
146 Patch by Todd Rinaldo
148 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches
149 Patch by Serge Demonchaux
151 NM/35 Bugzilla 39: Base64 decode bug fixes
152 Patch by Jakob Hirsch
154 NM/36 Bugzilla 909: Correct connect() call in dcc code
156 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling
158 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed
160 NM/39 Bugzilla 911: Fixed MakeLinks build script
166 TK/01 Add preliminary DKIM support. Currently requires a forked version of
167 ALT-N's libdkim that I have put here:
168 http://duncanthrax.net/exim-experimental/
170 Note to Michael Haardt: I had to rename some vars in sieve.c. They
171 were called 'true' and it seems that C99 defines that as a reserved
172 keyword to be used with 'bool' variable types. That means you could
173 not include C99-style headers which use bools without triggering
174 build errors in sieve.c.
176 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
177 as mailq or other aliases. Changed the --help handling significantly
178 to do whats expected. exim_usage() emits usage/help information.
180 SC/01 Added the -bylocaldomain option to eximstats.
182 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
184 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
186 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
192 PH/01 Another patch from the Sieve maintainer.
194 PH/02 When an IPv6 address is converted to a string for single-key lookup
195 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
196 dots are used instead of colons so that keys in lsearch files need not
197 contain colons. This was done some time before quoting was made available
198 in lsearch files. However, iplsearch files do require colons in IPv6 keys
199 (notated using the quote facility) so as to distinguish them from IPv4
200 keys. This meant that lookups for IP addresses in host lists did not work
201 for iplsearch lookups.
203 This has been fixed by arranging for IPv6 addresses to be expressed with
204 colons if the lookup type is iplsearch. This is not incompatible, because
205 previously such lookups could never work.
207 The situation is now rather anomolous, since one *can* have colons in
208 ordinary lsearch keys. However, making the change in all cases is
209 incompatible and would probably break a number of configurations.
211 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
214 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
215 conversion specification without a maximum field width, thereby enabling
216 a rogue spamd server to cause a buffer overflow. While nobody in their
217 right mind would setup Exim to query an untrusted spamd server, an
218 attacker that gains access to a server running spamd could potentially
219 exploit this vulnerability to run arbitrary code as the Exim user.
221 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
222 $primary_hostname instead of what libspf2 thinks the hosts name is.
224 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
225 a directory entry by the name of the lookup key. Previously, if a
226 symlink pointed to a non-existing file or a file in a directory that
227 Exim lacked permissions to read, a lookup for a key matching that
228 symlink would fail. Now it is enough that a matching directory entry
229 exists, symlink or not. (Bugzilla 503.)
231 PH/03 The body_linecount and body_zerocount variables are now exported in the
234 PH/04 Added the $dnslist_matched variable.
236 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
237 This means they are set thereafter only if the connection becomes
240 PH/06 Added the client_condition to authenticators so that some can be skipped
241 by clients under certain conditions.
243 PH/07 The error message for a badly-placed control=no_multiline_responses left
244 "_responses" off the end of the name.
246 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
248 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
249 (without spaces) instead of just copying the configuration text.
251 PH/10 Added the /noupdate option to the ratelimit ACL condition.
253 PH/11 Added $max_received_linelength.
255 PH/12 Added +ignore_defer and +include_defer to host lists.
257 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
258 way in which PCRE > 7.0 is built.
260 PH/14 Implemented queue_only_load_latch.
262 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
263 MAIL command. The effect was to mangle the value on 64-bit systems.
265 PH/16 Another patch from the Sieve maintainer.
267 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
269 PH/18 If a system quota error occurred while trying to create the file for
270 a maildir delivery, the message "Mailbox is full" was not appended to the
271 bounce if the delivery eventually timed out. Change 4.67/27 below applied
272 only to a quota excession during the actual writing of the file.
274 PH/19 It seems that peer DN values may contain newlines (and other non-printing
275 characters?) which causes problems in log lines. The DN values are now
276 passed through string_printing() before being added to log lines.
278 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
279 and InterBase are left for another time.)
281 PH/21 Added message_body_newlines option.
283 PH/22 Guard against possible overflow in moan_check_errorcopy().
285 PH/23 POSIX allows open() to be a macro; guard against that.
287 PH/24 If the recipient of an error message contained an @ in the local part
288 (suitably quoted, of course), incorrect values were put in $domain and
289 $local_part during the evaluation of errors_copy.
295 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
296 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
299 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
300 issue a MAIL command.
302 PH/02 In an ACL statement such as
304 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
306 if a client was not listed at all, or was listed with a value other than
307 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
308 the condition was not true (as it should be), so access was not denied.
309 The bug was that the ! inversion was incorrectly passed on to the second
310 item. This has been fixed.
312 PH/03 Added additional dnslists conditions == and =& which are different from
313 = and & when the dns lookup returns more than one IP address.
315 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
316 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
318 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
319 FSYNC, which compiles an option called disable_fsync that allows for
320 bypassing fsync(). The documentation is heavily laced with warnings.
322 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
324 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
325 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
326 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
327 including adding "make clean"; (3) Added -fPIC when compiling the test
328 dynamically loaded module, to get rid of a warning.
330 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
331 message fails, move_frozen_messages = true and ignore_bounce_errors_after
332 = 0s. The bug is otherwise harmless.
334 PH/07 There was a bug in the dovecot authenticator such that the value of
335 $auth1 could be overwritten, and so not correctly preserved, after a
336 successful authentication. This usually meant that the value preserved by
337 the server_setid option was incorrect.
339 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
341 PH/09 Installed PCRE release 7.0.
343 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
344 run for batched SMTP input. It is now run at the start of every message
345 in the batch. While fixing this I discovered that the process information
346 (output by running exiwhat) was not always getting set for -bs and -bS
347 input. This is fixed, and it now also says "batched" for BSMTP.
349 PH/11 Added control=no_pipelining.
351 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
352 patch, slightly modified), and move the expansion of helo_data till after
353 the connection is made in the smtp transport (so it can use these
356 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
358 PH/14 Added log_selector = +pid.
360 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
362 PH/16 Add ${if forany and ${if forall.
364 PH/17 Added dsn_from option to vary the From: line in DSNs.
366 PH/18 Flush SMTP output before performing a callout, unless control =
367 no_callout_flush is set.
369 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
370 was true (the default) a successful delivery failed to delete the retry
371 item, thus causing premature timeout of the address. The bug is now
374 PH/20 Added hosts_avoid_pipelining to the smtp transport.
376 PH/21 Long custom messages for fakedefer and fakereject are now split up
377 into multiline reponses in the same way that messages for "deny" and
378 other ACL rejections are.
380 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
381 with slight modification.
383 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
384 draft, changing the syntax and factoring some duplicate code".
386 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
387 for deliveries of the second and subsequent messages over the same SMTP
390 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
391 ${reduce, with only minor "tidies".
393 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
395 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
396 expansion side effects.
398 PH/27 When a message times out after an over-quota error from an Exim-imposed
399 quota, the bounce message says "mailbox is full". This message was not
400 being given when it was a system quota that was exceeded. It now should
403 MH/03 Made $recipients available in local_scan(). local_scan() already has
404 better access to the recipient list through recipients_list[], but
405 $recipients can be useful in postmaster-provided expansion strings.
407 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
408 in the case of a MAIL command with additional options following the
409 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
410 were accidentally chopped off.
412 PH/29 SMTP synchronization checks are implemented when a command is read -
413 there is a check that no more input is waiting when there shouldn't be
414 any. However, for some commands, a delay in an ACL can mean that it is
415 some time before the response is written. In this time, more input might
416 arrive, invalidly. So now there are extra checks after an ACL has run for
417 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
418 pipelining has not been advertised.
420 PH/30 MH's patch to allow iscntrl() characters to be list separators.
422 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
423 returned in the SMTP response when smtp_return_error_details was false.
426 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
427 instead of the C I/O that was originally supplied, because problems were
430 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
431 Exim which did not show up earlier: it was assuming that a call to
432 SSL_CTX_set_info_callback() might give an error value. In fact, there is
433 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
434 was a macro that became an assignment, so it seemed to work. This has
435 changed to a proper function call with a void return, hence the compile
436 error. Exim's code has been fixed.
438 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
441 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
443 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
445 PH/37 If a message is not accepted after it has had an id assigned (e.g.
446 because it turns out to be too big or there is a timeout) there is no
447 "Completed" line in the log. When some messages of this type were
448 selected by exigrep, they were listed as "not completed". Others were
449 picked up by some special patterns. I have improved the selection
450 criteria to be more general.
452 PH/38 The host_find_failed option in the manualroute router can now be set
453 to "ignore", to completely ignore a host whose IP address cannot be
454 found. If all hosts are ignored, the behaviour is controlled by the new
455 host_all_ignored option.
457 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
458 homing or because of multiple MX records with /mx) generated more than
459 one IP address, and the following item turned out to be the local host,
460 all the secondary addresses of the first item were incorrectly removed
461 from the list, along with the local host and any following hosts (which
462 is what is supposed to happen).
464 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
465 whoever called Exim into the -H file. In the case of the daemon it was
466 behaving confusingly. When first started, it used values for whoever
467 started the daemon, but after a SIGHUP it used the Exim user (because it
468 calls itself on a restart). I have changed the code so that it now always
471 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
472 message are rejected with the same error (e.g. no authentication or bad
473 sender address), and a DATA command is nevertheless sent (as can happen
474 with PIPELINING or a stupid MUA), the error message that was given to the
475 RCPT commands is included in the rejection of the DATA command. This is
476 intended to be helpful for MUAs that show only the final error to their
479 PH/42 Another patch from the Sieve maintainer.
481 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
482 Eximstats - Fixed some broken HTML links and added missing column headers
484 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
485 columns for Rejects, Temp Rejects, Ham, and Spam rows.
487 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
489 PH/43 Yet another patch from the Sieve maintainer.
491 PH/44 I found a way to check for a TCP/IP connection going away before sending
492 the response to the final '.' that terminates a message, but only in the
493 case where the client has not sent further data following the '.'
494 (unfortunately, this is allowed). However, in many cases there won't be
495 any further data because there won't be any more messages to send. A call
496 to select() can be used: if it shows that the input is "ready", there is
497 either input waiting, or the socket has been closed. An attempt to read
498 the next input character can distinguish the two cases. Previously, Exim
499 would have sent an OK response which the client would never have see.
500 This could lead to message repetition. This fix should cure that, at
501 least in a lot of common cases.
503 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
504 advertised in response to EHLO.
510 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
511 fixed by 4.65/MH/01 (is this a record?) are fixed:
513 (i) An empty string was always treated as zero by the numeric comparison
514 operators. This behaviour has been restored.
516 (ii) It is documented that the numeric comparison operators always treat
517 their arguments as decimal numbers. This was broken in that numbers
518 starting with 0 were being interpreted as octal.
520 While fixing these problems I realized that there was another issue that
521 hadn't been noticed. Values of message_size_limit (both the global option
522 and the transport option) were treated as octal if they started with 0.
523 The documentation was vague. These values are now always treated as
524 decimal, and I will make that clear in the documentation.
530 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
531 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
534 MH/01 Don't check that the operands of numeric comparison operators are
535 integers when their expansion is in "skipping" mode (fixes bug
536 introduced by 4.64-PH/07).
538 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
539 child addresses, Exim now panics and dies. Previously, because the count
540 is held in a short int, deliveries were likely to be lost. As such a
541 large number of recipients for a single message is ridiculous
542 (performance will be very, very poor), I have chosen to impose a limit
543 rather than extend the field.
549 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
550 leftover -K file (the existence of which was triggered by #402).
551 While we were at it, introduced process PID as part of the -K
552 filename. This should rule out race conditions when creating
555 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
556 processing considerably. Previous code took too long for large mails,
557 triggering a timeout which in turn triggers #401.
559 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
560 in the DK code in transports.c. sendfile() is not really portable,
561 hence the _LINUX specificness.
563 TF/01 In the add_headers option to the mail command in an Exim filter,
564 there was a bug that Exim would claim a syntax error in any
565 header after the first one which had an odd number of characters
568 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
569 callout verification, Exim cached a "reject" for the entire domain. This
570 is correct for most verifications, but it is not correct for a recipient
571 verification with use_sender or use_postmaster set, because in that case
572 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
573 case of MAIL FROM:<> rejection from other early rejections (e.g.
574 rejection of HELO). When verifying a recipient using a non-null MAIL
575 address, the cache is ignored if it shows MAIL FROM:<> rejection.
576 Whatever the result of the callout, the value of the domain cache is
577 left unchanged (for any other kind of callout, getting as far as trying
578 RCPT means that the domain itself is ok).
580 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
583 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
584 manifest itself as EPIPE rather than ECONNECT. When tidying away a
585 session, the daemon ignores ECONNECT errors and logs others; it now
586 ignores EPIPE as well.
588 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
589 (quoted-printable decoding).
591 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
592 later the small subsequent patch to fix an introduced bug.
594 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
596 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
598 PH/08 An error is now given if message_size_limit is specified negative.
600 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
601 to be given (somewhat) arbitrary names.
603 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
606 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
607 miscellaneous code fixes
609 PH/10 Added the log_reject_target ACL modifier to specify where to log
612 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
613 hostname. This is wrong, because it relates to the incoming message (and
614 probably the interface on which it is arriving) and not to the outgoing
615 callout (which could be using a different interface). This has been
616 changed to use the value of the helo_data option from the smtp transport
617 instead - this is what is used when a message is actually being sent. If
618 there is no remote transport (possible with a router that sets up host
619 addresses), $smtp_active_hostname is used.
621 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
622 tweaks were necessary in order to get it to work (see also 21 below):
623 (a) The code assumed that strncpy() returns a negative number on buffer
624 overflow, which isn't the case. Replaced with Exim's string_format()
626 (b) There were several signed/unsigned issues. I just did the minimum
627 hacking in of casts. There is scope for a larger refactoring.
628 (c) The code used strcasecmp() which is not a standard C function.
629 Replaced with Exim's strcmpic() function.
630 (d) The code set only $1; it now sets $auth1 as well.
631 (e) A simple test gave the error "authentication client didn't specify
632 service in request". It would seem that Dovecot has changed its
633 interface. Fortunately there's a specification; I followed it and
634 changed what the client sends and it appears to be working now.
636 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
639 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
640 address A is aliased to B and C, where B exists and C does not. Without
641 -v the output is "A verified" because verification stops after a
642 successful redirection if more than one address is generated. However,
643 with -v the child addresses are also verified. Exim was outputting "A
644 failed to verify" and then showing the successful verification for C,
645 with its parentage. It now outputs "B failed to verify", showing B's
646 parentage before showing the successful verification of C.
648 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
649 look up a TXT record in a specific list after matching in a combined
652 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
653 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
654 they consult the DNS. I had assumed they would set it the way they
655 wanted; and indeed my experiments on Linux seem to show that in some
656 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
657 To be on the safe side, however, I have now made the interface to
658 host_find_byname() similar to host_find_bydns(), with an argument
659 containing the DNS resolver options. The host_find_byname() function now
660 sets these options at its start, just as host_find_bydns() does. The smtp
661 transport options dns_qualify_single and dns_search_parents are passed to
662 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
663 of host_find_byname() use the default settings of RES_DEFNAMES
664 (qualify_single) but not RES_DNSRCH (search_parents).
666 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
667 spool_read_header() do less string testing, by means of a preliminary
668 switch on the second character of optional "-foo" lines. (This is
669 overdue, caused by the large number of possibilities that now exist.
670 Originally there were few.) While I was there, I also converted the
671 str(n)cmp tests so they don't re-test the leading "-" and the first
672 character, in the hope this might squeeze out yet more improvement.
674 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
675 flag allowing group syntax was set by the header_syntax check but not
676 turned off, possible causing trouble later; (2) The flag was not being
677 set at all for the header_verify test, causing "group"-style headers to
678 be rejected. I have now set it in this case, and also caused header_
679 verify to ignore an empty address taken from a group. While doing this, I
680 came across some other cases where the code for allowing group syntax
681 while scanning a header line wasn't quite right (mostly, not resetting
682 the flag correctly in the right place). These bugs could have caused
683 trouble for malformed header lines. I hope it is now all correct.
685 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
686 with the "reply" argument non-NULL. The code, however (which originally
687 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
688 but it didn't always do it. This confused somebody who was copying the
689 code for some other use. I have removed all the tests.
691 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
692 feature that was used to support insecure browsers during the U.S. crypto
693 embargo. It requires special client support, and Exim is probably the
694 only MTA that supported it -- and would never use it because real RSA is
695 always available. This code has been removed, because it had the bad
696 effect of slowing Exim down by computing (never used) parameters for the
697 RSA_EXPORT functionality.
699 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
700 authenticator to fail if there's a tab character in the incoming data
701 (there should never be unless someone is messing about, as it's supposed
702 to be base64-encoded). Also added, on Timo's advice, the "secured" option
703 if the connection is using TLS or if the remote IP is the same as the
704 local IP, and the "valid-client-cert option" if a client certificate has
707 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
708 authenticators. This can be used for authorization after authentication
709 succeeds. (In the case of plaintext, it servers for both authentication
712 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
713 if any retry times were supplied.
715 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
716 connection, where there is no client IP address to check. In this
717 situation, the verify now always succeeds.
719 PH/25 Applied John Jetmore's -Mset patch.
721 PH/26 Added -bem to be like -Mset, but loading a message from a file.
723 PH/27 In a string expansion for a processed (not raw) header when multiple
724 headers of the same name were present, leading whitespace was being
725 removed from all of them, but trailing whitespace was being removed only
726 from the last one. Now trailing whitespace is removed from each header
727 before concatenation. Completely empty headers in a concatenation (as
730 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
731 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
733 PH/29 [Removed. This was a change that I later backed out, and forgot to
734 correct the ChangeLog entry (that I had efficiently created) before
735 committing the later change.]
737 PH/30 Exim was sometimes attempting to deliver messages that had suffered
738 address errors (4xx response to RCPT) over the same connection as other
739 messages routed to the same hosts. Such deliveries are always "forced",
740 so retry times are not inspected. This resulted in far too many retries
741 for the affected addresses. The effect occurred only when there were more
742 hosts than the hosts_max_try setting in the smtp transport when it had
743 the 4xx errors. Those hosts that it had tried were not added to the list
744 of hosts for which the message was waiting, so if all were tried, there
745 was no problem. Two fixes have been applied:
747 (i) If there are any address or message errors in an SMTP delivery, none
748 of the hosts (tried or untried) are now added to the list of hosts
749 for which the message is waiting, so the message should not be a
750 candidate for sending over the same connection that was used for a
751 successful delivery of some other message. This seems entirely
752 reasonable: after all the message is NOT "waiting for some host".
753 This is so "obvious" that I'm not sure why it wasn't done
754 previously. Hope I haven't missed anything, but it can't do any
755 harm, as the worst effect is to miss an optimization.
757 (ii) If, despite (i), such a delivery is accidentally attempted, the
758 routing retry time is respected, so at least it doesn't keep
759 hammering the server.
761 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
762 in ${readsocket because some servers need this prod.
764 PH/32 Added some extra debug output when updating a wait-xxx database.
766 PH/33 The hint "could be header name not terminated by colon", which has been
767 given for certain expansion errors for a long time, was not being given
768 for the ${if def:h_colon_omitted{... case.
770 PH/34 The spec says: "With one important exception, whenever a domain list is
771 being scanned, $domain contains the subject domain." There was at least
772 one case where this was not true.
774 PH/35 The error "getsockname() failed: connection reset by peer" was being
775 written to the panic log as well as the main log, but it isn't really
776 panic-worthy as it just means the connection died rather early on. I have
777 removed the panic log writing for the ECONNRESET error when getsockname()
780 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
781 runs only) independently of the message's sender address. This meant
782 that, if the 4xx error was in fact related to the sender, a different
783 message to the same recipient with a different sender could confuse
784 things. In particualar, this can happen when sending to a greylisting
785 server, but other circumstances could also provoke similar problems.
786 I have changed the default so that the retry time for these errors is now
787 based a combination of the sender and recipient addresses. This change
788 can be overridden by setting address_retry_include_sender=false in the
791 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
792 remote server are returned as part of bounce messages. This was not
793 happening for LMTP over a pipe (the lmtp transport), but now it is the
794 same for both kinds of LMTP.
796 PH/38 Despite being documented as not happening, Exim was rewriting addresses
797 in header lines that were in fact CNAMEs. This is no longer the case.
799 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
800 and queue runs started by the daemon processed all messages. This has
801 been fixed so that -R and -S can now usefully be given with -q<time>.
803 PH/40 Import PCRE release 6.7 (fixes some bugs).
805 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
807 PH/42 Give an error if -q is specified more than once.
809 PH/43 Renamed the variables $interface_address and $interface_port as
810 $received_ip_address and $received_port, to make it clear that these
811 values apply to message reception, and not to the outgoing interface when
812 a message is delivered. (The old names remain recognized, of course.)
814 PH/44 There was no timeout on the connect() call when using a Unix domain
815 socket in the ${readsocket expansion. There now is.
817 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
818 be meaningful with "accept".
820 SC/01 Eximstats V1.43
821 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
823 SC/02 Eximstats V1.44
824 Use a glob alias rather than an array ref in the generated
825 parser. This improves both readability and performance.
827 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
828 Collect SpamAssassin and rejection statistics.
829 Don't display local sender or destination tables unless
830 there is data to show.
831 Added average volumes into the top table text output.
833 SC/04 Eximstats V1.46
834 Collect data on the number of addresses (recipients)
835 as well as the number of messages.
837 SC/05 Eximstats V1.47
838 Added 'Message too big' to the list of mail rejection
839 reasons (thanks to Marco Gaiarin).
841 SC/06 Eximstats V1.48
842 Mainlog lines which have GMT offsets and are too short to
843 have a flag are now skipped.
845 SC/07 Eximstats V1.49 (Alain Williams)
846 Added the -emptyok flag.
848 SC/08 Eximstats V1.50
849 Fixes for obtaining the IP address from reject messages.
851 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
852 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
853 whitesspace changes from 4.64-PH/27
855 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
858 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
859 are found, allow negative numbers in numeric criteria)
861 JJ/06 exipick.20061117.2, added new $message_body_missing variable
863 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
864 to match changes made in 4.64-PH/43
866 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
868 PH/47 Put in an explicit test for a DNS lookup of an address record where the
869 "domain" is actually an IP address, and force a failure. This locks out
870 those revolvers/nameservers that support "A-for-A" lookups, in
871 contravention of the specifications.
873 PH/48 When a host name was looked up from an IP address, and the subsequent
874 forward lookup of the name timed out, the host name was left in
875 $sender_host_name, contrary to the specification.
877 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
878 restricted to single-key lookups, Exim was not diagnosing an error if
879 * or *@ was used with a query-style lookup.
881 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
883 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
884 long ago, but noone interested enough thought of it. Let's just say that
885 the "1.1" means that there are some new functions that weren't there at
886 some point in the past.
888 PH/51 Error processing for expansion failure of helo_data from an smtp
889 transport during callout processing was broken.
891 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
892 tested/used via the -bh/-bhc/-bs options.
894 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
895 bug, fixed in subsequent PCRE releases).
897 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
898 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
900 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
906 SC/01 Use a glob alias rather than an array ref in eximstats generated
907 parser. This improves both readability and performance.
909 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
910 Don't display local sender or destination tables in eximstats unless
911 there is data to show.
912 Added average volumes into the eximstats top table text output.
914 SC/03 Collect data on the number of addresses (recipients) as well
915 as the number of messages in eximstats.
917 TF/01 Correct an error in the documentation for the redirect router. Exim
918 does (usually) call initgroups() when daemonizing.
920 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
921 with consistent privilege compared to when running as a daemon.
923 TF/03 Note in the spec that $authenticated_id is not set for local
924 submissions from trusted users.
926 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
927 Thanks to Dean Brooks <dean@iglou.com> for the patch.
929 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
930 by adding some example configuration directives to the default
931 configuration file. A little bit of work is required to uncomment the
932 directives and define how usernames and passwords are checked, but
933 there is now a framework to start from.
935 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
936 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
937 without this. I don't know how relevant this is to other LDAP libraries.
939 PH/02 Add the verb name to the "unknown ACL verb" error.
941 PH/03 Magnus Holmgren's patch for filter_prepend_home.
943 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
945 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
946 directory not expanded when it should be if an expanded home directory
947 was set for the address (which is overridden by the transport).
949 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
952 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
953 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
954 because it is too late at that time, and has no effect.
956 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
957 security issue with \' (bugzilla #107). I could not use the
958 PQescapeStringConn() function, because it needs a PGconn value as one of
961 PH/08 When testing addresses using -bt, indicate those final addresses that
962 are duplicates that would not cause an additional delivery. At least one
963 person was confused, thinking that -bt output corresponded to deliveries.
964 (Suppressing duplicates isn't a good idea as you lose the information
965 about possibly different redirections that led to the duplicates.)
967 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
968 systems where poll() doesn't work, in particular OS X.
970 PH/10 Added more information to debugging output for retry time not reached.
972 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
973 operations in malware.c.
975 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
978 PH/13 If write_rejectlog was set false when logging was sent to syslog with
979 syslog_duplication set false, log lines that would normally be written
980 both the the main log and to the reject log were not written to syslog at
983 PH/14 In the default configuration, change the use of "message" in ACL warn
984 statements to "add_header".
986 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
987 not followed by a command (e.g. "seen endif").
989 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
990 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
993 PH/17 Added extra conditions to the default value of delay_warning_condition
997 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
998 { match{$h_precedence:}{(?i)bulk|list|junk} } \
999 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
1002 The Auto-Submitted: and various List- headers are standardised, whereas I
1003 don't think Precedence: ever was.
1005 PH/18 Refactored debugging code in route_finduser() to show more information,
1006 in particular, the error code if getpwnam() issues one.
1008 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1009 This is apparently needed in addition to the PH/07 change above to avoid
1010 any possible encoding problems.
1012 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1013 but not after initializing Perl.
1015 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1016 output them only if debugging. By default they are written stderr,
1017 apparently, which is not desirable.
1019 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1022 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1025 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1027 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1028 authenticated or an ident call has been made. Suppress the default
1029 values for $authenticated_id and $authenticated_sender (but permit -oMai
1030 and -oMas) when testing with -bh.
1032 PH/24 Re-jigged the order of the tests in the default configuration so that the
1033 tests for valid domains and recipients precede the DNS black list and CSA
1034 tests, on the grounds that those ones are more expensive.
1036 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1037 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1038 This bug exists in every version of Exim that I still have, right back to
1041 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1042 However, an attempt to turn on case-sensitivity in a regex key by
1043 including (?-i) didn't work because the subject string was already
1044 lowercased, and the effects were non-intuitive. It turns out that a
1045 one-line patch can be used to allow (?-i) to work as expected.
1051 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1052 other effects) broke the use of negated acl sub-conditions.
1054 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1057 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1058 "Deny" causes Exim to reject the incoming connection with a 554 error.
1059 Unfortunately, if there is a major crisis, such as a disk failure,
1060 tcp-wrappers gives "deny", whereas what one would like would be some
1061 kind of temporary error. A kludge has been added to help with this.
1062 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1063 554 error is used if errno is still zero or contains ENOENT (which occurs
1064 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1067 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1069 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1070 errors. Otherwise a message that provokes a temporary error (when other
1071 messages do not) can cause a whole host to time out.
1073 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1074 addresses were routed directly to files or pipes from a redirect router.
1075 File deliveries just didn't batch; pipe deliveries might have suffered
1078 PH/06 A failure to get a lock for a hints database would erroneously always say
1079 "Failed to get write lock", even when it was really a read lock.
1081 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1082 of 0600. This has been changed to use the value of the lockfile_mode
1083 option (which defaults to 0600).
1085 PH/08 Applied small patch from the Sieve maintainer.
1087 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1088 folder from quota calculations, a direct delivery into this folder messed
1089 up the contents of the maildirsize file. This was because the regex was
1090 used only to exclude .Trash (or whatever) when the size of the mailbox
1091 was calculated. There was no check that a delivery was happening into an
1092 excluded directory. This bug has been fixed by ignoring all quota
1093 processing for deliveries into excluded directories.
1095 PH/10 Added the maildirfolder_create_regex option to appendfile.
1101 PH/01 The code for finding all the local interface addresses on a FreeBSD
1102 system running IPv6 was broken. This may well have applied to all BSD
1103 systems, as well as to others that have similar system calls. The broken
1104 code found IPv4 interfaces correctly, but gave incorrect values for the
1105 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1106 that it would not match correctly against @[] and not recognize the IPv6
1109 PH/02 The ipliteral router was not recognizing addresses of the form user@
1110 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1112 PH/03 Added disable_ipv6.
1114 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1115 decoded form, because it is most often used to construct To: headers
1116 lines in autoreplies, and the decoded form may well be syntactically
1117 invalid. However, $reply_address has leading white space removed, and all
1118 newlines turned into spaces so that the autoreply transport does not
1121 PH/05 If group was specified without a user on a router, and no group or user
1122 was specified on a transport, the group from the router was ignored.
1124 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1125 for visible compile-time settings that can be used to change these
1126 numbers, for those that want even more. Backwards compatibility with old
1127 spool files has been maintained. However, going back to a previous Exim
1128 release will lost any variables that are in spool files.
1130 PH/07 Two small changes when running in the test harness: increase delay when
1131 passing a TCP/IP connection to a new process, in case the original
1132 process has to generate a bounce, and remove special handling of
1133 127.0.0.2 (sic), which is no longer necessary.
1135 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1136 be the same on different OS.
1138 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1141 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1142 whether --show-vars was specified or not
1144 JJ/02 exipick: Added support for new ACL variable spool format introduced
1147 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1148 syntactically invalid From: or Reply-to: line, and a filter used this to
1149 generate an autoreply, and therefore failed to obtain an address for the
1150 autoreply, Exim could try to deliver to a non-existent relative file
1151 name, causing unrelated and misleading errors. What now happens is that
1152 it logs this as a hard delivery error, but does not attempt to create a
1155 PH/11 The exinext utility has a -C option for testing purposes, but although
1156 the given file was scanned by exinext itself; it wasn't being passed on
1157 when Exim was called.
1159 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1160 an end-of-file indication when reading a command response.
1162 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1163 compiled. In many other places in Exim, IPv6 addresses are always
1164 recognized, so I have changed this. It also means that IPv4 domain
1165 literals of the form [IPV4:n.n.n.n] are now always recognized.
1167 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1168 used if the router is not running as root, for example, when verifying at
1169 ACL time, or when using -bh. The debugging output from this situation was
1170 non-existent - all you got was a failure to exec. I have made two
1173 (a) Failures to set uid/gid, the current directory, or a process leader
1174 in a subprocess such as that created by queryprogram now generate
1175 suitable debugging ouput when -d is set.
1177 (b) The queryprogram router detects when it is not running as root,
1178 outputs suitable debugging information if -d is set, and then runs
1179 the subprocess without attempting to change uid/gid.
1181 PH/15 Minor change to Makefile for building test_host (undocumented testing
1184 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1185 additional section of a DNS packet that returns MX or SRV records.
1186 Instead, it always explicitly searches for A/AAAA records. This avoids
1187 major problems that occur when a DNS server includes only records of one
1188 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1189 fixed another bug: if SRV records were looked up and the corresponding
1190 address records were *not* found in the additional section, the port
1191 values from the SRV records were lost.
1193 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1194 using the correct key (the original address) when searching the retry
1195 rules in order to find which one to use for generating the retry hint.
1197 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1198 adding the default one. Similarly, if it contains a Reply-To: header, the
1199 errors_reply_to option, if set, is not used.
1201 PH/19 When calculating a retry time, Exim used to measure the "time since
1202 failure" by looking at the "first failed" field in the retry record. Now
1203 it does not use this if it is later than than the arrival time of the
1204 message. Instead it uses the arrival time. This makes for better
1205 behaviour in cases where some deliveries succeed, thus re-setting the
1206 "first failed" field. An example is a quota failure for a huge message
1207 when small messages continue to be delivered. Without this change, the
1208 "time since failure" will always be short, possible causing more frequent
1209 delivery attempts for the huge message than are intended.
1210 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1212 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1213 $1, $2, $3) because the numerical variables can be reset during some
1214 expansion items (e.g. "match"), thereby losing the authentication data.
1216 PH/21 Make -bV show the size of off_t variables so that the test suite can
1217 decide whether to run tests for quotas > 2G.
1219 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1220 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1221 filecount value is greater than 2G or if a quota value is greater than 2G
1222 on a system where the size of off_t is not greater than 4, a panic error
1225 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1226 never match. The debug and -bh output now contains an explicit error
1227 message indicating a malformed IPv4 address or mask.
1229 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1230 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1231 PH/23 above applies.
1233 PH/25 Do not write to syslog when running in the test harness. The only
1234 occasion when this arises is a failure to open the main or panic logs
1235 (for which there is an explicit test).
1237 PH/26 Added the /no_tell option to "control=freeze".
1239 PH/27 If a host name lookup failed very early in a connection, for example, if
1240 the IP address matched host_lookup and the reverse lookup yielded a name
1241 that did not have a forward lookup, an error message of the form "no IP
1242 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1243 could be logged. Now it outputs the IP address instead of "NULL".
1245 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1246 allows the sender and the authenticated sender to be set when
1247 submitting a message from within Exim. Since child_open_exim() is
1248 documented for local_scan(), the new function should be too.
1250 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1251 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1252 results in an empty string is now treated as unset.
1254 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1256 PH/31 Added sender_verify_fail logging option.
1258 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1259 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1260 tidied the source and removed it altogether.
1262 PH/33 When a queue run was abandoned because the load average was too high, a
1263 log line was always written; now it is written only if the queue_run log
1264 selector is set. In addition, the log line for abandonment now contains
1265 information about the queue run such as the pid. This is always present
1266 in "start" and "stop" lines but was omitted from the "abandon" line.
1268 PH/34 Omit spaces between a header name and the colon in the error message that
1269 is given when verify = headers_syntax fails (if there are lots of them,
1270 the message gets confusing).
1272 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1273 names, as there are now some PTR records that contain slashes. This check
1274 is only to protect against broken name servers that fall over on strange
1275 characters, so the fact that it applies to all lookups doesn't matter.
1277 PH/36 Now that the new test suite is complete, we can remove some of the
1278 special code in Exim that was needed for the old test suite. For example,
1279 sorting DNS records because real resolvers return them in an arbitrary
1280 order. The new test suite's fake resolver always returns records in the
1283 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1284 bounces) except when queue_only is set, to avoid logging races between
1285 the different processes.
1287 PH/38 Panic-die if .include specifies a non-absolute path.
1289 PH/39 A tweak to the "H" retry rule from its user.
1291 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1292 a label. They prevented compilation on older perls.
1294 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1295 a warning to be raised on newish perls.
1297 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1298 on queue. Changes to match documented behaviour of showing count of
1299 messages matching specified criteria.
1301 PH/40 Changed the default ident timeout from 30s to 5s.
1303 PH/41 Added support for the use of login_cap features, on those BSD systems
1304 that have them, for controlling the resources used by pipe deliveries.
1306 PH/42 The content-scanning code uses fopen() to create files in which to put
1307 message data. Previously it was not paying any attention to the mode of
1308 the files. Exim runs with umask(0) because the rest of the code creates
1309 files with open(), and sets the required mode explicitly. Thus, these
1310 files were ending up world-writeable. This was not a big issue, because,
1311 being within the spool directory, they were not world-accessible. I have
1312 created a function called modefopen, which takes an additional mode
1313 argument. It sets umask(777), creates the file, chmods it to the required
1314 mode, then resets the umask. All the relevant calls to fopen() in the
1315 content scanning code have been changed to use this function.
1317 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1318 to 24 hours. This avoids potential overflow problems when processing G
1319 and H retry rules. I suspect nobody ever tinkers with this value.
1321 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1323 PH/45 When the plaintext authenticator is running as a client, the server's
1324 challenges are checked to ensure they are valid base64 strings. By
1325 default, the authentication attempt is cancelled if an invalid string is
1326 received. Setting client_ignore_invalid_base64 true ignores these errors.
1327 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1328 they are received. Thus, the responses can be made to depend on the
1329 challenges. If an invalid string is ignored, an empty string is placed in
1332 PH/46 Messages that are created by the autoreply transport now contains a
1333 References: header, in accordance with RFCs 2822 and 3834.
1335 PH/47 Added authenticated_sender_force to the smtp transport.
1337 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1339 PH/49 Installed latest patch from the Sieve maintainer.
1341 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1342 was also set, the appendfile transport was unnecessarily scanning a
1343 directory of message files (e.g. for maildir delivery) to find the count
1344 of files (along with the size), even though it did not need this
1345 information. It now does the scan only if it needs to find either the
1346 size of the count of files.
1348 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1350 PH/52 Two bugs concerned with error handling when the smtp transport is
1353 (i) Exim was not creating retry information for temporary errors given
1354 for individual recipients after the DATA command when the smtp transport
1355 was used in LMTP mode. This meant that they could be retried too
1356 frequently, and not timed out correctly.
1358 (ii) Exim was setting the flag that allows error details to be returned
1359 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1360 recipients that were returned after the DATA command.
1362 PH/53 This is related to PH/52, but is more general: for any failing address,
1363 when detailed error information was permitted to be returned to the
1364 sender, but the error was temporary, then after the final timeout, only
1365 "retry timeout exceeded" was returned. Now it returns the full error as
1366 well as "retry timeout exceeded".
1368 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1369 do this, and (what is worse) MTAs that accept it.
1371 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1372 will now be deprecated.
1374 PH/56 New os.c-cygwin from the Cygwin maintainer.
1376 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1377 formats (previously only available in exim formats via -bpr, -bpru,
1378 and -bpra. Now also available in native and exiqgrep formats)
1380 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1381 with very large, slow to parse queues
1383 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1385 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1387 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1388 responses to authentication challenges, though it was showing the
1389 challenges; (ii) I've removed the CR characters from the debug output for
1392 PH/58 Allow for the insertion of a newline as well as a space when a string
1393 is turned into more than one encoded-word during RFC 2047 encoding. The
1394 Sieve code now uses this.
1396 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1397 data_4xx, lost_connection, tls_required.
1399 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1400 message was being sent as an SMTP response.
1402 PH/61 Add -l and -k options to exicyclog.
1404 PH/62 When verifying, if an address was redirected to one new address, so that
1405 verification continued, and the new address failed or deferred after
1406 having set something in $address_data, the value of $address_data was not
1407 passed back to the ACL. This was different to the case when no
1408 redirection occurred. The value is now passed back in both cases.
1410 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1411 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1412 use login_cap.h, so on its own it isn't the distinguishing feature. The
1413 new name refers directly to the setclassresources() function.
1415 PH/65 Added configuration files for NetBSD3.
1417 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1419 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1420 is preferred over IPv4.
1422 PH/68 The bounce_return_message and bounce_return_body options were not being
1423 honoured for bounces generated during the reception of non-SMTP messages.
1424 In particular, this applied to messages rejected by the ACL. This bug has
1425 been fixed. However, if bounce_return_message is true and bounce_return_
1426 body is false, the headers that are returned for a non-SMTP message
1427 include only those that have been read before the error was detected.
1428 (In the case of an ACL rejection, they have all been read.)
1430 PH/69 The HTML version of the specification is now built in a directory called
1431 spec_html instead of spec.html, because the latter looks like a path with
1432 a MIME-type, and this confuses some software.
1434 PH/70 Catch two compiler warnings in sieve.c.
1436 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1437 function verify_get_ident() calls ip_connect() to connect a socket, but
1438 if the "connect()" function timed out, ip_connect() used to close the
1439 socket. However, verify_get_ident() also closes the socket later, and in
1440 between Exim writes to the log, which may get opened at this point. When
1441 the socket was closed in ip_connect(), the log could get the same file
1442 descriptor number as the socket. This naturally causes chaos. The fix is
1443 not to close the socket in ip_connect(); the socket should be closed by
1444 the function that creates it. There was only one place in the code where
1445 this was missing, in the iplookup router, which I don't think anybody now
1446 uses, but I've fixed it anyway.
1448 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1449 well as to direct DNS lookups. Otherwise the handling of names in host
1450 lists is inconsistent and therefore confusing.
1456 PH/01 Two changes to the default runtime configuration:
1458 (1) Move the checks for relay_from_hosts and authenticated clients from
1459 after to before the (commented out) DNS black list checks.
1461 (2) Add control=submission to the relay_from_hosts and authenticated
1462 clients checks, on the grounds that messages accepted by these
1463 statements are most likely to be submissions.
1465 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1467 (1) Generate an error if the third argument for the ${prvs expansion is
1470 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1473 (3) Reset the variables that are obtained from the first argument of
1474 ${prvscheck and used in the second argument before leaving the code,
1475 because their memory is reclaimed, so using them afterwards may do
1478 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1479 one (it's much easier than Tom thought :-).
1481 (5) Because of (4), we can now allow for the use of $prvscheck_result
1482 inside the third argument.
1484 PH/03 For some reason, the default setting of PATH when running a command from
1485 a pipe transport was just "/usr/bin". I have changed it to
1488 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1489 anything to be listed in the output from -bV.
1491 PH/05 When a filter generated an autoreply, the entire To: header line was
1492 quoted in the delivery log line, like this:
1494 => >A.N.Other <ano@some.domain> <original@ddress> ...
1496 This has been changed so that it extracts the operative address. There
1497 may be more than one such address. If so, they are comma-separated, like
1500 => >ano@some.domain,ona@other.domain <original@ddress> ...
1502 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1503 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1504 not being looked up in the rDNS to get a host name, Exim was showing the
1505 IP address twice in Received: lines, even though the IP addresses were
1506 identical. For example:
1508 Received: from [1.2.3.4] (helo=[1.2.3.4])
1510 However, if the real host name was known, it was omitting the HELO data
1511 if it matched the actual IP address. This has been tidied up so that it
1512 doesn't show the same IP address twice.
1514 PH/07 When both +timestamp and +memory debugging was on, the value given by
1515 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1516 called by the debug printing, thereby overwriting the timestamp buffer.
1517 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1519 PH/08 When the original message was included in an autoreply transport, it
1520 always said "this is a copy of the message, including all the headers",
1521 even if body_only or headers_only was set. It now gives an appropriate
1524 PH/09 Applied a patch from the Sieve maintainer which:
1526 o fixes some comments
1527 o adds the (disabled) notify extension core
1528 o adds some debug output for the result of if/elsif tests
1529 o points to the current vacation draft in the documentation
1530 and documents the missing references header update
1534 o fixes a bug in processing the envelope test (when testing
1535 multiple envelope elements, the last element determinted the
1538 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1539 Electronic Mail") by including:
1541 Auto-submitted: auto-generated
1543 in the messages that it generates (bounce messages and others, such as
1544 warnings). In the case of bounce messages for non-SMTP mesages, there was
1545 also a typo: it was using "Auto_submitted" (underscore instead of
1546 hyphen). Since every message generated by Exim is necessarily in response
1547 to another message, thes have all been changed to:
1549 Auto-Submitted: auto-replied
1551 in accordance with these statements in the RFC:
1553 The auto-replied keyword:
1555 - SHOULD be used on messages sent in direct response to another
1556 message by an automatic process,
1558 - MUST NOT be used on manually-generated messages,
1560 - MAY be used on Delivery Status Notifications (DSNs) and Message
1561 Disposition Notifications (MDNs),
1563 - MUST NOT be used on messages generated by automatic or periodic
1564 processes, except for messages which are automatic responses to
1567 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1568 to the default Received: header definition.
1570 PH/12 Added log selector acl_warn_skipped (default on).
1572 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1573 variables because (a) they are in the wrong storage pool and (b) even if
1574 they were copied, it wouldn't work properly because of the caching.
1576 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1577 checking when decoding. Apparently there are clients that generate
1578 overlong encoded strings. Why am I not surprised?
1580 PH/15 If the first argument of "${if match_address" was not empty, but did not
1581 contain an "@" character, Exim crashed. Now it writes a panic log message
1582 and treats the condition as false.
1584 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1586 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1587 extension "envelope-auth". The code is finished and in agreement with
1588 other implementations, but there is no documentation so far and in fact,
1589 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1590 not changing the active code.
1592 Print executed "if" and "elsif" statements when debugging is used. This
1593 helps a great deal to understand what a filter does.
1595 Document more things not specified clearly in RFC3028. I had all this
1596 sorted out, when out of a sudden new issues came to my mind. Oops."
1598 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1601 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1602 canonical form (as documented). However, after a host name lookup from
1603 the IP address, check_host() was doing a simple string comparison with
1604 addresses acquired from the DNS when checking that the found name did
1605 have the original IP as one of its addresses. Since any found IPv6
1606 addresses are likely to be in abbreviated form, the comparison could
1607 fail. Luckily, there already exists a function for doing the comparison
1608 by converting both addresses to binary, so now that is used instead of
1609 the text comparison.
1611 PH/20 There was another similar case to PH/19, when a complete host name was
1612 given in a host list; looking up its IP address could give an abbreviated
1613 form, whereas the current host's name might or might not be abbreviated.
1614 The same fix has been applied.
1620 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1621 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1624 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1625 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1627 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1629 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1630 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1631 Also, for "Auto-Submitted", treat anything other than "no" as
1632 non-personal, in accordance with RFC 3834. (Previously it treated
1633 anything starting "auto-" as non-personal.)
1635 TF/01 The control=submission/name=... option had a problem with syntax
1636 errors if the name included a slash character. The /name= option
1637 now slurps the rest of the string, so it can include any characters
1638 but it must come last in the list of options (after /sender_retain
1641 PH/05 Some modifications to the interface to the fake nameserver for the new
1649 TK/01 Added the "success_on_redirect" address verification option. See
1650 NewStuff for rationale and an example.
1652 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1654 PH/02 Patch to exigrep to allow it to work on syslog lines.
1656 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1657 fread() to skip over the body file's header line, because in Cygwin the
1658 header line is locked and is inaccessible.
1660 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1661 co-exist for some time) to make it clear that it is the Exim ID that is
1662 referenced, not the Message-ID: header line.
1664 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1665 string_format() function, because snprintf() does not exist on all
1668 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1669 ${stat: expansion item.
1671 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1672 protocol synchronization error", to keep the pedants happy.
1674 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1675 well as for IRIX systems, when gcc is being used. See the host.c source
1678 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1680 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1683 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1684 transport and to the smtp transport in LMTP mode.
1686 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1688 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1689 run a filter in a subprocess. This could lead to confusion in subsequent
1690 lookups in the parent process. There should also be a search_tidyup() at
1691 the end of the subprocess.
1693 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1694 only if the host matched helo_try_verify_hosts, which caused the
1695 verification to occur when the EHLO/HELO command was issued. The ACL just
1696 tested the remembered result. Now, if a previous verification attempt has
1697 not happened, "verify = helo" does it there and then.
1699 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1701 TK/03 Fix log output including CR from clamd.
1703 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1704 exist provoked a memory error which could cause a segfault.
1706 PH/15 Installed PCRE 6.2
1708 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1710 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1711 of the problem. Specifically, suggested +O2 rather than +O1 for the
1714 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1716 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1717 record, the port was not correctly logged when the outgoing_port log
1718 selector was set (it logged the transort's default port).
1720 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1721 fallback_hosts, and "hosts" in the smtp transport.
1723 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1724 host errors such as "Connection refused".
1726 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1727 authentication with radiusclient 0.4.9:
1729 - Error returned from rc_read_config was caught wrongly
1730 - Username/password not passed on to radius server due to wrong length.
1732 The presumption is that some radiusclient API changes for 4.51/PH/17
1733 were not taken care of correctly. The code is still untested by me (my
1734 Linux distribution still has 0.3.2 of radiusclient), but it was
1735 contributed by a Radius user.
1737 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1738 expanding the "port" option of the smtp transport.
1740 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1741 while reading a MIME header. Thanks to Tom Hughes for a patch.
1743 PH/24 Include config.h inside local_scan.h so that configuration settings are
1746 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1747 that in an ACL for RCPT (for example), you can examine exactly what was
1750 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1751 commands, but it was not correctly comparing the address with the actual
1752 client host address. Thus, it would show the EHLO address in Received:
1753 header lines when this was not necessary.
1755 PH/27 Added the % operator to ${eval:}.
1757 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1758 it should be ignoring failures (because with -C, for example, it has lost
1759 privilege). It wasn't ignoring creation failures other than "already
1762 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1765 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1766 because an input line was too long, either on its own, or by virtue of
1767 too many continuations, the temporary file was not being removed, and the
1768 return code was incorrect.
1770 PH/31 Missing "BOOL" in function definition in filtertest.c.
1772 PH/32 Applied Sieve patches from the maintainer.
1774 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1776 PH/33 Added "verify = not_blind".
1778 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1779 Local/Makefile (with some defaults set). These are used in built scripts
1780 such as exicyclog, but they have never been used in the exim_install
1781 script (though there are many overriding facilities there). I have
1782 arranged that the exim_install script now takes note of these two
1785 PH/35 Installed configuration files for Dragonfly.
1787 PH/36 When a locally submitted message by a trusted user did not contain a
1788 From: header, and the sender address was obtained from -f or from an SMTP
1789 MAIL command, and the trusted user did not use -F to supply a sender
1790 name, $originator_name was incorrectly used when constructing a From:
1791 header. Furthermore, $originator_name was used for submission mode
1792 messages from external hosts without From: headers in a similar way,
1793 which is clearly wrong.
1795 PH/37 Added control=suppress_local_fixups.
1797 PH/38 When log_selector = +received_sender was set, and the addition of the
1798 sender made the log line's construction buffer exactly full, or one byte
1799 less than full, an overflow happened when the terminating "\n" was
1802 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1803 when the result of a list match is failure because a DNS lookup failed.
1805 PH/40 RM_COMMAND is now used in the building process.
1807 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1808 the "build-* directories that it finds.
1810 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1811 address in a domain literal was a prefix of an interface address.
1813 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1814 when verifying a sender address, unless rewrite_headers is false.
1816 PH/44 Wrote a long comment about why errors_to addresses are verified as
1817 recipients, not senders.
1819 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1820 the ratelimit ACL was added.
1822 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1824 PH/46 Added extra information about PostgreSQL errors to the error string.
1826 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1827 avoiding the need to install special zones in a real server. This is
1828 backwards compatible; if it can't find the fake resolver, it drops back.
1829 Thus, both old and new test suites can be run.
1831 TF/02 Added util/ratelimit.pl
1833 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1834 clock is set back in time.
1836 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1837 Candler <B.Candler@pobox.com>.
1839 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1840 OK for addresses that are the result of redirections.
1842 PH/48 A number of further additions for the benefit of the new test suite,
1843 including a fake gethostbyname() that interfaces to the fake DNS resolver
1846 TF/06 The fix for widen_domains has also been applied to qualify_single and
1847 search_parents which are the other dnslookup options that can cause
1850 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1853 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1855 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1856 service" if it encountered an MX record with an empty target hostname.
1857 The message is now "an MX or SRV record indicated no SMTP service".
1859 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1860 if the DNS of the sending site is misconfigured. This is quite a
1861 common situation. This change restores the behaviour of treating a
1862 helo verification defer as a failure.
1864 PH/51 If self=fail was set on a router, the bounce message did not include the
1865 actual error message.
1871 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1873 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1874 command itself ended in error, the underlying message about the transport
1875 filter timeout was being overwritten with the pipe command error. Now the
1876 underlying error message should be appended to the second error message.
1878 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1880 PH/02 Reduce the amount of output that "make" produces by default. Full output
1881 can still be requested.
1883 PH/03 The warning log line about a condition test deferring for a "warn" verb
1884 was being output only once per connection, rather than after each
1885 occurrence (because it was using the same function as for successful
1886 "warn" verbs). This seems wrong, so I have changed it.
1888 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1889 it should not have, which might have caused a crash in the right
1890 circumstances, but probably never did.
1892 PH/04 Installed a modified version of Tony Finch's patch to make submission
1893 mode fix the return path as well as the Sender: header line, and to
1894 add a /name= option so that you can make the user's friendly name appear
1897 TF/03 Added the control = fakedefer ACL modifier.
1899 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1900 Mark Lowes for thorough testing.
1902 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1904 TK/03 Merged latest SRS patch from Miles Wilton.
1906 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1907 with the definition in sysexits.h (which is #included earlier).
1908 Fortunately, Exim does not actually use EX_OK. The code used to try to
1909 preserve the sysexits.h value, by assumimg that macro definitions were
1910 scanned for macro replacements. I have been disabused of this notion,
1911 so now the code just undefines EX_OK before #including unistd.h.
1913 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1914 in the smtp transport. When a block could not be written in a single
1915 write() function, the timeout was being re-applied to each part-write.
1916 This seems wrong - if the receiver was accepting one byte at a time it
1917 would take for ever. The timeout is now adjusted when this happens. It
1918 doesn't have to be particularly precise.
1920 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1921 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1923 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1924 without a domain if the check to <postmaster@domain> fails.
1926 SC/01 Eximstats: added -xls and the ability to specify output files
1927 (patch written by Frank Heydlauf).
1929 SC/02 Eximstats: use FileHandles for outputing results.
1931 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1933 SC/04 Eximstats: fixed display of large numbers with -nvr option
1935 SC/05 Eximstats: fixed merging of reports with empty tables.
1937 SC/06 Eximstats: added the -include_original_destination flag
1939 SC/07 Eximstats: removed tabs and trailing whitespace.
1941 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1943 TK/06 MBOX spool code: Add real "From " MBOX separator line
1944 so the .eml file is really in mbox format (even though
1945 most programs do not really care). Patch from Alex Miller.
1947 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1948 The latter is generated from $received_to and is only set if the
1949 message has one envelope recipient. SA can use these headers,
1950 obviously out-of-the-box. Patch from Alex Miller.
1952 PH/08 The ${def test on a variable was returning false if the variable's
1953 value was "0", contrary to what the specification has always said!
1954 The result should be true unless the variable is empty.
1956 PH/09 The syntax error of a character other than { following "${if
1957 def:variable_name" (after optional whitespace) was not being diagnosed.
1958 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1959 accidental colon was present, for example, could give incorrect results.
1961 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1962 result is used (not including appendfile, where other changes are about
1965 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1966 This involved changing a lot of size variables from int to off_t. It
1967 should work with maildirs and everything.
1969 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1970 spamd dying while we are connected to it.
1972 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1975 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1976 function for BDB 4.3.
1978 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1980 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1981 the sources for PCRE have been re-arranged and more files are now
1984 PH/15 The code I had for printing potentially long long variables in PH/11
1985 above was not the best (it lost precision). The length of off_t variables
1986 is now inspected at build time, and an appropriate printing format (%ld
1987 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1988 to be "long long int" or "long int". This is needed for the internal
1989 formatting function string_vformat().
1991 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1992 the configuration file to be ":syslog", then the script "guesses" where
1993 the logs files are, rather than using the compiled in default. In our
1994 case the guess is not the same as the compiled default, so the script
1995 suddenly stopped working when I started to use syslog. The patch checks
1996 to see if log_file_path is "". If so, it attempts to read it from exim
1997 with no configuration file to get the compiled in version, before it
1998 falls back to the previous guessing code."
2000 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
2001 implementing BATV in an Exim configuration. See NewStuff for the gory
2004 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2005 Makefile that are specific to HP-UX.
2007 PH/18 If the "use_postmaster" option was set for a recipient callout together
2008 with the "random" option, the postmaster address was used as the MAIL
2009 FROM address for the random test, but not for the subsequent recipient
2010 test. It is now used for both.
2012 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2013 patch removes a few documentation additions to RFC 3028, because the
2014 latest draft now contains them. It adds the new en;ascii-case comparator
2015 and a new error check for 8bit text in MIME parts. Comparator and
2016 require names are now matched exactly. I enabled the subaddress
2017 extension, but it is not well tested yet (read: it works for me)."
2019 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2020 rework some of the code of TK/09 above to avoid the hardwired use of
2021 "%lld" and "long long". Replaced the call to snprintf() with a call to
2024 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2025 records point to non-existent hosts", "retry timeout exceeded", and
2026 "retry time not reached for any host after a long failure period".
2028 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2029 experimental DomainKeys support:
2031 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2032 (2) On an error such as an illegally used "control", the wrong name for
2033 the control was given.
2035 These problems did NOT occur unless DomainKeys support was compiled.
2037 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2039 PH/24 Added ${if match_ip condition.
2041 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2042 databases so that it will be absolutely obvious if a crash occurs in the
2043 DB library. This is a regular occurrence (often caused by mis-matched
2046 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2047 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2048 picked up on a user's system that detects such things. There doesn't seem
2049 to be a gcc warning option for this - only an attribute that has to be
2050 put on the function's prototype. It seems that in Fedora Core 4 they have
2051 set this on a number of new functions. No doubt there will be more in due
2054 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2055 specify a transport. However, if an address that was verified by such a
2056 router was the subject of a callout, Exim crashed because it tried to
2057 read the rcpt_include_affixes from the non-existent transport. Now it
2058 just assumes that the setting of that option is false. This bug was
2059 introduced by 4.51/PH/31.
2061 PH/28 Changed -d+all to exclude +memory, because that information is very
2062 rarely of interest, but it makes the output a lot bigger. People tend to
2063 do -d+all out of habit.
2065 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2066 code in os-type was giving problems when libc.so lives in lib64, like on
2069 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2070 aren't the modern standard, and it seems that some systems' include files
2071 don't always have them. Exim was already checking for some of the newer
2072 ones like T_AAAA, and defining it itself. I've added checks for all the
2073 record types that Exim uses.
2075 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2076 not automatically generating a new one, as it is supposed to. This
2077 prevented TLS from working. If the file did exist, but contained invalid
2078 data, a new version was generated, as expected. It was only the case of a
2079 non-existent file that was broken.
2081 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2082 with a change in libdomainkeys > 0.64.
2084 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2085 from DNS. If the selector record carries the flag, it now has
2086 precedence over the domain-wide flag.
2088 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2090 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2091 the use of a port name that isn't defined in /etc/services) occurred, the
2092 message was deferred as in a normal delivery, and thus remained on the
2093 spool, instead of being failed because of the mua_wrapper setting. This
2094 is now fixed, and I tidied up some of the mua_wrapper messages at the
2097 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2098 the messages in a hash of arrays rather than using individual hashes.
2099 This is a bit cleaner and results in dramatic memory savings, albeit
2100 at a slight CPU cost.
2102 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2103 as requested by Marc Sherman.
2105 SC/10 Eximstats: added histograms for user specified patterns as requested
2108 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2110 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2111 fopen() in the content-scanning modules that did not already have it.
2117 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2118 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2120 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2122 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2124 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2125 to test Sieve filters that use "vacation".
2127 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2128 that changes the way the GnuTLS parameters are stored in the cache file.
2129 The new format can be generated externally. For backward compatibility,
2130 if the data in the cache doesn't make sense, Exim assumes it has read an
2131 old-format file, and it generates new data and writes a new file. This
2132 means that you can't go back to an older release without removing the
2135 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2136 work if there are any delivery delays because "one_time" forces the
2137 parent to be marked "delivered", so its unseen clone is never tried
2138 again. For this reason, Exim now forbids the simultaneous setting of
2141 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2142 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2143 entry if you want to know the details. The fix, however, neglected to
2144 consider the case when local delivery batching is involved. The test for
2145 "previously delivered" was not happening when checking to see if an
2146 address could be batched with a previous (undelivered) one; under
2147 certain circumstances this could lead to multiple deliveries to the same
2150 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2151 in its include files, and this causes problems building Exim.
2153 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2154 header_syntax) but Exim was just ignoring anything given after a slash.
2155 In particular, this caused confusion with an attempt to use "verify =
2156 reverse_host_lookup/defer_ok". An error is now given when options are
2157 supplied for verify items that do not have them. (Maybe reverse_host_
2158 lookup should have a defer_ok option, but that's a different point.)
2160 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2161 defined by RFC 821) to 2048, because there were problems with some AUTH
2162 commands, and RFC 1869 says the size should be increased for extended
2163 SMTP commands that take arguments.
2165 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2168 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2169 "unknown" error; now it says that the functionality isn't in the binary.
2171 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2172 an address' error message when a string expansion fails (syntax or
2173 whatever). Otherwise the password may appear in the log. Following change
2174 PH/42 below, there is no longer a chance of it appearing in a bounce
2177 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2179 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2180 of its addresses was ever tried. (Bugzilla bug #2.)
2182 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2183 the result incorrectly in the debug output. (It correctly added a newline
2184 to what was transported.)
2186 TF/01 Added $received_time.
2188 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2189 commented out examples of how to interface to a virus scanner and to
2190 SpamAssassin. Also added commented examples of av_scanner and
2191 spamd_address settings.
2193 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2194 and controls are allowed in which ACLs. There were a couple of minor
2195 errors. Some of the entries in the conditions table (which is a table of
2196 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2197 negation of where the condition IS allowed.
2199 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2201 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2202 header file does not have a version number, so I've had to invent a new
2203 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2204 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2205 radiusclient), but it was contributed by a Radius user.
2207 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2208 files or directories, for OpenSSL.
2210 PH/19 When an Exim process that is running as root has to create an Exim log
2211 file, it does so in a subprocess that runs as exim:exim so as to get the
2212 ownership right at creation (otherwise, other Exim processes might see
2213 the file with the wrong ownership). There was no test for failure of this
2214 fork() call, which would lead to the process getting stuck as it waited
2215 for a non-existent subprocess. Forks do occasionally fail when resources
2216 run out. I reviewed all the other calls to fork(); they all seem to check
2219 PH/20 When checking for unexpected SMTP input at connect time (before writing
2220 the banner), Exim was not dealing correctly with a non-positive return
2221 from the read() function. If the client had disconnected by this time,
2222 the result was a log entry for a synchronization error with an empty
2223 string after "input=" when read() returned zero. If read() returned -1
2224 (an event I could not check), uninitialized data bytes were printed.
2225 There were reports of junk text (parts of files, etc) appearing after
2228 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2230 PH/22 Added support for macro redefinition, and (re)definition in between
2231 driver and ACL definitions.
2233 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2234 forgetting to use the resulting value; it was using the unexpanded value.
2236 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2237 hadn't been configured. The fix is from Juergen Kreileder, who
2238 understands it better than I do:
2240 "Here's what I see happening with three configured cyrus_sasl
2241 authenticators configured (plain, login, cram-md5):
2243 On startup auth_cyrus_sasl_init() gets called for each of these.
2244 This means three calls to sasl_listmech() without a specified mech_list.
2245 => SASL tests which mechs of all available mechs actually work
2246 => three warnings about OTP not working
2247 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2249 With the patch, sasl_listmech() also gets called three times. But now
2250 SASL's mech_list option is set to the server_mech specified in the the
2251 authenticator. Or in other words, the answer from sasl_listmech()
2252 gets limited to just the mech you're testing for (which is different
2254 => the return list contains just 'plain' or 'login', 'cram-md5' or
2255 nothing depending on the value of ob->server_mech.
2257 I've just tested the patch: Authentication still works fine,
2258 unavailable mechs specified in the exim configuration are still
2259 caught, and the auth.log warnings about OTP are gone."
2261 PH/25 When debugging is enabled, the contents of the command line are added
2262 to the debugging output, even when log_selector=+arguments is not
2265 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2266 answer is "GNU", and only if the return is "GNU/something" is the answer
2269 PH/27 $acl_verify_message is now set immediately after the failure of a
2270 verification in an ACL, and so is available in subsequent modifiers. In
2271 particular, the message can be preserved by coding like this:
2273 warn !verify = sender
2274 set acl_m0 = $acl_verify_message
2276 Previously, $acl_verify_message was set only while expanding "message"
2277 and "log_message" when a very denied access.
2279 PH/28 Modified OS/os.c-Linux with
2281 -#ifndef OS_LOAD_AVERAGE
2282 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2284 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2285 nomenclature these days.)
2287 PH/29 Installed patch from the Sieve maintainer that adds the options
2288 sieve_useraddress and sieve_subaddress to the redirect router.
2290 PH/30 In these circumstances:
2291 . Two addresses routed to the same list of hosts;
2292 . First host does not offer TLS;
2293 . First host accepts first address;
2294 . First host gives temporary error to second address;
2295 . Second host offers TLS and a TLS session is established;
2296 . Second host accepts second address.
2297 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2298 and peerdn, if requested) that were in fact used only for the second
2301 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2302 attention to any local part prefix or suffix that was matched by the
2303 router that accepted the address. It now behaves in the same way as it
2304 does for delivery: the affixes are removed from the local part unless
2305 rcpt_include_affixes is set on the transport.
2307 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2308 timeout during the DATA phase of an incoming message.
2310 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2311 applied a patch sanctioned by the Sieve maintainer.
2313 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2314 the uid or gid is negative. A case of a negative gid caused this to be
2315 noticed. The fix allows for either to be negative.
2317 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2318 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2321 PH/36 The change PH/12 above was broken. Fixed it.
2323 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2324 the grounds that routing the same address twice would always produce the
2325 same answer. This might have been true once, but it is certainly no
2326 longer true now. Routing a child address may depend on the previous
2327 routing that produced that child. Some complicated redirection strategies
2328 went wrong when messages had multiple recipients, and made Exim's
2329 behaviour dependent on the order in which the addresses were given.
2331 I have moved the duplicate checking until after the routing is complete.
2332 Exim scans the addresses that are assigned to local and remote
2333 transports, and removes any duplicates. This means that more work will be
2334 done, as duplicates will always all be routed, but duplicates are
2335 presumably rare, so I don't expect this is of any significance.
2337 For deliveries to pipes, files, and autoreplies, the duplicate checking
2338 still happens during the routing process, since they are not going to be
2341 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2342 It corrects a timeout issue with spamd. This is Ian's comment: "The
2343 background is that sometimes spamd either never reads data from a
2344 connection it has accepted, or it never writes response data. The exiscan
2345 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2346 blindly assumes that writes won't block so it may never time out."
2348 PH/39 Allow G after quota size as well as K and M.
2350 PH/40 The value set for $authenticated_id in an authenticator may not contain
2351 binary zeroes or newlines because the value is written to log lines and
2352 to spool files. There was no check on this. Now the value is run through
2353 the string_printing() function so that such characters are converted to
2354 printable escape sequences.
2356 PH/41 $message_linecount is a new variable that contains the total number of
2357 lines in the message. Compare $body_linecount, which is the count for the
2360 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2361 bounce and delay warning messages, except in certain special cases, which
2364 (a) An SMTP error message from a remote host;
2365 (b) A message specified in a :fail: redirection;
2366 (c) A message specified in a "fail" command in a system filter;
2367 (d) A message specified in a FAIL return from the queryprogram router;
2368 (e) A message specified by the cannot_route_message router option.
2370 In these cases only, Exim does include the error details in bounce and
2371 warning messages. There are also a few cases where bland messages such
2372 as "unrouteable address" or "local delivery error" are given.
2374 PH/43 $value is now also set for the "else" part of a ${run expansion.
2376 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2377 being worked on, but at least Exim now implements the latest version to
2380 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2381 process to complete was treated as a delivery failure, a timeout while
2382 writing the message to the pipe was logged, but erroneously treated as a
2383 successful delivery. Such timeouts include transport filter timeouts. For
2384 consistency with the overall process timeout, these timeouts are now
2385 treated as errors, giving rise to delivery failures by default. However,
2386 there is now a new Boolean option for the pipe transport called
2387 timeout_defer, which, if set TRUE, converts the failures into defers for
2388 both kinds of timeout. A transport filter timeout is now identified in
2391 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2392 systems where "make" and "gmake" are different, calling "gmake" at top
2393 level broke things. I've arranged for the value of $(MAKE) to be passed
2394 from the Makefile to this script so that it can call the same version of
2398 A note about Exim versions 4.44 and 4.50
2399 ----------------------------------------
2401 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2402 changes of various kinds. As a consequence, a big documentation update was
2403 needed. This delayed the release for rather longer than seemed good, especially
2404 in the light of a couple of (minor) security issues. Therefore, the changes
2405 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2406 release. So 4.44 and 4.50 are in effect two different branches that both start
2409 I have left the 4.50 change log unchanged; it contains all the changes since
2410 4.43. The change log for 4.44 is below; many of its items are identical to
2411 those for 4.50. This seems to be the most sensible way to preserve the
2412 historical information.
2418 1. Minor wording change to the doc/README.SIEVE file.
2420 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2421 computation of the current number of files was incorrect.
2423 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2424 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2427 4. Give more explanation in the error message when the command for a transport
2428 filter fails to execute.
2430 5. There are several places where Exim runs a non-Exim command in a
2431 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2432 was being done only for the command run by the queryprogram router. It is
2433 now done for all such subprocesses. The other cases are: ${run, transport
2434 filters, and the commands run by the lmtp and pipe transports.
2436 6. Added CONFIGURE_GROUP build-time option.
2438 7. Some older OS have a limit of 256 on the maximum number of file
2439 descriptors. Exim was using setrlimit() to set 1000 as a large value
2440 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2441 systems. I've change it so that if it can't get 1000, it tries for 256.
2443 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2444 was an oversight, and furthermore, ever since the addition of extra
2445 controls (e.g. 4.43/32), the checks on when to allow different forms of
2446 "control" were broken. There should now be diagnostics for all cases when a
2447 control that does not make sense is encountered.
2449 9. Added the /retain_sender option to "control=submission".
2451 10. $recipients is now available in the predata ACL (oversight).
2453 11. Tidy the search cache before the fork to do a delivery from a message
2454 received from the command line. Otherwise the child will trigger a lookup
2455 failure and thereby defer the delivery if it tries to use (for example) a
2456 cached ldap connection that the parent has called unbind on.
2458 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2459 of $address_data from the recipient verification was clobbered by the
2460 sender verification.
2462 13. The value of address_data from a sender verification is now available in
2463 $sender_address_data in subsequent conditions in the ACL statement.
2465 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2467 15. Added a new option "connect=<time>" to callout options, to set a different
2470 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2471 was its contents. (It was OK if the option was not defined at all.)
2473 17. A "Completed" log line is now written for messages that are removed from
2474 the spool by the -Mrm option.
2476 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2477 information about exactly what failed.
2479 19. Added -dd to debug only the daemon process.
2481 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2482 handles timeouts, both on the server side and network timeouts. Renamed the
2483 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2485 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2486 It is now set to "smtps".
2488 22. $host_address is now set to the target address during the checking of
2489 ignore_target_hosts.
2491 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2492 being passed; this would have caused $sender_host_name to have been used if
2493 matching the list had actually called for a host name (not very likely,
2494 since this list is usually IP addresses). A host name is now passed as
2497 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2498 code that specifies a non-restarting handler (typically sigaction() in
2499 modern systems) in an attempt to fix a rare and obscure crash bug.
2501 25. Narrowed the window for a race in the daemon that could cause it to ignore
2502 SIGCHLD signals. This is not a major problem, because they are used only to
2503 wake it up if nothing else does.
2505 26. A malformed maildirsize file could cause Exim to calculate negative values
2506 for the mailbox size or file count. Odd effects could occur as a result.
2507 The maildirsize information is now recalculated if the size or filecount
2510 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2511 support for a long time. Removed HAVE_SYS_VFS_H.
2513 28. Installed the latest version of exipick from John Jetmore.
2515 29. In an address list, if the pattern was not a regular expression, an empty
2516 subject address (from a bounce message) matched only if the pattern was an
2517 empty string. Non-empty patterns were not even tested. This was the wrong
2518 because it is perfectly reasonable to use an empty address as part of a
2519 database query. An empty address is now tested by patterns that are
2520 lookups. However, all the other forms of pattern expect the subject to
2521 contain a local part and a domain, and therefore, for them, an empty
2522 address still always fails if the pattern is not itself empty.
2524 30. Exim went into a mad DNS loop when attempting to do a callout where the
2525 host was specified on an smtp transport, and looking it up yielded more
2526 than one IP address.
2528 31. Re-factored the code for checking spool and log partition space into a
2529 function that finds that data and another that does the check. The former
2530 is then used to implement four new variables: $spool_space, $log_space,
2531 $spool_inodes, and $log_inodes.
2533 32. The RFC2047 encoding function was originally intended for short strings
2534 such as real names; it was not keeping to the 75-character limit for
2535 encoded words that the RFC imposes. It now respects the limit, and
2536 generates multiple encoded words if necessary. To be on the safe side, I
2537 have increased the buffer size for the ${rfc2047: expansion operator from
2540 33. It is now permitted to omit both strings after an "if" condition; if the
2541 condition is true, the result is "true". As before, when the second string
2542 is omitted, a false condition yields an empty string. This makes it less
2543 cumbersome to write custom ACL and router conditions.
2545 34. Failure to deliver a bounce message always caused it to be frozen, even if
2546 there was an errors_to setting on the router. The errors_to setting is now
2549 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2550 canonical form (fully expanded) before being placed in
2551 $sender_host_address.
2553 36. The table in the code that translates DNS record types into text (T_A to
2554 "A" for instance) was missing entries for NS and CNAME. It is just possible
2555 that this could have caused confusion if both these types were looked up
2556 for the same domain, because the text type is used as part of Exim's
2557 per-process caching. But the chance of anyone hitting this buglet seems
2560 37. The dnsdb lookup has been extended in a number of ways.
2562 (1) There is a new type, "zns", which walks up the domain tree until it
2563 finds some nameserver records. It should be used with care.
2565 (2) There is a new type, "mxh", which is like "mx" except that it returns
2566 just the host names, not the priorities.
2568 (3) It is now possible to give a list of domains (or IP addresses) to be
2569 looked up. The behaviour when one of the lookups defers can be
2570 controlled by a keyword.
2572 (4) It is now possible to specify the separator character for use when
2573 multiple records are returned.
2575 38. The dnslists ACL condition has been extended: it is now possible to supply
2576 a list of IP addresses and/or domains to be looked up in a particular DNS
2579 39. Added log_selector=+queue_time_overall.
2581 40. When running the queue in the test harness, wait just a tad after forking a
2582 delivery process, to get repeatability of debugging output.
2584 41. Include certificate and key file names in error message when GnuTLS fails
2585 to set them up, because the GnuTLS error message doesn't include the name
2586 of the failing file when there is a problem reading it.
2588 42. Allow both -bf and -bF in the same test run.
2590 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2592 44. The "Exiscan patch" is now merged into the mainline Exim source.
2594 45. Sometimes the final signoff response after QUIT could fail to get
2595 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2596 < 0 before doing a fflush(). This bug looks as though it goes back to the
2597 introduction of TLS in release 3.20, but "sometimes" must have been rare
2598 because the tests only now provoked it.
2600 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2601 (this can affect the format of dates).
2603 47. exim_tidydb, when checking for the continued existence of a message for
2604 which it has found a message-specific retry record, was not finding
2605 messages that were in split spool directories. Consequently, it was
2606 deleting retry records that should have stayed in existence.
2608 48. Steve fixed some bugs in eximstats.
2610 49. The SPA authentication driver was not abandoning authentication and moving
2611 on to the next authenticator when an expansion was forced to fail,
2612 contradicting the general specification for all authenticators. Instead it
2613 was generating a temporary error. It now behaves as specified.
2615 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2616 (the order specifies the preference for clients). The order is now AES256,
2617 AES128, 3DES, ARCFOUR128.
2619 51. Small patch to Sieve code - explicitly set From: when generating an
2622 52. Exim crashed if a remote delivery caused a very long error message to be
2623 recorded - for instance if somebody sent an entire SpamAssassin report back
2624 as a large number of 550 error lines. This bug was coincidentally fixed by
2625 increasing the size of one of Exim's internal buffers (big_buffer) that
2626 happened as part of the Exiscan merge. However, to be on the safe side, I
2627 have made the code more robust (and fixed the comments that describe what
2630 53. Now that there can be additional text after "Completed" in log lines (if
2631 the queue_time_overall log selector is set), a one-byte patch to exigrep
2632 was needed to allow it to recognize "Completed" as not the last thing in
2635 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2636 patch that reportedly fixes this has been added. I am not expert enough to
2637 create a test for it. This is what the patch creator wrote:
2639 "I found a little strange behaviour of ldap code when working with
2640 Windows 2003 AD Domain, where users was placed in more than one
2641 Organization Units. When I tried to give exim partial DN, the exit code
2642 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2643 But simultaneously result of request was absolutely normal ldap result,
2644 so I produce this patch..."
2646 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2647 REFERENCE, so I have modified the code to exclude the patch when that macro
2650 55. Some experimental protocols are using DNS PTR records for new purposes. The
2651 keys for these records are domain names, not reversed IP addresses. The
2652 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2653 leaves it alone. Component reversal etc. now happens only for IP addresses.
2656 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2658 57. Double the size of the debug message buffer (to 2048) so that more of very
2659 long debug lines gets shown.
2661 58. The exicyclog utility now does better if the number of log files to keep
2662 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2664 59. Two changes related to the smtp_active_hostname option:
2666 (1) $smtp_active_hostname is now available as a variable.
2667 (2) The default for smtp_banner uses $smtp_active_hostname instead
2668 of $primary_hostname.
2670 60. The host_aton() function is supposed to be passed a string that is known
2671 to be a valid IP address. However, in the case of IPv6 addresses, it was
2672 not checking this. This is a hostage to fortune. Exim now panics and dies
2673 if the condition is not met. A case was found where this could be provoked
2674 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2675 components; fortuitously, this particular loophole had already been fixed
2676 by change 4.50/55 above.
2678 If there are any other similar loopholes, the new check in host_aton()
2679 itself should stop them being exploited. The report I received stated that
2680 data on the command line could provoke the exploit when Exim was running as
2681 exim, but did not say which command line option was involved. All I could
2682 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2683 running as the user.
2686 61. There was a buffer overflow vulnerability in the SPA authentication code
2687 (which came originally from the Samba project). I have added a test to the
2688 spa_base64_to_bits() function which I hope fixes it.
2691 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2692 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2694 63. The daemon start-up calls getloadavg() while still root for those OS that
2695 need the first call to be done as root, but it missed one case: when
2696 deliver_queue_load_max is set with deliver_drop_privilege. This is
2697 necessary for the benefit of the queue runner, because there is no re-exec
2698 when deliver_drop_privilege is set.
2700 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2701 This has been fixed.
2703 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2704 was in use, was not putting the data itself into the right store pool;
2705 consequently, it could be overwritten for a subsequent message in the same
2706 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2709 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2711 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2712 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2713 the function were treating the return as a boolean value, which happened to
2714 work because 0=false and not-0=true, but is not correct code.
2716 68. The host_aton() function was not handling scoped IPv6 addresses (those
2717 with, for example, "%eth0" on the end) correctly.
2719 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2720 negated items (that is, ~something) in unsigned ints. Some compilers
2721 apparently mutter when there is no cast.
2723 70. If an address verification called from an ACL failed, and did not produce a
2724 user-specific message (i.e. there was only a "system" message), nothing was
2725 put in $acl_verify_message. In this situation, it now puts the system
2728 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2729 session; change 4.31/43 added the unwanted input to the log line - except
2730 that it did not do this in the start of session case. It now does.
2732 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2733 This is wrong and can cause the other end to generate a synchronization
2734 error if it is another Exim or anything else that does the synchronization
2735 check. A QUIT command is no longer sent after a timeout.
2737 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2738 during host lookups.
2740 74. The defer_ok option of callout verification was not working if it was used
2741 when verifying addresses in header lines, that is, for this case:
2743 verify = header_sender/callout=defer_ok
2745 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2746 those file descriptors could be used for SMTP connections. If anything
2747 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2748 could be sent to the SMTP client, causing chaos. The daemon now opens
2749 stdin, stdout, and stderr to /dev/null when it puts itself into the
2752 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2753 log by default. The user can override this with suitable Perl magic.
2755 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2756 the log message when discard triggers, was not working for the DATA ACL or
2757 for the non-SMTP ACL.
2759 78. Error message wording change in sieve.c.
2761 79. If smtp_accept_max_per_host was set, the number of connections could be
2762 restricted to fewer than expected, because the daemon was trying to set up
2763 a new connection before checking whether the processes handling previous
2764 connections had finished. The check for completed processes is now done
2765 earlier. On busy systems, this bug wouldn't be noticed because something
2766 else would have woken the daemon, and it would have reaped the completed
2769 80. If a message was submitted locally by a user whose login name contained one
2770 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2771 It caused a spool format error. I have fixed the spool reading code. A
2772 related problem was that the "from" clause in the Received: line became
2773 illegal because of the space(s). It is now covered by ${quote_local_part.
2775 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2778 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2780 83. Patch from Sieve maintainer to fix unterminated string problem in
2781 "vacation" handling.
2783 84. Some minor changes to the Linux configuration files to help with other
2784 OS variants using glibc.
2786 85. One more patch for Sieve to update vacation handling to latest spec.
2789 ----------------------------------------------------
2790 See the note above about the 4.44 and 4.50 releases.
2791 ----------------------------------------------------
2797 1. Change 4.43/35 introduced a bug that caused file counts to be
2798 incorrectly computed when quota_filecount was set in an appendfile
2801 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2802 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2805 3. Give more explanation in the error message when the command for a transport
2806 filter fails to execute.
2808 4. There are several places where Exim runs a non-Exim command in a
2809 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2810 was being done only for the command run by the queryprogram router. It is
2811 now done for all such subprocesses. The other cases are: ${run, transport
2812 filters, and the commands run by the lmtp and pipe transports.
2814 5. Some older OS have a limit of 256 on the maximum number of file
2815 descriptors. Exim was using setrlimit() to set 1000 as a large value
2816 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2817 systems. I've change it so that if it can't get 1000, it tries for 256.
2819 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2820 was an oversight, and furthermore, ever since the addition of extra
2821 controls (e.g. 4.43/32), the checks on when to allow different forms of
2822 "control" were broken. There should now be diagnostics for all cases when a
2823 control that does not make sense is encountered.
2825 7. $recipients is now available in the predata ACL (oversight).
2827 8. Tidy the search cache before the fork to do a delivery from a message
2828 received from the command line. Otherwise the child will trigger a lookup
2829 failure and thereby defer the delivery if it tries to use (for example) a
2830 cached ldap connection that the parent has called unbind on.
2832 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2833 of $address_data from the recipient verification was clobbered by the
2834 sender verification.
2836 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2837 was its contents. (It was OK if the option was not defined at all.)
2839 11. A "Completed" log line is now written for messages that are removed from
2840 the spool by the -Mrm option.
2842 12. $host_address is now set to the target address during the checking of
2843 ignore_target_hosts.
2845 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2846 being passed; this would have caused $sender_host_name to have been used if
2847 matching the list had actually called for a host name (not very likely,
2848 since this list is usually IP addresses). A host name is now passed as
2851 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2852 code that specifies a non-restarting handler (typically sigaction() in
2853 modern systems) in an attempt to fix a rare and obscure crash bug.
2855 15. Narrowed the window for a race in the daemon that could cause it to ignore
2856 SIGCHLD signals. This is not a major problem, because they are used only to
2857 wake it up if nothing else does.
2859 16. A malformed maildirsize file could cause Exim to calculate negative values
2860 for the mailbox size or file count. Odd effects could occur as a result.
2861 The maildirsize information is now recalculated if the size or filecount
2864 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2865 support for a long time. Removed HAVE_SYS_VFS_H.
2867 18. Updated exipick to current release from John Jetmore.
2869 19. Allow an empty sender to be matched against a lookup in an address list.
2870 Previously the only cases considered were a regular expression, or an
2873 20. Exim went into a mad DNS lookup loop when doing a callout where the
2874 host was specified on the transport, if the DNS lookup yielded more than
2877 21. The RFC2047 encoding function was originally intended for short strings
2878 such as real names; it was not keeping to the 75-character limit for
2879 encoded words that the RFC imposes. It now respects the limit, and
2880 generates multiple encoded words if necessary. To be on the safe side, I
2881 have increased the buffer size for the ${rfc2047: expansion operator from
2884 22. Failure to deliver a bounce message always caused it to be frozen, even if
2885 there was an errors_to setting on the router. The errors_to setting is now
2888 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2889 canonical form (fully expanded) before being placed in
2890 $sender_host_address.
2892 24. Updated eximstats to version 1.33
2894 25. Include certificate and key file names in error message when GnuTLS fails
2895 to set them up, because the GnuTLS error message doesn't include the name
2896 of the failing file when there is a problem reading it.
2898 26. Expand error message when OpenSSL has problems setting up cert/key files.
2901 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2902 (this can affect the format of dates).
2904 28. exim_tidydb, when checking for the continued existence of a message for
2905 which it has found a message-specific retry record, was not finding
2906 messages that were in split spool directories. Consequently, it was
2907 deleting retry records that should have stayed in existence.
2909 29. eximstats updated to version 1.35
2910 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2911 1.35 - bugfix such that pie charts by volume are generated correctly
2913 30. The SPA authentication driver was not abandoning authentication and moving
2914 on to the next authenticator when an expansion was forced to fail,
2915 contradicting the general specification for all authenticators. Instead it
2916 was generating a temporary error. It now behaves as specified.
2918 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2919 (the order specifies the preference for clients). The order is now AES256,
2920 AES128, 3DES, ARCFOUR128.
2922 31. Small patch to Sieve code - explicitly set From: when generating an
2925 32. Exim crashed if a remote delivery caused a very long error message to be
2926 recorded - for instance if somebody sent an entire SpamAssassin report back
2927 as a large number of 550 error lines. This bug was coincidentally fixed by
2928 increasing the size of one of Exim's internal buffers (big_buffer) that
2929 happened as part of the Exiscan merge. However, to be on the safe side, I
2930 have made the code more robust (and fixed the comments that describe what
2933 33. Some experimental protocols are using DNS PTR records for new purposes. The
2934 keys for these records are domain names, not reversed IP addresses. The
2935 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2936 leaves it alone. Component reversal etc. now happens only for IP addresses.
2939 34. The host_aton() function is supposed to be passed a string that is known
2940 to be a valid IP address. However, in the case of IPv6 addresses, it was
2941 not checking this. This is a hostage to fortune. Exim now panics and dies
2942 if the condition is not met. A case was found where this could be provoked
2943 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2944 components; fortuitously, this particular loophole had already been fixed
2945 by change 4.50/55 or 4.44/33 above.
2947 If there are any other similar loopholes, the new check in host_aton()
2948 itself should stop them being exploited. The report I received stated that
2949 data on the command line could provoke the exploit when Exim was running as
2950 exim, but did not say which command line option was involved. All I could
2951 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2952 running as the user.
2955 35. There was a buffer overflow vulnerability in the SPA authentication code
2956 (which came originally from the Samba project). I have added a test to the
2957 spa_base64_to_bits() function which I hope fixes it.
2960 36. The daemon start-up calls getloadavg() while still root for those OS that
2961 need the first call to be done as root, but it missed one case: when
2962 deliver_queue_load_max is set with deliver_drop_privilege. This is
2963 necessary for the benefit of the queue runner, because there is no re-exec
2964 when deliver_drop_privilege is set.
2966 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2967 was in use, was not putting the data itself into the right store pool;
2968 consequently, it could be overwritten for a subsequent message in the same
2969 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2972 38. Sometimes the final signoff response after QUIT could fail to get
2973 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2974 < 0 before doing a fflush(). This bug looks as though it goes back to the
2975 introduction of TLS in release 3.20, but "sometimes" must have been rare
2976 because the tests only now provoked it.
2982 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2983 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2984 Now it returns an integer. A number of calls were still expecting a T/F
2985 return. Fortuitously, in all cases, the tests worked in OK situations,
2986 which is the norm. However, things would have gone wrong on any write
2987 failures on the smtp file descriptor. This function is used when sending
2988 messages over SMTP and also when doing verify callouts.
2990 2. When Exim is called to do synchronous delivery of a locally submitted
2991 message (the -odf or -odi options), it no longer closes stderr before doing
2994 3. Implemented the mua_wrapper option.
2996 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2998 5. Implemented the functions header_remove(), header_testname(),
2999 header_add_at_position(), and receive_remove_recipient(), and exported them
3002 6. If an ACL "warn" statement specified the addition of headers, Exim already
3003 inserted X-ACL-Warn: at the start if there was no header name. However, it
3004 was not making this test for the second and subsequent header lines if
3005 there were newlines in the string. This meant that an invalid header could
3006 be inserted if Exim was badly configured.
3008 7. Allow an ACL "warn" statement to add header lines at the start or after all
3009 the Received: headers, as well as at the end.
3011 8. Added the rcpt_4xx retry error code.
3013 9. Added postmaster_mailfrom=xxx to callout verification option.
3015 10. Added mailfrom=xxxx to the callout verification option, for verify=
3018 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3019 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3021 12. Inserted some casts to stop certain compilers warning when using pointer
3022 differences as field lengths or precisions in printf-type calls (mostly
3023 affecting debugging statements).
3025 13. Added optional readline() support for -be (dynamically loaded).
3027 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3028 same clock tick as a message's arrival, so that its received time was the
3029 same as the "first fail" time on the retry record, and that message
3030 remained on the queue past the ultimate address timeout, every queue runner
3031 would try a delivery (because it was past the ultimate address timeout) but
3032 after another failure, the ultimate address timeout, which should have then
3033 bounced the address, did not kick in. This was a "< instead of <=" error;
3034 in most cases the first failure would have been in the next clock tick
3035 after the received time, and all would be well.
3037 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3038 being recognized when the domain list was tested by the match_domain
3039 condition in an expansion string.
3041 16. Added the ${str2b64: operator.
3043 17. Exim was always calling setrlimit() to set a large limit for the number of
3044 processes, without checking whether the existing limit was already
3045 adequate. (It did check for the limit on file descriptors.) Furthermore,
3046 errors from getrlimit() and setrlimit() were being ignored. Now they are
3047 logged to the main and panic logs, but Exim does carry on, to try to do its
3048 job under whatever limits there are.
3050 18. Imported PCRE 5.0.
3052 19. Trivial typo in log message " temporarily refused connection" (the leading
3055 20. If the log selector return_path_on_delivery was set and an address was
3056 redirected to /dev/null, the delivery process crashed because it assumed
3057 that a return path would always be set for a "successful" delivery. In this
3058 case, the whole delivery is bypassed as an optimization, and therefore no
3061 21. Internal re-arrangement: the function for sending a challenge and reading
3062 a response while authentication was assuming a zero-terminated challenge
3063 string. It's now changed to take a pointer and a length, to allow for
3064 binary data in such strings.
3066 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3068 23. Exim was not respecting finduser_retries when seeking the login of the
3069 uid under which it was called; it was always trying 10 times. (The default
3070 setting of finduser_retries is zero.) Also, it was sleeping after the final
3071 failure, which is pointless.
3073 24. Implemented tls_on_connect_ports.
3075 25. Implemented acl_smtp_predata.
3077 26. If the domain in control=submission is set empty, Exim assumes that the
3078 authenticated id is a complete email address when it generates From: or
3079 Sender: header lines.
3081 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3082 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3083 chown and chgrp in /bin and hostname in /usr/bin.
3085 28. Exim was keeping the "process log" file open after each use, just as it
3086 does for the main log. This opens the possibility of it remaining open for
3087 long periods when the USR1 signal hits a daemon. Occasional processlog
3088 errors were reported, that could have been caused by this. Anyway, it seems
3089 much more sensible not to leave this file open at all, so that is what now
3092 29. The long-running daemon process does not normally write to the log once it
3093 has entered its main loop, and it closes the log before doing so. This is
3094 so that log files can straightforwardly be renamed and moved. However,
3095 there are a couple of unusual error situations where the daemon does write
3096 log entries, and I had neglected to close the log afterwards.
3098 30. The text of an SMTP error response that was received during a remote
3099 delivery was being truncated at 512 bytes. This is too short for some of
3100 the long messages that one sometimes sees. I've increased the limit to
3103 31. It is now possible to make retry rules that apply only when a message has a
3104 specific sender, in particular, an empty sender.
3106 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3107 it possible to be selective about when SMTP synchronization is enforced.
3109 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3111 32. Implemented hosts_connection_nolog.
3113 33. Added an ACL for QUIT.
3115 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3118 35. Added mailbox_size and mailbox_filecount to appendfile.
3120 36. Added control = no_multiline_responses to ACLs.
3122 37. There was a bug in the logic of the code that waits for the clock to tick
3123 in the case where the clock went backwards by a substantial amount such
3124 that the microsecond fraction of "now" was more than the microsecond
3125 fraction of "then" (but the whole seconds number was less).
3127 38. Added support for the libradius Radius client library this is found on
3128 FreeBSD (previously only the radiusclient library was supported).
3134 1. When certain lookups returned multiple values in the form name=value, the
3135 quoting of the values was not always being done properly. Specifically:
3136 (a) If the value started with a double quote, but contained no whitespace,
3138 (b) If the value contained whitespace other than a space character (i.e.
3139 tabs or newlines or carriage returns) it was not quoted.
3140 This fix has been applied to the mysql and pgsql lookups by writing a
3141 separate quoting function and calling it from the lookup code. The fix
3142 should probably also be applied to nisplus, ibase and oracle lookups, but
3143 since I cannot test any of those, I have not disturbed their existing code.
3145 2. A hit in the callout cache for a specific address caused a log line with no
3146 reason for rejecting RCPT. Now it says "Previous (cached) callout
3147 verification failure".
3149 3. There was an off-by-one bug in the queryprogram router. An over-long
3150 return line was truncated at 256 instead of 255 characters, thereby
3151 overflowing its buffer with the terminating zero. As well as fixing this, I
3152 have increased the buffer size to 1024 (and made a note to document this).
3154 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3155 when Exim is waiting for an SMTP response from a remote server, Exim
3156 restarts its select() call on the socket, thereby resetting its timeout.
3157 This is not a problem when such interrupts are rare. Somebody set up a cron
3158 job to run exiwhat every 2 minutes, which is less than the normal select()
3159 timeout (5 or 10 minutes). This meant that the select() timeout never
3160 kicked in because it was always reset. I have fixed this by comparing the
3161 time when an interrupt arrives with the time at the start of the first call
3162 to select(). If more time than the timeout has elapsed, the interrupt is
3163 treated as a timeout.
3165 5. Some internal re-factoring in preparation for the addition of Sieve
3166 extensions (by MH). In particular, the "personal" test is moved to a
3167 separate function, and given an option for scanning Cc: and Bcc: (which is
3168 not set for Exim filters).
3170 6. When Exim created an email address using the login of the caller as the
3171 local part (e.g. when creating a From: or Sender: header line), it was not
3172 quoting the local part when it contained special characters such as @.
3174 7. Installed new OpenBSD configuration files.
3176 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3177 try to make them clearer.
3179 9. Callout options, other than the timeout value, were being ignored when
3180 verifying sender addresses in header lines. For example, when using
3182 verify = header_sender/callout=no_cache
3184 the cache was (incorrectly) being used.
3186 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3187 only the Cygwin environment.
3189 11. When return_path_on_delivery was set as a log selector, if different remote
3190 addresses in the same message used different return paths and parallel
3191 remote delivery occurred, the wrong values would sometimes be logged.
3192 (Whenever a remote delivery process finished, the return path value from
3193 the most recently started remote delivery process was logged.)
3195 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3196 lines when AUTH and/or TLS are in use. This is the "received protocol"
3197 field. Exim used to use "asmtp" for authenticated SMTP, without any
3198 indication (in the protocol name) for TLS use. Now it follows the RFC and
3199 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3200 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3201 names appear in log lines as well as in Received: header lines.
3203 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3204 extensions, and comparison tests, and to fix some bugs.
3206 14. Changes to the "personal" filter test:
3208 (1) The test was buggy in that it was just doing the equivalent of
3209 "contains" tests on header lines. For example, if a user's address was
3210 anne@some.where, the "personal" test would incorrectly be true for
3212 To: susanne@some.where
3214 This test is now done by extracting each address from the header in turn,
3215 and checking the entire address. Other tests that are part of "personal"
3216 are now done using regular expressions (for example, to check local parts
3217 of addresses in From: header lines).
3219 (2) The list of non-personal local parts in From: addresses has been
3220 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3221 taken from the Sieve specification recommendations.
3223 (3) If the message contains any header line starting with "List-" it is
3224 treated as non-personal.
3226 (4) The test for "circular" in the Subject: header line has been removed
3227 because it now seems ill-conceived.
3229 15. Minor typos in src/EDITME comments corrected.
3231 16. Installed latest exipick from John Jetmore.
3233 17. If headers_add on a router specified a text string that was too long for
3234 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3235 of string_sprintf() is now avoided.
3237 18. $message_body_size was not set (it was always zero) when running the DATA
3238 ACL and the local_scan() function.
3240 19. For the "mail" command in an Exim filter, no default was being set for
3241 the once_repeat time, causing a random time value to be used if "once" was
3242 specified. (If the value happened to be <= 0, no repeat happened.) The
3243 default is now 0s, meaning "never repeat". The "vacation" command was OK
3244 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3245 (I found it when inspecting the code).
3247 20. There is now an overall timeout for performing a callout verification. It
3248 defaults to 4 times the callout timeout, which applies to individual SMTP
3249 commands during the callout. The overall timeout applies when there is more
3250 than one host that can be tried. The timeout is checked before trying the
3251 next host. This prevents very long delays if there are a large number of
3252 hosts and all are timing out (e.g. when the network connections are timing
3253 out). The value of the overall timeout can be changed by specifying an
3254 additional sub-option for "callout", called "maxwait". For example:
3256 verify = sender/callout=5s,maxwait=20s
3258 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3259 to the end before writing, but this should make it even safer).
3261 22. Exim was forgetting that it had advertised PIPELINING for the second and
3262 subsequent messages on an SMTP connection. It was also not resetting its
3263 memory on STARTTLS and an internal HELO.
3265 23. When Exim logs an SMTP synchronization error within a session, it now
3266 records whether PIPELINING has been advertised or not.
3268 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3269 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3270 rather than long int.
3272 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3274 26. Added the never_mail option to autoreply.
3280 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3281 crash if the getsockname() call failed; this can happen if a connection is
3282 closed very soon after it is established. The problem was simply in the
3283 order in which certain operations were done, causing Exim to try to write
3284 to the SMTP stream before it had set up the file descriptor. The bug has
3285 been fixed by making things happen in the correct order.
3291 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3292 before the connection was closed, thus losing the rejection response.
3294 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3295 some early Solaris releases, but causes trouble in current releases where
3296 socklen_t is defined.
3298 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3301 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3304 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3305 that releasing the top of it at the end releases what was used for sub-
3306 expansions (unless the block got too big). However, discard this block if
3307 the first thing is a variable or header, so that we can use its block when
3308 it is dynamic (useful for very large $message_headers, for example).
3310 6. Lookups now cache *every* query, not just the most recent. A new, separate
3311 store pool is used for this. It can be recovered when all lookup caches are
3312 flushed. Lookups now release memory at the end of their result strings.
3313 This has involved some general refactoring of the lookup sources.
3315 7. Some code has been added to the store_xxx() functions to reduce the amount
3316 of flapping under certain conditions.
3318 8. log_incoming_interface used to affect only the <= reception log lines. Now
3319 it causes the local interface and port to be added to several more SMTP log
3320 lines, for example "SMTP connection from", and rejection lines.
3322 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3324 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3326 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3327 could be overwritten at the end of the current message (or the start of a
3328 new message if it was set in a HELO ACL). The value is now preserved for
3329 the duration of the SMTP connection.
3331 12. If a transport had a headers_rewrite setting, and a matching header line
3332 contained an unqualified address, that address was qualified, even if it
3333 did not match any rewriting rules. The underlying bug was that the values
3334 of the flags that permit the existence of unqualified sender and recipient
3335 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3336 non-local messages, and by -bnq for local messages) were not being
3337 preserved with the message after it was received.
3339 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3340 "next input=" as part of the text comprising the host identity instead of
3341 the correct text. The code was using the same buffer for two different
3342 strings. However, depending on which order the printing function evaluated
3343 its arguments, the bug did not always show up. Under Linux, for example, my
3344 test suite worked just fine.
3346 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3347 doesn't work with some older versions of Perl. It has been changed to "my",
3348 which in any case is probably the better facility to use.
3350 15. A really picky compiler found some instances of statements for creating
3351 error messages that either had too many or two few arguments for the format
3354 16. The size of the buffer for calls to the DNS resolver has been increased
3355 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3356 for addresses that have a lot of PTR records. This alleviates a problem; it
3357 does not fully solve it.
3359 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3360 buffer now truncates the list and logs the incident, which is the same
3361 action as happens when Exim is looking up a host name and its aliases.
3362 Previously in this situation something unpredictable would happen;
3363 sometimes it was "internal error: store_reset failed".
3365 18. If a server dropped the connection unexpectedly when an Exim client was
3366 using GnuTLS and trying to read a response, the client delivery process
3367 crashed while trying to generate an error log message.
3369 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3370 string, for example:
3372 warn message = H1: something\nH2: something
3374 the text was added as a single header line from Exim's point of view
3375 though it ended up OK in the delivered message. However, searching for the
3376 second and subsequent header lines using $h_h2: did not work. This has been
3377 fixed. Similarly, if a system filter added multiple headers in this way,
3378 the routers could not see them.
3380 20. Expanded the error message when iplsearch is called with an invalid key to
3381 suggest using net-iplsearch in a host list.
3383 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3384 ACLs are no longer actually imposed (and a message to that effect is
3387 22. If a "gecos" field in a passwd entry contained escaped characters, in
3388 particular, if it contained a \" sequence, Exim got it wrong when building
3389 a From: or a Sender: header from that name. A second bug also caused
3390 incorrect handling when an unquoted " was present following a character
3391 that needed quoting.
3393 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3394 was not being matched caselessly.
3396 24. Arranged for all hyphens in the exim.8 source to be escaped with
3399 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3400 properly. Recipient callout cache records were still being keyed to include
3401 the sender, even when use_sender was set false. This led to far more
3402 callouts that were necessary. The sender is no longer included in the key
3403 when use_sender is false.
3405 26. Added "control = submission" modifier to ACLs.
3407 27. Added the ${base62d: operator to decode base 62 numbers.
3409 28. dnsdb lookups can now access SRV records.
3411 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3412 the configuration file.
3414 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3415 (-v) mode. This makes the output for a verbose queue run more intelligible.
3417 31. Added a use_postmaster feature to recipient callouts.
3419 32. Added the $body_zerocount variable, containing the number of binary zero
3420 bytes in the message body.
3422 33. The time of last modification of the "new" subdirectory is now used as the
3423 "mailbox time last read" when there is a quota error for a maildir
3426 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3428 35. Added +ignore_unknown as a special item in host lists.
3430 36. Code for decoding IPv6 addresses in host lists is now included, even if
3431 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3432 address was recognized as an IP address, but was then not correctly decoded
3433 into binary, causing unexpected and incorrect effects when compared with
3440 1. Very minor rewording of debugging text in manualroute to say "list of
3441 hosts" instead of "hostlist".
3443 2. If verify=header_syntax was set, and a header line with an unqualified
3444 address (no domain) and a large number of spaces between the end of the
3445 name and the colon was received, the reception process suffered a buffer
3446 overflow, and (when I tested it) crashed. This was caused by some obsolete
3447 code that should have been removed. The fix is to remove it!
3449 3. When running in the test harness, delay a bit after writing a bounce
3450 message to get a bit more predictability in the log output.
3452 4. Added a call to search_tidyup() just before forking a reception process. In
3453 theory, someone could use a lookup in the expansion of smtp_accept_max_
3454 per_host which, without the tidyup, could leave open a database connection.
3456 5. Added the variables $recipient_data and $sender_data which get set from a
3457 lookup success in an ACL "recipients" or "senders" condition, or a router
3458 "senders" option, similar to $domain_data and $local_part_data.
3460 6. Moved the writing of debug_print from before to after the "senders" test
3463 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3464 problems for message scanning, either using a data ACL, or using
3465 local_scan() because the Received: header was not generated till after they
3466 were called (in order to set the time as the time of reception completion).
3467 I have revised the way this works. The header is now generated after the
3468 body is received, but before the ACL or local_scan() are called. After they
3469 are run, the timestamp in the header is updated.
3475 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3476 before starting a queue runner without re-exec. This happened only when
3477 deliver_drop_privilege was set or when the Exim user was set to root. The
3478 effect of the bug was that timeouts during subsequent deliveries caused
3479 crashes instead of being properly handled. The handler is now left at its
3480 default (and expected) setting.
3482 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3483 message, again when deliver_drop_privilege is set or Exim is run as root.
3484 The bug described in (1) was not present in this case, but the tidying up
3485 of the other signals was missing. I have made the two cases consistent.
3487 3. The ignore_target_hosts setting on a manualroute router was being ignored
3488 for hosts that were looked up using the /MX notation.
3490 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3493 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3494 operated on the sender address. After changing the $sender_address to <>
3495 for the sender address verify, Exim was re-instated it as the original
3496 (before rewriting) address, but remembering that it had rewritten it, so it
3497 wasn't rewriting it again. This bug also had the effect of breaking the
3498 sender address verification caching when the sender address was rewritten.
3500 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3501 This has been changed so that if the ip literal address matches
3502 ignore_target_hosts, the router declines.
3504 7. Added expansion conditions match_domain, match_address, and match_local_
3505 part (NOT match_host).
3507 8. The placeholder for the Received: header didn't have a length field set.
3509 9. Added code to Exim itself and to exim_lock to test for a specific race
3510 condition that could lead to file corruption when using MBX delivery. The
3511 issue is with the lockfile that is created in /tmp. If this file is removed
3512 after a process has opened it but before that process has acquired a lock,
3513 there is the potential for a second process to recreate the file and also
3514 acquire a lock. This could lead to two Exim processes writing to the file
3515 at the same time. The added code performs the same test as UW imapd; it
3516 checks after acquiring the lock that its file descriptor still refers to
3517 the same named file.
3519 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3520 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3523 11. Added the smtp_active_hostname option. If used, this will typically be made
3524 to depend on the incoming interface address. Because $interface_address is
3525 not set up until the daemon has forked a reception process, error responses
3526 that can happen earlier (such as "too many connections") no longer contain
3529 12. If an expansion in a condition on a "warn" statement fails because a lookup
3530 defers, the "warn" statement is abandoned, and the next ACL statement is
3531 processed. Previously this caused the whole ACL to be aborted.
3533 13. Added the iplsearch lookup type.
3535 14. Added ident_timeout as a log selector.
3537 15. Added tls_certificate_verified as a log selector.
3539 16. Added a global option tls_require_ciphers (compare the smtp transport
3540 option of the same name). This controls incoming TLS connections.
3542 17. I finally figured out how to make tls_require_ciphers do a similar thing
3543 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3544 before starting the TLS session.
3546 18. Tabs are now shown as \t in -bP output.
3548 19. If the log selector return_path_on_delivery was set, Exim crashed when
3549 bouncing a message because it had too many Received: header lines.
3551 20. If two routers both had headers_remove settings, and the first one included
3552 a superfluous trailing colon, the final name in the first list and the
3553 first name in the second list were incorrectly joined into one item (with a
3554 colon in the middle).
3560 1. Added -C and -D options to the exinext utility, mainly to make it easier
3561 to include in the automated testing, but these could be helpful when
3562 multiple configurations are in use.
3564 2. The exinext utility was not formatting the output nicely when there was
3565 an alternate port involved in the retry record key, nor when there was a
3566 message id as well (for retries that were specific to a specific message
3567 and a specific host). It was also confused by IPv6 addresses, because of
3568 the additional colons they contain. I have fixed the IPv4 problem, and
3569 patched it up to do a reasonable job for IPv6.
3571 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3572 delivery, the log line now contains "pipelined" if PIPELINING was used.
3574 4. An SMTP transport process used to panic and die if the bind() call to set
3575 an explicit outgoing interface failed. This has been changed; it is now
3576 treated in the same way as a connect() failure.
3578 5. A reference to $sender_host_name in the part of a conditional expansion
3579 that was being skipped was still causing a DNS lookup. This no longer
3582 6. The def: expansion condition was not recognizing references to header lines
3583 that used bh_ and bheader_.
3585 7. Added the _cache feature to named lists.
3587 8. The code for checking quota_filecount in the appendfile transport was
3588 allowing one more file than it should have been.
3590 9. For compatibility with Sendmail, the command line option
3598 and sets the incoming protocol and host name (for trusted callers). The
3599 host name and its colon can be omitted when only the protocol is to be set.
3600 Note the Exim already has two private options, -pd and -ps, that refer to
3601 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3602 "s", but I don't think that's a major issue.
3604 10. A number of refactoring changes to the code, none of which should affect
3607 (a) The number of logging options was getting close to filling up the
3608 32-bit word that was used as a bit map. I have split them into two classes:
3609 those that are passed in the argument to log_write(), and those that are
3610 only ever tested independently outside of that function. These are now in
3611 separate 32-bit words, so there is plenty of room for expansion again.
3612 There is no change in the user interface or the logging behaviour.
3614 (b) When building, for example, log lines, the code previously used a
3615 macro that called string_cat() twice, in order to add two strings. This is
3616 not really sufficiently general. Furthermore, there was one instance where
3617 it was actually wrong because one of the argument was used twice, and in
3618 one call a function was used. (As it happened, calling the function twice
3619 did not affect the overall behaviour.) The macro has been replaced by a
3620 function that can join an arbitrary number of extra strings onto a growing
3623 (c) The code for expansion conditions now uses a table and a binary chop
3624 instead of a serial search (which was left over from when there were very
3625 few conditions). Also, it now recognizes conditions like "pam" even when
3626 the relevant support is not compiled in: a suitably worded error message is
3627 given if an attempt is made to use such a condition.
3629 11. Added ${time_interval:xxxxx}.
3631 12. A bug was causing one of the ddress fields not to be passed back correctly
3632 from remote delivery subprocesses. The field in question was not being
3633 subsequently used, so this caused to problems in practice.
3635 13. Added new log selectors queue_time and deliver_time.
3637 14. Might have fixed a bug in maildirsizefile handling that threw up
3638 "unexpected character" debug warnings, and recalculated the data
3639 unnecessarily. In any case, I expanded the warning message to give more
3642 15. Added the message "Restricted characters in address" to the statements in
3643 the default ACL that block characters like @ and % in local parts.
3645 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3646 Three changes have been made:
3648 (a) There was a serious bug; a negative response to MAIL caused the whole
3649 recipient domain to be cached as invalid, thereby blocking all messages
3650 to all local parts at the same domain, from all senders. This bug has
3651 been fixed. The domain is no longer cached after a negative response to
3652 MAIL if the sender used is not empty.
3654 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3657 (c) A new callout option, "use_sender" has been added for people who want
3658 the modified behaviour.
3664 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3667 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3668 indeed breaks things for older releases.
3670 3. Added additional logging to the case where there is a problem reading data
3671 from a filter that is running in a subprocess using a pipe, in order to
3672 try to track down a specific problem.
3674 4. Testing facility fudge: when running in the test harness and attempting
3675 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3676 getting "No route to host". Convert this to a timeout.
3678 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3681 6. Some OS don't have socklen_t but use size_t instead. This affects the
3682 fifth argument of getsockopt() amongst other things. This is now
3683 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3684 can be set for individual OS. I have set it for SunOS5, OSF1, and
3685 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3686 some earlier ones do not.
3688 7. Change 4.30/15 was not doing the test caselessly.
3690 8. The standard form for an IPv6 address literal was being rejected by address
3691 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3692 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3693 this, as well as the form without the "IPv6" on the front (but only when
3694 address literals are enabled, of course).
3696 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3698 10. Exim crashed if a message with an empty sender address specified by -f
3699 encountered a router with an errors_to setting. This could be provoked only
3700 by a command such as
3704 where an empty string was supplied; "<>" did not hit this bug.
3706 11. Installed PCRE release 4.5.
3708 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3709 remained set. It is now erased.
3711 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3712 times from message ids (which are base 36 rather than the normal 62).
3714 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3715 were being counted as actual protocol errors, and logged if the log
3716 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3717 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3718 and DATA following a set of rejected RCPTs do not count as protocol errors.
3719 In other words, Exim assumes they were pipelined, though this may not
3720 actually be the case. Of course, in all cases the client gets an
3721 appropriate error code.
3723 15. If a lookup fails in an ACL condition, a message about the failure may
3724 be available; it is used if testing the ACL cannot continue, because most
3725 such messages specify what the cause of the deferral is. However, some
3726 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3727 that caused an old message to be retained and used if a later statement
3728 caused a defer, replacing the real cause of the deferral.
3730 16. If an IP address had so many PTR records that the DNS lookup buffer
3731 was not large enough to hold them, Exim could crash while trying to process
3732 the truncated data. It now detects and logs this case.
3734 17. Further to 4.21/58, another change has been made: if (and only if) the
3735 first line of a message (the first header line) ends with CRLF, a bare LF
3736 in a subsequent header line has a space inserted after it, so as not to
3737 terminate the header.
3739 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3740 unnecessarily, used atoi() instead of strtol(), and didn't check the
3741 termination when getting file sizes from file names by regex.
3743 19. Completely re-implemented the support for maildirsize files, in the light
3744 of a number of problems with the previous contributed implementation
3745 (4.30/29). In particular:
3747 . If the quota is zero, the maildirsize file is maintained, but no quota is
3750 . If the maildir directory does not exist, it is created before any attempt
3751 to write a maildirsize file.
3753 . The quota value in the file is just a cache; if the quota is changed in
3754 the transport, the new value overrides.
3756 . A regular expression is available for excluding directories from the
3759 20. The autoreply transport checks the characters in options that define the
3760 message's headers; it allows continued headers, but it was checking with
3761 isspace() after an embedded newline instead of explicitly looking for a
3764 21. If all the "regular" hosts to which an address was routed had passed their
3765 expiry times, and had not reached their retry times, the address was
3766 bounced, even if fallback hosts were defined. Now Exim should go on to try
3769 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3770 equivalent code in the SMTP transport. Some hosts send humungous responses
3771 to HELO/EHLO, more than 1024 it seems.
3773 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3774 gives compiler warnings in some environments. I've now done it "properly",
3777 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3778 (because of problems with the built-in one) was declared to return uschar *
3779 instead of char *, causing compiler failure.
3781 25. Fixed a file descriptor leak when processing alias/forward files.
3783 26. Fixed a minor format string issue in dbfn.c.
3785 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3787 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3788 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3791 29. When a sender address is verified, it is cached, to save repeating the test
3792 when there is more than one recipient in a message. However, when the
3793 verification involves a callout, it is possible for different callout
3794 options to be set for different recipients. It is too complicated to keep
3795 track of this in the cache, so now Exim always runs a verification when a
3796 callout is required, relying on the callout cache for the optimization.
3797 The overhead is duplication of the address routing, but this should not be
3800 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3801 to be verified with callout=postmaster, and the main callout worked but the
3802 postmaster check failed, the verification correctly failed. However, if a
3803 subsequent RCPT command asked for sender verification *without* the
3804 postmaster check, incorrect caching caused this verification also to fail,
3807 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3808 it was not caching the DNS options (qualify_single, search_parents) that
3809 were used when the lookup failed. A subsequent lookup with different
3810 options therefore always gave the same answer, though there were cases
3811 where it should not have. (Example: a "domains = !$mx_any" option on a
3812 dnslookup router: the "domains" option is always processed without any
3813 widening, but the router might have qualify_single set.) Now Exim uses the
3814 cached value only when the same options are set.
3816 32. Added John Jetmore's "exipick" utility to the distribution.
3818 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3819 than a timeout (e.g. a certificate is required, and is not provided), an
3820 Exim server now closes the connection immediately. Previously it waited for
3821 the client to close - but if the client is SSL, it seems that they each
3822 wait for each other, leading to a delay before one of them times out.
3824 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3825 maintained 0.8.x compatibility because I don't think many are using it, and
3826 it is clearly obsolete.
3828 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3831 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3832 client certificate was expired. A simple patch fixes this, though I don't
3833 understand the full logic of why the verify callback is called multiple
3836 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3837 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3838 which causes problems with some clients (such as the Certicom SSL Plus
3839 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3840 disables the coutermeasure allowing Eudora to connect."
3842 38. Exim was not checking that a write() to a log file succeeded. This could
3843 lead to Bad Things if a log got too big, in particular if it hit a file
3844 size limit. Exim now panics and dies if it cannot write to a log file, just
3845 as it does if it cannot open a log file.
3847 39. Modified OS/Makefile-Linux so that it now contains
3849 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3851 The two -D definitions ensure that Exim is compiled with large file
3852 support, which makes it possible to handle log files that are bigger than
3855 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3856 instance) a domain was checked against a named list that involved a lookup,
3857 causing $domain_data to be set, then another domain was checked against the
3858 same list, then the first domain was re-checked, the value of $domain_data
3859 after the final check could be wrong. In particular, if the second check
3860 failed, it could be set empty. This bug probably also applied to
3863 41. The strip_trailing_dot option was not being applied to the address given
3864 with the -f command-line option.
3866 42. The code for reading a message's header from the spool was incrementing
3867 $received_count, but never initializing it. This meant that the value was
3868 incorrect (doubled) while delivering a message in the same process in which
3869 it was received. In the most common configuration of Exim, this never
3870 happens - a fresh exec is done - but it can happen when
3871 deliver_drop_privilege is set.
3873 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3874 it now includes up to 150 characters of the unexpected data in the log
3877 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3878 and building data strings. The size of both of these buffers was 10 000
3879 bytes - far larger than anybody would *ever* want, thought I. Needless to
3880 say, somebody hit the limit. I have increased the maximum line length to
3881 20 000 and the maximum data length of concatenated lines to 100 000. I have
3882 also fixed two bugs, because there was no checking on these buffers. Tsk,
3883 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3884 buffer is too small.
3886 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3887 lsearch lookups. Now it does.
3889 46. When parsing a route_list item in a manualroute router, a fixed-length
3890 buffer was used for the list of hosts. I made this 1024 bytes long,
3891 thinking that nobody would ever have a list of hosts that long. Wrong.
3892 Somebody had a whole pile of complicated expansion conditions, and the
3893 string was silently truncated, leading to an expansion error. It turns out
3894 that it is easier to change to an unlimited length (owing to other changes
3895 that have happened since this code was originally written) than to build
3896 structure for giving a limitation error. The length of the item that
3897 expands into the list of hosts is now unlimited.
3899 47. The lsearch lookup could not handle data where the length of text line was
3900 more than 4095 characters. Such lines were truncated, leading to shortened
3901 data being returned. It should now handle lines of any length.
3903 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3904 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3907 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3908 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3909 no longer happens in comment lines. A list of replacements is now placed
3910 at the head of all of the source files, except those whose only change is
3911 to replace PERL_COMMAND in the very first #! line.
3913 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3914 messages on the queue, with a bottom-up merge sort, using code contributed
3915 by Michael Haardt. This should make operations like -bp somewhat faster on
3916 large queues. It won't affect queue runners, except when queue_run_in_order
3919 51. Installed eximstats 1.31 in the distribution.
3921 52. Added support for SRV lookups to the dnslookup router.
3923 53. If an ACL referred to $message_body or $message_body_end, the value was not
3924 reset for any messages that followed in the same SMTP session.
3926 54. The store-handling optimization for building very long strings was not
3927 differentiating between the different store pools. I don't think this
3928 actually made any difference in practice, but I've tidied it.
3930 55. While running the routers to verify a sender address, $sender_address
3931 was still set to the sender address. This is wrong, because when routing to
3932 send a bounce to the sender, it would be empty. Therefore, I have changed
3933 it so that, while verifying a sender address, $sender_address is set to <>.
3934 (There is no change to what happens when verifying a recipient address.)
3936 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3937 target A or AAAA records (if not already returned) without resetting the
3938 qualify_single or search_parents options of the DNS resolver. These are
3939 inappropriate in this case because the targets of MX and SRV records must
3940 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3941 target that, when qualified, matched something in the local domain. These
3942 two options are now turned off when doing these lookups.
3944 57. It seems that at least some releases of Reiserfs (which does not have the
3945 concept of a fixed number of inodes) returns zero and not -1 for the
3946 number of available inodes. This interacted badly with check_spool_inodes,
3947 which assumed that -1 was the "no such thing" setting. What I have done is
3948 to check that the total number of inodes is greater than zero before doing
3949 the test of how many are available.
3951 58. When a "warn" ACL statement has a log_message modifier, the message is
3952 remembered, and not repeated. This is to avoid a lot of repetition when a
3953 message has many recipients that cause the same warning to be written.
3954 Howewer, Exim was preserving the list of already written lines for an
3955 entire SMTP session, which doesn't seem right. The memory is now reset if a
3956 new message is started.
3958 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3959 debugging output unless log_rewrite was also set.
3961 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3962 of (int)(handle) when we know that handle contains (void *)(-1).
3964 61. The Exim daemon panic-logs an error return when it closes the incoming
3965 connection. However "connection reset by peer" seems to be common, and
3966 isn't really an error worthy of noting specially, so that particular error
3969 62. When Exim is trying to find all the local interfaces, it used to panic and
3970 die if the ioctl to get the interface flags failed. However, it seems that
3971 on at least one OS (Solaris 9) it is possible to have an interface that is
3972 included in the list of interfaces, but for which you get a failure error
3973 for this call. This happens when the interface is not "plumbed" into a
3974 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3975 failure of the "get flags" call assumes that the interface is down.
3977 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3978 makes life easier for people who are doing arithmetic on fields extracted
3979 from dates, where you often get leading zeros that should not be
3980 interpreted as octal.
3982 64. Added qualify_domain to the redirect router, to override the global
3985 65. If a pathologically long header line contained very many addresses (the
3986 report of this problem mentioned 10 000) and each of them was rewritten,
3987 Exim could use up a very large amount of memory. (It kept on making new
3988 copies of the header line as it rewrote, and never released the old ones.)
3989 At the expense of a bit more processing, the header rewriting function has
3990 been changed so that it no longer eats memory in this way.
3992 66. The generation of the Received: header has been moved from the time that a
3993 message starts to be received, to the time that it finishes. The timestamp
3994 in the Received: header should now be very close to that of the <= log
3995 line. There are two side-effects of this change:
3997 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3998 logged header lines no longer include the local Received: line, because
3999 it has not yet been created. The same applies to a copy of the message
4000 that is returned to a non-SMTP sender when a message is rejected.
4002 (b) When a filter file is tested using -bf, no additional Received: header
4003 is added to the test message. After some thought, I decided that this
4006 This change does not affect the value of $received_for. It is still set
4007 after address rewriting, but before local_scan() is called.
4009 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4011 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4012 gave an unhelpful panic error message, and a defer error. I have managed to
4013 change this behaviour so that it now rejects any supplied certificate,
4014 which seems right, as the list of acceptable certificates is empty.
4016 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4017 gave an unhelpful defer error. I have not managed to make this reject any
4018 supplied certificates, but the error message it gives is "no certificate
4019 supplied", which is not helpful.
4021 70. exigrep's output now also includes lines that are not associated with any
4022 message, but which match the given pattern. Implemented by a patch from
4023 Martin Sluka, which also tidied up the Perl a bit.
4025 71. Recipient callout verification, like sender verification, was using <> in
4026 the MAIL FROM command. This isn't really the right thing, since the actual
4027 sender may affect whether the remote host accepts the recipient or not. I
4028 have changed it to use the actual sender in the callout; this means that
4029 the cache record is now keyed on a recipient/sender pair, not just the
4030 recipient address. There doesn't seem to be a real danger of callout loops,
4031 since a callout by the remote host to check the sender would use <>.
4032 [SEE ABOVE: changed after hitting problems.]
4034 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4035 temporary errors. However, in the case of such a code being given after
4036 the end of a data transmission (i.e. after ".") Exim was failing to write
4037 a retry record for the message. (Yes, there was some broken host that was
4038 actually sending 8xx at this point.)
4040 73. An unknown lookup type in a host list could cause Exim to panic-die when
4041 the list was checked. (An example that provoked this was putting <; in the
4042 middle of a list instead of at the start.) If this happened during a DATA
4043 ACL check, a -D file could be left lying around. This kind of configuration
4044 error no longer causes Exim to die; instead it causes a defer errror. The
4045 incident is still logged to the main and panic logs.
4047 74. Buglet left over from Exim 3 conversion. The message "too many messages
4048 in one connection" was written to the rejectlog but not the mainlog, except
4049 when address rewriting (yes!) was being logged.
4051 75. Added write_rejectlog option.
4053 76. When a system filter was run not as root (that is, when system_filter_user
4054 was set), the values of the $n variables were not being returned to the
4055 main process; thus, they were not subsequently available in the $sn
4058 77. Added +return_path_on_delivery log selector.
4060 78. A connection timeout was being treated differently from recipients deferred
4061 when testing hosts_max_try with a message that was older than the host's
4062 retry timeout. (The host should not be counted, thus allowing all hosts to
4063 be tried at least once before bouncing.) This may have been the cause of an
4064 occasionally reported bug whereby a message would remain on the queue
4065 longer than the retry timeout, but would be bounced if a delivery was
4066 forced. I say "may" because I never totally pinned down the problem;
4067 setting up timeout/retry tests is difficult. See also the next item.
4069 79. The ultimate address timeout was not being applied to errors that involved
4070 a combination of host plus message (for example, a timeout on a MAIL
4071 command). When an address resolved to a number of possible hosts, and they
4072 were not all tried for each delivery (e.g. because of hosts_max_try), a
4073 message could remain on the queue longer than the retry timeout.
4075 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4078 81. Fixed an obscure SMTP outgoing bug which required at least the following
4079 conditions: (a) there was another message waiting for the same server;
4080 (b) the server returned 5xx to all RCPT commands in the first message so
4081 that the message was not completed; (c) the server dropped the connection
4082 or gave a negative response to the RSET that Exim sends to abort the
4083 transaction. The observed case was a dropped connection after DATA that had
4084 been sent in pipelining mode. That is, the server had advertised PIPELINING
4085 but was not implementing it correctly. The effect of the bug was incorrect
4086 behaviour, such as trying another host, and this could lead to a crash.
4092 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4093 and daemon.c were passed as pointers to ints; they should have been
4094 pointers to socklen_t variables (which are typically unsigned ints).
4096 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4099 3. Fixed a really odd bug that affected only the testing scheme; patching a
4100 certain fixed string in the binary changed the value of another string that
4101 happened to be identical to the end of the original first string.
4103 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4104 name", it returns that address as the IP address. On some operating
4105 systems (e.g. Solaris), it also passes back the IP address string as the
4106 "host name". However, on others (e.g. Linux), it passes back an empty
4107 string. Exim wasn't checking for this, and was changing the host name to an
4108 empty string, assuming it had been canonicized.
4110 5. Although rare, it is permitted to have more than one PTR record for a given
4111 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4112 all the names associated with an address, because they do in Solaris.
4113 However, it seems that they do not in Linux for data that comes from the
4114 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4115 I found this out when I moved to a new Linux workstation and tried to run
4116 the Exim test suite.
4118 To get round this problem I have changed the code so that it now does its
4119 own call to the DNS to look up PTR records when searching for a host name.
4120 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4121 addresses that are only in /etc/hosts are still found.
4123 This behaviour is, however, controlled by an option called host_lookup_
4124 order, which defaults to "bydns:byaddr". If people want to use the other
4125 order, or indeed, just use one or the other means of lookup, they can
4126 specify it in this variable.
4128 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4129 some operating systems, this comes back from gethostbyaddr() as an empty
4130 string, and this is what Exim used to test for. However, it seems that in
4131 other systems, "." is yielded. Exim now tests for this case too.
4133 7. The values of check_spool_space and check_log_space are now held internally
4134 as a number of kilobytes instead of an absolute number of bytes. If a
4135 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4136 kilobyte. This means that much larger values can be stored.
4138 8. Exim monitor: an attempt to get the action menu when not actually pointing
4139 at a message produces an empty menu entitled "No message selected". This
4140 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4141 no entries in it ("Shell widget menu has zero width and/or height"). So I
4142 have added a single, blank menu entry in this case.
4144 9. Added ${quote_local_part.
4146 10. MIME decoding is now applied to the contents of Subject: header lines when
4149 11. Now that a reference to $sender_host_address automatically causes a reverse
4150 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4151 host lookup before query-style lookups in lists that might use this
4152 variable. This has therefore been abolished, and the "net-" prefix is no
4153 longer necessary for query-style lookups.
4155 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4156 have been a typo for LFLAGS, so it has been changed.
4158 13. The install script calls Exim with "-C /dev/null" in order to find the
4159 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4160 to be output. Howeve, since Exim outputs its version number before the
4161 error, it didn't break the script. It just looked ugly. I fixed this by
4162 always allowing "-C /dev/null" if the caller is root.
4164 14. Ignore overlarge ACL variable number when reading spool file - insurance
4165 against a later release with more variables having written the file.
4167 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4168 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4169 this, as well as the form without the "IPv6" on the front.
4171 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4172 OS/Makefile-Darwin file.
4174 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4175 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4177 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4178 variable. However, some OS use ints for inodes. I've added cast to long int
4179 to get rid of the compiler warning.
4181 19. I had forgotten to lock out "/../" in configuration file names when
4182 ALT_CONFIG_PREFIX was set.
4184 20. Routers used for verification do not need to specify transports. However,
4185 if such a router generated a host list, and callout was configured, Exim
4186 crashed, because it could not find a port number from the (non-existent)
4187 transport. It now assumes port 25 in this circumstance.
4189 21. Added the -t option to exigrep.
4191 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4192 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4193 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4194 them is set, LOOKUP_LSEARCH is forced.
4196 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4198 24. Added sender and host information to the "rejected by local_scan()" log
4199 line; previously there was no indication of these.
4201 25. Added .include_if_exists.
4203 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4204 Linux. It turns out that not all file systems support this. Apparently some
4205 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4206 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4207 not supported on the file descriptor, is now ignored when Exim is trying to
4208 sync a directory. This applies only to Linux.
4210 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4212 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4213 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4216 29. Added support for maildirsize files from supplied patch (modified a bit).
4218 30. The use of :fail: followed by an empty string could lead Exim to respond to
4219 sender verification failures with (e.g.):
4221 550 Verification failed for <xxx>
4222 550 Sender verify failed
4224 where the first response line was missing the '-' that indicates it is not
4225 the final line of the response.
4227 31. The loop for finding the name of the user that called Exim had a hardwired
4228 limit of 10; it now uses the value of finduser_retries, which is used for
4229 all other user lookups.
4231 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4234 33. Exim was neglecting to zero errno before one call of strtol() when
4235 expanding a string and expecting an integer value. On some systems this
4236 resulted in spurious "integer overflow" errors. Also, it was casting the
4237 result into an int without checking.
4239 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4240 did not work. The code looks as if it has *never* worked, though it appears
4241 to have been documented since at least releast 1.62. I have made it work.
4243 35. The "timeout_DNS" error in retry rules, also documented since at least
4244 1.62, also never worked. As it isn't clear exactly what this means, and
4245 clearly it isn't a major issue, I have abolished the feature by treating it
4246 as "timeout", and writing a warning to the main and panic logs.
4248 36. The display of retry rules for -brt wasn't always showing the error code
4251 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4252 timeout_connect_A, timeout_connect_MX.
4254 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4255 to the empty sender.
4257 39. The daemon was not analysing the content of -oX till after it had closed
4258 stderr and disconnected from the controlling terminal. This meant that any
4259 syntax errors were only noted on the panic log, and the return code from
4260 the command was 0. By re-arranging the code a little, I've made the
4261 decoding happen first, so such errors now appear on stderr, and the return
4262 code is 1. However, the actual setting up of the sockets still happens in
4263 the disconnected process, so errors there are still only recorded on the
4266 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4267 connections (as happens on some IP stacks) was logged at start up time as
4268 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4269 it from "IPv6 and IPv4", which means that two separate sockets are being
4272 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4273 says whether AF_INET or AF_INET6 was passed as an argument.
4275 42. Exiwhat output was messed up when time zones were included in log
4278 43. Exiwhat now gives more information about the daemon's listening ports,
4279 and whether -tls-on-connect was used.
4281 44. The "port" option of the smtp transport is now expanded.
4283 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4284 silently ignored. Now an error message is written to the main and panic
4287 46. There's a new ACL modifier called "logwrite" which writes to a log file
4288 as soon as it is encountered.
4290 47. Added $local_user_uid and $local_user_gid at routing time.
4292 48. Exim crashed when trying to verify a sender address that was being
4295 49. Exim was recognizing only a space character after ".include". It now also
4296 recognizes a tab character.
4298 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4299 extracting the relevant information from the specification. The man page no
4300 longer contains scrambled data for the -d option, and I've added a section
4301 at the front about calling Exim under different names.
4303 51. Added "extra_headers" argument to the "mail" command in filter files.
4305 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4308 53. Installed eximstats 1.29.
4310 54. Added transport_filter_timeout as a generic transport option.
4312 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4313 Cc: header lines. This was required by RFC 822, but it not required by RFC
4316 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4317 incoming messages that did not have them. Now it does so only if the
4318 message originates locally, that is, if there is no associated remote host
4319 address. When Resent- header lines are present, this applies to the Resent-
4320 lines rather than the non-Resent- lines.
4322 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4323 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4325 58. Messages for configuration errors now include the name of the main
4326 configuration file - useful now that there may be more than one file in a
4327 list (.included file names were always shown).
4329 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4330 for those rare installations that do not start the daemon as root or run it
4331 setuid root. I've cut out the call to initgroups() if the daemon is not
4334 60. The Exim user and group can now be bound into the binary as text strings
4335 that are looked up at the start of Exim's processing.
4337 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4339 62. Added $mailstore_basename variable.
4341 63. Installed patch to sieve.c from Michael Haardt.
4343 64. When Exim failed to open the panic log after failing to open the main log,
4344 the original message it was trying to log was written to stderr and debug
4345 output, but if they were not available (the usual case in production), it
4346 was lost. Now it is written to syslog before the two lines that record the
4347 failures to open the logs.
4349 65. Users' Exim filters run in subprocesses under the user's uid. It is
4350 possible for a "deliver" command or an alias in a "personal" command to
4351 provoke an address rewrite. If logging of address rewriting is configured,
4352 this fails because the process is not running as root or exim. There may be
4353 a better way of dealing with this, but for the moment (because 4.30 needs
4354 to be released), I have disabled address rewrite logging when running a
4355 filter in a non-root, non-exim process.
4361 1. The buildconfig auxiliary program wasn't quoting the value set for
4362 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4363 not defined. This bug was masked in 4.22 by the effect that was fixed in
4366 2. Some messages that were rejected after a message id was allocated were
4367 shown as "incomplete" by exigrep. It no longer does this for messages that
4368 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4370 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4371 have allow_domain_literals set, the ID did not get logged in the <= line.
4372 Domain literals are now always recognized in Message-ID: header lines.
4374 4. The first argument for a ${extract expansion item is the key name or field
4375 number. Leading and trailing spaces in this item were not being ignored,
4376 causing some misleading effects.
4378 5. When deliver_drop_privilege was set, single queue runner processes started
4379 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4380 same command in the process it spins off) were not dropping privilege.
4382 6. When the daemon running as "exim" started a queue runner, it always
4383 re-executed Exim in the spun-off process. This is a waste of effort when
4384 deliver_drop_privilege is set. The new process now just calls the
4385 queue-runner function directly.
4391 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4394 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4395 ignored. Though the use of -oP was forcing the writing of a pid file, it
4396 was always written to the default place.
4398 3. If the message "no IP address found for host xxxx" is generated during
4399 incoming verification, it is now followed by identification of the incoming
4400 connection (so you can more easily find what provoked it).
4402 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4404 5. Added some features to "harden" Exim a bit more against certain attacks:
4406 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4407 be put in Local/Makefile. This is like the never_users runtime option,
4408 but it cannot be overridden. The default setting is "root".
4410 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4411 prefix string with which any file named in a -C command line option
4414 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4415 is retained for -C and -D only if the caller of Exim is root. Without
4416 it, the exim user may also use -C and -D and retain privilege.
4418 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4419 command line option is disabled.
4421 6. Macro names set by the -D option must start with an upper case letter, just
4422 like macro names defined in the configuration file.
4424 7. Added "dereference=" facility to LDAP.
4426 8. Two instances of the typo "uknown" in the source files are fixed.
4428 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4429 the Configure-Makefile script screwed up while processing it.
4431 10. Incorporated PCRE 4.4.
4433 11. The SMTP synchronization check was not operating right at the start of an
4434 SMTP session. For example, it could not catch a HELO sent before the client
4435 waited for the greeting. There is now a check for outstanding input at the
4436 point when the greeting is written. Because of the duplex, asynchronous
4437 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4438 way, but not yet received, when the check is performed.
4440 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4441 on TCP/IP sockets, because this apparently causes some broken clients to
4444 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4445 unchanged) from the Cygwin maintainer.
4447 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4448 was supported instead of testing for mbx. Effectively a typo.
4450 15. The spa authenticator server code was not checking that the input it
4451 received was valid base64.
4453 16. The debug output line for the "set" modifier in ACLs was not showing the
4454 name of the variable that was being set.
4456 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4458 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4459 lookup on its own. Something else was needed to trigger the lookup. For
4460 example, a match in host_lookup or the need for a host name in a host list.
4461 Now, if $sender_host_name is referenced and the host name has not yet been
4462 looked up, a lookup is performed. If the lookup fails, the variable remains
4463 empty, and $host_lookup_failed is set to "1".
4465 19. Added "eqi" as a case-independent comparison operator.
4467 20. The saslauthd authentication condition could segfault if neither service
4468 nor realm was specified.
4470 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4471 error message that was logged was misleading, and incoming SMTP
4472 connections were dropped. The message is now more accurate, and temporary
4473 errors are given to SMTP connections.
4475 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4476 (except RSET) with a 421 error, until QUIT is received. However, it was
4477 failing to send a response to QUIT.
4479 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4480 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4481 ACL is now run after the helo_try_verify_hosts code.
4483 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4484 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4485 case-independent, so other case variants are also recognized). Apparently
4486 some systems use these upper case variants.
4488 25. If more than two messages were waiting for the same host, and a transport
4489 filter was specified for the transport, Exim sent two messages over the
4490 same TCP/IP connection, and then failed with "socket operation on non-
4491 socket" when it tried to send the third.
4493 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4495 27. The extern definition of crypt16() in expand.c was not being excluded when
4496 the OS had its own crypt16() function.
4498 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4499 as a preferred synonym for return_size_limit, both as an option and as an
4502 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4504 30. Changed the default configuration ACL to relax the local part checking rule
4505 for addresses that are not in any local domains. For these addresses,
4506 slashes and pipe symbols are allowed within local parts, but the sequence
4507 /../ is explicitly forbidden.
4509 31. SPA server authentication was not clearing the challenge buffer before
4512 32. log_message in a "warn" ACL statement was writing to the reject log as
4513 well as to the main log, which contradicts the documentation and doesn't
4514 seem right (because no rejection is happening). So I have stopped it.
4516 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4517 However, I am unable to do any testing of this.
4519 34. Fixed an infelicity in the appendfile transport. When checking directories
4520 for a mailbox, to see if any needed to be created, it was accidentally
4521 using path names with one or more superfluous leading slashes; tracing
4522 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4524 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4525 contents are added to the log line that is written for every discarded
4526 recipient. (Previously a log_message setting was ignored.)
4528 36. The ${quote: operator now quotes the string if it is empty.
4530 37. The install script runs exim in order to find its version number. If for
4531 some reason other than non-existence or emptiness, which it checks, it
4532 could not run './exim', it was installing it with an empty version number,
4533 i.e. as "exim-". This error state is now caught, and the installation is
4536 38. An argument was missing from the function that creates an error message
4537 when Exim fails to connect to the socket for saslauthd authentication.
4538 This could cause Exim to crash, or give a corrupted message.
4540 39. Added isip, isip4, and isip6 to ${if conditions.
4542 40. The ACL variables $acl_xx are now saved with the message, and can be
4543 accessed later in routers, transports, and filters.
4545 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4546 strings in the file are not string-expanded.
4548 42. If a MAIL command specified a SIZE value that was too large to fit into an
4549 int variable, the check against message_size_limit failed. Such values are
4550 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4551 day this will have to be increased, but I don't think I want to be around
4552 when emails are that large.
4559 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4560 iconv() is not standard in FreeBSD.
4562 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4563 IPv6 enabled. The observed symptom was a segmentation fault on return from
4564 the function os_common_find_running_interfaces() in src/os.c.
4566 3. In the check_special_case() function in daemon.c I had used "errno" as an
4567 argument name, which causes warnings on some systems. This was basically a
4568 typo, since it was named "eno" in the comments!
4570 4. The code that waits for the clock to tick (at a resolution of some fraction
4571 of a second) so as to ensure message-id uniqueness was always waiting for
4572 at least one whole tick, when it could have waited for less. [This is
4573 almost certainly not relevant at current processor speeds, where it is
4574 unlikely to ever wait at all. But we try to future-proof.]
4576 5. The function that sleeps for a time interval that includes fractions of a
4577 second contained a race. It did not block SIGALRM between setting the
4578 timer, and suspending (a couple of lines later). If the interval was short
4579 and the sigsuspend() was delayed until after it had expired, the suspension
4580 never ended. On busy systems this could lead to processes getting stuck for
4583 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4584 process, before it forks any delivery processes. The open lookup caching
4585 mechanism meant that the open file or database connection was passed into
4586 the delivery process. The problem was that delivery processes always tidy
4587 up cached lookup data. This could cause a problem for the next delivery
4588 process started by the queue runner, because the external queue runner
4589 process does not know about the closure. So the next delivery process
4590 still has data in the lookup cache. In the case of a file lookup, there was
4591 no problem because closing a file descriptor in a subprocess doesn't affect
4592 the parent. However, if the lookup was caching a connection to a database,
4593 the connection was closed, and the second delivery process was likely to
4594 see errors such as "PGSQL: query failed: server closed the connection
4595 unexpectedly". The problem has been fixed by closing all cached lookups
4596 in a queue runner before running a delivery process.
4598 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4599 seem to have the "const" qualifier which it has on other OS. I've
4602 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4603 *of the same type* (client or server) with the same public name that an
4604 error should be diagnosed.
4606 9. When Exim looked up a host name for an IP address, but failed to find the
4607 original IP address when looking up the host name (a safety check), it
4608 output the message "<ip address> does not match any IP for NULL", which was
4609 confusing, to say the least. The bug was that the host name should have
4610 appeared instead of "NULL".
4612 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4613 user that is built into the binary, and the -C or -D options is used, root
4614 privilege is dropped before the configuration file is read. In addition,
4615 logging is switched to stderr instead of the normal log files. If the
4616 configuration then re-defines the Exim user, the unprivileged environment
4617 is probably not what is expected, so Exim logs a panic warning message (but
4620 However, if deliver_drop_privilege is set, the unprivileged state may well
4621 be exactly what is intended, so the warning has been cut out in that case,
4622 and Exim is allowed to try to write to its normal log files.
4628 1. smtp_return_error_details was not giving details for temporary sender
4629 or receiver verification errors.
4631 2. Diagnose a configuration error if two authenticators have the same public
4634 3. Exim used not to create the message log file for a message until the first
4635 delivery attempt. This could be confusing when incoming messages were held
4636 for policy or load reasons. The message log file is now created at the time
4637 the message is received, and an initial "Received" line is written to it.
4639 4. The automatically generated man page for command line options had a minor
4640 bug that caused no ill effects; however, a more serious problem was that
4641 the procedure for building the man page automatically didn't always
4642 operate. Consequently, release 4.20 contains an out-of-date version. This
4643 shouldn't happen again.
4645 5. When building Exim with embedded Perl support, the script that builds the
4646 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4647 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4649 6. The freeze_tell option was not being used for messages that were frozen on
4650 arrival, either by an ACL or by local_scan().
4652 7. Added the smtp_incomplete_transaction log selector.
4654 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4655 was accepting AUTH without a new EHLO.
4657 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4658 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4661 10. Logging of TCP/IP connections (when configured) now happens in the main
4662 daemon process instead of the child process, so that the TCP/IP connection
4663 count is more accurate (but it can never be perfect).
4665 11. The use of "drop" in a nested ACL was not being handled correctly in the
4666 outer ACL. Now, if condition failure induced by the nested "drop" causes
4667 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4668 or "require"), the connection is dropped.
4670 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4671 that yield "discard" can now be used with an "accept" or a "discard" verb,
4672 but an error is generated for any others (because I can't see a useful way
4673 to define what should happen).
4675 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4676 are now processed in the same way as lines in the Exim configuration file.
4677 In particular, continuation lines are supported.
4679 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4681 15. Added -ti meaning -t -i.
4683 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4684 domains, and warn by logging if others are found.
4686 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4687 returned from GIFCONF when Exim is trying to find the list of interfaces on
4688 a host. The code in os.c has been modified to copy each ifreq to an aligned
4689 structure in all cases.
4691 Also, in some cases, the returned ifreq's were being copied to a 'struct
4692 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4693 means the last couple of bytes of an IPv6 address could be chopped if the
4694 ifreq contained only a normal sockaddr (14 bytes storage).
4696 18. Named domain lists were not supported in the hosts_treat_as_local option.
4697 An entry such as +xxxx was not recognized, and was treated as a literal
4700 19. Ensure that header lines added by a DATA ACL are included in the reject log
4701 if the ACL subsequently rejects the message.
4703 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4704 MD5 (which is deprecated).
4706 21. When testing a filter file using -bf, Exim was writing a message when it
4707 took the sender from a "From " line in the message, but it was not doing so
4708 when it took $return_path from a Return-Path: header line. It now does.
4710 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4711 with a valid header line field name (a series of printing characters
4712 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4714 23. Changed "disc" in the source to "disk" to conform to the documentation and
4715 the book and for uniformity.
4717 24. Ignore Sendmail's -Ooption=value command line item.
4719 25. When execve() failed while trying to run a command in a pipe transport,
4720 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4721 could be confused with a return value of 69 from the command itself. This
4722 has been changed to 127, the value the shell returns if it is asked to run
4723 a non-existent command. The wording for the related log line suggests a
4724 non-existent command as the problem.
4726 26. If received_header_text expands to an empty string, do not add a Received:
4727 header line to the message. (Well, it adds a token one on the spool, but
4728 marks it "old" so that it doesn't get used or transmitted.)
4730 27. Installed eximstats 1.28 (addition of -nt option).
4732 28. There was no check for failure on the call to getsockname() in the daemon
4733 code. This can fail if there is a shortage of resources on the system, with
4734 ENOMEM, for example. A temporary error is now given on failure.
4736 29. Contrary to the C standard, it seems that in some environments, the
4737 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4738 program. Exim now does this explicitly; it affects the formatting of
4739 timestamps using strftime().
4741 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4742 complaints. I've now initialized all the variables, to avoid this.
4744 32. Header lines added by a system filter were not being "seen" during
4745 transport-time rewrites.
4747 33. The info_callback() function passed to OpenSSL is set up with type void
4748 (*)(SSL *, int, int), as described somewhere. However, when calling the
4749 function (actually a macro) that sets it up, the type void(*)() is
4750 expected. I've put in a cast to prevent warnings from picky compilers.
4752 34. If a DNS black list lookup found a CNAME record, but there were no A
4753 records associated with the domain it pointed at, Exim crashed.
4755 35. If a DNS black list lookup returned more than one A record, Exim ignored
4756 all but the first. It now scans all returned addresses if a particular IP
4757 value is being sought. In this situation, the contents of the
4758 $dnslist_value variable are a list of all the addresses, separated by a
4761 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4762 to accept a host name and all its aliases if the forward lookup for any of
4763 them yielded the IP address of the incoming connection. Now it accepts only
4764 those names whose forward lookup yields the correct IP address. Any other
4765 names are discarded. This closes a loophole whereby a rogue DNS
4766 administrator could create reverse DNS records to break through a
4767 wildcarded host restriction in an ACL.
4769 37. If a user filter or a system filter that ran in a subprocess used any of
4770 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4771 the wrong values were passed to the pipe command ($thisaddress had the
4772 value of $0, $0 had the value of $1, etc). This bug was introduced by
4773 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4775 38. Improved the line breaking for long SMTP error messages from ACLs.
4776 Previously, if there was no break point between 40 and 75 characters, Exim
4777 left the rest of the message alone. Two changes have been made: (a) I've
4778 reduced the minimum length to 35 characters; (b) if it can't find a break
4779 point between 35 and 75 characters, it looks ahead and uses the first one
4780 that it finds. This may give the occasional overlong line, but at least the
4781 remaining text gets split now.
4783 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4784 file descriptors might be low, and that setting 1000 would always raise it.
4785 It turns out that in some environments, the limit is already over 1000 and
4786 that lowering it causes trouble. So now Exim takes care not to decrease it.
4788 40. When delivering a message, the value of $return_path is set to $sender_
4789 address at the start of routing (routers may change the value). By an
4790 oversight, this default was not being set up when an address was tested by
4791 -bt or -bv, which affected the outcome if any router or filter referred to
4794 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4795 log only when "message" or "log_message" are set. However, if one of the
4796 conditions was an address verification, or a call to a nested ACL, the
4797 messages generated by the underlying test were being passed through. This
4798 no longer happens. The underlying message is available in $acl_verify_
4799 message for both "message" and "log_message" expansions, so it can be
4800 passed through if needed.
4802 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4803 new expansion $bh_ to give the encoded byte string without charset
4804 translation. Translation happens only if iconv() is available; HAVE_ICONV
4805 indicates this at build time. HEADERS_CHARSET gives the charset to
4806 translate to; headers_charset can change it in the configuration, and
4807 "headers charset" can change it in an individual filter file.
4809 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4810 that creates RFC 2047 encoded "words" labels them as that charset instead
4811 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4812 expansion operator; (ii) when Exim creates a From: line for a local
4813 message; (iii) when a header line is rewritten to include a "phrase" part.
4815 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4816 buggy, causing it to skip the first lines of messages whose message ID
4817 ended in 'D'. This would not have bitten before Exim release 4.14, because
4818 message IDs were unlikely to end in 'D' before then. The effect was to have
4819 incorrect size information for certain domains.
4821 45. #include "config.h" was missing at the start of the crypt16.c module. This
4822 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4825 46. If there was a timeout during a "random" callout check, Exim treated it as
4826 a failure of the random address, and carried on sending RSET and the real
4827 address. If the delay was just some slowness somewhere, the response to the
4828 original RCPT would be taken as a response to RSET and so on, causing
4829 mayhem of various kinds.
4831 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4832 when I implemented it. It didn't allow for the fact that some option values
4833 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4834 the right test for positive values.
4836 48. Domain names in DNS records are case-independent. Exim always looks them up
4837 in lower case. Some resolvers return domain names in exactly the case they
4838 appear in the zone file, that is, they may contain uppercase letters. Not
4839 all resolvers do this - some return always lower case. Exim was treating a
4840 change of case by a resolver as a change of domain, similar to a widening
4841 of a domain abbreviation. This triggered its re-routing code and so it was
4842 trying to route what was effectively the same domain again. This normally
4843 caused routing to fail (because the router wouldn't handle the domain
4844 twice). Now Exim checks for this case specially, and just changes the
4845 casing of the domain that it ultimately uses when it transmits the message
4848 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4851 50. If a filter generated a file delivery with a non-absolute name (possible if
4852 no home directory exists for the router), the forbid_file option was not
4855 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4856 the existing equality matching.
4858 52. Exim was using ints instead of ino_t variables in some places where it was
4859 dealing with inode numbers.
4861 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4862 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4863 called TMPDIR, and if it finds it is different, it changes its value.
4865 54. The smtp_printf() function is now made available to local_scan() so
4866 additional output lines can be written before returning. There is also an
4867 smtp_fflush() function to enable the detection of a dropped connection.
4868 The variables smtp_input and smtp_batched_input are exported to
4871 55. Changed the default runtime configuration: the message "Unknown user"
4872 has been removed from the ACL, and instead placed on the localuser router,
4873 using the cannot_route_message feature. This means that any verification
4874 failures that generate their own messages won't get overridden. Similarly,
4875 the "Unrouteable address" message that was in the ACL for unverifiable
4876 relay addresses has also been removed.
4878 56. Added hosts_avoid_esmtp to the smtp transport.
4880 57. The exicyclog script was not checking for the esoteric option
4881 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4882 will work only if exicyclog is run under the appropriate euid.
4884 58. Following a discussion on the list, the rules by which Exim recognises line
4885 endings on incoming messages have been changed. The -dropcr and drop_cr
4886 options are now no-ops, retained only for backwards compatibility. The
4887 following line terminators are recognized: LF CRLF CR. However, special
4888 processing applies to CR:
4890 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4891 nor a local message in the state where . is a terminator.
4893 (ii) If a bare CR is encountered in a header line, an extra space is added
4894 after the line terminator so as not to end the header. The reasoning
4895 behind this is that bare CRs in header lines are most likely either
4896 to be mistakes, or people trying to play silly games.
4898 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4899 was being incorrectly given as 18 bytes larger than it should have been.
4900 This is a VOB (very old bug).
4902 60. This may never have affected anything current, but just in case it has:
4903 When the local host is found other than at the start of a list of hosts,
4904 the local host, those with the same MX, and any that follow, are discarded.
4905 When the list in question was part of a longer list of hosts, the following
4906 hosts (not currently being processed) were also being discarded. This no
4907 longer happens. I'm not sure if this situation could ever has previously
4910 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4913 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4914 header. This is something that is recommended in a new Internet Draft, and
4915 is something that is documented as being done by Sendmail. There are two
4916 possible values. For messages generated by the autoreply transport, Exim
4919 Auto-Submitted: auto-replied
4921 whereas for all other generated messages (e.g. bounces) it adds
4923 Auto-Submitted: auto-generated
4925 63. The "personal" condition in filters now includes a test for the
4926 Auto-Submitted: header. If it contains the string "auto-" the message it
4927 not considered personal.
4929 64. Added rcpt_include_affixes as a generic transport option.
4931 65. Added queue_only_override (default true).
4933 66. Added the syslog_duplication option.
4935 67. If what should have been the first header line of a message consisted of
4936 a space followed by a colon, Exim was mis-interpreting it as a header line.
4937 It isn't of course - it is syntactically invalid and should therefore be
4938 treated as the start of the message body. The misbehaviour could have
4939 caused a number of strange effects, including loss of data in subsequent
4940 header lines, and spool format errors.
4942 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4943 client host had authenticated. This control can now be exercised by an ACL
4944 for more flexibility.
4946 69. By default, callouts do not happen when testing with -bh. There is now a
4947 variant, -bhc, which does actually run the callout code, including
4948 consulting and updating the callout cache.
4950 70. Added support for saslauthd authentication, courtesy of Alexander
4953 71. If statvfs() failed on the spool or log directories while checking their
4954 size for availability, Exim confusingly gave the error "space shortage".
4955 Furthermore, in debugging mode it crashed with a floating point exception.
4956 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4957 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4958 serious problem, Exim now writes to the main and panic logs when this
4959 happens, with details of the failure. It then refuses to accept the
4960 incoming message, giving the message "spool directory problem" or "log
4961 directory problem" with a 421 code for SMTP messages.
4963 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4964 0, 1, and 2 exist, because some OS complain for execs without them (see
4965 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4966 descriptors. However, the code omitted to check that the open succeeded,
4967 causing mysterious errors if for some reason the permissions on /dev/null
4968 got screwed. Now Exim writes a message to the main and panic logs, and
4969 bombs out if it can't open /dev/null.
4971 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4972 interact so that it is all more flexible. It is supposed to remain
4973 backwards compatible. Also added extra_local_interfaces.
4975 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4976 to bomb out with an assertion failure - to the client this appears as a
4977 connection drop. This problem occurs in the part of the code that was taken
4978 from the Samba project. Fortunately, the assertion is in a very simple
4979 function, so I have fixed this by reproducing the function inline in the
4980 one place where it is called, and arranging for authentication to fail
4981 instead of killing the process with assert().
4983 75. The SPA client code was not working when the server requested OEM rather
4984 than Unicode encoding.
4986 76. Added code to make require_files with a specific uid setting more usable in
4987 the case where statting the file as root fails - usually a non-root-mounted
4988 NFS file system. When this happens and the failure is EACCES, Exim now
4989 forks a subprocess and does the per-uid checking as the relevant uid.
4991 77. Added process_log_path.
4993 78. If log_file_path was not explicitly set, a setting of check_log_space or
4994 check_log_inodes was ignored.
4996 79. If a space check for the spool or log partitions fails, the incident is now
4997 logged. Of course, in the latter case the data may get lost...
4999 80. Added the %p formatting code to string_format() so that it can be used to
5000 print addresses in debug_print(). Adjusted all the address printing in the
5001 debugging in store.c to use %p rather than %d.
5003 81. There was a concern that a line of code in smtp_in.c could overflow a
5004 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5005 initially expressed, the concern was not well-founded, because trailing
5006 spaces are removed early. However, if the trailing spaces were followed by
5007 a NULL, they did not get removed, so the overflow was possible. Two fixes
5010 (a) I re-wrote the offending code in a cleaner fashion.
5011 (b) If an incoming SMTP command contains a NULL character, it is rejected
5014 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5015 runs initgroups(), so that if the Exim user is in any additional groups,
5016 they will be used during message reception.
5022 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob