1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.591 2009/11/12 13:02:16 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/36 Bugzilla 910: Correct issue with relaxed/simple handling
162 TK/01 Add preliminary DKIM support. Currently requires a forked version of
163 ALT-N's libdkim that I have put here:
164 http://duncanthrax.net/exim-experimental/
166 Note to Michael Haardt: I had to rename some vars in sieve.c. They
167 were called 'true' and it seems that C99 defines that as a reserved
168 keyword to be used with 'bool' variable types. That means you could
169 not include C99-style headers which use bools without triggering
170 build errors in sieve.c.
172 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
173 as mailq or other aliases. Changed the --help handling significantly
174 to do whats expected. exim_usage() emits usage/help information.
176 SC/01 Added the -bylocaldomain option to eximstats.
178 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
180 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
182 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
188 PH/01 Another patch from the Sieve maintainer.
190 PH/02 When an IPv6 address is converted to a string for single-key lookup
191 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
192 dots are used instead of colons so that keys in lsearch files need not
193 contain colons. This was done some time before quoting was made available
194 in lsearch files. However, iplsearch files do require colons in IPv6 keys
195 (notated using the quote facility) so as to distinguish them from IPv4
196 keys. This meant that lookups for IP addresses in host lists did not work
197 for iplsearch lookups.
199 This has been fixed by arranging for IPv6 addresses to be expressed with
200 colons if the lookup type is iplsearch. This is not incompatible, because
201 previously such lookups could never work.
203 The situation is now rather anomolous, since one *can* have colons in
204 ordinary lsearch keys. However, making the change in all cases is
205 incompatible and would probably break a number of configurations.
207 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
210 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
211 conversion specification without a maximum field width, thereby enabling
212 a rogue spamd server to cause a buffer overflow. While nobody in their
213 right mind would setup Exim to query an untrusted spamd server, an
214 attacker that gains access to a server running spamd could potentially
215 exploit this vulnerability to run arbitrary code as the Exim user.
217 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
218 $primary_hostname instead of what libspf2 thinks the hosts name is.
220 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
221 a directory entry by the name of the lookup key. Previously, if a
222 symlink pointed to a non-existing file or a file in a directory that
223 Exim lacked permissions to read, a lookup for a key matching that
224 symlink would fail. Now it is enough that a matching directory entry
225 exists, symlink or not. (Bugzilla 503.)
227 PH/03 The body_linecount and body_zerocount variables are now exported in the
230 PH/04 Added the $dnslist_matched variable.
232 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
233 This means they are set thereafter only if the connection becomes
236 PH/06 Added the client_condition to authenticators so that some can be skipped
237 by clients under certain conditions.
239 PH/07 The error message for a badly-placed control=no_multiline_responses left
240 "_responses" off the end of the name.
242 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
244 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
245 (without spaces) instead of just copying the configuration text.
247 PH/10 Added the /noupdate option to the ratelimit ACL condition.
249 PH/11 Added $max_received_linelength.
251 PH/12 Added +ignore_defer and +include_defer to host lists.
253 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
254 way in which PCRE > 7.0 is built.
256 PH/14 Implemented queue_only_load_latch.
258 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
259 MAIL command. The effect was to mangle the value on 64-bit systems.
261 PH/16 Another patch from the Sieve maintainer.
263 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
265 PH/18 If a system quota error occurred while trying to create the file for
266 a maildir delivery, the message "Mailbox is full" was not appended to the
267 bounce if the delivery eventually timed out. Change 4.67/27 below applied
268 only to a quota excession during the actual writing of the file.
270 PH/19 It seems that peer DN values may contain newlines (and other non-printing
271 characters?) which causes problems in log lines. The DN values are now
272 passed through string_printing() before being added to log lines.
274 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
275 and InterBase are left for another time.)
277 PH/21 Added message_body_newlines option.
279 PH/22 Guard against possible overflow in moan_check_errorcopy().
281 PH/23 POSIX allows open() to be a macro; guard against that.
283 PH/24 If the recipient of an error message contained an @ in the local part
284 (suitably quoted, of course), incorrect values were put in $domain and
285 $local_part during the evaluation of errors_copy.
291 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
292 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
295 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
296 issue a MAIL command.
298 PH/02 In an ACL statement such as
300 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
302 if a client was not listed at all, or was listed with a value other than
303 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
304 the condition was not true (as it should be), so access was not denied.
305 The bug was that the ! inversion was incorrectly passed on to the second
306 item. This has been fixed.
308 PH/03 Added additional dnslists conditions == and =& which are different from
309 = and & when the dns lookup returns more than one IP address.
311 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
312 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
314 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
315 FSYNC, which compiles an option called disable_fsync that allows for
316 bypassing fsync(). The documentation is heavily laced with warnings.
318 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
320 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
321 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
322 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
323 including adding "make clean"; (3) Added -fPIC when compiling the test
324 dynamically loaded module, to get rid of a warning.
326 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
327 message fails, move_frozen_messages = true and ignore_bounce_errors_after
328 = 0s. The bug is otherwise harmless.
330 PH/07 There was a bug in the dovecot authenticator such that the value of
331 $auth1 could be overwritten, and so not correctly preserved, after a
332 successful authentication. This usually meant that the value preserved by
333 the server_setid option was incorrect.
335 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
337 PH/09 Installed PCRE release 7.0.
339 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
340 run for batched SMTP input. It is now run at the start of every message
341 in the batch. While fixing this I discovered that the process information
342 (output by running exiwhat) was not always getting set for -bs and -bS
343 input. This is fixed, and it now also says "batched" for BSMTP.
345 PH/11 Added control=no_pipelining.
347 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
348 patch, slightly modified), and move the expansion of helo_data till after
349 the connection is made in the smtp transport (so it can use these
352 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
354 PH/14 Added log_selector = +pid.
356 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
358 PH/16 Add ${if forany and ${if forall.
360 PH/17 Added dsn_from option to vary the From: line in DSNs.
362 PH/18 Flush SMTP output before performing a callout, unless control =
363 no_callout_flush is set.
365 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
366 was true (the default) a successful delivery failed to delete the retry
367 item, thus causing premature timeout of the address. The bug is now
370 PH/20 Added hosts_avoid_pipelining to the smtp transport.
372 PH/21 Long custom messages for fakedefer and fakereject are now split up
373 into multiline reponses in the same way that messages for "deny" and
374 other ACL rejections are.
376 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
377 with slight modification.
379 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
380 draft, changing the syntax and factoring some duplicate code".
382 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
383 for deliveries of the second and subsequent messages over the same SMTP
386 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
387 ${reduce, with only minor "tidies".
389 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
391 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
392 expansion side effects.
394 PH/27 When a message times out after an over-quota error from an Exim-imposed
395 quota, the bounce message says "mailbox is full". This message was not
396 being given when it was a system quota that was exceeded. It now should
399 MH/03 Made $recipients available in local_scan(). local_scan() already has
400 better access to the recipient list through recipients_list[], but
401 $recipients can be useful in postmaster-provided expansion strings.
403 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
404 in the case of a MAIL command with additional options following the
405 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
406 were accidentally chopped off.
408 PH/29 SMTP synchronization checks are implemented when a command is read -
409 there is a check that no more input is waiting when there shouldn't be
410 any. However, for some commands, a delay in an ACL can mean that it is
411 some time before the response is written. In this time, more input might
412 arrive, invalidly. So now there are extra checks after an ACL has run for
413 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
414 pipelining has not been advertised.
416 PH/30 MH's patch to allow iscntrl() characters to be list separators.
418 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
419 returned in the SMTP response when smtp_return_error_details was false.
422 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
423 instead of the C I/O that was originally supplied, because problems were
426 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
427 Exim which did not show up earlier: it was assuming that a call to
428 SSL_CTX_set_info_callback() might give an error value. In fact, there is
429 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
430 was a macro that became an assignment, so it seemed to work. This has
431 changed to a proper function call with a void return, hence the compile
432 error. Exim's code has been fixed.
434 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
437 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
439 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
441 PH/37 If a message is not accepted after it has had an id assigned (e.g.
442 because it turns out to be too big or there is a timeout) there is no
443 "Completed" line in the log. When some messages of this type were
444 selected by exigrep, they were listed as "not completed". Others were
445 picked up by some special patterns. I have improved the selection
446 criteria to be more general.
448 PH/38 The host_find_failed option in the manualroute router can now be set
449 to "ignore", to completely ignore a host whose IP address cannot be
450 found. If all hosts are ignored, the behaviour is controlled by the new
451 host_all_ignored option.
453 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
454 homing or because of multiple MX records with /mx) generated more than
455 one IP address, and the following item turned out to be the local host,
456 all the secondary addresses of the first item were incorrectly removed
457 from the list, along with the local host and any following hosts (which
458 is what is supposed to happen).
460 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
461 whoever called Exim into the -H file. In the case of the daemon it was
462 behaving confusingly. When first started, it used values for whoever
463 started the daemon, but after a SIGHUP it used the Exim user (because it
464 calls itself on a restart). I have changed the code so that it now always
467 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
468 message are rejected with the same error (e.g. no authentication or bad
469 sender address), and a DATA command is nevertheless sent (as can happen
470 with PIPELINING or a stupid MUA), the error message that was given to the
471 RCPT commands is included in the rejection of the DATA command. This is
472 intended to be helpful for MUAs that show only the final error to their
475 PH/42 Another patch from the Sieve maintainer.
477 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
478 Eximstats - Fixed some broken HTML links and added missing column headers
480 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
481 columns for Rejects, Temp Rejects, Ham, and Spam rows.
483 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
485 PH/43 Yet another patch from the Sieve maintainer.
487 PH/44 I found a way to check for a TCP/IP connection going away before sending
488 the response to the final '.' that terminates a message, but only in the
489 case where the client has not sent further data following the '.'
490 (unfortunately, this is allowed). However, in many cases there won't be
491 any further data because there won't be any more messages to send. A call
492 to select() can be used: if it shows that the input is "ready", there is
493 either input waiting, or the socket has been closed. An attempt to read
494 the next input character can distinguish the two cases. Previously, Exim
495 would have sent an OK response which the client would never have see.
496 This could lead to message repetition. This fix should cure that, at
497 least in a lot of common cases.
499 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
500 advertised in response to EHLO.
506 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
507 fixed by 4.65/MH/01 (is this a record?) are fixed:
509 (i) An empty string was always treated as zero by the numeric comparison
510 operators. This behaviour has been restored.
512 (ii) It is documented that the numeric comparison operators always treat
513 their arguments as decimal numbers. This was broken in that numbers
514 starting with 0 were being interpreted as octal.
516 While fixing these problems I realized that there was another issue that
517 hadn't been noticed. Values of message_size_limit (both the global option
518 and the transport option) were treated as octal if they started with 0.
519 The documentation was vague. These values are now always treated as
520 decimal, and I will make that clear in the documentation.
526 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
527 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
530 MH/01 Don't check that the operands of numeric comparison operators are
531 integers when their expansion is in "skipping" mode (fixes bug
532 introduced by 4.64-PH/07).
534 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
535 child addresses, Exim now panics and dies. Previously, because the count
536 is held in a short int, deliveries were likely to be lost. As such a
537 large number of recipients for a single message is ridiculous
538 (performance will be very, very poor), I have chosen to impose a limit
539 rather than extend the field.
545 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
546 leftover -K file (the existence of which was triggered by #402).
547 While we were at it, introduced process PID as part of the -K
548 filename. This should rule out race conditions when creating
551 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
552 processing considerably. Previous code took too long for large mails,
553 triggering a timeout which in turn triggers #401.
555 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
556 in the DK code in transports.c. sendfile() is not really portable,
557 hence the _LINUX specificness.
559 TF/01 In the add_headers option to the mail command in an Exim filter,
560 there was a bug that Exim would claim a syntax error in any
561 header after the first one which had an odd number of characters
564 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
565 callout verification, Exim cached a "reject" for the entire domain. This
566 is correct for most verifications, but it is not correct for a recipient
567 verification with use_sender or use_postmaster set, because in that case
568 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
569 case of MAIL FROM:<> rejection from other early rejections (e.g.
570 rejection of HELO). When verifying a recipient using a non-null MAIL
571 address, the cache is ignored if it shows MAIL FROM:<> rejection.
572 Whatever the result of the callout, the value of the domain cache is
573 left unchanged (for any other kind of callout, getting as far as trying
574 RCPT means that the domain itself is ok).
576 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
579 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
580 manifest itself as EPIPE rather than ECONNECT. When tidying away a
581 session, the daemon ignores ECONNECT errors and logs others; it now
582 ignores EPIPE as well.
584 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
585 (quoted-printable decoding).
587 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
588 later the small subsequent patch to fix an introduced bug.
590 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
592 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
594 PH/08 An error is now given if message_size_limit is specified negative.
596 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
597 to be given (somewhat) arbitrary names.
599 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
602 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
603 miscellaneous code fixes
605 PH/10 Added the log_reject_target ACL modifier to specify where to log
608 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
609 hostname. This is wrong, because it relates to the incoming message (and
610 probably the interface on which it is arriving) and not to the outgoing
611 callout (which could be using a different interface). This has been
612 changed to use the value of the helo_data option from the smtp transport
613 instead - this is what is used when a message is actually being sent. If
614 there is no remote transport (possible with a router that sets up host
615 addresses), $smtp_active_hostname is used.
617 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
618 tweaks were necessary in order to get it to work (see also 21 below):
619 (a) The code assumed that strncpy() returns a negative number on buffer
620 overflow, which isn't the case. Replaced with Exim's string_format()
622 (b) There were several signed/unsigned issues. I just did the minimum
623 hacking in of casts. There is scope for a larger refactoring.
624 (c) The code used strcasecmp() which is not a standard C function.
625 Replaced with Exim's strcmpic() function.
626 (d) The code set only $1; it now sets $auth1 as well.
627 (e) A simple test gave the error "authentication client didn't specify
628 service in request". It would seem that Dovecot has changed its
629 interface. Fortunately there's a specification; I followed it and
630 changed what the client sends and it appears to be working now.
632 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
635 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
636 address A is aliased to B and C, where B exists and C does not. Without
637 -v the output is "A verified" because verification stops after a
638 successful redirection if more than one address is generated. However,
639 with -v the child addresses are also verified. Exim was outputting "A
640 failed to verify" and then showing the successful verification for C,
641 with its parentage. It now outputs "B failed to verify", showing B's
642 parentage before showing the successful verification of C.
644 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
645 look up a TXT record in a specific list after matching in a combined
648 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
649 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
650 they consult the DNS. I had assumed they would set it the way they
651 wanted; and indeed my experiments on Linux seem to show that in some
652 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
653 To be on the safe side, however, I have now made the interface to
654 host_find_byname() similar to host_find_bydns(), with an argument
655 containing the DNS resolver options. The host_find_byname() function now
656 sets these options at its start, just as host_find_bydns() does. The smtp
657 transport options dns_qualify_single and dns_search_parents are passed to
658 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
659 of host_find_byname() use the default settings of RES_DEFNAMES
660 (qualify_single) but not RES_DNSRCH (search_parents).
662 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
663 spool_read_header() do less string testing, by means of a preliminary
664 switch on the second character of optional "-foo" lines. (This is
665 overdue, caused by the large number of possibilities that now exist.
666 Originally there were few.) While I was there, I also converted the
667 str(n)cmp tests so they don't re-test the leading "-" and the first
668 character, in the hope this might squeeze out yet more improvement.
670 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
671 flag allowing group syntax was set by the header_syntax check but not
672 turned off, possible causing trouble later; (2) The flag was not being
673 set at all for the header_verify test, causing "group"-style headers to
674 be rejected. I have now set it in this case, and also caused header_
675 verify to ignore an empty address taken from a group. While doing this, I
676 came across some other cases where the code for allowing group syntax
677 while scanning a header line wasn't quite right (mostly, not resetting
678 the flag correctly in the right place). These bugs could have caused
679 trouble for malformed header lines. I hope it is now all correct.
681 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
682 with the "reply" argument non-NULL. The code, however (which originally
683 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
684 but it didn't always do it. This confused somebody who was copying the
685 code for some other use. I have removed all the tests.
687 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
688 feature that was used to support insecure browsers during the U.S. crypto
689 embargo. It requires special client support, and Exim is probably the
690 only MTA that supported it -- and would never use it because real RSA is
691 always available. This code has been removed, because it had the bad
692 effect of slowing Exim down by computing (never used) parameters for the
693 RSA_EXPORT functionality.
695 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
696 authenticator to fail if there's a tab character in the incoming data
697 (there should never be unless someone is messing about, as it's supposed
698 to be base64-encoded). Also added, on Timo's advice, the "secured" option
699 if the connection is using TLS or if the remote IP is the same as the
700 local IP, and the "valid-client-cert option" if a client certificate has
703 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
704 authenticators. This can be used for authorization after authentication
705 succeeds. (In the case of plaintext, it servers for both authentication
708 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
709 if any retry times were supplied.
711 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
712 connection, where there is no client IP address to check. In this
713 situation, the verify now always succeeds.
715 PH/25 Applied John Jetmore's -Mset patch.
717 PH/26 Added -bem to be like -Mset, but loading a message from a file.
719 PH/27 In a string expansion for a processed (not raw) header when multiple
720 headers of the same name were present, leading whitespace was being
721 removed from all of them, but trailing whitespace was being removed only
722 from the last one. Now trailing whitespace is removed from each header
723 before concatenation. Completely empty headers in a concatenation (as
726 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
727 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
729 PH/29 [Removed. This was a change that I later backed out, and forgot to
730 correct the ChangeLog entry (that I had efficiently created) before
731 committing the later change.]
733 PH/30 Exim was sometimes attempting to deliver messages that had suffered
734 address errors (4xx response to RCPT) over the same connection as other
735 messages routed to the same hosts. Such deliveries are always "forced",
736 so retry times are not inspected. This resulted in far too many retries
737 for the affected addresses. The effect occurred only when there were more
738 hosts than the hosts_max_try setting in the smtp transport when it had
739 the 4xx errors. Those hosts that it had tried were not added to the list
740 of hosts for which the message was waiting, so if all were tried, there
741 was no problem. Two fixes have been applied:
743 (i) If there are any address or message errors in an SMTP delivery, none
744 of the hosts (tried or untried) are now added to the list of hosts
745 for which the message is waiting, so the message should not be a
746 candidate for sending over the same connection that was used for a
747 successful delivery of some other message. This seems entirely
748 reasonable: after all the message is NOT "waiting for some host".
749 This is so "obvious" that I'm not sure why it wasn't done
750 previously. Hope I haven't missed anything, but it can't do any
751 harm, as the worst effect is to miss an optimization.
753 (ii) If, despite (i), such a delivery is accidentally attempted, the
754 routing retry time is respected, so at least it doesn't keep
755 hammering the server.
757 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
758 in ${readsocket because some servers need this prod.
760 PH/32 Added some extra debug output when updating a wait-xxx database.
762 PH/33 The hint "could be header name not terminated by colon", which has been
763 given for certain expansion errors for a long time, was not being given
764 for the ${if def:h_colon_omitted{... case.
766 PH/34 The spec says: "With one important exception, whenever a domain list is
767 being scanned, $domain contains the subject domain." There was at least
768 one case where this was not true.
770 PH/35 The error "getsockname() failed: connection reset by peer" was being
771 written to the panic log as well as the main log, but it isn't really
772 panic-worthy as it just means the connection died rather early on. I have
773 removed the panic log writing for the ECONNRESET error when getsockname()
776 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
777 runs only) independently of the message's sender address. This meant
778 that, if the 4xx error was in fact related to the sender, a different
779 message to the same recipient with a different sender could confuse
780 things. In particualar, this can happen when sending to a greylisting
781 server, but other circumstances could also provoke similar problems.
782 I have changed the default so that the retry time for these errors is now
783 based a combination of the sender and recipient addresses. This change
784 can be overridden by setting address_retry_include_sender=false in the
787 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
788 remote server are returned as part of bounce messages. This was not
789 happening for LMTP over a pipe (the lmtp transport), but now it is the
790 same for both kinds of LMTP.
792 PH/38 Despite being documented as not happening, Exim was rewriting addresses
793 in header lines that were in fact CNAMEs. This is no longer the case.
795 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
796 and queue runs started by the daemon processed all messages. This has
797 been fixed so that -R and -S can now usefully be given with -q<time>.
799 PH/40 Import PCRE release 6.7 (fixes some bugs).
801 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
803 PH/42 Give an error if -q is specified more than once.
805 PH/43 Renamed the variables $interface_address and $interface_port as
806 $received_ip_address and $received_port, to make it clear that these
807 values apply to message reception, and not to the outgoing interface when
808 a message is delivered. (The old names remain recognized, of course.)
810 PH/44 There was no timeout on the connect() call when using a Unix domain
811 socket in the ${readsocket expansion. There now is.
813 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
814 be meaningful with "accept".
816 SC/01 Eximstats V1.43
817 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
819 SC/02 Eximstats V1.44
820 Use a glob alias rather than an array ref in the generated
821 parser. This improves both readability and performance.
823 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
824 Collect SpamAssassin and rejection statistics.
825 Don't display local sender or destination tables unless
826 there is data to show.
827 Added average volumes into the top table text output.
829 SC/04 Eximstats V1.46
830 Collect data on the number of addresses (recipients)
831 as well as the number of messages.
833 SC/05 Eximstats V1.47
834 Added 'Message too big' to the list of mail rejection
835 reasons (thanks to Marco Gaiarin).
837 SC/06 Eximstats V1.48
838 Mainlog lines which have GMT offsets and are too short to
839 have a flag are now skipped.
841 SC/07 Eximstats V1.49 (Alain Williams)
842 Added the -emptyok flag.
844 SC/08 Eximstats V1.50
845 Fixes for obtaining the IP address from reject messages.
847 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
848 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
849 whitesspace changes from 4.64-PH/27
851 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
854 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
855 are found, allow negative numbers in numeric criteria)
857 JJ/06 exipick.20061117.2, added new $message_body_missing variable
859 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
860 to match changes made in 4.64-PH/43
862 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
864 PH/47 Put in an explicit test for a DNS lookup of an address record where the
865 "domain" is actually an IP address, and force a failure. This locks out
866 those revolvers/nameservers that support "A-for-A" lookups, in
867 contravention of the specifications.
869 PH/48 When a host name was looked up from an IP address, and the subsequent
870 forward lookup of the name timed out, the host name was left in
871 $sender_host_name, contrary to the specification.
873 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
874 restricted to single-key lookups, Exim was not diagnosing an error if
875 * or *@ was used with a query-style lookup.
877 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
879 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
880 long ago, but noone interested enough thought of it. Let's just say that
881 the "1.1" means that there are some new functions that weren't there at
882 some point in the past.
884 PH/51 Error processing for expansion failure of helo_data from an smtp
885 transport during callout processing was broken.
887 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
888 tested/used via the -bh/-bhc/-bs options.
890 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
891 bug, fixed in subsequent PCRE releases).
893 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
894 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
896 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
902 SC/01 Use a glob alias rather than an array ref in eximstats generated
903 parser. This improves both readability and performance.
905 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
906 Don't display local sender or destination tables in eximstats unless
907 there is data to show.
908 Added average volumes into the eximstats top table text output.
910 SC/03 Collect data on the number of addresses (recipients) as well
911 as the number of messages in eximstats.
913 TF/01 Correct an error in the documentation for the redirect router. Exim
914 does (usually) call initgroups() when daemonizing.
916 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
917 with consistent privilege compared to when running as a daemon.
919 TF/03 Note in the spec that $authenticated_id is not set for local
920 submissions from trusted users.
922 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
923 Thanks to Dean Brooks <dean@iglou.com> for the patch.
925 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
926 by adding some example configuration directives to the default
927 configuration file. A little bit of work is required to uncomment the
928 directives and define how usernames and passwords are checked, but
929 there is now a framework to start from.
931 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
932 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
933 without this. I don't know how relevant this is to other LDAP libraries.
935 PH/02 Add the verb name to the "unknown ACL verb" error.
937 PH/03 Magnus Holmgren's patch for filter_prepend_home.
939 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
941 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
942 directory not expanded when it should be if an expanded home directory
943 was set for the address (which is overridden by the transport).
945 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
948 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
949 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
950 because it is too late at that time, and has no effect.
952 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
953 security issue with \' (bugzilla #107). I could not use the
954 PQescapeStringConn() function, because it needs a PGconn value as one of
957 PH/08 When testing addresses using -bt, indicate those final addresses that
958 are duplicates that would not cause an additional delivery. At least one
959 person was confused, thinking that -bt output corresponded to deliveries.
960 (Suppressing duplicates isn't a good idea as you lose the information
961 about possibly different redirections that led to the duplicates.)
963 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
964 systems where poll() doesn't work, in particular OS X.
966 PH/10 Added more information to debugging output for retry time not reached.
968 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
969 operations in malware.c.
971 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
974 PH/13 If write_rejectlog was set false when logging was sent to syslog with
975 syslog_duplication set false, log lines that would normally be written
976 both the the main log and to the reject log were not written to syslog at
979 PH/14 In the default configuration, change the use of "message" in ACL warn
980 statements to "add_header".
982 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
983 not followed by a command (e.g. "seen endif").
985 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
986 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
989 PH/17 Added extra conditions to the default value of delay_warning_condition
993 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
994 { match{$h_precedence:}{(?i)bulk|list|junk} } \
995 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
998 The Auto-Submitted: and various List- headers are standardised, whereas I
999 don't think Precedence: ever was.
1001 PH/18 Refactored debugging code in route_finduser() to show more information,
1002 in particular, the error code if getpwnam() issues one.
1004 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1005 This is apparently needed in addition to the PH/07 change above to avoid
1006 any possible encoding problems.
1008 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1009 but not after initializing Perl.
1011 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1012 output them only if debugging. By default they are written stderr,
1013 apparently, which is not desirable.
1015 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1018 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1021 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1023 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1024 authenticated or an ident call has been made. Suppress the default
1025 values for $authenticated_id and $authenticated_sender (but permit -oMai
1026 and -oMas) when testing with -bh.
1028 PH/24 Re-jigged the order of the tests in the default configuration so that the
1029 tests for valid domains and recipients precede the DNS black list and CSA
1030 tests, on the grounds that those ones are more expensive.
1032 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1033 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1034 This bug exists in every version of Exim that I still have, right back to
1037 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1038 However, an attempt to turn on case-sensitivity in a regex key by
1039 including (?-i) didn't work because the subject string was already
1040 lowercased, and the effects were non-intuitive. It turns out that a
1041 one-line patch can be used to allow (?-i) to work as expected.
1047 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1048 other effects) broke the use of negated acl sub-conditions.
1050 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1053 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1054 "Deny" causes Exim to reject the incoming connection with a 554 error.
1055 Unfortunately, if there is a major crisis, such as a disk failure,
1056 tcp-wrappers gives "deny", whereas what one would like would be some
1057 kind of temporary error. A kludge has been added to help with this.
1058 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1059 554 error is used if errno is still zero or contains ENOENT (which occurs
1060 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1063 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1065 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1066 errors. Otherwise a message that provokes a temporary error (when other
1067 messages do not) can cause a whole host to time out.
1069 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1070 addresses were routed directly to files or pipes from a redirect router.
1071 File deliveries just didn't batch; pipe deliveries might have suffered
1074 PH/06 A failure to get a lock for a hints database would erroneously always say
1075 "Failed to get write lock", even when it was really a read lock.
1077 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1078 of 0600. This has been changed to use the value of the lockfile_mode
1079 option (which defaults to 0600).
1081 PH/08 Applied small patch from the Sieve maintainer.
1083 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1084 folder from quota calculations, a direct delivery into this folder messed
1085 up the contents of the maildirsize file. This was because the regex was
1086 used only to exclude .Trash (or whatever) when the size of the mailbox
1087 was calculated. There was no check that a delivery was happening into an
1088 excluded directory. This bug has been fixed by ignoring all quota
1089 processing for deliveries into excluded directories.
1091 PH/10 Added the maildirfolder_create_regex option to appendfile.
1097 PH/01 The code for finding all the local interface addresses on a FreeBSD
1098 system running IPv6 was broken. This may well have applied to all BSD
1099 systems, as well as to others that have similar system calls. The broken
1100 code found IPv4 interfaces correctly, but gave incorrect values for the
1101 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1102 that it would not match correctly against @[] and not recognize the IPv6
1105 PH/02 The ipliteral router was not recognizing addresses of the form user@
1106 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1108 PH/03 Added disable_ipv6.
1110 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1111 decoded form, because it is most often used to construct To: headers
1112 lines in autoreplies, and the decoded form may well be syntactically
1113 invalid. However, $reply_address has leading white space removed, and all
1114 newlines turned into spaces so that the autoreply transport does not
1117 PH/05 If group was specified without a user on a router, and no group or user
1118 was specified on a transport, the group from the router was ignored.
1120 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1121 for visible compile-time settings that can be used to change these
1122 numbers, for those that want even more. Backwards compatibility with old
1123 spool files has been maintained. However, going back to a previous Exim
1124 release will lost any variables that are in spool files.
1126 PH/07 Two small changes when running in the test harness: increase delay when
1127 passing a TCP/IP connection to a new process, in case the original
1128 process has to generate a bounce, and remove special handling of
1129 127.0.0.2 (sic), which is no longer necessary.
1131 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1132 be the same on different OS.
1134 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1137 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1138 whether --show-vars was specified or not
1140 JJ/02 exipick: Added support for new ACL variable spool format introduced
1143 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1144 syntactically invalid From: or Reply-to: line, and a filter used this to
1145 generate an autoreply, and therefore failed to obtain an address for the
1146 autoreply, Exim could try to deliver to a non-existent relative file
1147 name, causing unrelated and misleading errors. What now happens is that
1148 it logs this as a hard delivery error, but does not attempt to create a
1151 PH/11 The exinext utility has a -C option for testing purposes, but although
1152 the given file was scanned by exinext itself; it wasn't being passed on
1153 when Exim was called.
1155 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1156 an end-of-file indication when reading a command response.
1158 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1159 compiled. In many other places in Exim, IPv6 addresses are always
1160 recognized, so I have changed this. It also means that IPv4 domain
1161 literals of the form [IPV4:n.n.n.n] are now always recognized.
1163 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1164 used if the router is not running as root, for example, when verifying at
1165 ACL time, or when using -bh. The debugging output from this situation was
1166 non-existent - all you got was a failure to exec. I have made two
1169 (a) Failures to set uid/gid, the current directory, or a process leader
1170 in a subprocess such as that created by queryprogram now generate
1171 suitable debugging ouput when -d is set.
1173 (b) The queryprogram router detects when it is not running as root,
1174 outputs suitable debugging information if -d is set, and then runs
1175 the subprocess without attempting to change uid/gid.
1177 PH/15 Minor change to Makefile for building test_host (undocumented testing
1180 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1181 additional section of a DNS packet that returns MX or SRV records.
1182 Instead, it always explicitly searches for A/AAAA records. This avoids
1183 major problems that occur when a DNS server includes only records of one
1184 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1185 fixed another bug: if SRV records were looked up and the corresponding
1186 address records were *not* found in the additional section, the port
1187 values from the SRV records were lost.
1189 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1190 using the correct key (the original address) when searching the retry
1191 rules in order to find which one to use for generating the retry hint.
1193 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1194 adding the default one. Similarly, if it contains a Reply-To: header, the
1195 errors_reply_to option, if set, is not used.
1197 PH/19 When calculating a retry time, Exim used to measure the "time since
1198 failure" by looking at the "first failed" field in the retry record. Now
1199 it does not use this if it is later than than the arrival time of the
1200 message. Instead it uses the arrival time. This makes for better
1201 behaviour in cases where some deliveries succeed, thus re-setting the
1202 "first failed" field. An example is a quota failure for a huge message
1203 when small messages continue to be delivered. Without this change, the
1204 "time since failure" will always be short, possible causing more frequent
1205 delivery attempts for the huge message than are intended.
1206 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1208 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1209 $1, $2, $3) because the numerical variables can be reset during some
1210 expansion items (e.g. "match"), thereby losing the authentication data.
1212 PH/21 Make -bV show the size of off_t variables so that the test suite can
1213 decide whether to run tests for quotas > 2G.
1215 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1216 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1217 filecount value is greater than 2G or if a quota value is greater than 2G
1218 on a system where the size of off_t is not greater than 4, a panic error
1221 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1222 never match. The debug and -bh output now contains an explicit error
1223 message indicating a malformed IPv4 address or mask.
1225 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1226 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1227 PH/23 above applies.
1229 PH/25 Do not write to syslog when running in the test harness. The only
1230 occasion when this arises is a failure to open the main or panic logs
1231 (for which there is an explicit test).
1233 PH/26 Added the /no_tell option to "control=freeze".
1235 PH/27 If a host name lookup failed very early in a connection, for example, if
1236 the IP address matched host_lookup and the reverse lookup yielded a name
1237 that did not have a forward lookup, an error message of the form "no IP
1238 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1239 could be logged. Now it outputs the IP address instead of "NULL".
1241 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1242 allows the sender and the authenticated sender to be set when
1243 submitting a message from within Exim. Since child_open_exim() is
1244 documented for local_scan(), the new function should be too.
1246 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1247 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1248 results in an empty string is now treated as unset.
1250 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1252 PH/31 Added sender_verify_fail logging option.
1254 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1255 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1256 tidied the source and removed it altogether.
1258 PH/33 When a queue run was abandoned because the load average was too high, a
1259 log line was always written; now it is written only if the queue_run log
1260 selector is set. In addition, the log line for abandonment now contains
1261 information about the queue run such as the pid. This is always present
1262 in "start" and "stop" lines but was omitted from the "abandon" line.
1264 PH/34 Omit spaces between a header name and the colon in the error message that
1265 is given when verify = headers_syntax fails (if there are lots of them,
1266 the message gets confusing).
1268 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1269 names, as there are now some PTR records that contain slashes. This check
1270 is only to protect against broken name servers that fall over on strange
1271 characters, so the fact that it applies to all lookups doesn't matter.
1273 PH/36 Now that the new test suite is complete, we can remove some of the
1274 special code in Exim that was needed for the old test suite. For example,
1275 sorting DNS records because real resolvers return them in an arbitrary
1276 order. The new test suite's fake resolver always returns records in the
1279 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1280 bounces) except when queue_only is set, to avoid logging races between
1281 the different processes.
1283 PH/38 Panic-die if .include specifies a non-absolute path.
1285 PH/39 A tweak to the "H" retry rule from its user.
1287 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1288 a label. They prevented compilation on older perls.
1290 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1291 a warning to be raised on newish perls.
1293 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1294 on queue. Changes to match documented behaviour of showing count of
1295 messages matching specified criteria.
1297 PH/40 Changed the default ident timeout from 30s to 5s.
1299 PH/41 Added support for the use of login_cap features, on those BSD systems
1300 that have them, for controlling the resources used by pipe deliveries.
1302 PH/42 The content-scanning code uses fopen() to create files in which to put
1303 message data. Previously it was not paying any attention to the mode of
1304 the files. Exim runs with umask(0) because the rest of the code creates
1305 files with open(), and sets the required mode explicitly. Thus, these
1306 files were ending up world-writeable. This was not a big issue, because,
1307 being within the spool directory, they were not world-accessible. I have
1308 created a function called modefopen, which takes an additional mode
1309 argument. It sets umask(777), creates the file, chmods it to the required
1310 mode, then resets the umask. All the relevant calls to fopen() in the
1311 content scanning code have been changed to use this function.
1313 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1314 to 24 hours. This avoids potential overflow problems when processing G
1315 and H retry rules. I suspect nobody ever tinkers with this value.
1317 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1319 PH/45 When the plaintext authenticator is running as a client, the server's
1320 challenges are checked to ensure they are valid base64 strings. By
1321 default, the authentication attempt is cancelled if an invalid string is
1322 received. Setting client_ignore_invalid_base64 true ignores these errors.
1323 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1324 they are received. Thus, the responses can be made to depend on the
1325 challenges. If an invalid string is ignored, an empty string is placed in
1328 PH/46 Messages that are created by the autoreply transport now contains a
1329 References: header, in accordance with RFCs 2822 and 3834.
1331 PH/47 Added authenticated_sender_force to the smtp transport.
1333 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1335 PH/49 Installed latest patch from the Sieve maintainer.
1337 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1338 was also set, the appendfile transport was unnecessarily scanning a
1339 directory of message files (e.g. for maildir delivery) to find the count
1340 of files (along with the size), even though it did not need this
1341 information. It now does the scan only if it needs to find either the
1342 size of the count of files.
1344 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1346 PH/52 Two bugs concerned with error handling when the smtp transport is
1349 (i) Exim was not creating retry information for temporary errors given
1350 for individual recipients after the DATA command when the smtp transport
1351 was used in LMTP mode. This meant that they could be retried too
1352 frequently, and not timed out correctly.
1354 (ii) Exim was setting the flag that allows error details to be returned
1355 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1356 recipients that were returned after the DATA command.
1358 PH/53 This is related to PH/52, but is more general: for any failing address,
1359 when detailed error information was permitted to be returned to the
1360 sender, but the error was temporary, then after the final timeout, only
1361 "retry timeout exceeded" was returned. Now it returns the full error as
1362 well as "retry timeout exceeded".
1364 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1365 do this, and (what is worse) MTAs that accept it.
1367 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1368 will now be deprecated.
1370 PH/56 New os.c-cygwin from the Cygwin maintainer.
1372 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1373 formats (previously only available in exim formats via -bpr, -bpru,
1374 and -bpra. Now also available in native and exiqgrep formats)
1376 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1377 with very large, slow to parse queues
1379 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1381 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1383 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1384 responses to authentication challenges, though it was showing the
1385 challenges; (ii) I've removed the CR characters from the debug output for
1388 PH/58 Allow for the insertion of a newline as well as a space when a string
1389 is turned into more than one encoded-word during RFC 2047 encoding. The
1390 Sieve code now uses this.
1392 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1393 data_4xx, lost_connection, tls_required.
1395 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1396 message was being sent as an SMTP response.
1398 PH/61 Add -l and -k options to exicyclog.
1400 PH/62 When verifying, if an address was redirected to one new address, so that
1401 verification continued, and the new address failed or deferred after
1402 having set something in $address_data, the value of $address_data was not
1403 passed back to the ACL. This was different to the case when no
1404 redirection occurred. The value is now passed back in both cases.
1406 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1407 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1408 use login_cap.h, so on its own it isn't the distinguishing feature. The
1409 new name refers directly to the setclassresources() function.
1411 PH/65 Added configuration files for NetBSD3.
1413 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1415 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1416 is preferred over IPv4.
1418 PH/68 The bounce_return_message and bounce_return_body options were not being
1419 honoured for bounces generated during the reception of non-SMTP messages.
1420 In particular, this applied to messages rejected by the ACL. This bug has
1421 been fixed. However, if bounce_return_message is true and bounce_return_
1422 body is false, the headers that are returned for a non-SMTP message
1423 include only those that have been read before the error was detected.
1424 (In the case of an ACL rejection, they have all been read.)
1426 PH/69 The HTML version of the specification is now built in a directory called
1427 spec_html instead of spec.html, because the latter looks like a path with
1428 a MIME-type, and this confuses some software.
1430 PH/70 Catch two compiler warnings in sieve.c.
1432 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1433 function verify_get_ident() calls ip_connect() to connect a socket, but
1434 if the "connect()" function timed out, ip_connect() used to close the
1435 socket. However, verify_get_ident() also closes the socket later, and in
1436 between Exim writes to the log, which may get opened at this point. When
1437 the socket was closed in ip_connect(), the log could get the same file
1438 descriptor number as the socket. This naturally causes chaos. The fix is
1439 not to close the socket in ip_connect(); the socket should be closed by
1440 the function that creates it. There was only one place in the code where
1441 this was missing, in the iplookup router, which I don't think anybody now
1442 uses, but I've fixed it anyway.
1444 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1445 well as to direct DNS lookups. Otherwise the handling of names in host
1446 lists is inconsistent and therefore confusing.
1452 PH/01 Two changes to the default runtime configuration:
1454 (1) Move the checks for relay_from_hosts and authenticated clients from
1455 after to before the (commented out) DNS black list checks.
1457 (2) Add control=submission to the relay_from_hosts and authenticated
1458 clients checks, on the grounds that messages accepted by these
1459 statements are most likely to be submissions.
1461 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1463 (1) Generate an error if the third argument for the ${prvs expansion is
1466 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1469 (3) Reset the variables that are obtained from the first argument of
1470 ${prvscheck and used in the second argument before leaving the code,
1471 because their memory is reclaimed, so using them afterwards may do
1474 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1475 one (it's much easier than Tom thought :-).
1477 (5) Because of (4), we can now allow for the use of $prvscheck_result
1478 inside the third argument.
1480 PH/03 For some reason, the default setting of PATH when running a command from
1481 a pipe transport was just "/usr/bin". I have changed it to
1484 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1485 anything to be listed in the output from -bV.
1487 PH/05 When a filter generated an autoreply, the entire To: header line was
1488 quoted in the delivery log line, like this:
1490 => >A.N.Other <ano@some.domain> <original@ddress> ...
1492 This has been changed so that it extracts the operative address. There
1493 may be more than one such address. If so, they are comma-separated, like
1496 => >ano@some.domain,ona@other.domain <original@ddress> ...
1498 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1499 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1500 not being looked up in the rDNS to get a host name, Exim was showing the
1501 IP address twice in Received: lines, even though the IP addresses were
1502 identical. For example:
1504 Received: from [1.2.3.4] (helo=[1.2.3.4])
1506 However, if the real host name was known, it was omitting the HELO data
1507 if it matched the actual IP address. This has been tidied up so that it
1508 doesn't show the same IP address twice.
1510 PH/07 When both +timestamp and +memory debugging was on, the value given by
1511 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1512 called by the debug printing, thereby overwriting the timestamp buffer.
1513 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1515 PH/08 When the original message was included in an autoreply transport, it
1516 always said "this is a copy of the message, including all the headers",
1517 even if body_only or headers_only was set. It now gives an appropriate
1520 PH/09 Applied a patch from the Sieve maintainer which:
1522 o fixes some comments
1523 o adds the (disabled) notify extension core
1524 o adds some debug output for the result of if/elsif tests
1525 o points to the current vacation draft in the documentation
1526 and documents the missing references header update
1530 o fixes a bug in processing the envelope test (when testing
1531 multiple envelope elements, the last element determinted the
1534 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1535 Electronic Mail") by including:
1537 Auto-submitted: auto-generated
1539 in the messages that it generates (bounce messages and others, such as
1540 warnings). In the case of bounce messages for non-SMTP mesages, there was
1541 also a typo: it was using "Auto_submitted" (underscore instead of
1542 hyphen). Since every message generated by Exim is necessarily in response
1543 to another message, thes have all been changed to:
1545 Auto-Submitted: auto-replied
1547 in accordance with these statements in the RFC:
1549 The auto-replied keyword:
1551 - SHOULD be used on messages sent in direct response to another
1552 message by an automatic process,
1554 - MUST NOT be used on manually-generated messages,
1556 - MAY be used on Delivery Status Notifications (DSNs) and Message
1557 Disposition Notifications (MDNs),
1559 - MUST NOT be used on messages generated by automatic or periodic
1560 processes, except for messages which are automatic responses to
1563 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1564 to the default Received: header definition.
1566 PH/12 Added log selector acl_warn_skipped (default on).
1568 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1569 variables because (a) they are in the wrong storage pool and (b) even if
1570 they were copied, it wouldn't work properly because of the caching.
1572 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1573 checking when decoding. Apparently there are clients that generate
1574 overlong encoded strings. Why am I not surprised?
1576 PH/15 If the first argument of "${if match_address" was not empty, but did not
1577 contain an "@" character, Exim crashed. Now it writes a panic log message
1578 and treats the condition as false.
1580 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1582 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1583 extension "envelope-auth". The code is finished and in agreement with
1584 other implementations, but there is no documentation so far and in fact,
1585 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1586 not changing the active code.
1588 Print executed "if" and "elsif" statements when debugging is used. This
1589 helps a great deal to understand what a filter does.
1591 Document more things not specified clearly in RFC3028. I had all this
1592 sorted out, when out of a sudden new issues came to my mind. Oops."
1594 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1597 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1598 canonical form (as documented). However, after a host name lookup from
1599 the IP address, check_host() was doing a simple string comparison with
1600 addresses acquired from the DNS when checking that the found name did
1601 have the original IP as one of its addresses. Since any found IPv6
1602 addresses are likely to be in abbreviated form, the comparison could
1603 fail. Luckily, there already exists a function for doing the comparison
1604 by converting both addresses to binary, so now that is used instead of
1605 the text comparison.
1607 PH/20 There was another similar case to PH/19, when a complete host name was
1608 given in a host list; looking up its IP address could give an abbreviated
1609 form, whereas the current host's name might or might not be abbreviated.
1610 The same fix has been applied.
1616 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1617 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1620 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1621 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1623 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1625 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1626 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1627 Also, for "Auto-Submitted", treat anything other than "no" as
1628 non-personal, in accordance with RFC 3834. (Previously it treated
1629 anything starting "auto-" as non-personal.)
1631 TF/01 The control=submission/name=... option had a problem with syntax
1632 errors if the name included a slash character. The /name= option
1633 now slurps the rest of the string, so it can include any characters
1634 but it must come last in the list of options (after /sender_retain
1637 PH/05 Some modifications to the interface to the fake nameserver for the new
1645 TK/01 Added the "success_on_redirect" address verification option. See
1646 NewStuff for rationale and an example.
1648 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1650 PH/02 Patch to exigrep to allow it to work on syslog lines.
1652 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1653 fread() to skip over the body file's header line, because in Cygwin the
1654 header line is locked and is inaccessible.
1656 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1657 co-exist for some time) to make it clear that it is the Exim ID that is
1658 referenced, not the Message-ID: header line.
1660 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1661 string_format() function, because snprintf() does not exist on all
1664 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1665 ${stat: expansion item.
1667 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1668 protocol synchronization error", to keep the pedants happy.
1670 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1671 well as for IRIX systems, when gcc is being used. See the host.c source
1674 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1676 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1679 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1680 transport and to the smtp transport in LMTP mode.
1682 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1684 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1685 run a filter in a subprocess. This could lead to confusion in subsequent
1686 lookups in the parent process. There should also be a search_tidyup() at
1687 the end of the subprocess.
1689 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1690 only if the host matched helo_try_verify_hosts, which caused the
1691 verification to occur when the EHLO/HELO command was issued. The ACL just
1692 tested the remembered result. Now, if a previous verification attempt has
1693 not happened, "verify = helo" does it there and then.
1695 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1697 TK/03 Fix log output including CR from clamd.
1699 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1700 exist provoked a memory error which could cause a segfault.
1702 PH/15 Installed PCRE 6.2
1704 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1706 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1707 of the problem. Specifically, suggested +O2 rather than +O1 for the
1710 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1712 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1713 record, the port was not correctly logged when the outgoing_port log
1714 selector was set (it logged the transort's default port).
1716 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1717 fallback_hosts, and "hosts" in the smtp transport.
1719 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1720 host errors such as "Connection refused".
1722 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1723 authentication with radiusclient 0.4.9:
1725 - Error returned from rc_read_config was caught wrongly
1726 - Username/password not passed on to radius server due to wrong length.
1728 The presumption is that some radiusclient API changes for 4.51/PH/17
1729 were not taken care of correctly. The code is still untested by me (my
1730 Linux distribution still has 0.3.2 of radiusclient), but it was
1731 contributed by a Radius user.
1733 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1734 expanding the "port" option of the smtp transport.
1736 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1737 while reading a MIME header. Thanks to Tom Hughes for a patch.
1739 PH/24 Include config.h inside local_scan.h so that configuration settings are
1742 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1743 that in an ACL for RCPT (for example), you can examine exactly what was
1746 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1747 commands, but it was not correctly comparing the address with the actual
1748 client host address. Thus, it would show the EHLO address in Received:
1749 header lines when this was not necessary.
1751 PH/27 Added the % operator to ${eval:}.
1753 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1754 it should be ignoring failures (because with -C, for example, it has lost
1755 privilege). It wasn't ignoring creation failures other than "already
1758 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1761 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1762 because an input line was too long, either on its own, or by virtue of
1763 too many continuations, the temporary file was not being removed, and the
1764 return code was incorrect.
1766 PH/31 Missing "BOOL" in function definition in filtertest.c.
1768 PH/32 Applied Sieve patches from the maintainer.
1770 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1772 PH/33 Added "verify = not_blind".
1774 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1775 Local/Makefile (with some defaults set). These are used in built scripts
1776 such as exicyclog, but they have never been used in the exim_install
1777 script (though there are many overriding facilities there). I have
1778 arranged that the exim_install script now takes note of these two
1781 PH/35 Installed configuration files for Dragonfly.
1783 PH/36 When a locally submitted message by a trusted user did not contain a
1784 From: header, and the sender address was obtained from -f or from an SMTP
1785 MAIL command, and the trusted user did not use -F to supply a sender
1786 name, $originator_name was incorrectly used when constructing a From:
1787 header. Furthermore, $originator_name was used for submission mode
1788 messages from external hosts without From: headers in a similar way,
1789 which is clearly wrong.
1791 PH/37 Added control=suppress_local_fixups.
1793 PH/38 When log_selector = +received_sender was set, and the addition of the
1794 sender made the log line's construction buffer exactly full, or one byte
1795 less than full, an overflow happened when the terminating "\n" was
1798 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1799 when the result of a list match is failure because a DNS lookup failed.
1801 PH/40 RM_COMMAND is now used in the building process.
1803 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1804 the "build-* directories that it finds.
1806 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1807 address in a domain literal was a prefix of an interface address.
1809 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1810 when verifying a sender address, unless rewrite_headers is false.
1812 PH/44 Wrote a long comment about why errors_to addresses are verified as
1813 recipients, not senders.
1815 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1816 the ratelimit ACL was added.
1818 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1820 PH/46 Added extra information about PostgreSQL errors to the error string.
1822 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1823 avoiding the need to install special zones in a real server. This is
1824 backwards compatible; if it can't find the fake resolver, it drops back.
1825 Thus, both old and new test suites can be run.
1827 TF/02 Added util/ratelimit.pl
1829 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1830 clock is set back in time.
1832 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1833 Candler <B.Candler@pobox.com>.
1835 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1836 OK for addresses that are the result of redirections.
1838 PH/48 A number of further additions for the benefit of the new test suite,
1839 including a fake gethostbyname() that interfaces to the fake DNS resolver
1842 TF/06 The fix for widen_domains has also been applied to qualify_single and
1843 search_parents which are the other dnslookup options that can cause
1846 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1849 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1851 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1852 service" if it encountered an MX record with an empty target hostname.
1853 The message is now "an MX or SRV record indicated no SMTP service".
1855 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1856 if the DNS of the sending site is misconfigured. This is quite a
1857 common situation. This change restores the behaviour of treating a
1858 helo verification defer as a failure.
1860 PH/51 If self=fail was set on a router, the bounce message did not include the
1861 actual error message.
1867 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1869 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1870 command itself ended in error, the underlying message about the transport
1871 filter timeout was being overwritten with the pipe command error. Now the
1872 underlying error message should be appended to the second error message.
1874 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1876 PH/02 Reduce the amount of output that "make" produces by default. Full output
1877 can still be requested.
1879 PH/03 The warning log line about a condition test deferring for a "warn" verb
1880 was being output only once per connection, rather than after each
1881 occurrence (because it was using the same function as for successful
1882 "warn" verbs). This seems wrong, so I have changed it.
1884 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1885 it should not have, which might have caused a crash in the right
1886 circumstances, but probably never did.
1888 PH/04 Installed a modified version of Tony Finch's patch to make submission
1889 mode fix the return path as well as the Sender: header line, and to
1890 add a /name= option so that you can make the user's friendly name appear
1893 TF/03 Added the control = fakedefer ACL modifier.
1895 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1896 Mark Lowes for thorough testing.
1898 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1900 TK/03 Merged latest SRS patch from Miles Wilton.
1902 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1903 with the definition in sysexits.h (which is #included earlier).
1904 Fortunately, Exim does not actually use EX_OK. The code used to try to
1905 preserve the sysexits.h value, by assumimg that macro definitions were
1906 scanned for macro replacements. I have been disabused of this notion,
1907 so now the code just undefines EX_OK before #including unistd.h.
1909 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1910 in the smtp transport. When a block could not be written in a single
1911 write() function, the timeout was being re-applied to each part-write.
1912 This seems wrong - if the receiver was accepting one byte at a time it
1913 would take for ever. The timeout is now adjusted when this happens. It
1914 doesn't have to be particularly precise.
1916 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1917 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1919 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1920 without a domain if the check to <postmaster@domain> fails.
1922 SC/01 Eximstats: added -xls and the ability to specify output files
1923 (patch written by Frank Heydlauf).
1925 SC/02 Eximstats: use FileHandles for outputing results.
1927 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1929 SC/04 Eximstats: fixed display of large numbers with -nvr option
1931 SC/05 Eximstats: fixed merging of reports with empty tables.
1933 SC/06 Eximstats: added the -include_original_destination flag
1935 SC/07 Eximstats: removed tabs and trailing whitespace.
1937 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1939 TK/06 MBOX spool code: Add real "From " MBOX separator line
1940 so the .eml file is really in mbox format (even though
1941 most programs do not really care). Patch from Alex Miller.
1943 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1944 The latter is generated from $received_to and is only set if the
1945 message has one envelope recipient. SA can use these headers,
1946 obviously out-of-the-box. Patch from Alex Miller.
1948 PH/08 The ${def test on a variable was returning false if the variable's
1949 value was "0", contrary to what the specification has always said!
1950 The result should be true unless the variable is empty.
1952 PH/09 The syntax error of a character other than { following "${if
1953 def:variable_name" (after optional whitespace) was not being diagnosed.
1954 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1955 accidental colon was present, for example, could give incorrect results.
1957 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1958 result is used (not including appendfile, where other changes are about
1961 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1962 This involved changing a lot of size variables from int to off_t. It
1963 should work with maildirs and everything.
1965 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1966 spamd dying while we are connected to it.
1968 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1971 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1972 function for BDB 4.3.
1974 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1976 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1977 the sources for PCRE have been re-arranged and more files are now
1980 PH/15 The code I had for printing potentially long long variables in PH/11
1981 above was not the best (it lost precision). The length of off_t variables
1982 is now inspected at build time, and an appropriate printing format (%ld
1983 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1984 to be "long long int" or "long int". This is needed for the internal
1985 formatting function string_vformat().
1987 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1988 the configuration file to be ":syslog", then the script "guesses" where
1989 the logs files are, rather than using the compiled in default. In our
1990 case the guess is not the same as the compiled default, so the script
1991 suddenly stopped working when I started to use syslog. The patch checks
1992 to see if log_file_path is "". If so, it attempts to read it from exim
1993 with no configuration file to get the compiled in version, before it
1994 falls back to the previous guessing code."
1996 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1997 implementing BATV in an Exim configuration. See NewStuff for the gory
2000 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
2001 Makefile that are specific to HP-UX.
2003 PH/18 If the "use_postmaster" option was set for a recipient callout together
2004 with the "random" option, the postmaster address was used as the MAIL
2005 FROM address for the random test, but not for the subsequent recipient
2006 test. It is now used for both.
2008 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2009 patch removes a few documentation additions to RFC 3028, because the
2010 latest draft now contains them. It adds the new en;ascii-case comparator
2011 and a new error check for 8bit text in MIME parts. Comparator and
2012 require names are now matched exactly. I enabled the subaddress
2013 extension, but it is not well tested yet (read: it works for me)."
2015 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2016 rework some of the code of TK/09 above to avoid the hardwired use of
2017 "%lld" and "long long". Replaced the call to snprintf() with a call to
2020 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2021 records point to non-existent hosts", "retry timeout exceeded", and
2022 "retry time not reached for any host after a long failure period".
2024 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2025 experimental DomainKeys support:
2027 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2028 (2) On an error such as an illegally used "control", the wrong name for
2029 the control was given.
2031 These problems did NOT occur unless DomainKeys support was compiled.
2033 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2035 PH/24 Added ${if match_ip condition.
2037 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2038 databases so that it will be absolutely obvious if a crash occurs in the
2039 DB library. This is a regular occurrence (often caused by mis-matched
2042 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2043 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2044 picked up on a user's system that detects such things. There doesn't seem
2045 to be a gcc warning option for this - only an attribute that has to be
2046 put on the function's prototype. It seems that in Fedora Core 4 they have
2047 set this on a number of new functions. No doubt there will be more in due
2050 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2051 specify a transport. However, if an address that was verified by such a
2052 router was the subject of a callout, Exim crashed because it tried to
2053 read the rcpt_include_affixes from the non-existent transport. Now it
2054 just assumes that the setting of that option is false. This bug was
2055 introduced by 4.51/PH/31.
2057 PH/28 Changed -d+all to exclude +memory, because that information is very
2058 rarely of interest, but it makes the output a lot bigger. People tend to
2059 do -d+all out of habit.
2061 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2062 code in os-type was giving problems when libc.so lives in lib64, like on
2065 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2066 aren't the modern standard, and it seems that some systems' include files
2067 don't always have them. Exim was already checking for some of the newer
2068 ones like T_AAAA, and defining it itself. I've added checks for all the
2069 record types that Exim uses.
2071 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2072 not automatically generating a new one, as it is supposed to. This
2073 prevented TLS from working. If the file did exist, but contained invalid
2074 data, a new version was generated, as expected. It was only the case of a
2075 non-existent file that was broken.
2077 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2078 with a change in libdomainkeys > 0.64.
2080 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2081 from DNS. If the selector record carries the flag, it now has
2082 precedence over the domain-wide flag.
2084 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2086 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2087 the use of a port name that isn't defined in /etc/services) occurred, the
2088 message was deferred as in a normal delivery, and thus remained on the
2089 spool, instead of being failed because of the mua_wrapper setting. This
2090 is now fixed, and I tidied up some of the mua_wrapper messages at the
2093 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2094 the messages in a hash of arrays rather than using individual hashes.
2095 This is a bit cleaner and results in dramatic memory savings, albeit
2096 at a slight CPU cost.
2098 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2099 as requested by Marc Sherman.
2101 SC/10 Eximstats: added histograms for user specified patterns as requested
2104 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2106 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2107 fopen() in the content-scanning modules that did not already have it.
2113 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2114 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2116 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2118 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2120 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2121 to test Sieve filters that use "vacation".
2123 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2124 that changes the way the GnuTLS parameters are stored in the cache file.
2125 The new format can be generated externally. For backward compatibility,
2126 if the data in the cache doesn't make sense, Exim assumes it has read an
2127 old-format file, and it generates new data and writes a new file. This
2128 means that you can't go back to an older release without removing the
2131 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2132 work if there are any delivery delays because "one_time" forces the
2133 parent to be marked "delivered", so its unseen clone is never tried
2134 again. For this reason, Exim now forbids the simultaneous setting of
2137 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2138 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2139 entry if you want to know the details. The fix, however, neglected to
2140 consider the case when local delivery batching is involved. The test for
2141 "previously delivered" was not happening when checking to see if an
2142 address could be batched with a previous (undelivered) one; under
2143 certain circumstances this could lead to multiple deliveries to the same
2146 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2147 in its include files, and this causes problems building Exim.
2149 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2150 header_syntax) but Exim was just ignoring anything given after a slash.
2151 In particular, this caused confusion with an attempt to use "verify =
2152 reverse_host_lookup/defer_ok". An error is now given when options are
2153 supplied for verify items that do not have them. (Maybe reverse_host_
2154 lookup should have a defer_ok option, but that's a different point.)
2156 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2157 defined by RFC 821) to 2048, because there were problems with some AUTH
2158 commands, and RFC 1869 says the size should be increased for extended
2159 SMTP commands that take arguments.
2161 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2164 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2165 "unknown" error; now it says that the functionality isn't in the binary.
2167 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2168 an address' error message when a string expansion fails (syntax or
2169 whatever). Otherwise the password may appear in the log. Following change
2170 PH/42 below, there is no longer a chance of it appearing in a bounce
2173 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2175 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2176 of its addresses was ever tried. (Bugzilla bug #2.)
2178 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2179 the result incorrectly in the debug output. (It correctly added a newline
2180 to what was transported.)
2182 TF/01 Added $received_time.
2184 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2185 commented out examples of how to interface to a virus scanner and to
2186 SpamAssassin. Also added commented examples of av_scanner and
2187 spamd_address settings.
2189 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2190 and controls are allowed in which ACLs. There were a couple of minor
2191 errors. Some of the entries in the conditions table (which is a table of
2192 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2193 negation of where the condition IS allowed.
2195 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2197 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2198 header file does not have a version number, so I've had to invent a new
2199 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2200 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2201 radiusclient), but it was contributed by a Radius user.
2203 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2204 files or directories, for OpenSSL.
2206 PH/19 When an Exim process that is running as root has to create an Exim log
2207 file, it does so in a subprocess that runs as exim:exim so as to get the
2208 ownership right at creation (otherwise, other Exim processes might see
2209 the file with the wrong ownership). There was no test for failure of this
2210 fork() call, which would lead to the process getting stuck as it waited
2211 for a non-existent subprocess. Forks do occasionally fail when resources
2212 run out. I reviewed all the other calls to fork(); they all seem to check
2215 PH/20 When checking for unexpected SMTP input at connect time (before writing
2216 the banner), Exim was not dealing correctly with a non-positive return
2217 from the read() function. If the client had disconnected by this time,
2218 the result was a log entry for a synchronization error with an empty
2219 string after "input=" when read() returned zero. If read() returned -1
2220 (an event I could not check), uninitialized data bytes were printed.
2221 There were reports of junk text (parts of files, etc) appearing after
2224 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2226 PH/22 Added support for macro redefinition, and (re)definition in between
2227 driver and ACL definitions.
2229 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2230 forgetting to use the resulting value; it was using the unexpanded value.
2232 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2233 hadn't been configured. The fix is from Juergen Kreileder, who
2234 understands it better than I do:
2236 "Here's what I see happening with three configured cyrus_sasl
2237 authenticators configured (plain, login, cram-md5):
2239 On startup auth_cyrus_sasl_init() gets called for each of these.
2240 This means three calls to sasl_listmech() without a specified mech_list.
2241 => SASL tests which mechs of all available mechs actually work
2242 => three warnings about OTP not working
2243 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2245 With the patch, sasl_listmech() also gets called three times. But now
2246 SASL's mech_list option is set to the server_mech specified in the the
2247 authenticator. Or in other words, the answer from sasl_listmech()
2248 gets limited to just the mech you're testing for (which is different
2250 => the return list contains just 'plain' or 'login', 'cram-md5' or
2251 nothing depending on the value of ob->server_mech.
2253 I've just tested the patch: Authentication still works fine,
2254 unavailable mechs specified in the exim configuration are still
2255 caught, and the auth.log warnings about OTP are gone."
2257 PH/25 When debugging is enabled, the contents of the command line are added
2258 to the debugging output, even when log_selector=+arguments is not
2261 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2262 answer is "GNU", and only if the return is "GNU/something" is the answer
2265 PH/27 $acl_verify_message is now set immediately after the failure of a
2266 verification in an ACL, and so is available in subsequent modifiers. In
2267 particular, the message can be preserved by coding like this:
2269 warn !verify = sender
2270 set acl_m0 = $acl_verify_message
2272 Previously, $acl_verify_message was set only while expanding "message"
2273 and "log_message" when a very denied access.
2275 PH/28 Modified OS/os.c-Linux with
2277 -#ifndef OS_LOAD_AVERAGE
2278 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2280 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2281 nomenclature these days.)
2283 PH/29 Installed patch from the Sieve maintainer that adds the options
2284 sieve_useraddress and sieve_subaddress to the redirect router.
2286 PH/30 In these circumstances:
2287 . Two addresses routed to the same list of hosts;
2288 . First host does not offer TLS;
2289 . First host accepts first address;
2290 . First host gives temporary error to second address;
2291 . Second host offers TLS and a TLS session is established;
2292 . Second host accepts second address.
2293 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2294 and peerdn, if requested) that were in fact used only for the second
2297 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2298 attention to any local part prefix or suffix that was matched by the
2299 router that accepted the address. It now behaves in the same way as it
2300 does for delivery: the affixes are removed from the local part unless
2301 rcpt_include_affixes is set on the transport.
2303 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2304 timeout during the DATA phase of an incoming message.
2306 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2307 applied a patch sanctioned by the Sieve maintainer.
2309 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2310 the uid or gid is negative. A case of a negative gid caused this to be
2311 noticed. The fix allows for either to be negative.
2313 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2314 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2317 PH/36 The change PH/12 above was broken. Fixed it.
2319 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2320 the grounds that routing the same address twice would always produce the
2321 same answer. This might have been true once, but it is certainly no
2322 longer true now. Routing a child address may depend on the previous
2323 routing that produced that child. Some complicated redirection strategies
2324 went wrong when messages had multiple recipients, and made Exim's
2325 behaviour dependent on the order in which the addresses were given.
2327 I have moved the duplicate checking until after the routing is complete.
2328 Exim scans the addresses that are assigned to local and remote
2329 transports, and removes any duplicates. This means that more work will be
2330 done, as duplicates will always all be routed, but duplicates are
2331 presumably rare, so I don't expect this is of any significance.
2333 For deliveries to pipes, files, and autoreplies, the duplicate checking
2334 still happens during the routing process, since they are not going to be
2337 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2338 It corrects a timeout issue with spamd. This is Ian's comment: "The
2339 background is that sometimes spamd either never reads data from a
2340 connection it has accepted, or it never writes response data. The exiscan
2341 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2342 blindly assumes that writes won't block so it may never time out."
2344 PH/39 Allow G after quota size as well as K and M.
2346 PH/40 The value set for $authenticated_id in an authenticator may not contain
2347 binary zeroes or newlines because the value is written to log lines and
2348 to spool files. There was no check on this. Now the value is run through
2349 the string_printing() function so that such characters are converted to
2350 printable escape sequences.
2352 PH/41 $message_linecount is a new variable that contains the total number of
2353 lines in the message. Compare $body_linecount, which is the count for the
2356 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2357 bounce and delay warning messages, except in certain special cases, which
2360 (a) An SMTP error message from a remote host;
2361 (b) A message specified in a :fail: redirection;
2362 (c) A message specified in a "fail" command in a system filter;
2363 (d) A message specified in a FAIL return from the queryprogram router;
2364 (e) A message specified by the cannot_route_message router option.
2366 In these cases only, Exim does include the error details in bounce and
2367 warning messages. There are also a few cases where bland messages such
2368 as "unrouteable address" or "local delivery error" are given.
2370 PH/43 $value is now also set for the "else" part of a ${run expansion.
2372 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2373 being worked on, but at least Exim now implements the latest version to
2376 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2377 process to complete was treated as a delivery failure, a timeout while
2378 writing the message to the pipe was logged, but erroneously treated as a
2379 successful delivery. Such timeouts include transport filter timeouts. For
2380 consistency with the overall process timeout, these timeouts are now
2381 treated as errors, giving rise to delivery failures by default. However,
2382 there is now a new Boolean option for the pipe transport called
2383 timeout_defer, which, if set TRUE, converts the failures into defers for
2384 both kinds of timeout. A transport filter timeout is now identified in
2387 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2388 systems where "make" and "gmake" are different, calling "gmake" at top
2389 level broke things. I've arranged for the value of $(MAKE) to be passed
2390 from the Makefile to this script so that it can call the same version of
2394 A note about Exim versions 4.44 and 4.50
2395 ----------------------------------------
2397 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2398 changes of various kinds. As a consequence, a big documentation update was
2399 needed. This delayed the release for rather longer than seemed good, especially
2400 in the light of a couple of (minor) security issues. Therefore, the changes
2401 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2402 release. So 4.44 and 4.50 are in effect two different branches that both start
2405 I have left the 4.50 change log unchanged; it contains all the changes since
2406 4.43. The change log for 4.44 is below; many of its items are identical to
2407 those for 4.50. This seems to be the most sensible way to preserve the
2408 historical information.
2414 1. Minor wording change to the doc/README.SIEVE file.
2416 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2417 computation of the current number of files was incorrect.
2419 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2420 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2423 4. Give more explanation in the error message when the command for a transport
2424 filter fails to execute.
2426 5. There are several places where Exim runs a non-Exim command in a
2427 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2428 was being done only for the command run by the queryprogram router. It is
2429 now done for all such subprocesses. The other cases are: ${run, transport
2430 filters, and the commands run by the lmtp and pipe transports.
2432 6. Added CONFIGURE_GROUP build-time option.
2434 7. Some older OS have a limit of 256 on the maximum number of file
2435 descriptors. Exim was using setrlimit() to set 1000 as a large value
2436 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2437 systems. I've change it so that if it can't get 1000, it tries for 256.
2439 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2440 was an oversight, and furthermore, ever since the addition of extra
2441 controls (e.g. 4.43/32), the checks on when to allow different forms of
2442 "control" were broken. There should now be diagnostics for all cases when a
2443 control that does not make sense is encountered.
2445 9. Added the /retain_sender option to "control=submission".
2447 10. $recipients is now available in the predata ACL (oversight).
2449 11. Tidy the search cache before the fork to do a delivery from a message
2450 received from the command line. Otherwise the child will trigger a lookup
2451 failure and thereby defer the delivery if it tries to use (for example) a
2452 cached ldap connection that the parent has called unbind on.
2454 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2455 of $address_data from the recipient verification was clobbered by the
2456 sender verification.
2458 13. The value of address_data from a sender verification is now available in
2459 $sender_address_data in subsequent conditions in the ACL statement.
2461 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2463 15. Added a new option "connect=<time>" to callout options, to set a different
2466 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2467 was its contents. (It was OK if the option was not defined at all.)
2469 17. A "Completed" log line is now written for messages that are removed from
2470 the spool by the -Mrm option.
2472 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2473 information about exactly what failed.
2475 19. Added -dd to debug only the daemon process.
2477 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2478 handles timeouts, both on the server side and network timeouts. Renamed the
2479 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2481 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2482 It is now set to "smtps".
2484 22. $host_address is now set to the target address during the checking of
2485 ignore_target_hosts.
2487 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2488 being passed; this would have caused $sender_host_name to have been used if
2489 matching the list had actually called for a host name (not very likely,
2490 since this list is usually IP addresses). A host name is now passed as
2493 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2494 code that specifies a non-restarting handler (typically sigaction() in
2495 modern systems) in an attempt to fix a rare and obscure crash bug.
2497 25. Narrowed the window for a race in the daemon that could cause it to ignore
2498 SIGCHLD signals. This is not a major problem, because they are used only to
2499 wake it up if nothing else does.
2501 26. A malformed maildirsize file could cause Exim to calculate negative values
2502 for the mailbox size or file count. Odd effects could occur as a result.
2503 The maildirsize information is now recalculated if the size or filecount
2506 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2507 support for a long time. Removed HAVE_SYS_VFS_H.
2509 28. Installed the latest version of exipick from John Jetmore.
2511 29. In an address list, if the pattern was not a regular expression, an empty
2512 subject address (from a bounce message) matched only if the pattern was an
2513 empty string. Non-empty patterns were not even tested. This was the wrong
2514 because it is perfectly reasonable to use an empty address as part of a
2515 database query. An empty address is now tested by patterns that are
2516 lookups. However, all the other forms of pattern expect the subject to
2517 contain a local part and a domain, and therefore, for them, an empty
2518 address still always fails if the pattern is not itself empty.
2520 30. Exim went into a mad DNS loop when attempting to do a callout where the
2521 host was specified on an smtp transport, and looking it up yielded more
2522 than one IP address.
2524 31. Re-factored the code for checking spool and log partition space into a
2525 function that finds that data and another that does the check. The former
2526 is then used to implement four new variables: $spool_space, $log_space,
2527 $spool_inodes, and $log_inodes.
2529 32. The RFC2047 encoding function was originally intended for short strings
2530 such as real names; it was not keeping to the 75-character limit for
2531 encoded words that the RFC imposes. It now respects the limit, and
2532 generates multiple encoded words if necessary. To be on the safe side, I
2533 have increased the buffer size for the ${rfc2047: expansion operator from
2536 33. It is now permitted to omit both strings after an "if" condition; if the
2537 condition is true, the result is "true". As before, when the second string
2538 is omitted, a false condition yields an empty string. This makes it less
2539 cumbersome to write custom ACL and router conditions.
2541 34. Failure to deliver a bounce message always caused it to be frozen, even if
2542 there was an errors_to setting on the router. The errors_to setting is now
2545 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2546 canonical form (fully expanded) before being placed in
2547 $sender_host_address.
2549 36. The table in the code that translates DNS record types into text (T_A to
2550 "A" for instance) was missing entries for NS and CNAME. It is just possible
2551 that this could have caused confusion if both these types were looked up
2552 for the same domain, because the text type is used as part of Exim's
2553 per-process caching. But the chance of anyone hitting this buglet seems
2556 37. The dnsdb lookup has been extended in a number of ways.
2558 (1) There is a new type, "zns", which walks up the domain tree until it
2559 finds some nameserver records. It should be used with care.
2561 (2) There is a new type, "mxh", which is like "mx" except that it returns
2562 just the host names, not the priorities.
2564 (3) It is now possible to give a list of domains (or IP addresses) to be
2565 looked up. The behaviour when one of the lookups defers can be
2566 controlled by a keyword.
2568 (4) It is now possible to specify the separator character for use when
2569 multiple records are returned.
2571 38. The dnslists ACL condition has been extended: it is now possible to supply
2572 a list of IP addresses and/or domains to be looked up in a particular DNS
2575 39. Added log_selector=+queue_time_overall.
2577 40. When running the queue in the test harness, wait just a tad after forking a
2578 delivery process, to get repeatability of debugging output.
2580 41. Include certificate and key file names in error message when GnuTLS fails
2581 to set them up, because the GnuTLS error message doesn't include the name
2582 of the failing file when there is a problem reading it.
2584 42. Allow both -bf and -bF in the same test run.
2586 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2588 44. The "Exiscan patch" is now merged into the mainline Exim source.
2590 45. Sometimes the final signoff response after QUIT could fail to get
2591 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2592 < 0 before doing a fflush(). This bug looks as though it goes back to the
2593 introduction of TLS in release 3.20, but "sometimes" must have been rare
2594 because the tests only now provoked it.
2596 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2597 (this can affect the format of dates).
2599 47. exim_tidydb, when checking for the continued existence of a message for
2600 which it has found a message-specific retry record, was not finding
2601 messages that were in split spool directories. Consequently, it was
2602 deleting retry records that should have stayed in existence.
2604 48. Steve fixed some bugs in eximstats.
2606 49. The SPA authentication driver was not abandoning authentication and moving
2607 on to the next authenticator when an expansion was forced to fail,
2608 contradicting the general specification for all authenticators. Instead it
2609 was generating a temporary error. It now behaves as specified.
2611 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2612 (the order specifies the preference for clients). The order is now AES256,
2613 AES128, 3DES, ARCFOUR128.
2615 51. Small patch to Sieve code - explicitly set From: when generating an
2618 52. Exim crashed if a remote delivery caused a very long error message to be
2619 recorded - for instance if somebody sent an entire SpamAssassin report back
2620 as a large number of 550 error lines. This bug was coincidentally fixed by
2621 increasing the size of one of Exim's internal buffers (big_buffer) that
2622 happened as part of the Exiscan merge. However, to be on the safe side, I
2623 have made the code more robust (and fixed the comments that describe what
2626 53. Now that there can be additional text after "Completed" in log lines (if
2627 the queue_time_overall log selector is set), a one-byte patch to exigrep
2628 was needed to allow it to recognize "Completed" as not the last thing in
2631 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2632 patch that reportedly fixes this has been added. I am not expert enough to
2633 create a test for it. This is what the patch creator wrote:
2635 "I found a little strange behaviour of ldap code when working with
2636 Windows 2003 AD Domain, where users was placed in more than one
2637 Organization Units. When I tried to give exim partial DN, the exit code
2638 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2639 But simultaneously result of request was absolutely normal ldap result,
2640 so I produce this patch..."
2642 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2643 REFERENCE, so I have modified the code to exclude the patch when that macro
2646 55. Some experimental protocols are using DNS PTR records for new purposes. The
2647 keys for these records are domain names, not reversed IP addresses. The
2648 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2649 leaves it alone. Component reversal etc. now happens only for IP addresses.
2652 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2654 57. Double the size of the debug message buffer (to 2048) so that more of very
2655 long debug lines gets shown.
2657 58. The exicyclog utility now does better if the number of log files to keep
2658 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2660 59. Two changes related to the smtp_active_hostname option:
2662 (1) $smtp_active_hostname is now available as a variable.
2663 (2) The default for smtp_banner uses $smtp_active_hostname instead
2664 of $primary_hostname.
2666 60. The host_aton() function is supposed to be passed a string that is known
2667 to be a valid IP address. However, in the case of IPv6 addresses, it was
2668 not checking this. This is a hostage to fortune. Exim now panics and dies
2669 if the condition is not met. A case was found where this could be provoked
2670 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2671 components; fortuitously, this particular loophole had already been fixed
2672 by change 4.50/55 above.
2674 If there are any other similar loopholes, the new check in host_aton()
2675 itself should stop them being exploited. The report I received stated that
2676 data on the command line could provoke the exploit when Exim was running as
2677 exim, but did not say which command line option was involved. All I could
2678 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2679 running as the user.
2682 61. There was a buffer overflow vulnerability in the SPA authentication code
2683 (which came originally from the Samba project). I have added a test to the
2684 spa_base64_to_bits() function which I hope fixes it.
2687 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2688 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2690 63. The daemon start-up calls getloadavg() while still root for those OS that
2691 need the first call to be done as root, but it missed one case: when
2692 deliver_queue_load_max is set with deliver_drop_privilege. This is
2693 necessary for the benefit of the queue runner, because there is no re-exec
2694 when deliver_drop_privilege is set.
2696 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2697 This has been fixed.
2699 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2700 was in use, was not putting the data itself into the right store pool;
2701 consequently, it could be overwritten for a subsequent message in the same
2702 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2705 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2707 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2708 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2709 the function were treating the return as a boolean value, which happened to
2710 work because 0=false and not-0=true, but is not correct code.
2712 68. The host_aton() function was not handling scoped IPv6 addresses (those
2713 with, for example, "%eth0" on the end) correctly.
2715 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2716 negated items (that is, ~something) in unsigned ints. Some compilers
2717 apparently mutter when there is no cast.
2719 70. If an address verification called from an ACL failed, and did not produce a
2720 user-specific message (i.e. there was only a "system" message), nothing was
2721 put in $acl_verify_message. In this situation, it now puts the system
2724 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2725 session; change 4.31/43 added the unwanted input to the log line - except
2726 that it did not do this in the start of session case. It now does.
2728 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2729 This is wrong and can cause the other end to generate a synchronization
2730 error if it is another Exim or anything else that does the synchronization
2731 check. A QUIT command is no longer sent after a timeout.
2733 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2734 during host lookups.
2736 74. The defer_ok option of callout verification was not working if it was used
2737 when verifying addresses in header lines, that is, for this case:
2739 verify = header_sender/callout=defer_ok
2741 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2742 those file descriptors could be used for SMTP connections. If anything
2743 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2744 could be sent to the SMTP client, causing chaos. The daemon now opens
2745 stdin, stdout, and stderr to /dev/null when it puts itself into the
2748 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2749 log by default. The user can override this with suitable Perl magic.
2751 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2752 the log message when discard triggers, was not working for the DATA ACL or
2753 for the non-SMTP ACL.
2755 78. Error message wording change in sieve.c.
2757 79. If smtp_accept_max_per_host was set, the number of connections could be
2758 restricted to fewer than expected, because the daemon was trying to set up
2759 a new connection before checking whether the processes handling previous
2760 connections had finished. The check for completed processes is now done
2761 earlier. On busy systems, this bug wouldn't be noticed because something
2762 else would have woken the daemon, and it would have reaped the completed
2765 80. If a message was submitted locally by a user whose login name contained one
2766 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2767 It caused a spool format error. I have fixed the spool reading code. A
2768 related problem was that the "from" clause in the Received: line became
2769 illegal because of the space(s). It is now covered by ${quote_local_part.
2771 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2774 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2776 83. Patch from Sieve maintainer to fix unterminated string problem in
2777 "vacation" handling.
2779 84. Some minor changes to the Linux configuration files to help with other
2780 OS variants using glibc.
2782 85. One more patch for Sieve to update vacation handling to latest spec.
2785 ----------------------------------------------------
2786 See the note above about the 4.44 and 4.50 releases.
2787 ----------------------------------------------------
2793 1. Change 4.43/35 introduced a bug that caused file counts to be
2794 incorrectly computed when quota_filecount was set in an appendfile
2797 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2798 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2801 3. Give more explanation in the error message when the command for a transport
2802 filter fails to execute.
2804 4. There are several places where Exim runs a non-Exim command in a
2805 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2806 was being done only for the command run by the queryprogram router. It is
2807 now done for all such subprocesses. The other cases are: ${run, transport
2808 filters, and the commands run by the lmtp and pipe transports.
2810 5. Some older OS have a limit of 256 on the maximum number of file
2811 descriptors. Exim was using setrlimit() to set 1000 as a large value
2812 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2813 systems. I've change it so that if it can't get 1000, it tries for 256.
2815 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2816 was an oversight, and furthermore, ever since the addition of extra
2817 controls (e.g. 4.43/32), the checks on when to allow different forms of
2818 "control" were broken. There should now be diagnostics for all cases when a
2819 control that does not make sense is encountered.
2821 7. $recipients is now available in the predata ACL (oversight).
2823 8. Tidy the search cache before the fork to do a delivery from a message
2824 received from the command line. Otherwise the child will trigger a lookup
2825 failure and thereby defer the delivery if it tries to use (for example) a
2826 cached ldap connection that the parent has called unbind on.
2828 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2829 of $address_data from the recipient verification was clobbered by the
2830 sender verification.
2832 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2833 was its contents. (It was OK if the option was not defined at all.)
2835 11. A "Completed" log line is now written for messages that are removed from
2836 the spool by the -Mrm option.
2838 12. $host_address is now set to the target address during the checking of
2839 ignore_target_hosts.
2841 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2842 being passed; this would have caused $sender_host_name to have been used if
2843 matching the list had actually called for a host name (not very likely,
2844 since this list is usually IP addresses). A host name is now passed as
2847 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2848 code that specifies a non-restarting handler (typically sigaction() in
2849 modern systems) in an attempt to fix a rare and obscure crash bug.
2851 15. Narrowed the window for a race in the daemon that could cause it to ignore
2852 SIGCHLD signals. This is not a major problem, because they are used only to
2853 wake it up if nothing else does.
2855 16. A malformed maildirsize file could cause Exim to calculate negative values
2856 for the mailbox size or file count. Odd effects could occur as a result.
2857 The maildirsize information is now recalculated if the size or filecount
2860 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2861 support for a long time. Removed HAVE_SYS_VFS_H.
2863 18. Updated exipick to current release from John Jetmore.
2865 19. Allow an empty sender to be matched against a lookup in an address list.
2866 Previously the only cases considered were a regular expression, or an
2869 20. Exim went into a mad DNS lookup loop when doing a callout where the
2870 host was specified on the transport, if the DNS lookup yielded more than
2873 21. The RFC2047 encoding function was originally intended for short strings
2874 such as real names; it was not keeping to the 75-character limit for
2875 encoded words that the RFC imposes. It now respects the limit, and
2876 generates multiple encoded words if necessary. To be on the safe side, I
2877 have increased the buffer size for the ${rfc2047: expansion operator from
2880 22. Failure to deliver a bounce message always caused it to be frozen, even if
2881 there was an errors_to setting on the router. The errors_to setting is now
2884 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2885 canonical form (fully expanded) before being placed in
2886 $sender_host_address.
2888 24. Updated eximstats to version 1.33
2890 25. Include certificate and key file names in error message when GnuTLS fails
2891 to set them up, because the GnuTLS error message doesn't include the name
2892 of the failing file when there is a problem reading it.
2894 26. Expand error message when OpenSSL has problems setting up cert/key files.
2897 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2898 (this can affect the format of dates).
2900 28. exim_tidydb, when checking for the continued existence of a message for
2901 which it has found a message-specific retry record, was not finding
2902 messages that were in split spool directories. Consequently, it was
2903 deleting retry records that should have stayed in existence.
2905 29. eximstats updated to version 1.35
2906 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2907 1.35 - bugfix such that pie charts by volume are generated correctly
2909 30. The SPA authentication driver was not abandoning authentication and moving
2910 on to the next authenticator when an expansion was forced to fail,
2911 contradicting the general specification for all authenticators. Instead it
2912 was generating a temporary error. It now behaves as specified.
2914 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2915 (the order specifies the preference for clients). The order is now AES256,
2916 AES128, 3DES, ARCFOUR128.
2918 31. Small patch to Sieve code - explicitly set From: when generating an
2921 32. Exim crashed if a remote delivery caused a very long error message to be
2922 recorded - for instance if somebody sent an entire SpamAssassin report back
2923 as a large number of 550 error lines. This bug was coincidentally fixed by
2924 increasing the size of one of Exim's internal buffers (big_buffer) that
2925 happened as part of the Exiscan merge. However, to be on the safe side, I
2926 have made the code more robust (and fixed the comments that describe what
2929 33. Some experimental protocols are using DNS PTR records for new purposes. The
2930 keys for these records are domain names, not reversed IP addresses. The
2931 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2932 leaves it alone. Component reversal etc. now happens only for IP addresses.
2935 34. The host_aton() function is supposed to be passed a string that is known
2936 to be a valid IP address. However, in the case of IPv6 addresses, it was
2937 not checking this. This is a hostage to fortune. Exim now panics and dies
2938 if the condition is not met. A case was found where this could be provoked
2939 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2940 components; fortuitously, this particular loophole had already been fixed
2941 by change 4.50/55 or 4.44/33 above.
2943 If there are any other similar loopholes, the new check in host_aton()
2944 itself should stop them being exploited. The report I received stated that
2945 data on the command line could provoke the exploit when Exim was running as
2946 exim, but did not say which command line option was involved. All I could
2947 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2948 running as the user.
2951 35. There was a buffer overflow vulnerability in the SPA authentication code
2952 (which came originally from the Samba project). I have added a test to the
2953 spa_base64_to_bits() function which I hope fixes it.
2956 36. The daemon start-up calls getloadavg() while still root for those OS that
2957 need the first call to be done as root, but it missed one case: when
2958 deliver_queue_load_max is set with deliver_drop_privilege. This is
2959 necessary for the benefit of the queue runner, because there is no re-exec
2960 when deliver_drop_privilege is set.
2962 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2963 was in use, was not putting the data itself into the right store pool;
2964 consequently, it could be overwritten for a subsequent message in the same
2965 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2968 38. Sometimes the final signoff response after QUIT could fail to get
2969 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2970 < 0 before doing a fflush(). This bug looks as though it goes back to the
2971 introduction of TLS in release 3.20, but "sometimes" must have been rare
2972 because the tests only now provoked it.
2978 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2979 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2980 Now it returns an integer. A number of calls were still expecting a T/F
2981 return. Fortuitously, in all cases, the tests worked in OK situations,
2982 which is the norm. However, things would have gone wrong on any write
2983 failures on the smtp file descriptor. This function is used when sending
2984 messages over SMTP and also when doing verify callouts.
2986 2. When Exim is called to do synchronous delivery of a locally submitted
2987 message (the -odf or -odi options), it no longer closes stderr before doing
2990 3. Implemented the mua_wrapper option.
2992 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2994 5. Implemented the functions header_remove(), header_testname(),
2995 header_add_at_position(), and receive_remove_recipient(), and exported them
2998 6. If an ACL "warn" statement specified the addition of headers, Exim already
2999 inserted X-ACL-Warn: at the start if there was no header name. However, it
3000 was not making this test for the second and subsequent header lines if
3001 there were newlines in the string. This meant that an invalid header could
3002 be inserted if Exim was badly configured.
3004 7. Allow an ACL "warn" statement to add header lines at the start or after all
3005 the Received: headers, as well as at the end.
3007 8. Added the rcpt_4xx retry error code.
3009 9. Added postmaster_mailfrom=xxx to callout verification option.
3011 10. Added mailfrom=xxxx to the callout verification option, for verify=
3014 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3015 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3017 12. Inserted some casts to stop certain compilers warning when using pointer
3018 differences as field lengths or precisions in printf-type calls (mostly
3019 affecting debugging statements).
3021 13. Added optional readline() support for -be (dynamically loaded).
3023 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3024 same clock tick as a message's arrival, so that its received time was the
3025 same as the "first fail" time on the retry record, and that message
3026 remained on the queue past the ultimate address timeout, every queue runner
3027 would try a delivery (because it was past the ultimate address timeout) but
3028 after another failure, the ultimate address timeout, which should have then
3029 bounced the address, did not kick in. This was a "< instead of <=" error;
3030 in most cases the first failure would have been in the next clock tick
3031 after the received time, and all would be well.
3033 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3034 being recognized when the domain list was tested by the match_domain
3035 condition in an expansion string.
3037 16. Added the ${str2b64: operator.
3039 17. Exim was always calling setrlimit() to set a large limit for the number of
3040 processes, without checking whether the existing limit was already
3041 adequate. (It did check for the limit on file descriptors.) Furthermore,
3042 errors from getrlimit() and setrlimit() were being ignored. Now they are
3043 logged to the main and panic logs, but Exim does carry on, to try to do its
3044 job under whatever limits there are.
3046 18. Imported PCRE 5.0.
3048 19. Trivial typo in log message " temporarily refused connection" (the leading
3051 20. If the log selector return_path_on_delivery was set and an address was
3052 redirected to /dev/null, the delivery process crashed because it assumed
3053 that a return path would always be set for a "successful" delivery. In this
3054 case, the whole delivery is bypassed as an optimization, and therefore no
3057 21. Internal re-arrangement: the function for sending a challenge and reading
3058 a response while authentication was assuming a zero-terminated challenge
3059 string. It's now changed to take a pointer and a length, to allow for
3060 binary data in such strings.
3062 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3064 23. Exim was not respecting finduser_retries when seeking the login of the
3065 uid under which it was called; it was always trying 10 times. (The default
3066 setting of finduser_retries is zero.) Also, it was sleeping after the final
3067 failure, which is pointless.
3069 24. Implemented tls_on_connect_ports.
3071 25. Implemented acl_smtp_predata.
3073 26. If the domain in control=submission is set empty, Exim assumes that the
3074 authenticated id is a complete email address when it generates From: or
3075 Sender: header lines.
3077 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3078 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3079 chown and chgrp in /bin and hostname in /usr/bin.
3081 28. Exim was keeping the "process log" file open after each use, just as it
3082 does for the main log. This opens the possibility of it remaining open for
3083 long periods when the USR1 signal hits a daemon. Occasional processlog
3084 errors were reported, that could have been caused by this. Anyway, it seems
3085 much more sensible not to leave this file open at all, so that is what now
3088 29. The long-running daemon process does not normally write to the log once it
3089 has entered its main loop, and it closes the log before doing so. This is
3090 so that log files can straightforwardly be renamed and moved. However,
3091 there are a couple of unusual error situations where the daemon does write
3092 log entries, and I had neglected to close the log afterwards.
3094 30. The text of an SMTP error response that was received during a remote
3095 delivery was being truncated at 512 bytes. This is too short for some of
3096 the long messages that one sometimes sees. I've increased the limit to
3099 31. It is now possible to make retry rules that apply only when a message has a
3100 specific sender, in particular, an empty sender.
3102 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3103 it possible to be selective about when SMTP synchronization is enforced.
3105 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3107 32. Implemented hosts_connection_nolog.
3109 33. Added an ACL for QUIT.
3111 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3114 35. Added mailbox_size and mailbox_filecount to appendfile.
3116 36. Added control = no_multiline_responses to ACLs.
3118 37. There was a bug in the logic of the code that waits for the clock to tick
3119 in the case where the clock went backwards by a substantial amount such
3120 that the microsecond fraction of "now" was more than the microsecond
3121 fraction of "then" (but the whole seconds number was less).
3123 38. Added support for the libradius Radius client library this is found on
3124 FreeBSD (previously only the radiusclient library was supported).
3130 1. When certain lookups returned multiple values in the form name=value, the
3131 quoting of the values was not always being done properly. Specifically:
3132 (a) If the value started with a double quote, but contained no whitespace,
3134 (b) If the value contained whitespace other than a space character (i.e.
3135 tabs or newlines or carriage returns) it was not quoted.
3136 This fix has been applied to the mysql and pgsql lookups by writing a
3137 separate quoting function and calling it from the lookup code. The fix
3138 should probably also be applied to nisplus, ibase and oracle lookups, but
3139 since I cannot test any of those, I have not disturbed their existing code.
3141 2. A hit in the callout cache for a specific address caused a log line with no
3142 reason for rejecting RCPT. Now it says "Previous (cached) callout
3143 verification failure".
3145 3. There was an off-by-one bug in the queryprogram router. An over-long
3146 return line was truncated at 256 instead of 255 characters, thereby
3147 overflowing its buffer with the terminating zero. As well as fixing this, I
3148 have increased the buffer size to 1024 (and made a note to document this).
3150 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3151 when Exim is waiting for an SMTP response from a remote server, Exim
3152 restarts its select() call on the socket, thereby resetting its timeout.
3153 This is not a problem when such interrupts are rare. Somebody set up a cron
3154 job to run exiwhat every 2 minutes, which is less than the normal select()
3155 timeout (5 or 10 minutes). This meant that the select() timeout never
3156 kicked in because it was always reset. I have fixed this by comparing the
3157 time when an interrupt arrives with the time at the start of the first call
3158 to select(). If more time than the timeout has elapsed, the interrupt is
3159 treated as a timeout.
3161 5. Some internal re-factoring in preparation for the addition of Sieve
3162 extensions (by MH). In particular, the "personal" test is moved to a
3163 separate function, and given an option for scanning Cc: and Bcc: (which is
3164 not set for Exim filters).
3166 6. When Exim created an email address using the login of the caller as the
3167 local part (e.g. when creating a From: or Sender: header line), it was not
3168 quoting the local part when it contained special characters such as @.
3170 7. Installed new OpenBSD configuration files.
3172 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3173 try to make them clearer.
3175 9. Callout options, other than the timeout value, were being ignored when
3176 verifying sender addresses in header lines. For example, when using
3178 verify = header_sender/callout=no_cache
3180 the cache was (incorrectly) being used.
3182 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3183 only the Cygwin environment.
3185 11. When return_path_on_delivery was set as a log selector, if different remote
3186 addresses in the same message used different return paths and parallel
3187 remote delivery occurred, the wrong values would sometimes be logged.
3188 (Whenever a remote delivery process finished, the return path value from
3189 the most recently started remote delivery process was logged.)
3191 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3192 lines when AUTH and/or TLS are in use. This is the "received protocol"
3193 field. Exim used to use "asmtp" for authenticated SMTP, without any
3194 indication (in the protocol name) for TLS use. Now it follows the RFC and
3195 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3196 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3197 names appear in log lines as well as in Received: header lines.
3199 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3200 extensions, and comparison tests, and to fix some bugs.
3202 14. Changes to the "personal" filter test:
3204 (1) The test was buggy in that it was just doing the equivalent of
3205 "contains" tests on header lines. For example, if a user's address was
3206 anne@some.where, the "personal" test would incorrectly be true for
3208 To: susanne@some.where
3210 This test is now done by extracting each address from the header in turn,
3211 and checking the entire address. Other tests that are part of "personal"
3212 are now done using regular expressions (for example, to check local parts
3213 of addresses in From: header lines).
3215 (2) The list of non-personal local parts in From: addresses has been
3216 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3217 taken from the Sieve specification recommendations.
3219 (3) If the message contains any header line starting with "List-" it is
3220 treated as non-personal.
3222 (4) The test for "circular" in the Subject: header line has been removed
3223 because it now seems ill-conceived.
3225 15. Minor typos in src/EDITME comments corrected.
3227 16. Installed latest exipick from John Jetmore.
3229 17. If headers_add on a router specified a text string that was too long for
3230 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3231 of string_sprintf() is now avoided.
3233 18. $message_body_size was not set (it was always zero) when running the DATA
3234 ACL and the local_scan() function.
3236 19. For the "mail" command in an Exim filter, no default was being set for
3237 the once_repeat time, causing a random time value to be used if "once" was
3238 specified. (If the value happened to be <= 0, no repeat happened.) The
3239 default is now 0s, meaning "never repeat". The "vacation" command was OK
3240 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3241 (I found it when inspecting the code).
3243 20. There is now an overall timeout for performing a callout verification. It
3244 defaults to 4 times the callout timeout, which applies to individual SMTP
3245 commands during the callout. The overall timeout applies when there is more
3246 than one host that can be tried. The timeout is checked before trying the
3247 next host. This prevents very long delays if there are a large number of
3248 hosts and all are timing out (e.g. when the network connections are timing
3249 out). The value of the overall timeout can be changed by specifying an
3250 additional sub-option for "callout", called "maxwait". For example:
3252 verify = sender/callout=5s,maxwait=20s
3254 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3255 to the end before writing, but this should make it even safer).
3257 22. Exim was forgetting that it had advertised PIPELINING for the second and
3258 subsequent messages on an SMTP connection. It was also not resetting its
3259 memory on STARTTLS and an internal HELO.
3261 23. When Exim logs an SMTP synchronization error within a session, it now
3262 records whether PIPELINING has been advertised or not.
3264 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3265 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3266 rather than long int.
3268 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3270 26. Added the never_mail option to autoreply.
3276 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3277 crash if the getsockname() call failed; this can happen if a connection is
3278 closed very soon after it is established. The problem was simply in the
3279 order in which certain operations were done, causing Exim to try to write
3280 to the SMTP stream before it had set up the file descriptor. The bug has
3281 been fixed by making things happen in the correct order.
3287 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3288 before the connection was closed, thus losing the rejection response.
3290 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3291 some early Solaris releases, but causes trouble in current releases where
3292 socklen_t is defined.
3294 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3297 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3300 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3301 that releasing the top of it at the end releases what was used for sub-
3302 expansions (unless the block got too big). However, discard this block if
3303 the first thing is a variable or header, so that we can use its block when
3304 it is dynamic (useful for very large $message_headers, for example).
3306 6. Lookups now cache *every* query, not just the most recent. A new, separate
3307 store pool is used for this. It can be recovered when all lookup caches are
3308 flushed. Lookups now release memory at the end of their result strings.
3309 This has involved some general refactoring of the lookup sources.
3311 7. Some code has been added to the store_xxx() functions to reduce the amount
3312 of flapping under certain conditions.
3314 8. log_incoming_interface used to affect only the <= reception log lines. Now
3315 it causes the local interface and port to be added to several more SMTP log
3316 lines, for example "SMTP connection from", and rejection lines.
3318 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3320 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3322 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3323 could be overwritten at the end of the current message (or the start of a
3324 new message if it was set in a HELO ACL). The value is now preserved for
3325 the duration of the SMTP connection.
3327 12. If a transport had a headers_rewrite setting, and a matching header line
3328 contained an unqualified address, that address was qualified, even if it
3329 did not match any rewriting rules. The underlying bug was that the values
3330 of the flags that permit the existence of unqualified sender and recipient
3331 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3332 non-local messages, and by -bnq for local messages) were not being
3333 preserved with the message after it was received.
3335 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3336 "next input=" as part of the text comprising the host identity instead of
3337 the correct text. The code was using the same buffer for two different
3338 strings. However, depending on which order the printing function evaluated
3339 its arguments, the bug did not always show up. Under Linux, for example, my
3340 test suite worked just fine.
3342 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3343 doesn't work with some older versions of Perl. It has been changed to "my",
3344 which in any case is probably the better facility to use.
3346 15. A really picky compiler found some instances of statements for creating
3347 error messages that either had too many or two few arguments for the format
3350 16. The size of the buffer for calls to the DNS resolver has been increased
3351 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3352 for addresses that have a lot of PTR records. This alleviates a problem; it
3353 does not fully solve it.
3355 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3356 buffer now truncates the list and logs the incident, which is the same
3357 action as happens when Exim is looking up a host name and its aliases.
3358 Previously in this situation something unpredictable would happen;
3359 sometimes it was "internal error: store_reset failed".
3361 18. If a server dropped the connection unexpectedly when an Exim client was
3362 using GnuTLS and trying to read a response, the client delivery process
3363 crashed while trying to generate an error log message.
3365 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3366 string, for example:
3368 warn message = H1: something\nH2: something
3370 the text was added as a single header line from Exim's point of view
3371 though it ended up OK in the delivered message. However, searching for the
3372 second and subsequent header lines using $h_h2: did not work. This has been
3373 fixed. Similarly, if a system filter added multiple headers in this way,
3374 the routers could not see them.
3376 20. Expanded the error message when iplsearch is called with an invalid key to
3377 suggest using net-iplsearch in a host list.
3379 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3380 ACLs are no longer actually imposed (and a message to that effect is
3383 22. If a "gecos" field in a passwd entry contained escaped characters, in
3384 particular, if it contained a \" sequence, Exim got it wrong when building
3385 a From: or a Sender: header from that name. A second bug also caused
3386 incorrect handling when an unquoted " was present following a character
3387 that needed quoting.
3389 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3390 was not being matched caselessly.
3392 24. Arranged for all hyphens in the exim.8 source to be escaped with
3395 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3396 properly. Recipient callout cache records were still being keyed to include
3397 the sender, even when use_sender was set false. This led to far more
3398 callouts that were necessary. The sender is no longer included in the key
3399 when use_sender is false.
3401 26. Added "control = submission" modifier to ACLs.
3403 27. Added the ${base62d: operator to decode base 62 numbers.
3405 28. dnsdb lookups can now access SRV records.
3407 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3408 the configuration file.
3410 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3411 (-v) mode. This makes the output for a verbose queue run more intelligible.
3413 31. Added a use_postmaster feature to recipient callouts.
3415 32. Added the $body_zerocount variable, containing the number of binary zero
3416 bytes in the message body.
3418 33. The time of last modification of the "new" subdirectory is now used as the
3419 "mailbox time last read" when there is a quota error for a maildir
3422 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3424 35. Added +ignore_unknown as a special item in host lists.
3426 36. Code for decoding IPv6 addresses in host lists is now included, even if
3427 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3428 address was recognized as an IP address, but was then not correctly decoded
3429 into binary, causing unexpected and incorrect effects when compared with
3436 1. Very minor rewording of debugging text in manualroute to say "list of
3437 hosts" instead of "hostlist".
3439 2. If verify=header_syntax was set, and a header line with an unqualified
3440 address (no domain) and a large number of spaces between the end of the
3441 name and the colon was received, the reception process suffered a buffer
3442 overflow, and (when I tested it) crashed. This was caused by some obsolete
3443 code that should have been removed. The fix is to remove it!
3445 3. When running in the test harness, delay a bit after writing a bounce
3446 message to get a bit more predictability in the log output.
3448 4. Added a call to search_tidyup() just before forking a reception process. In
3449 theory, someone could use a lookup in the expansion of smtp_accept_max_
3450 per_host which, without the tidyup, could leave open a database connection.
3452 5. Added the variables $recipient_data and $sender_data which get set from a
3453 lookup success in an ACL "recipients" or "senders" condition, or a router
3454 "senders" option, similar to $domain_data and $local_part_data.
3456 6. Moved the writing of debug_print from before to after the "senders" test
3459 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3460 problems for message scanning, either using a data ACL, or using
3461 local_scan() because the Received: header was not generated till after they
3462 were called (in order to set the time as the time of reception completion).
3463 I have revised the way this works. The header is now generated after the
3464 body is received, but before the ACL or local_scan() are called. After they
3465 are run, the timestamp in the header is updated.
3471 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3472 before starting a queue runner without re-exec. This happened only when
3473 deliver_drop_privilege was set or when the Exim user was set to root. The
3474 effect of the bug was that timeouts during subsequent deliveries caused
3475 crashes instead of being properly handled. The handler is now left at its
3476 default (and expected) setting.
3478 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3479 message, again when deliver_drop_privilege is set or Exim is run as root.
3480 The bug described in (1) was not present in this case, but the tidying up
3481 of the other signals was missing. I have made the two cases consistent.
3483 3. The ignore_target_hosts setting on a manualroute router was being ignored
3484 for hosts that were looked up using the /MX notation.
3486 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3489 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3490 operated on the sender address. After changing the $sender_address to <>
3491 for the sender address verify, Exim was re-instated it as the original
3492 (before rewriting) address, but remembering that it had rewritten it, so it
3493 wasn't rewriting it again. This bug also had the effect of breaking the
3494 sender address verification caching when the sender address was rewritten.
3496 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3497 This has been changed so that if the ip literal address matches
3498 ignore_target_hosts, the router declines.
3500 7. Added expansion conditions match_domain, match_address, and match_local_
3501 part (NOT match_host).
3503 8. The placeholder for the Received: header didn't have a length field set.
3505 9. Added code to Exim itself and to exim_lock to test for a specific race
3506 condition that could lead to file corruption when using MBX delivery. The
3507 issue is with the lockfile that is created in /tmp. If this file is removed
3508 after a process has opened it but before that process has acquired a lock,
3509 there is the potential for a second process to recreate the file and also
3510 acquire a lock. This could lead to two Exim processes writing to the file
3511 at the same time. The added code performs the same test as UW imapd; it
3512 checks after acquiring the lock that its file descriptor still refers to
3513 the same named file.
3515 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3516 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3519 11. Added the smtp_active_hostname option. If used, this will typically be made
3520 to depend on the incoming interface address. Because $interface_address is
3521 not set up until the daemon has forked a reception process, error responses
3522 that can happen earlier (such as "too many connections") no longer contain
3525 12. If an expansion in a condition on a "warn" statement fails because a lookup
3526 defers, the "warn" statement is abandoned, and the next ACL statement is
3527 processed. Previously this caused the whole ACL to be aborted.
3529 13. Added the iplsearch lookup type.
3531 14. Added ident_timeout as a log selector.
3533 15. Added tls_certificate_verified as a log selector.
3535 16. Added a global option tls_require_ciphers (compare the smtp transport
3536 option of the same name). This controls incoming TLS connections.
3538 17. I finally figured out how to make tls_require_ciphers do a similar thing
3539 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3540 before starting the TLS session.
3542 18. Tabs are now shown as \t in -bP output.
3544 19. If the log selector return_path_on_delivery was set, Exim crashed when
3545 bouncing a message because it had too many Received: header lines.
3547 20. If two routers both had headers_remove settings, and the first one included
3548 a superfluous trailing colon, the final name in the first list and the
3549 first name in the second list were incorrectly joined into one item (with a
3550 colon in the middle).
3556 1. Added -C and -D options to the exinext utility, mainly to make it easier
3557 to include in the automated testing, but these could be helpful when
3558 multiple configurations are in use.
3560 2. The exinext utility was not formatting the output nicely when there was
3561 an alternate port involved in the retry record key, nor when there was a
3562 message id as well (for retries that were specific to a specific message
3563 and a specific host). It was also confused by IPv6 addresses, because of
3564 the additional colons they contain. I have fixed the IPv4 problem, and
3565 patched it up to do a reasonable job for IPv6.
3567 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3568 delivery, the log line now contains "pipelined" if PIPELINING was used.
3570 4. An SMTP transport process used to panic and die if the bind() call to set
3571 an explicit outgoing interface failed. This has been changed; it is now
3572 treated in the same way as a connect() failure.
3574 5. A reference to $sender_host_name in the part of a conditional expansion
3575 that was being skipped was still causing a DNS lookup. This no longer
3578 6. The def: expansion condition was not recognizing references to header lines
3579 that used bh_ and bheader_.
3581 7. Added the _cache feature to named lists.
3583 8. The code for checking quota_filecount in the appendfile transport was
3584 allowing one more file than it should have been.
3586 9. For compatibility with Sendmail, the command line option
3594 and sets the incoming protocol and host name (for trusted callers). The
3595 host name and its colon can be omitted when only the protocol is to be set.
3596 Note the Exim already has two private options, -pd and -ps, that refer to
3597 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3598 "s", but I don't think that's a major issue.
3600 10. A number of refactoring changes to the code, none of which should affect
3603 (a) The number of logging options was getting close to filling up the
3604 32-bit word that was used as a bit map. I have split them into two classes:
3605 those that are passed in the argument to log_write(), and those that are
3606 only ever tested independently outside of that function. These are now in
3607 separate 32-bit words, so there is plenty of room for expansion again.
3608 There is no change in the user interface or the logging behaviour.
3610 (b) When building, for example, log lines, the code previously used a
3611 macro that called string_cat() twice, in order to add two strings. This is
3612 not really sufficiently general. Furthermore, there was one instance where
3613 it was actually wrong because one of the argument was used twice, and in
3614 one call a function was used. (As it happened, calling the function twice
3615 did not affect the overall behaviour.) The macro has been replaced by a
3616 function that can join an arbitrary number of extra strings onto a growing
3619 (c) The code for expansion conditions now uses a table and a binary chop
3620 instead of a serial search (which was left over from when there were very
3621 few conditions). Also, it now recognizes conditions like "pam" even when
3622 the relevant support is not compiled in: a suitably worded error message is
3623 given if an attempt is made to use such a condition.
3625 11. Added ${time_interval:xxxxx}.
3627 12. A bug was causing one of the ddress fields not to be passed back correctly
3628 from remote delivery subprocesses. The field in question was not being
3629 subsequently used, so this caused to problems in practice.
3631 13. Added new log selectors queue_time and deliver_time.
3633 14. Might have fixed a bug in maildirsizefile handling that threw up
3634 "unexpected character" debug warnings, and recalculated the data
3635 unnecessarily. In any case, I expanded the warning message to give more
3638 15. Added the message "Restricted characters in address" to the statements in
3639 the default ACL that block characters like @ and % in local parts.
3641 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3642 Three changes have been made:
3644 (a) There was a serious bug; a negative response to MAIL caused the whole
3645 recipient domain to be cached as invalid, thereby blocking all messages
3646 to all local parts at the same domain, from all senders. This bug has
3647 been fixed. The domain is no longer cached after a negative response to
3648 MAIL if the sender used is not empty.
3650 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3653 (c) A new callout option, "use_sender" has been added for people who want
3654 the modified behaviour.
3660 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3663 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3664 indeed breaks things for older releases.
3666 3. Added additional logging to the case where there is a problem reading data
3667 from a filter that is running in a subprocess using a pipe, in order to
3668 try to track down a specific problem.
3670 4. Testing facility fudge: when running in the test harness and attempting
3671 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3672 getting "No route to host". Convert this to a timeout.
3674 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3677 6. Some OS don't have socklen_t but use size_t instead. This affects the
3678 fifth argument of getsockopt() amongst other things. This is now
3679 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3680 can be set for individual OS. I have set it for SunOS5, OSF1, and
3681 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3682 some earlier ones do not.
3684 7. Change 4.30/15 was not doing the test caselessly.
3686 8. The standard form for an IPv6 address literal was being rejected by address
3687 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3688 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3689 this, as well as the form without the "IPv6" on the front (but only when
3690 address literals are enabled, of course).
3692 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3694 10. Exim crashed if a message with an empty sender address specified by -f
3695 encountered a router with an errors_to setting. This could be provoked only
3696 by a command such as
3700 where an empty string was supplied; "<>" did not hit this bug.
3702 11. Installed PCRE release 4.5.
3704 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3705 remained set. It is now erased.
3707 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3708 times from message ids (which are base 36 rather than the normal 62).
3710 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3711 were being counted as actual protocol errors, and logged if the log
3712 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3713 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3714 and DATA following a set of rejected RCPTs do not count as protocol errors.
3715 In other words, Exim assumes they were pipelined, though this may not
3716 actually be the case. Of course, in all cases the client gets an
3717 appropriate error code.
3719 15. If a lookup fails in an ACL condition, a message about the failure may
3720 be available; it is used if testing the ACL cannot continue, because most
3721 such messages specify what the cause of the deferral is. However, some
3722 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3723 that caused an old message to be retained and used if a later statement
3724 caused a defer, replacing the real cause of the deferral.
3726 16. If an IP address had so many PTR records that the DNS lookup buffer
3727 was not large enough to hold them, Exim could crash while trying to process
3728 the truncated data. It now detects and logs this case.
3730 17. Further to 4.21/58, another change has been made: if (and only if) the
3731 first line of a message (the first header line) ends with CRLF, a bare LF
3732 in a subsequent header line has a space inserted after it, so as not to
3733 terminate the header.
3735 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3736 unnecessarily, used atoi() instead of strtol(), and didn't check the
3737 termination when getting file sizes from file names by regex.
3739 19. Completely re-implemented the support for maildirsize files, in the light
3740 of a number of problems with the previous contributed implementation
3741 (4.30/29). In particular:
3743 . If the quota is zero, the maildirsize file is maintained, but no quota is
3746 . If the maildir directory does not exist, it is created before any attempt
3747 to write a maildirsize file.
3749 . The quota value in the file is just a cache; if the quota is changed in
3750 the transport, the new value overrides.
3752 . A regular expression is available for excluding directories from the
3755 20. The autoreply transport checks the characters in options that define the
3756 message's headers; it allows continued headers, but it was checking with
3757 isspace() after an embedded newline instead of explicitly looking for a
3760 21. If all the "regular" hosts to which an address was routed had passed their
3761 expiry times, and had not reached their retry times, the address was
3762 bounced, even if fallback hosts were defined. Now Exim should go on to try
3765 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3766 equivalent code in the SMTP transport. Some hosts send humungous responses
3767 to HELO/EHLO, more than 1024 it seems.
3769 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3770 gives compiler warnings in some environments. I've now done it "properly",
3773 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3774 (because of problems with the built-in one) was declared to return uschar *
3775 instead of char *, causing compiler failure.
3777 25. Fixed a file descriptor leak when processing alias/forward files.
3779 26. Fixed a minor format string issue in dbfn.c.
3781 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3783 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3784 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3787 29. When a sender address is verified, it is cached, to save repeating the test
3788 when there is more than one recipient in a message. However, when the
3789 verification involves a callout, it is possible for different callout
3790 options to be set for different recipients. It is too complicated to keep
3791 track of this in the cache, so now Exim always runs a verification when a
3792 callout is required, relying on the callout cache for the optimization.
3793 The overhead is duplication of the address routing, but this should not be
3796 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3797 to be verified with callout=postmaster, and the main callout worked but the
3798 postmaster check failed, the verification correctly failed. However, if a
3799 subsequent RCPT command asked for sender verification *without* the
3800 postmaster check, incorrect caching caused this verification also to fail,
3803 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3804 it was not caching the DNS options (qualify_single, search_parents) that
3805 were used when the lookup failed. A subsequent lookup with different
3806 options therefore always gave the same answer, though there were cases
3807 where it should not have. (Example: a "domains = !$mx_any" option on a
3808 dnslookup router: the "domains" option is always processed without any
3809 widening, but the router might have qualify_single set.) Now Exim uses the
3810 cached value only when the same options are set.
3812 32. Added John Jetmore's "exipick" utility to the distribution.
3814 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3815 than a timeout (e.g. a certificate is required, and is not provided), an
3816 Exim server now closes the connection immediately. Previously it waited for
3817 the client to close - but if the client is SSL, it seems that they each
3818 wait for each other, leading to a delay before one of them times out.
3820 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3821 maintained 0.8.x compatibility because I don't think many are using it, and
3822 it is clearly obsolete.
3824 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3827 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3828 client certificate was expired. A simple patch fixes this, though I don't
3829 understand the full logic of why the verify callback is called multiple
3832 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3833 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3834 which causes problems with some clients (such as the Certicom SSL Plus
3835 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3836 disables the coutermeasure allowing Eudora to connect."
3838 38. Exim was not checking that a write() to a log file succeeded. This could
3839 lead to Bad Things if a log got too big, in particular if it hit a file
3840 size limit. Exim now panics and dies if it cannot write to a log file, just
3841 as it does if it cannot open a log file.
3843 39. Modified OS/Makefile-Linux so that it now contains
3845 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3847 The two -D definitions ensure that Exim is compiled with large file
3848 support, which makes it possible to handle log files that are bigger than
3851 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3852 instance) a domain was checked against a named list that involved a lookup,
3853 causing $domain_data to be set, then another domain was checked against the
3854 same list, then the first domain was re-checked, the value of $domain_data
3855 after the final check could be wrong. In particular, if the second check
3856 failed, it could be set empty. This bug probably also applied to
3859 41. The strip_trailing_dot option was not being applied to the address given
3860 with the -f command-line option.
3862 42. The code for reading a message's header from the spool was incrementing
3863 $received_count, but never initializing it. This meant that the value was
3864 incorrect (doubled) while delivering a message in the same process in which
3865 it was received. In the most common configuration of Exim, this never
3866 happens - a fresh exec is done - but it can happen when
3867 deliver_drop_privilege is set.
3869 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3870 it now includes up to 150 characters of the unexpected data in the log
3873 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3874 and building data strings. The size of both of these buffers was 10 000
3875 bytes - far larger than anybody would *ever* want, thought I. Needless to
3876 say, somebody hit the limit. I have increased the maximum line length to
3877 20 000 and the maximum data length of concatenated lines to 100 000. I have
3878 also fixed two bugs, because there was no checking on these buffers. Tsk,
3879 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3880 buffer is too small.
3882 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3883 lsearch lookups. Now it does.
3885 46. When parsing a route_list item in a manualroute router, a fixed-length
3886 buffer was used for the list of hosts. I made this 1024 bytes long,
3887 thinking that nobody would ever have a list of hosts that long. Wrong.
3888 Somebody had a whole pile of complicated expansion conditions, and the
3889 string was silently truncated, leading to an expansion error. It turns out
3890 that it is easier to change to an unlimited length (owing to other changes
3891 that have happened since this code was originally written) than to build
3892 structure for giving a limitation error. The length of the item that
3893 expands into the list of hosts is now unlimited.
3895 47. The lsearch lookup could not handle data where the length of text line was
3896 more than 4095 characters. Such lines were truncated, leading to shortened
3897 data being returned. It should now handle lines of any length.
3899 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3900 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3903 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3904 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3905 no longer happens in comment lines. A list of replacements is now placed
3906 at the head of all of the source files, except those whose only change is
3907 to replace PERL_COMMAND in the very first #! line.
3909 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3910 messages on the queue, with a bottom-up merge sort, using code contributed
3911 by Michael Haardt. This should make operations like -bp somewhat faster on
3912 large queues. It won't affect queue runners, except when queue_run_in_order
3915 51. Installed eximstats 1.31 in the distribution.
3917 52. Added support for SRV lookups to the dnslookup router.
3919 53. If an ACL referred to $message_body or $message_body_end, the value was not
3920 reset for any messages that followed in the same SMTP session.
3922 54. The store-handling optimization for building very long strings was not
3923 differentiating between the different store pools. I don't think this
3924 actually made any difference in practice, but I've tidied it.
3926 55. While running the routers to verify a sender address, $sender_address
3927 was still set to the sender address. This is wrong, because when routing to
3928 send a bounce to the sender, it would be empty. Therefore, I have changed
3929 it so that, while verifying a sender address, $sender_address is set to <>.
3930 (There is no change to what happens when verifying a recipient address.)
3932 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3933 target A or AAAA records (if not already returned) without resetting the
3934 qualify_single or search_parents options of the DNS resolver. These are
3935 inappropriate in this case because the targets of MX and SRV records must
3936 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3937 target that, when qualified, matched something in the local domain. These
3938 two options are now turned off when doing these lookups.
3940 57. It seems that at least some releases of Reiserfs (which does not have the
3941 concept of a fixed number of inodes) returns zero and not -1 for the
3942 number of available inodes. This interacted badly with check_spool_inodes,
3943 which assumed that -1 was the "no such thing" setting. What I have done is
3944 to check that the total number of inodes is greater than zero before doing
3945 the test of how many are available.
3947 58. When a "warn" ACL statement has a log_message modifier, the message is
3948 remembered, and not repeated. This is to avoid a lot of repetition when a
3949 message has many recipients that cause the same warning to be written.
3950 Howewer, Exim was preserving the list of already written lines for an
3951 entire SMTP session, which doesn't seem right. The memory is now reset if a
3952 new message is started.
3954 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3955 debugging output unless log_rewrite was also set.
3957 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3958 of (int)(handle) when we know that handle contains (void *)(-1).
3960 61. The Exim daemon panic-logs an error return when it closes the incoming
3961 connection. However "connection reset by peer" seems to be common, and
3962 isn't really an error worthy of noting specially, so that particular error
3965 62. When Exim is trying to find all the local interfaces, it used to panic and
3966 die if the ioctl to get the interface flags failed. However, it seems that
3967 on at least one OS (Solaris 9) it is possible to have an interface that is
3968 included in the list of interfaces, but for which you get a failure error
3969 for this call. This happens when the interface is not "plumbed" into a
3970 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3971 failure of the "get flags" call assumes that the interface is down.
3973 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3974 makes life easier for people who are doing arithmetic on fields extracted
3975 from dates, where you often get leading zeros that should not be
3976 interpreted as octal.
3978 64. Added qualify_domain to the redirect router, to override the global
3981 65. If a pathologically long header line contained very many addresses (the
3982 report of this problem mentioned 10 000) and each of them was rewritten,
3983 Exim could use up a very large amount of memory. (It kept on making new
3984 copies of the header line as it rewrote, and never released the old ones.)
3985 At the expense of a bit more processing, the header rewriting function has
3986 been changed so that it no longer eats memory in this way.
3988 66. The generation of the Received: header has been moved from the time that a
3989 message starts to be received, to the time that it finishes. The timestamp
3990 in the Received: header should now be very close to that of the <= log
3991 line. There are two side-effects of this change:
3993 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3994 logged header lines no longer include the local Received: line, because
3995 it has not yet been created. The same applies to a copy of the message
3996 that is returned to a non-SMTP sender when a message is rejected.
3998 (b) When a filter file is tested using -bf, no additional Received: header
3999 is added to the test message. After some thought, I decided that this
4002 This change does not affect the value of $received_for. It is still set
4003 after address rewriting, but before local_scan() is called.
4005 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4007 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4008 gave an unhelpful panic error message, and a defer error. I have managed to
4009 change this behaviour so that it now rejects any supplied certificate,
4010 which seems right, as the list of acceptable certificates is empty.
4012 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4013 gave an unhelpful defer error. I have not managed to make this reject any
4014 supplied certificates, but the error message it gives is "no certificate
4015 supplied", which is not helpful.
4017 70. exigrep's output now also includes lines that are not associated with any
4018 message, but which match the given pattern. Implemented by a patch from
4019 Martin Sluka, which also tidied up the Perl a bit.
4021 71. Recipient callout verification, like sender verification, was using <> in
4022 the MAIL FROM command. This isn't really the right thing, since the actual
4023 sender may affect whether the remote host accepts the recipient or not. I
4024 have changed it to use the actual sender in the callout; this means that
4025 the cache record is now keyed on a recipient/sender pair, not just the
4026 recipient address. There doesn't seem to be a real danger of callout loops,
4027 since a callout by the remote host to check the sender would use <>.
4028 [SEE ABOVE: changed after hitting problems.]
4030 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4031 temporary errors. However, in the case of such a code being given after
4032 the end of a data transmission (i.e. after ".") Exim was failing to write
4033 a retry record for the message. (Yes, there was some broken host that was
4034 actually sending 8xx at this point.)
4036 73. An unknown lookup type in a host list could cause Exim to panic-die when
4037 the list was checked. (An example that provoked this was putting <; in the
4038 middle of a list instead of at the start.) If this happened during a DATA
4039 ACL check, a -D file could be left lying around. This kind of configuration
4040 error no longer causes Exim to die; instead it causes a defer errror. The
4041 incident is still logged to the main and panic logs.
4043 74. Buglet left over from Exim 3 conversion. The message "too many messages
4044 in one connection" was written to the rejectlog but not the mainlog, except
4045 when address rewriting (yes!) was being logged.
4047 75. Added write_rejectlog option.
4049 76. When a system filter was run not as root (that is, when system_filter_user
4050 was set), the values of the $n variables were not being returned to the
4051 main process; thus, they were not subsequently available in the $sn
4054 77. Added +return_path_on_delivery log selector.
4056 78. A connection timeout was being treated differently from recipients deferred
4057 when testing hosts_max_try with a message that was older than the host's
4058 retry timeout. (The host should not be counted, thus allowing all hosts to
4059 be tried at least once before bouncing.) This may have been the cause of an
4060 occasionally reported bug whereby a message would remain on the queue
4061 longer than the retry timeout, but would be bounced if a delivery was
4062 forced. I say "may" because I never totally pinned down the problem;
4063 setting up timeout/retry tests is difficult. See also the next item.
4065 79. The ultimate address timeout was not being applied to errors that involved
4066 a combination of host plus message (for example, a timeout on a MAIL
4067 command). When an address resolved to a number of possible hosts, and they
4068 were not all tried for each delivery (e.g. because of hosts_max_try), a
4069 message could remain on the queue longer than the retry timeout.
4071 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4074 81. Fixed an obscure SMTP outgoing bug which required at least the following
4075 conditions: (a) there was another message waiting for the same server;
4076 (b) the server returned 5xx to all RCPT commands in the first message so
4077 that the message was not completed; (c) the server dropped the connection
4078 or gave a negative response to the RSET that Exim sends to abort the
4079 transaction. The observed case was a dropped connection after DATA that had
4080 been sent in pipelining mode. That is, the server had advertised PIPELINING
4081 but was not implementing it correctly. The effect of the bug was incorrect
4082 behaviour, such as trying another host, and this could lead to a crash.
4088 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4089 and daemon.c were passed as pointers to ints; they should have been
4090 pointers to socklen_t variables (which are typically unsigned ints).
4092 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4095 3. Fixed a really odd bug that affected only the testing scheme; patching a
4096 certain fixed string in the binary changed the value of another string that
4097 happened to be identical to the end of the original first string.
4099 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4100 name", it returns that address as the IP address. On some operating
4101 systems (e.g. Solaris), it also passes back the IP address string as the
4102 "host name". However, on others (e.g. Linux), it passes back an empty
4103 string. Exim wasn't checking for this, and was changing the host name to an
4104 empty string, assuming it had been canonicized.
4106 5. Although rare, it is permitted to have more than one PTR record for a given
4107 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4108 all the names associated with an address, because they do in Solaris.
4109 However, it seems that they do not in Linux for data that comes from the
4110 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4111 I found this out when I moved to a new Linux workstation and tried to run
4112 the Exim test suite.
4114 To get round this problem I have changed the code so that it now does its
4115 own call to the DNS to look up PTR records when searching for a host name.
4116 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4117 addresses that are only in /etc/hosts are still found.
4119 This behaviour is, however, controlled by an option called host_lookup_
4120 order, which defaults to "bydns:byaddr". If people want to use the other
4121 order, or indeed, just use one or the other means of lookup, they can
4122 specify it in this variable.
4124 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4125 some operating systems, this comes back from gethostbyaddr() as an empty
4126 string, and this is what Exim used to test for. However, it seems that in
4127 other systems, "." is yielded. Exim now tests for this case too.
4129 7. The values of check_spool_space and check_log_space are now held internally
4130 as a number of kilobytes instead of an absolute number of bytes. If a
4131 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4132 kilobyte. This means that much larger values can be stored.
4134 8. Exim monitor: an attempt to get the action menu when not actually pointing
4135 at a message produces an empty menu entitled "No message selected". This
4136 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4137 no entries in it ("Shell widget menu has zero width and/or height"). So I
4138 have added a single, blank menu entry in this case.
4140 9. Added ${quote_local_part.
4142 10. MIME decoding is now applied to the contents of Subject: header lines when
4145 11. Now that a reference to $sender_host_address automatically causes a reverse
4146 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4147 host lookup before query-style lookups in lists that might use this
4148 variable. This has therefore been abolished, and the "net-" prefix is no
4149 longer necessary for query-style lookups.
4151 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4152 have been a typo for LFLAGS, so it has been changed.
4154 13. The install script calls Exim with "-C /dev/null" in order to find the
4155 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4156 to be output. Howeve, since Exim outputs its version number before the
4157 error, it didn't break the script. It just looked ugly. I fixed this by
4158 always allowing "-C /dev/null" if the caller is root.
4160 14. Ignore overlarge ACL variable number when reading spool file - insurance
4161 against a later release with more variables having written the file.
4163 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4164 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4165 this, as well as the form without the "IPv6" on the front.
4167 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4168 OS/Makefile-Darwin file.
4170 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4171 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4173 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4174 variable. However, some OS use ints for inodes. I've added cast to long int
4175 to get rid of the compiler warning.
4177 19. I had forgotten to lock out "/../" in configuration file names when
4178 ALT_CONFIG_PREFIX was set.
4180 20. Routers used for verification do not need to specify transports. However,
4181 if such a router generated a host list, and callout was configured, Exim
4182 crashed, because it could not find a port number from the (non-existent)
4183 transport. It now assumes port 25 in this circumstance.
4185 21. Added the -t option to exigrep.
4187 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4188 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4189 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4190 them is set, LOOKUP_LSEARCH is forced.
4192 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4194 24. Added sender and host information to the "rejected by local_scan()" log
4195 line; previously there was no indication of these.
4197 25. Added .include_if_exists.
4199 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4200 Linux. It turns out that not all file systems support this. Apparently some
4201 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4202 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4203 not supported on the file descriptor, is now ignored when Exim is trying to
4204 sync a directory. This applies only to Linux.
4206 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4208 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4209 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4212 29. Added support for maildirsize files from supplied patch (modified a bit).
4214 30. The use of :fail: followed by an empty string could lead Exim to respond to
4215 sender verification failures with (e.g.):
4217 550 Verification failed for <xxx>
4218 550 Sender verify failed
4220 where the first response line was missing the '-' that indicates it is not
4221 the final line of the response.
4223 31. The loop for finding the name of the user that called Exim had a hardwired
4224 limit of 10; it now uses the value of finduser_retries, which is used for
4225 all other user lookups.
4227 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4230 33. Exim was neglecting to zero errno before one call of strtol() when
4231 expanding a string and expecting an integer value. On some systems this
4232 resulted in spurious "integer overflow" errors. Also, it was casting the
4233 result into an int without checking.
4235 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4236 did not work. The code looks as if it has *never* worked, though it appears
4237 to have been documented since at least releast 1.62. I have made it work.
4239 35. The "timeout_DNS" error in retry rules, also documented since at least
4240 1.62, also never worked. As it isn't clear exactly what this means, and
4241 clearly it isn't a major issue, I have abolished the feature by treating it
4242 as "timeout", and writing a warning to the main and panic logs.
4244 36. The display of retry rules for -brt wasn't always showing the error code
4247 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4248 timeout_connect_A, timeout_connect_MX.
4250 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4251 to the empty sender.
4253 39. The daemon was not analysing the content of -oX till after it had closed
4254 stderr and disconnected from the controlling terminal. This meant that any
4255 syntax errors were only noted on the panic log, and the return code from
4256 the command was 0. By re-arranging the code a little, I've made the
4257 decoding happen first, so such errors now appear on stderr, and the return
4258 code is 1. However, the actual setting up of the sockets still happens in
4259 the disconnected process, so errors there are still only recorded on the
4262 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4263 connections (as happens on some IP stacks) was logged at start up time as
4264 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4265 it from "IPv6 and IPv4", which means that two separate sockets are being
4268 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4269 says whether AF_INET or AF_INET6 was passed as an argument.
4271 42. Exiwhat output was messed up when time zones were included in log
4274 43. Exiwhat now gives more information about the daemon's listening ports,
4275 and whether -tls-on-connect was used.
4277 44. The "port" option of the smtp transport is now expanded.
4279 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4280 silently ignored. Now an error message is written to the main and panic
4283 46. There's a new ACL modifier called "logwrite" which writes to a log file
4284 as soon as it is encountered.
4286 47. Added $local_user_uid and $local_user_gid at routing time.
4288 48. Exim crashed when trying to verify a sender address that was being
4291 49. Exim was recognizing only a space character after ".include". It now also
4292 recognizes a tab character.
4294 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4295 extracting the relevant information from the specification. The man page no
4296 longer contains scrambled data for the -d option, and I've added a section
4297 at the front about calling Exim under different names.
4299 51. Added "extra_headers" argument to the "mail" command in filter files.
4301 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4304 53. Installed eximstats 1.29.
4306 54. Added transport_filter_timeout as a generic transport option.
4308 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4309 Cc: header lines. This was required by RFC 822, but it not required by RFC
4312 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4313 incoming messages that did not have them. Now it does so only if the
4314 message originates locally, that is, if there is no associated remote host
4315 address. When Resent- header lines are present, this applies to the Resent-
4316 lines rather than the non-Resent- lines.
4318 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4319 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4321 58. Messages for configuration errors now include the name of the main
4322 configuration file - useful now that there may be more than one file in a
4323 list (.included file names were always shown).
4325 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4326 for those rare installations that do not start the daemon as root or run it
4327 setuid root. I've cut out the call to initgroups() if the daemon is not
4330 60. The Exim user and group can now be bound into the binary as text strings
4331 that are looked up at the start of Exim's processing.
4333 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4335 62. Added $mailstore_basename variable.
4337 63. Installed patch to sieve.c from Michael Haardt.
4339 64. When Exim failed to open the panic log after failing to open the main log,
4340 the original message it was trying to log was written to stderr and debug
4341 output, but if they were not available (the usual case in production), it
4342 was lost. Now it is written to syslog before the two lines that record the
4343 failures to open the logs.
4345 65. Users' Exim filters run in subprocesses under the user's uid. It is
4346 possible for a "deliver" command or an alias in a "personal" command to
4347 provoke an address rewrite. If logging of address rewriting is configured,
4348 this fails because the process is not running as root or exim. There may be
4349 a better way of dealing with this, but for the moment (because 4.30 needs
4350 to be released), I have disabled address rewrite logging when running a
4351 filter in a non-root, non-exim process.
4357 1. The buildconfig auxiliary program wasn't quoting the value set for
4358 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4359 not defined. This bug was masked in 4.22 by the effect that was fixed in
4362 2. Some messages that were rejected after a message id was allocated were
4363 shown as "incomplete" by exigrep. It no longer does this for messages that
4364 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4366 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4367 have allow_domain_literals set, the ID did not get logged in the <= line.
4368 Domain literals are now always recognized in Message-ID: header lines.
4370 4. The first argument for a ${extract expansion item is the key name or field
4371 number. Leading and trailing spaces in this item were not being ignored,
4372 causing some misleading effects.
4374 5. When deliver_drop_privilege was set, single queue runner processes started
4375 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4376 same command in the process it spins off) were not dropping privilege.
4378 6. When the daemon running as "exim" started a queue runner, it always
4379 re-executed Exim in the spun-off process. This is a waste of effort when
4380 deliver_drop_privilege is set. The new process now just calls the
4381 queue-runner function directly.
4387 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4390 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4391 ignored. Though the use of -oP was forcing the writing of a pid file, it
4392 was always written to the default place.
4394 3. If the message "no IP address found for host xxxx" is generated during
4395 incoming verification, it is now followed by identification of the incoming
4396 connection (so you can more easily find what provoked it).
4398 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4400 5. Added some features to "harden" Exim a bit more against certain attacks:
4402 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4403 be put in Local/Makefile. This is like the never_users runtime option,
4404 but it cannot be overridden. The default setting is "root".
4406 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4407 prefix string with which any file named in a -C command line option
4410 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4411 is retained for -C and -D only if the caller of Exim is root. Without
4412 it, the exim user may also use -C and -D and retain privilege.
4414 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4415 command line option is disabled.
4417 6. Macro names set by the -D option must start with an upper case letter, just
4418 like macro names defined in the configuration file.
4420 7. Added "dereference=" facility to LDAP.
4422 8. Two instances of the typo "uknown" in the source files are fixed.
4424 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4425 the Configure-Makefile script screwed up while processing it.
4427 10. Incorporated PCRE 4.4.
4429 11. The SMTP synchronization check was not operating right at the start of an
4430 SMTP session. For example, it could not catch a HELO sent before the client
4431 waited for the greeting. There is now a check for outstanding input at the
4432 point when the greeting is written. Because of the duplex, asynchronous
4433 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4434 way, but not yet received, when the check is performed.
4436 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4437 on TCP/IP sockets, because this apparently causes some broken clients to
4440 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4441 unchanged) from the Cygwin maintainer.
4443 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4444 was supported instead of testing for mbx. Effectively a typo.
4446 15. The spa authenticator server code was not checking that the input it
4447 received was valid base64.
4449 16. The debug output line for the "set" modifier in ACLs was not showing the
4450 name of the variable that was being set.
4452 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4454 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4455 lookup on its own. Something else was needed to trigger the lookup. For
4456 example, a match in host_lookup or the need for a host name in a host list.
4457 Now, if $sender_host_name is referenced and the host name has not yet been
4458 looked up, a lookup is performed. If the lookup fails, the variable remains
4459 empty, and $host_lookup_failed is set to "1".
4461 19. Added "eqi" as a case-independent comparison operator.
4463 20. The saslauthd authentication condition could segfault if neither service
4464 nor realm was specified.
4466 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4467 error message that was logged was misleading, and incoming SMTP
4468 connections were dropped. The message is now more accurate, and temporary
4469 errors are given to SMTP connections.
4471 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4472 (except RSET) with a 421 error, until QUIT is received. However, it was
4473 failing to send a response to QUIT.
4475 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4476 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4477 ACL is now run after the helo_try_verify_hosts code.
4479 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4480 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4481 case-independent, so other case variants are also recognized). Apparently
4482 some systems use these upper case variants.
4484 25. If more than two messages were waiting for the same host, and a transport
4485 filter was specified for the transport, Exim sent two messages over the
4486 same TCP/IP connection, and then failed with "socket operation on non-
4487 socket" when it tried to send the third.
4489 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4491 27. The extern definition of crypt16() in expand.c was not being excluded when
4492 the OS had its own crypt16() function.
4494 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4495 as a preferred synonym for return_size_limit, both as an option and as an
4498 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4500 30. Changed the default configuration ACL to relax the local part checking rule
4501 for addresses that are not in any local domains. For these addresses,
4502 slashes and pipe symbols are allowed within local parts, but the sequence
4503 /../ is explicitly forbidden.
4505 31. SPA server authentication was not clearing the challenge buffer before
4508 32. log_message in a "warn" ACL statement was writing to the reject log as
4509 well as to the main log, which contradicts the documentation and doesn't
4510 seem right (because no rejection is happening). So I have stopped it.
4512 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4513 However, I am unable to do any testing of this.
4515 34. Fixed an infelicity in the appendfile transport. When checking directories
4516 for a mailbox, to see if any needed to be created, it was accidentally
4517 using path names with one or more superfluous leading slashes; tracing
4518 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4520 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4521 contents are added to the log line that is written for every discarded
4522 recipient. (Previously a log_message setting was ignored.)
4524 36. The ${quote: operator now quotes the string if it is empty.
4526 37. The install script runs exim in order to find its version number. If for
4527 some reason other than non-existence or emptiness, which it checks, it
4528 could not run './exim', it was installing it with an empty version number,
4529 i.e. as "exim-". This error state is now caught, and the installation is
4532 38. An argument was missing from the function that creates an error message
4533 when Exim fails to connect to the socket for saslauthd authentication.
4534 This could cause Exim to crash, or give a corrupted message.
4536 39. Added isip, isip4, and isip6 to ${if conditions.
4538 40. The ACL variables $acl_xx are now saved with the message, and can be
4539 accessed later in routers, transports, and filters.
4541 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4542 strings in the file are not string-expanded.
4544 42. If a MAIL command specified a SIZE value that was too large to fit into an
4545 int variable, the check against message_size_limit failed. Such values are
4546 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4547 day this will have to be increased, but I don't think I want to be around
4548 when emails are that large.
4555 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4556 iconv() is not standard in FreeBSD.
4558 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4559 IPv6 enabled. The observed symptom was a segmentation fault on return from
4560 the function os_common_find_running_interfaces() in src/os.c.
4562 3. In the check_special_case() function in daemon.c I had used "errno" as an
4563 argument name, which causes warnings on some systems. This was basically a
4564 typo, since it was named "eno" in the comments!
4566 4. The code that waits for the clock to tick (at a resolution of some fraction
4567 of a second) so as to ensure message-id uniqueness was always waiting for
4568 at least one whole tick, when it could have waited for less. [This is
4569 almost certainly not relevant at current processor speeds, where it is
4570 unlikely to ever wait at all. But we try to future-proof.]
4572 5. The function that sleeps for a time interval that includes fractions of a
4573 second contained a race. It did not block SIGALRM between setting the
4574 timer, and suspending (a couple of lines later). If the interval was short
4575 and the sigsuspend() was delayed until after it had expired, the suspension
4576 never ended. On busy systems this could lead to processes getting stuck for
4579 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4580 process, before it forks any delivery processes. The open lookup caching
4581 mechanism meant that the open file or database connection was passed into
4582 the delivery process. The problem was that delivery processes always tidy
4583 up cached lookup data. This could cause a problem for the next delivery
4584 process started by the queue runner, because the external queue runner
4585 process does not know about the closure. So the next delivery process
4586 still has data in the lookup cache. In the case of a file lookup, there was
4587 no problem because closing a file descriptor in a subprocess doesn't affect
4588 the parent. However, if the lookup was caching a connection to a database,
4589 the connection was closed, and the second delivery process was likely to
4590 see errors such as "PGSQL: query failed: server closed the connection
4591 unexpectedly". The problem has been fixed by closing all cached lookups
4592 in a queue runner before running a delivery process.
4594 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4595 seem to have the "const" qualifier which it has on other OS. I've
4598 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4599 *of the same type* (client or server) with the same public name that an
4600 error should be diagnosed.
4602 9. When Exim looked up a host name for an IP address, but failed to find the
4603 original IP address when looking up the host name (a safety check), it
4604 output the message "<ip address> does not match any IP for NULL", which was
4605 confusing, to say the least. The bug was that the host name should have
4606 appeared instead of "NULL".
4608 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4609 user that is built into the binary, and the -C or -D options is used, root
4610 privilege is dropped before the configuration file is read. In addition,
4611 logging is switched to stderr instead of the normal log files. If the
4612 configuration then re-defines the Exim user, the unprivileged environment
4613 is probably not what is expected, so Exim logs a panic warning message (but
4616 However, if deliver_drop_privilege is set, the unprivileged state may well
4617 be exactly what is intended, so the warning has been cut out in that case,
4618 and Exim is allowed to try to write to its normal log files.
4624 1. smtp_return_error_details was not giving details for temporary sender
4625 or receiver verification errors.
4627 2. Diagnose a configuration error if two authenticators have the same public
4630 3. Exim used not to create the message log file for a message until the first
4631 delivery attempt. This could be confusing when incoming messages were held
4632 for policy or load reasons. The message log file is now created at the time
4633 the message is received, and an initial "Received" line is written to it.
4635 4. The automatically generated man page for command line options had a minor
4636 bug that caused no ill effects; however, a more serious problem was that
4637 the procedure for building the man page automatically didn't always
4638 operate. Consequently, release 4.20 contains an out-of-date version. This
4639 shouldn't happen again.
4641 5. When building Exim with embedded Perl support, the script that builds the
4642 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4643 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4645 6. The freeze_tell option was not being used for messages that were frozen on
4646 arrival, either by an ACL or by local_scan().
4648 7. Added the smtp_incomplete_transaction log selector.
4650 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4651 was accepting AUTH without a new EHLO.
4653 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4654 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4657 10. Logging of TCP/IP connections (when configured) now happens in the main
4658 daemon process instead of the child process, so that the TCP/IP connection
4659 count is more accurate (but it can never be perfect).
4661 11. The use of "drop" in a nested ACL was not being handled correctly in the
4662 outer ACL. Now, if condition failure induced by the nested "drop" causes
4663 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4664 or "require"), the connection is dropped.
4666 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4667 that yield "discard" can now be used with an "accept" or a "discard" verb,
4668 but an error is generated for any others (because I can't see a useful way
4669 to define what should happen).
4671 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4672 are now processed in the same way as lines in the Exim configuration file.
4673 In particular, continuation lines are supported.
4675 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4677 15. Added -ti meaning -t -i.
4679 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4680 domains, and warn by logging if others are found.
4682 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4683 returned from GIFCONF when Exim is trying to find the list of interfaces on
4684 a host. The code in os.c has been modified to copy each ifreq to an aligned
4685 structure in all cases.
4687 Also, in some cases, the returned ifreq's were being copied to a 'struct
4688 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4689 means the last couple of bytes of an IPv6 address could be chopped if the
4690 ifreq contained only a normal sockaddr (14 bytes storage).
4692 18. Named domain lists were not supported in the hosts_treat_as_local option.
4693 An entry such as +xxxx was not recognized, and was treated as a literal
4696 19. Ensure that header lines added by a DATA ACL are included in the reject log
4697 if the ACL subsequently rejects the message.
4699 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4700 MD5 (which is deprecated).
4702 21. When testing a filter file using -bf, Exim was writing a message when it
4703 took the sender from a "From " line in the message, but it was not doing so
4704 when it took $return_path from a Return-Path: header line. It now does.
4706 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4707 with a valid header line field name (a series of printing characters
4708 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4710 23. Changed "disc" in the source to "disk" to conform to the documentation and
4711 the book and for uniformity.
4713 24. Ignore Sendmail's -Ooption=value command line item.
4715 25. When execve() failed while trying to run a command in a pipe transport,
4716 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4717 could be confused with a return value of 69 from the command itself. This
4718 has been changed to 127, the value the shell returns if it is asked to run
4719 a non-existent command. The wording for the related log line suggests a
4720 non-existent command as the problem.
4722 26. If received_header_text expands to an empty string, do not add a Received:
4723 header line to the message. (Well, it adds a token one on the spool, but
4724 marks it "old" so that it doesn't get used or transmitted.)
4726 27. Installed eximstats 1.28 (addition of -nt option).
4728 28. There was no check for failure on the call to getsockname() in the daemon
4729 code. This can fail if there is a shortage of resources on the system, with
4730 ENOMEM, for example. A temporary error is now given on failure.
4732 29. Contrary to the C standard, it seems that in some environments, the
4733 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4734 program. Exim now does this explicitly; it affects the formatting of
4735 timestamps using strftime().
4737 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4738 complaints. I've now initialized all the variables, to avoid this.
4740 32. Header lines added by a system filter were not being "seen" during
4741 transport-time rewrites.
4743 33. The info_callback() function passed to OpenSSL is set up with type void
4744 (*)(SSL *, int, int), as described somewhere. However, when calling the
4745 function (actually a macro) that sets it up, the type void(*)() is
4746 expected. I've put in a cast to prevent warnings from picky compilers.
4748 34. If a DNS black list lookup found a CNAME record, but there were no A
4749 records associated with the domain it pointed at, Exim crashed.
4751 35. If a DNS black list lookup returned more than one A record, Exim ignored
4752 all but the first. It now scans all returned addresses if a particular IP
4753 value is being sought. In this situation, the contents of the
4754 $dnslist_value variable are a list of all the addresses, separated by a
4757 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4758 to accept a host name and all its aliases if the forward lookup for any of
4759 them yielded the IP address of the incoming connection. Now it accepts only
4760 those names whose forward lookup yields the correct IP address. Any other
4761 names are discarded. This closes a loophole whereby a rogue DNS
4762 administrator could create reverse DNS records to break through a
4763 wildcarded host restriction in an ACL.
4765 37. If a user filter or a system filter that ran in a subprocess used any of
4766 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4767 the wrong values were passed to the pipe command ($thisaddress had the
4768 value of $0, $0 had the value of $1, etc). This bug was introduced by
4769 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4771 38. Improved the line breaking for long SMTP error messages from ACLs.
4772 Previously, if there was no break point between 40 and 75 characters, Exim
4773 left the rest of the message alone. Two changes have been made: (a) I've
4774 reduced the minimum length to 35 characters; (b) if it can't find a break
4775 point between 35 and 75 characters, it looks ahead and uses the first one
4776 that it finds. This may give the occasional overlong line, but at least the
4777 remaining text gets split now.
4779 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4780 file descriptors might be low, and that setting 1000 would always raise it.
4781 It turns out that in some environments, the limit is already over 1000 and
4782 that lowering it causes trouble. So now Exim takes care not to decrease it.
4784 40. When delivering a message, the value of $return_path is set to $sender_
4785 address at the start of routing (routers may change the value). By an
4786 oversight, this default was not being set up when an address was tested by
4787 -bt or -bv, which affected the outcome if any router or filter referred to
4790 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4791 log only when "message" or "log_message" are set. However, if one of the
4792 conditions was an address verification, or a call to a nested ACL, the
4793 messages generated by the underlying test were being passed through. This
4794 no longer happens. The underlying message is available in $acl_verify_
4795 message for both "message" and "log_message" expansions, so it can be
4796 passed through if needed.
4798 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4799 new expansion $bh_ to give the encoded byte string without charset
4800 translation. Translation happens only if iconv() is available; HAVE_ICONV
4801 indicates this at build time. HEADERS_CHARSET gives the charset to
4802 translate to; headers_charset can change it in the configuration, and
4803 "headers charset" can change it in an individual filter file.
4805 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4806 that creates RFC 2047 encoded "words" labels them as that charset instead
4807 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4808 expansion operator; (ii) when Exim creates a From: line for a local
4809 message; (iii) when a header line is rewritten to include a "phrase" part.
4811 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4812 buggy, causing it to skip the first lines of messages whose message ID
4813 ended in 'D'. This would not have bitten before Exim release 4.14, because
4814 message IDs were unlikely to end in 'D' before then. The effect was to have
4815 incorrect size information for certain domains.
4817 45. #include "config.h" was missing at the start of the crypt16.c module. This
4818 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4821 46. If there was a timeout during a "random" callout check, Exim treated it as
4822 a failure of the random address, and carried on sending RSET and the real
4823 address. If the delay was just some slowness somewhere, the response to the
4824 original RCPT would be taken as a response to RSET and so on, causing
4825 mayhem of various kinds.
4827 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4828 when I implemented it. It didn't allow for the fact that some option values
4829 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4830 the right test for positive values.
4832 48. Domain names in DNS records are case-independent. Exim always looks them up
4833 in lower case. Some resolvers return domain names in exactly the case they
4834 appear in the zone file, that is, they may contain uppercase letters. Not
4835 all resolvers do this - some return always lower case. Exim was treating a
4836 change of case by a resolver as a change of domain, similar to a widening
4837 of a domain abbreviation. This triggered its re-routing code and so it was
4838 trying to route what was effectively the same domain again. This normally
4839 caused routing to fail (because the router wouldn't handle the domain
4840 twice). Now Exim checks for this case specially, and just changes the
4841 casing of the domain that it ultimately uses when it transmits the message
4844 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4847 50. If a filter generated a file delivery with a non-absolute name (possible if
4848 no home directory exists for the router), the forbid_file option was not
4851 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4852 the existing equality matching.
4854 52. Exim was using ints instead of ino_t variables in some places where it was
4855 dealing with inode numbers.
4857 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4858 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4859 called TMPDIR, and if it finds it is different, it changes its value.
4861 54. The smtp_printf() function is now made available to local_scan() so
4862 additional output lines can be written before returning. There is also an
4863 smtp_fflush() function to enable the detection of a dropped connection.
4864 The variables smtp_input and smtp_batched_input are exported to
4867 55. Changed the default runtime configuration: the message "Unknown user"
4868 has been removed from the ACL, and instead placed on the localuser router,
4869 using the cannot_route_message feature. This means that any verification
4870 failures that generate their own messages won't get overridden. Similarly,
4871 the "Unrouteable address" message that was in the ACL for unverifiable
4872 relay addresses has also been removed.
4874 56. Added hosts_avoid_esmtp to the smtp transport.
4876 57. The exicyclog script was not checking for the esoteric option
4877 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4878 will work only if exicyclog is run under the appropriate euid.
4880 58. Following a discussion on the list, the rules by which Exim recognises line
4881 endings on incoming messages have been changed. The -dropcr and drop_cr
4882 options are now no-ops, retained only for backwards compatibility. The
4883 following line terminators are recognized: LF CRLF CR. However, special
4884 processing applies to CR:
4886 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4887 nor a local message in the state where . is a terminator.
4889 (ii) If a bare CR is encountered in a header line, an extra space is added
4890 after the line terminator so as not to end the header. The reasoning
4891 behind this is that bare CRs in header lines are most likely either
4892 to be mistakes, or people trying to play silly games.
4894 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4895 was being incorrectly given as 18 bytes larger than it should have been.
4896 This is a VOB (very old bug).
4898 60. This may never have affected anything current, but just in case it has:
4899 When the local host is found other than at the start of a list of hosts,
4900 the local host, those with the same MX, and any that follow, are discarded.
4901 When the list in question was part of a longer list of hosts, the following
4902 hosts (not currently being processed) were also being discarded. This no
4903 longer happens. I'm not sure if this situation could ever has previously
4906 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4909 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4910 header. This is something that is recommended in a new Internet Draft, and
4911 is something that is documented as being done by Sendmail. There are two
4912 possible values. For messages generated by the autoreply transport, Exim
4915 Auto-Submitted: auto-replied
4917 whereas for all other generated messages (e.g. bounces) it adds
4919 Auto-Submitted: auto-generated
4921 63. The "personal" condition in filters now includes a test for the
4922 Auto-Submitted: header. If it contains the string "auto-" the message it
4923 not considered personal.
4925 64. Added rcpt_include_affixes as a generic transport option.
4927 65. Added queue_only_override (default true).
4929 66. Added the syslog_duplication option.
4931 67. If what should have been the first header line of a message consisted of
4932 a space followed by a colon, Exim was mis-interpreting it as a header line.
4933 It isn't of course - it is syntactically invalid and should therefore be
4934 treated as the start of the message body. The misbehaviour could have
4935 caused a number of strange effects, including loss of data in subsequent
4936 header lines, and spool format errors.
4938 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4939 client host had authenticated. This control can now be exercised by an ACL
4940 for more flexibility.
4942 69. By default, callouts do not happen when testing with -bh. There is now a
4943 variant, -bhc, which does actually run the callout code, including
4944 consulting and updating the callout cache.
4946 70. Added support for saslauthd authentication, courtesy of Alexander
4949 71. If statvfs() failed on the spool or log directories while checking their
4950 size for availability, Exim confusingly gave the error "space shortage".
4951 Furthermore, in debugging mode it crashed with a floating point exception.
4952 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4953 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4954 serious problem, Exim now writes to the main and panic logs when this
4955 happens, with details of the failure. It then refuses to accept the
4956 incoming message, giving the message "spool directory problem" or "log
4957 directory problem" with a 421 code for SMTP messages.
4959 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4960 0, 1, and 2 exist, because some OS complain for execs without them (see
4961 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4962 descriptors. However, the code omitted to check that the open succeeded,
4963 causing mysterious errors if for some reason the permissions on /dev/null
4964 got screwed. Now Exim writes a message to the main and panic logs, and
4965 bombs out if it can't open /dev/null.
4967 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4968 interact so that it is all more flexible. It is supposed to remain
4969 backwards compatible. Also added extra_local_interfaces.
4971 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4972 to bomb out with an assertion failure - to the client this appears as a
4973 connection drop. This problem occurs in the part of the code that was taken
4974 from the Samba project. Fortunately, the assertion is in a very simple
4975 function, so I have fixed this by reproducing the function inline in the
4976 one place where it is called, and arranging for authentication to fail
4977 instead of killing the process with assert().
4979 75. The SPA client code was not working when the server requested OEM rather
4980 than Unicode encoding.
4982 76. Added code to make require_files with a specific uid setting more usable in
4983 the case where statting the file as root fails - usually a non-root-mounted
4984 NFS file system. When this happens and the failure is EACCES, Exim now
4985 forks a subprocess and does the per-uid checking as the relevant uid.
4987 77. Added process_log_path.
4989 78. If log_file_path was not explicitly set, a setting of check_log_space or
4990 check_log_inodes was ignored.
4992 79. If a space check for the spool or log partitions fails, the incident is now
4993 logged. Of course, in the latter case the data may get lost...
4995 80. Added the %p formatting code to string_format() so that it can be used to
4996 print addresses in debug_print(). Adjusted all the address printing in the
4997 debugging in store.c to use %p rather than %d.
4999 81. There was a concern that a line of code in smtp_in.c could overflow a
5000 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
5001 initially expressed, the concern was not well-founded, because trailing
5002 spaces are removed early. However, if the trailing spaces were followed by
5003 a NULL, they did not get removed, so the overflow was possible. Two fixes
5006 (a) I re-wrote the offending code in a cleaner fashion.
5007 (b) If an incoming SMTP command contains a NULL character, it is rejected
5010 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5011 runs initgroups(), so that if the Exim user is in any additional groups,
5012 they will be used during message reception.
5018 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob