1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.590 2009/11/12 08:34:23 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
160 TK/01 Add preliminary DKIM support. Currently requires a forked version of
161 ALT-N's libdkim that I have put here:
162 http://duncanthrax.net/exim-experimental/
164 Note to Michael Haardt: I had to rename some vars in sieve.c. They
165 were called 'true' and it seems that C99 defines that as a reserved
166 keyword to be used with 'bool' variable types. That means you could
167 not include C99-style headers which use bools without triggering
168 build errors in sieve.c.
170 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
171 as mailq or other aliases. Changed the --help handling significantly
172 to do whats expected. exim_usage() emits usage/help information.
174 SC/01 Added the -bylocaldomain option to eximstats.
176 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
178 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
180 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
186 PH/01 Another patch from the Sieve maintainer.
188 PH/02 When an IPv6 address is converted to a string for single-key lookup
189 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
190 dots are used instead of colons so that keys in lsearch files need not
191 contain colons. This was done some time before quoting was made available
192 in lsearch files. However, iplsearch files do require colons in IPv6 keys
193 (notated using the quote facility) so as to distinguish them from IPv4
194 keys. This meant that lookups for IP addresses in host lists did not work
195 for iplsearch lookups.
197 This has been fixed by arranging for IPv6 addresses to be expressed with
198 colons if the lookup type is iplsearch. This is not incompatible, because
199 previously such lookups could never work.
201 The situation is now rather anomolous, since one *can* have colons in
202 ordinary lsearch keys. However, making the change in all cases is
203 incompatible and would probably break a number of configurations.
205 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
208 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
209 conversion specification without a maximum field width, thereby enabling
210 a rogue spamd server to cause a buffer overflow. While nobody in their
211 right mind would setup Exim to query an untrusted spamd server, an
212 attacker that gains access to a server running spamd could potentially
213 exploit this vulnerability to run arbitrary code as the Exim user.
215 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
216 $primary_hostname instead of what libspf2 thinks the hosts name is.
218 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
219 a directory entry by the name of the lookup key. Previously, if a
220 symlink pointed to a non-existing file or a file in a directory that
221 Exim lacked permissions to read, a lookup for a key matching that
222 symlink would fail. Now it is enough that a matching directory entry
223 exists, symlink or not. (Bugzilla 503.)
225 PH/03 The body_linecount and body_zerocount variables are now exported in the
228 PH/04 Added the $dnslist_matched variable.
230 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
231 This means they are set thereafter only if the connection becomes
234 PH/06 Added the client_condition to authenticators so that some can be skipped
235 by clients under certain conditions.
237 PH/07 The error message for a badly-placed control=no_multiline_responses left
238 "_responses" off the end of the name.
240 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
242 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
243 (without spaces) instead of just copying the configuration text.
245 PH/10 Added the /noupdate option to the ratelimit ACL condition.
247 PH/11 Added $max_received_linelength.
249 PH/12 Added +ignore_defer and +include_defer to host lists.
251 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
252 way in which PCRE > 7.0 is built.
254 PH/14 Implemented queue_only_load_latch.
256 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
257 MAIL command. The effect was to mangle the value on 64-bit systems.
259 PH/16 Another patch from the Sieve maintainer.
261 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
263 PH/18 If a system quota error occurred while trying to create the file for
264 a maildir delivery, the message "Mailbox is full" was not appended to the
265 bounce if the delivery eventually timed out. Change 4.67/27 below applied
266 only to a quota excession during the actual writing of the file.
268 PH/19 It seems that peer DN values may contain newlines (and other non-printing
269 characters?) which causes problems in log lines. The DN values are now
270 passed through string_printing() before being added to log lines.
272 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
273 and InterBase are left for another time.)
275 PH/21 Added message_body_newlines option.
277 PH/22 Guard against possible overflow in moan_check_errorcopy().
279 PH/23 POSIX allows open() to be a macro; guard against that.
281 PH/24 If the recipient of an error message contained an @ in the local part
282 (suitably quoted, of course), incorrect values were put in $domain and
283 $local_part during the evaluation of errors_copy.
289 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
290 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
293 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
294 issue a MAIL command.
296 PH/02 In an ACL statement such as
298 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
300 if a client was not listed at all, or was listed with a value other than
301 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
302 the condition was not true (as it should be), so access was not denied.
303 The bug was that the ! inversion was incorrectly passed on to the second
304 item. This has been fixed.
306 PH/03 Added additional dnslists conditions == and =& which are different from
307 = and & when the dns lookup returns more than one IP address.
309 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
310 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
312 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
313 FSYNC, which compiles an option called disable_fsync that allows for
314 bypassing fsync(). The documentation is heavily laced with warnings.
316 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
318 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
319 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
320 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
321 including adding "make clean"; (3) Added -fPIC when compiling the test
322 dynamically loaded module, to get rid of a warning.
324 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
325 message fails, move_frozen_messages = true and ignore_bounce_errors_after
326 = 0s. The bug is otherwise harmless.
328 PH/07 There was a bug in the dovecot authenticator such that the value of
329 $auth1 could be overwritten, and so not correctly preserved, after a
330 successful authentication. This usually meant that the value preserved by
331 the server_setid option was incorrect.
333 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
335 PH/09 Installed PCRE release 7.0.
337 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
338 run for batched SMTP input. It is now run at the start of every message
339 in the batch. While fixing this I discovered that the process information
340 (output by running exiwhat) was not always getting set for -bs and -bS
341 input. This is fixed, and it now also says "batched" for BSMTP.
343 PH/11 Added control=no_pipelining.
345 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
346 patch, slightly modified), and move the expansion of helo_data till after
347 the connection is made in the smtp transport (so it can use these
350 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
352 PH/14 Added log_selector = +pid.
354 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
356 PH/16 Add ${if forany and ${if forall.
358 PH/17 Added dsn_from option to vary the From: line in DSNs.
360 PH/18 Flush SMTP output before performing a callout, unless control =
361 no_callout_flush is set.
363 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
364 was true (the default) a successful delivery failed to delete the retry
365 item, thus causing premature timeout of the address. The bug is now
368 PH/20 Added hosts_avoid_pipelining to the smtp transport.
370 PH/21 Long custom messages for fakedefer and fakereject are now split up
371 into multiline reponses in the same way that messages for "deny" and
372 other ACL rejections are.
374 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
375 with slight modification.
377 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
378 draft, changing the syntax and factoring some duplicate code".
380 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
381 for deliveries of the second and subsequent messages over the same SMTP
384 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
385 ${reduce, with only minor "tidies".
387 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
389 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
390 expansion side effects.
392 PH/27 When a message times out after an over-quota error from an Exim-imposed
393 quota, the bounce message says "mailbox is full". This message was not
394 being given when it was a system quota that was exceeded. It now should
397 MH/03 Made $recipients available in local_scan(). local_scan() already has
398 better access to the recipient list through recipients_list[], but
399 $recipients can be useful in postmaster-provided expansion strings.
401 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
402 in the case of a MAIL command with additional options following the
403 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
404 were accidentally chopped off.
406 PH/29 SMTP synchronization checks are implemented when a command is read -
407 there is a check that no more input is waiting when there shouldn't be
408 any. However, for some commands, a delay in an ACL can mean that it is
409 some time before the response is written. In this time, more input might
410 arrive, invalidly. So now there are extra checks after an ACL has run for
411 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
412 pipelining has not been advertised.
414 PH/30 MH's patch to allow iscntrl() characters to be list separators.
416 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
417 returned in the SMTP response when smtp_return_error_details was false.
420 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
421 instead of the C I/O that was originally supplied, because problems were
424 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
425 Exim which did not show up earlier: it was assuming that a call to
426 SSL_CTX_set_info_callback() might give an error value. In fact, there is
427 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
428 was a macro that became an assignment, so it seemed to work. This has
429 changed to a proper function call with a void return, hence the compile
430 error. Exim's code has been fixed.
432 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
435 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
437 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
439 PH/37 If a message is not accepted after it has had an id assigned (e.g.
440 because it turns out to be too big or there is a timeout) there is no
441 "Completed" line in the log. When some messages of this type were
442 selected by exigrep, they were listed as "not completed". Others were
443 picked up by some special patterns. I have improved the selection
444 criteria to be more general.
446 PH/38 The host_find_failed option in the manualroute router can now be set
447 to "ignore", to completely ignore a host whose IP address cannot be
448 found. If all hosts are ignored, the behaviour is controlled by the new
449 host_all_ignored option.
451 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
452 homing or because of multiple MX records with /mx) generated more than
453 one IP address, and the following item turned out to be the local host,
454 all the secondary addresses of the first item were incorrectly removed
455 from the list, along with the local host and any following hosts (which
456 is what is supposed to happen).
458 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
459 whoever called Exim into the -H file. In the case of the daemon it was
460 behaving confusingly. When first started, it used values for whoever
461 started the daemon, but after a SIGHUP it used the Exim user (because it
462 calls itself on a restart). I have changed the code so that it now always
465 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
466 message are rejected with the same error (e.g. no authentication or bad
467 sender address), and a DATA command is nevertheless sent (as can happen
468 with PIPELINING or a stupid MUA), the error message that was given to the
469 RCPT commands is included in the rejection of the DATA command. This is
470 intended to be helpful for MUAs that show only the final error to their
473 PH/42 Another patch from the Sieve maintainer.
475 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
476 Eximstats - Fixed some broken HTML links and added missing column headers
478 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
479 columns for Rejects, Temp Rejects, Ham, and Spam rows.
481 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
483 PH/43 Yet another patch from the Sieve maintainer.
485 PH/44 I found a way to check for a TCP/IP connection going away before sending
486 the response to the final '.' that terminates a message, but only in the
487 case where the client has not sent further data following the '.'
488 (unfortunately, this is allowed). However, in many cases there won't be
489 any further data because there won't be any more messages to send. A call
490 to select() can be used: if it shows that the input is "ready", there is
491 either input waiting, or the socket has been closed. An attempt to read
492 the next input character can distinguish the two cases. Previously, Exim
493 would have sent an OK response which the client would never have see.
494 This could lead to message repetition. This fix should cure that, at
495 least in a lot of common cases.
497 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
498 advertised in response to EHLO.
504 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
505 fixed by 4.65/MH/01 (is this a record?) are fixed:
507 (i) An empty string was always treated as zero by the numeric comparison
508 operators. This behaviour has been restored.
510 (ii) It is documented that the numeric comparison operators always treat
511 their arguments as decimal numbers. This was broken in that numbers
512 starting with 0 were being interpreted as octal.
514 While fixing these problems I realized that there was another issue that
515 hadn't been noticed. Values of message_size_limit (both the global option
516 and the transport option) were treated as octal if they started with 0.
517 The documentation was vague. These values are now always treated as
518 decimal, and I will make that clear in the documentation.
524 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
525 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
528 MH/01 Don't check that the operands of numeric comparison operators are
529 integers when their expansion is in "skipping" mode (fixes bug
530 introduced by 4.64-PH/07).
532 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
533 child addresses, Exim now panics and dies. Previously, because the count
534 is held in a short int, deliveries were likely to be lost. As such a
535 large number of recipients for a single message is ridiculous
536 (performance will be very, very poor), I have chosen to impose a limit
537 rather than extend the field.
543 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
544 leftover -K file (the existence of which was triggered by #402).
545 While we were at it, introduced process PID as part of the -K
546 filename. This should rule out race conditions when creating
549 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
550 processing considerably. Previous code took too long for large mails,
551 triggering a timeout which in turn triggers #401.
553 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
554 in the DK code in transports.c. sendfile() is not really portable,
555 hence the _LINUX specificness.
557 TF/01 In the add_headers option to the mail command in an Exim filter,
558 there was a bug that Exim would claim a syntax error in any
559 header after the first one which had an odd number of characters
562 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
563 callout verification, Exim cached a "reject" for the entire domain. This
564 is correct for most verifications, but it is not correct for a recipient
565 verification with use_sender or use_postmaster set, because in that case
566 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
567 case of MAIL FROM:<> rejection from other early rejections (e.g.
568 rejection of HELO). When verifying a recipient using a non-null MAIL
569 address, the cache is ignored if it shows MAIL FROM:<> rejection.
570 Whatever the result of the callout, the value of the domain cache is
571 left unchanged (for any other kind of callout, getting as far as trying
572 RCPT means that the domain itself is ok).
574 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
577 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
578 manifest itself as EPIPE rather than ECONNECT. When tidying away a
579 session, the daemon ignores ECONNECT errors and logs others; it now
580 ignores EPIPE as well.
582 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
583 (quoted-printable decoding).
585 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
586 later the small subsequent patch to fix an introduced bug.
588 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
590 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
592 PH/08 An error is now given if message_size_limit is specified negative.
594 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
595 to be given (somewhat) arbitrary names.
597 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
600 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
601 miscellaneous code fixes
603 PH/10 Added the log_reject_target ACL modifier to specify where to log
606 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
607 hostname. This is wrong, because it relates to the incoming message (and
608 probably the interface on which it is arriving) and not to the outgoing
609 callout (which could be using a different interface). This has been
610 changed to use the value of the helo_data option from the smtp transport
611 instead - this is what is used when a message is actually being sent. If
612 there is no remote transport (possible with a router that sets up host
613 addresses), $smtp_active_hostname is used.
615 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
616 tweaks were necessary in order to get it to work (see also 21 below):
617 (a) The code assumed that strncpy() returns a negative number on buffer
618 overflow, which isn't the case. Replaced with Exim's string_format()
620 (b) There were several signed/unsigned issues. I just did the minimum
621 hacking in of casts. There is scope for a larger refactoring.
622 (c) The code used strcasecmp() which is not a standard C function.
623 Replaced with Exim's strcmpic() function.
624 (d) The code set only $1; it now sets $auth1 as well.
625 (e) A simple test gave the error "authentication client didn't specify
626 service in request". It would seem that Dovecot has changed its
627 interface. Fortunately there's a specification; I followed it and
628 changed what the client sends and it appears to be working now.
630 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
633 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
634 address A is aliased to B and C, where B exists and C does not. Without
635 -v the output is "A verified" because verification stops after a
636 successful redirection if more than one address is generated. However,
637 with -v the child addresses are also verified. Exim was outputting "A
638 failed to verify" and then showing the successful verification for C,
639 with its parentage. It now outputs "B failed to verify", showing B's
640 parentage before showing the successful verification of C.
642 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
643 look up a TXT record in a specific list after matching in a combined
646 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
647 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
648 they consult the DNS. I had assumed they would set it the way they
649 wanted; and indeed my experiments on Linux seem to show that in some
650 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
651 To be on the safe side, however, I have now made the interface to
652 host_find_byname() similar to host_find_bydns(), with an argument
653 containing the DNS resolver options. The host_find_byname() function now
654 sets these options at its start, just as host_find_bydns() does. The smtp
655 transport options dns_qualify_single and dns_search_parents are passed to
656 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
657 of host_find_byname() use the default settings of RES_DEFNAMES
658 (qualify_single) but not RES_DNSRCH (search_parents).
660 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
661 spool_read_header() do less string testing, by means of a preliminary
662 switch on the second character of optional "-foo" lines. (This is
663 overdue, caused by the large number of possibilities that now exist.
664 Originally there were few.) While I was there, I also converted the
665 str(n)cmp tests so they don't re-test the leading "-" and the first
666 character, in the hope this might squeeze out yet more improvement.
668 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
669 flag allowing group syntax was set by the header_syntax check but not
670 turned off, possible causing trouble later; (2) The flag was not being
671 set at all for the header_verify test, causing "group"-style headers to
672 be rejected. I have now set it in this case, and also caused header_
673 verify to ignore an empty address taken from a group. While doing this, I
674 came across some other cases where the code for allowing group syntax
675 while scanning a header line wasn't quite right (mostly, not resetting
676 the flag correctly in the right place). These bugs could have caused
677 trouble for malformed header lines. I hope it is now all correct.
679 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
680 with the "reply" argument non-NULL. The code, however (which originally
681 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
682 but it didn't always do it. This confused somebody who was copying the
683 code for some other use. I have removed all the tests.
685 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
686 feature that was used to support insecure browsers during the U.S. crypto
687 embargo. It requires special client support, and Exim is probably the
688 only MTA that supported it -- and would never use it because real RSA is
689 always available. This code has been removed, because it had the bad
690 effect of slowing Exim down by computing (never used) parameters for the
691 RSA_EXPORT functionality.
693 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
694 authenticator to fail if there's a tab character in the incoming data
695 (there should never be unless someone is messing about, as it's supposed
696 to be base64-encoded). Also added, on Timo's advice, the "secured" option
697 if the connection is using TLS or if the remote IP is the same as the
698 local IP, and the "valid-client-cert option" if a client certificate has
701 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
702 authenticators. This can be used for authorization after authentication
703 succeeds. (In the case of plaintext, it servers for both authentication
706 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
707 if any retry times were supplied.
709 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
710 connection, where there is no client IP address to check. In this
711 situation, the verify now always succeeds.
713 PH/25 Applied John Jetmore's -Mset patch.
715 PH/26 Added -bem to be like -Mset, but loading a message from a file.
717 PH/27 In a string expansion for a processed (not raw) header when multiple
718 headers of the same name were present, leading whitespace was being
719 removed from all of them, but trailing whitespace was being removed only
720 from the last one. Now trailing whitespace is removed from each header
721 before concatenation. Completely empty headers in a concatenation (as
724 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
725 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
727 PH/29 [Removed. This was a change that I later backed out, and forgot to
728 correct the ChangeLog entry (that I had efficiently created) before
729 committing the later change.]
731 PH/30 Exim was sometimes attempting to deliver messages that had suffered
732 address errors (4xx response to RCPT) over the same connection as other
733 messages routed to the same hosts. Such deliveries are always "forced",
734 so retry times are not inspected. This resulted in far too many retries
735 for the affected addresses. The effect occurred only when there were more
736 hosts than the hosts_max_try setting in the smtp transport when it had
737 the 4xx errors. Those hosts that it had tried were not added to the list
738 of hosts for which the message was waiting, so if all were tried, there
739 was no problem. Two fixes have been applied:
741 (i) If there are any address or message errors in an SMTP delivery, none
742 of the hosts (tried or untried) are now added to the list of hosts
743 for which the message is waiting, so the message should not be a
744 candidate for sending over the same connection that was used for a
745 successful delivery of some other message. This seems entirely
746 reasonable: after all the message is NOT "waiting for some host".
747 This is so "obvious" that I'm not sure why it wasn't done
748 previously. Hope I haven't missed anything, but it can't do any
749 harm, as the worst effect is to miss an optimization.
751 (ii) If, despite (i), such a delivery is accidentally attempted, the
752 routing retry time is respected, so at least it doesn't keep
753 hammering the server.
755 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
756 in ${readsocket because some servers need this prod.
758 PH/32 Added some extra debug output when updating a wait-xxx database.
760 PH/33 The hint "could be header name not terminated by colon", which has been
761 given for certain expansion errors for a long time, was not being given
762 for the ${if def:h_colon_omitted{... case.
764 PH/34 The spec says: "With one important exception, whenever a domain list is
765 being scanned, $domain contains the subject domain." There was at least
766 one case where this was not true.
768 PH/35 The error "getsockname() failed: connection reset by peer" was being
769 written to the panic log as well as the main log, but it isn't really
770 panic-worthy as it just means the connection died rather early on. I have
771 removed the panic log writing for the ECONNRESET error when getsockname()
774 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
775 runs only) independently of the message's sender address. This meant
776 that, if the 4xx error was in fact related to the sender, a different
777 message to the same recipient with a different sender could confuse
778 things. In particualar, this can happen when sending to a greylisting
779 server, but other circumstances could also provoke similar problems.
780 I have changed the default so that the retry time for these errors is now
781 based a combination of the sender and recipient addresses. This change
782 can be overridden by setting address_retry_include_sender=false in the
785 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
786 remote server are returned as part of bounce messages. This was not
787 happening for LMTP over a pipe (the lmtp transport), but now it is the
788 same for both kinds of LMTP.
790 PH/38 Despite being documented as not happening, Exim was rewriting addresses
791 in header lines that were in fact CNAMEs. This is no longer the case.
793 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
794 and queue runs started by the daemon processed all messages. This has
795 been fixed so that -R and -S can now usefully be given with -q<time>.
797 PH/40 Import PCRE release 6.7 (fixes some bugs).
799 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
801 PH/42 Give an error if -q is specified more than once.
803 PH/43 Renamed the variables $interface_address and $interface_port as
804 $received_ip_address and $received_port, to make it clear that these
805 values apply to message reception, and not to the outgoing interface when
806 a message is delivered. (The old names remain recognized, of course.)
808 PH/44 There was no timeout on the connect() call when using a Unix domain
809 socket in the ${readsocket expansion. There now is.
811 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
812 be meaningful with "accept".
814 SC/01 Eximstats V1.43
815 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
817 SC/02 Eximstats V1.44
818 Use a glob alias rather than an array ref in the generated
819 parser. This improves both readability and performance.
821 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
822 Collect SpamAssassin and rejection statistics.
823 Don't display local sender or destination tables unless
824 there is data to show.
825 Added average volumes into the top table text output.
827 SC/04 Eximstats V1.46
828 Collect data on the number of addresses (recipients)
829 as well as the number of messages.
831 SC/05 Eximstats V1.47
832 Added 'Message too big' to the list of mail rejection
833 reasons (thanks to Marco Gaiarin).
835 SC/06 Eximstats V1.48
836 Mainlog lines which have GMT offsets and are too short to
837 have a flag are now skipped.
839 SC/07 Eximstats V1.49 (Alain Williams)
840 Added the -emptyok flag.
842 SC/08 Eximstats V1.50
843 Fixes for obtaining the IP address from reject messages.
845 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
846 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
847 whitesspace changes from 4.64-PH/27
849 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
852 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
853 are found, allow negative numbers in numeric criteria)
855 JJ/06 exipick.20061117.2, added new $message_body_missing variable
857 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
858 to match changes made in 4.64-PH/43
860 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
862 PH/47 Put in an explicit test for a DNS lookup of an address record where the
863 "domain" is actually an IP address, and force a failure. This locks out
864 those revolvers/nameservers that support "A-for-A" lookups, in
865 contravention of the specifications.
867 PH/48 When a host name was looked up from an IP address, and the subsequent
868 forward lookup of the name timed out, the host name was left in
869 $sender_host_name, contrary to the specification.
871 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
872 restricted to single-key lookups, Exim was not diagnosing an error if
873 * or *@ was used with a query-style lookup.
875 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
877 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
878 long ago, but noone interested enough thought of it. Let's just say that
879 the "1.1" means that there are some new functions that weren't there at
880 some point in the past.
882 PH/51 Error processing for expansion failure of helo_data from an smtp
883 transport during callout processing was broken.
885 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
886 tested/used via the -bh/-bhc/-bs options.
888 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
889 bug, fixed in subsequent PCRE releases).
891 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
892 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
894 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
900 SC/01 Use a glob alias rather than an array ref in eximstats generated
901 parser. This improves both readability and performance.
903 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
904 Don't display local sender or destination tables in eximstats unless
905 there is data to show.
906 Added average volumes into the eximstats top table text output.
908 SC/03 Collect data on the number of addresses (recipients) as well
909 as the number of messages in eximstats.
911 TF/01 Correct an error in the documentation for the redirect router. Exim
912 does (usually) call initgroups() when daemonizing.
914 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
915 with consistent privilege compared to when running as a daemon.
917 TF/03 Note in the spec that $authenticated_id is not set for local
918 submissions from trusted users.
920 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
921 Thanks to Dean Brooks <dean@iglou.com> for the patch.
923 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
924 by adding some example configuration directives to the default
925 configuration file. A little bit of work is required to uncomment the
926 directives and define how usernames and passwords are checked, but
927 there is now a framework to start from.
929 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
930 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
931 without this. I don't know how relevant this is to other LDAP libraries.
933 PH/02 Add the verb name to the "unknown ACL verb" error.
935 PH/03 Magnus Holmgren's patch for filter_prepend_home.
937 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
939 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
940 directory not expanded when it should be if an expanded home directory
941 was set for the address (which is overridden by the transport).
943 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
946 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
947 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
948 because it is too late at that time, and has no effect.
950 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
951 security issue with \' (bugzilla #107). I could not use the
952 PQescapeStringConn() function, because it needs a PGconn value as one of
955 PH/08 When testing addresses using -bt, indicate those final addresses that
956 are duplicates that would not cause an additional delivery. At least one
957 person was confused, thinking that -bt output corresponded to deliveries.
958 (Suppressing duplicates isn't a good idea as you lose the information
959 about possibly different redirections that led to the duplicates.)
961 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
962 systems where poll() doesn't work, in particular OS X.
964 PH/10 Added more information to debugging output for retry time not reached.
966 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
967 operations in malware.c.
969 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
972 PH/13 If write_rejectlog was set false when logging was sent to syslog with
973 syslog_duplication set false, log lines that would normally be written
974 both the the main log and to the reject log were not written to syslog at
977 PH/14 In the default configuration, change the use of "message" in ACL warn
978 statements to "add_header".
980 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
981 not followed by a command (e.g. "seen endif").
983 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
984 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
987 PH/17 Added extra conditions to the default value of delay_warning_condition
991 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
992 { match{$h_precedence:}{(?i)bulk|list|junk} } \
993 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
996 The Auto-Submitted: and various List- headers are standardised, whereas I
997 don't think Precedence: ever was.
999 PH/18 Refactored debugging code in route_finduser() to show more information,
1000 in particular, the error code if getpwnam() issues one.
1002 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
1003 This is apparently needed in addition to the PH/07 change above to avoid
1004 any possible encoding problems.
1006 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1007 but not after initializing Perl.
1009 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1010 output them only if debugging. By default they are written stderr,
1011 apparently, which is not desirable.
1013 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1016 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1019 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1021 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1022 authenticated or an ident call has been made. Suppress the default
1023 values for $authenticated_id and $authenticated_sender (but permit -oMai
1024 and -oMas) when testing with -bh.
1026 PH/24 Re-jigged the order of the tests in the default configuration so that the
1027 tests for valid domains and recipients precede the DNS black list and CSA
1028 tests, on the grounds that those ones are more expensive.
1030 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1031 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1032 This bug exists in every version of Exim that I still have, right back to
1035 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1036 However, an attempt to turn on case-sensitivity in a regex key by
1037 including (?-i) didn't work because the subject string was already
1038 lowercased, and the effects were non-intuitive. It turns out that a
1039 one-line patch can be used to allow (?-i) to work as expected.
1045 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1046 other effects) broke the use of negated acl sub-conditions.
1048 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1051 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1052 "Deny" causes Exim to reject the incoming connection with a 554 error.
1053 Unfortunately, if there is a major crisis, such as a disk failure,
1054 tcp-wrappers gives "deny", whereas what one would like would be some
1055 kind of temporary error. A kludge has been added to help with this.
1056 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1057 554 error is used if errno is still zero or contains ENOENT (which occurs
1058 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1061 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1063 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1064 errors. Otherwise a message that provokes a temporary error (when other
1065 messages do not) can cause a whole host to time out.
1067 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1068 addresses were routed directly to files or pipes from a redirect router.
1069 File deliveries just didn't batch; pipe deliveries might have suffered
1072 PH/06 A failure to get a lock for a hints database would erroneously always say
1073 "Failed to get write lock", even when it was really a read lock.
1075 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1076 of 0600. This has been changed to use the value of the lockfile_mode
1077 option (which defaults to 0600).
1079 PH/08 Applied small patch from the Sieve maintainer.
1081 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1082 folder from quota calculations, a direct delivery into this folder messed
1083 up the contents of the maildirsize file. This was because the regex was
1084 used only to exclude .Trash (or whatever) when the size of the mailbox
1085 was calculated. There was no check that a delivery was happening into an
1086 excluded directory. This bug has been fixed by ignoring all quota
1087 processing for deliveries into excluded directories.
1089 PH/10 Added the maildirfolder_create_regex option to appendfile.
1095 PH/01 The code for finding all the local interface addresses on a FreeBSD
1096 system running IPv6 was broken. This may well have applied to all BSD
1097 systems, as well as to others that have similar system calls. The broken
1098 code found IPv4 interfaces correctly, but gave incorrect values for the
1099 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1100 that it would not match correctly against @[] and not recognize the IPv6
1103 PH/02 The ipliteral router was not recognizing addresses of the form user@
1104 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1106 PH/03 Added disable_ipv6.
1108 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1109 decoded form, because it is most often used to construct To: headers
1110 lines in autoreplies, and the decoded form may well be syntactically
1111 invalid. However, $reply_address has leading white space removed, and all
1112 newlines turned into spaces so that the autoreply transport does not
1115 PH/05 If group was specified without a user on a router, and no group or user
1116 was specified on a transport, the group from the router was ignored.
1118 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1119 for visible compile-time settings that can be used to change these
1120 numbers, for those that want even more. Backwards compatibility with old
1121 spool files has been maintained. However, going back to a previous Exim
1122 release will lost any variables that are in spool files.
1124 PH/07 Two small changes when running in the test harness: increase delay when
1125 passing a TCP/IP connection to a new process, in case the original
1126 process has to generate a bounce, and remove special handling of
1127 127.0.0.2 (sic), which is no longer necessary.
1129 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1130 be the same on different OS.
1132 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1135 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1136 whether --show-vars was specified or not
1138 JJ/02 exipick: Added support for new ACL variable spool format introduced
1141 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1142 syntactically invalid From: or Reply-to: line, and a filter used this to
1143 generate an autoreply, and therefore failed to obtain an address for the
1144 autoreply, Exim could try to deliver to a non-existent relative file
1145 name, causing unrelated and misleading errors. What now happens is that
1146 it logs this as a hard delivery error, but does not attempt to create a
1149 PH/11 The exinext utility has a -C option for testing purposes, but although
1150 the given file was scanned by exinext itself; it wasn't being passed on
1151 when Exim was called.
1153 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1154 an end-of-file indication when reading a command response.
1156 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1157 compiled. In many other places in Exim, IPv6 addresses are always
1158 recognized, so I have changed this. It also means that IPv4 domain
1159 literals of the form [IPV4:n.n.n.n] are now always recognized.
1161 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1162 used if the router is not running as root, for example, when verifying at
1163 ACL time, or when using -bh. The debugging output from this situation was
1164 non-existent - all you got was a failure to exec. I have made two
1167 (a) Failures to set uid/gid, the current directory, or a process leader
1168 in a subprocess such as that created by queryprogram now generate
1169 suitable debugging ouput when -d is set.
1171 (b) The queryprogram router detects when it is not running as root,
1172 outputs suitable debugging information if -d is set, and then runs
1173 the subprocess without attempting to change uid/gid.
1175 PH/15 Minor change to Makefile for building test_host (undocumented testing
1178 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1179 additional section of a DNS packet that returns MX or SRV records.
1180 Instead, it always explicitly searches for A/AAAA records. This avoids
1181 major problems that occur when a DNS server includes only records of one
1182 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1183 fixed another bug: if SRV records were looked up and the corresponding
1184 address records were *not* found in the additional section, the port
1185 values from the SRV records were lost.
1187 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1188 using the correct key (the original address) when searching the retry
1189 rules in order to find which one to use for generating the retry hint.
1191 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1192 adding the default one. Similarly, if it contains a Reply-To: header, the
1193 errors_reply_to option, if set, is not used.
1195 PH/19 When calculating a retry time, Exim used to measure the "time since
1196 failure" by looking at the "first failed" field in the retry record. Now
1197 it does not use this if it is later than than the arrival time of the
1198 message. Instead it uses the arrival time. This makes for better
1199 behaviour in cases where some deliveries succeed, thus re-setting the
1200 "first failed" field. An example is a quota failure for a huge message
1201 when small messages continue to be delivered. Without this change, the
1202 "time since failure" will always be short, possible causing more frequent
1203 delivery attempts for the huge message than are intended.
1204 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1206 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1207 $1, $2, $3) because the numerical variables can be reset during some
1208 expansion items (e.g. "match"), thereby losing the authentication data.
1210 PH/21 Make -bV show the size of off_t variables so that the test suite can
1211 decide whether to run tests for quotas > 2G.
1213 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1214 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1215 filecount value is greater than 2G or if a quota value is greater than 2G
1216 on a system where the size of off_t is not greater than 4, a panic error
1219 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1220 never match. The debug and -bh output now contains an explicit error
1221 message indicating a malformed IPv4 address or mask.
1223 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1224 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1225 PH/23 above applies.
1227 PH/25 Do not write to syslog when running in the test harness. The only
1228 occasion when this arises is a failure to open the main or panic logs
1229 (for which there is an explicit test).
1231 PH/26 Added the /no_tell option to "control=freeze".
1233 PH/27 If a host name lookup failed very early in a connection, for example, if
1234 the IP address matched host_lookup and the reverse lookup yielded a name
1235 that did not have a forward lookup, an error message of the form "no IP
1236 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1237 could be logged. Now it outputs the IP address instead of "NULL".
1239 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1240 allows the sender and the authenticated sender to be set when
1241 submitting a message from within Exim. Since child_open_exim() is
1242 documented for local_scan(), the new function should be too.
1244 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1245 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1246 results in an empty string is now treated as unset.
1248 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1250 PH/31 Added sender_verify_fail logging option.
1252 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1253 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1254 tidied the source and removed it altogether.
1256 PH/33 When a queue run was abandoned because the load average was too high, a
1257 log line was always written; now it is written only if the queue_run log
1258 selector is set. In addition, the log line for abandonment now contains
1259 information about the queue run such as the pid. This is always present
1260 in "start" and "stop" lines but was omitted from the "abandon" line.
1262 PH/34 Omit spaces between a header name and the colon in the error message that
1263 is given when verify = headers_syntax fails (if there are lots of them,
1264 the message gets confusing).
1266 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1267 names, as there are now some PTR records that contain slashes. This check
1268 is only to protect against broken name servers that fall over on strange
1269 characters, so the fact that it applies to all lookups doesn't matter.
1271 PH/36 Now that the new test suite is complete, we can remove some of the
1272 special code in Exim that was needed for the old test suite. For example,
1273 sorting DNS records because real resolvers return them in an arbitrary
1274 order. The new test suite's fake resolver always returns records in the
1277 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1278 bounces) except when queue_only is set, to avoid logging races between
1279 the different processes.
1281 PH/38 Panic-die if .include specifies a non-absolute path.
1283 PH/39 A tweak to the "H" retry rule from its user.
1285 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1286 a label. They prevented compilation on older perls.
1288 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1289 a warning to be raised on newish perls.
1291 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1292 on queue. Changes to match documented behaviour of showing count of
1293 messages matching specified criteria.
1295 PH/40 Changed the default ident timeout from 30s to 5s.
1297 PH/41 Added support for the use of login_cap features, on those BSD systems
1298 that have them, for controlling the resources used by pipe deliveries.
1300 PH/42 The content-scanning code uses fopen() to create files in which to put
1301 message data. Previously it was not paying any attention to the mode of
1302 the files. Exim runs with umask(0) because the rest of the code creates
1303 files with open(), and sets the required mode explicitly. Thus, these
1304 files were ending up world-writeable. This was not a big issue, because,
1305 being within the spool directory, they were not world-accessible. I have
1306 created a function called modefopen, which takes an additional mode
1307 argument. It sets umask(777), creates the file, chmods it to the required
1308 mode, then resets the umask. All the relevant calls to fopen() in the
1309 content scanning code have been changed to use this function.
1311 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1312 to 24 hours. This avoids potential overflow problems when processing G
1313 and H retry rules. I suspect nobody ever tinkers with this value.
1315 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1317 PH/45 When the plaintext authenticator is running as a client, the server's
1318 challenges are checked to ensure they are valid base64 strings. By
1319 default, the authentication attempt is cancelled if an invalid string is
1320 received. Setting client_ignore_invalid_base64 true ignores these errors.
1321 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1322 they are received. Thus, the responses can be made to depend on the
1323 challenges. If an invalid string is ignored, an empty string is placed in
1326 PH/46 Messages that are created by the autoreply transport now contains a
1327 References: header, in accordance with RFCs 2822 and 3834.
1329 PH/47 Added authenticated_sender_force to the smtp transport.
1331 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1333 PH/49 Installed latest patch from the Sieve maintainer.
1335 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1336 was also set, the appendfile transport was unnecessarily scanning a
1337 directory of message files (e.g. for maildir delivery) to find the count
1338 of files (along with the size), even though it did not need this
1339 information. It now does the scan only if it needs to find either the
1340 size of the count of files.
1342 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1344 PH/52 Two bugs concerned with error handling when the smtp transport is
1347 (i) Exim was not creating retry information for temporary errors given
1348 for individual recipients after the DATA command when the smtp transport
1349 was used in LMTP mode. This meant that they could be retried too
1350 frequently, and not timed out correctly.
1352 (ii) Exim was setting the flag that allows error details to be returned
1353 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1354 recipients that were returned after the DATA command.
1356 PH/53 This is related to PH/52, but is more general: for any failing address,
1357 when detailed error information was permitted to be returned to the
1358 sender, but the error was temporary, then after the final timeout, only
1359 "retry timeout exceeded" was returned. Now it returns the full error as
1360 well as "retry timeout exceeded".
1362 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1363 do this, and (what is worse) MTAs that accept it.
1365 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1366 will now be deprecated.
1368 PH/56 New os.c-cygwin from the Cygwin maintainer.
1370 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1371 formats (previously only available in exim formats via -bpr, -bpru,
1372 and -bpra. Now also available in native and exiqgrep formats)
1374 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1375 with very large, slow to parse queues
1377 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1379 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1381 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1382 responses to authentication challenges, though it was showing the
1383 challenges; (ii) I've removed the CR characters from the debug output for
1386 PH/58 Allow for the insertion of a newline as well as a space when a string
1387 is turned into more than one encoded-word during RFC 2047 encoding. The
1388 Sieve code now uses this.
1390 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1391 data_4xx, lost_connection, tls_required.
1393 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1394 message was being sent as an SMTP response.
1396 PH/61 Add -l and -k options to exicyclog.
1398 PH/62 When verifying, if an address was redirected to one new address, so that
1399 verification continued, and the new address failed or deferred after
1400 having set something in $address_data, the value of $address_data was not
1401 passed back to the ACL. This was different to the case when no
1402 redirection occurred. The value is now passed back in both cases.
1404 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1405 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1406 use login_cap.h, so on its own it isn't the distinguishing feature. The
1407 new name refers directly to the setclassresources() function.
1409 PH/65 Added configuration files for NetBSD3.
1411 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1413 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1414 is preferred over IPv4.
1416 PH/68 The bounce_return_message and bounce_return_body options were not being
1417 honoured for bounces generated during the reception of non-SMTP messages.
1418 In particular, this applied to messages rejected by the ACL. This bug has
1419 been fixed. However, if bounce_return_message is true and bounce_return_
1420 body is false, the headers that are returned for a non-SMTP message
1421 include only those that have been read before the error was detected.
1422 (In the case of an ACL rejection, they have all been read.)
1424 PH/69 The HTML version of the specification is now built in a directory called
1425 spec_html instead of spec.html, because the latter looks like a path with
1426 a MIME-type, and this confuses some software.
1428 PH/70 Catch two compiler warnings in sieve.c.
1430 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1431 function verify_get_ident() calls ip_connect() to connect a socket, but
1432 if the "connect()" function timed out, ip_connect() used to close the
1433 socket. However, verify_get_ident() also closes the socket later, and in
1434 between Exim writes to the log, which may get opened at this point. When
1435 the socket was closed in ip_connect(), the log could get the same file
1436 descriptor number as the socket. This naturally causes chaos. The fix is
1437 not to close the socket in ip_connect(); the socket should be closed by
1438 the function that creates it. There was only one place in the code where
1439 this was missing, in the iplookup router, which I don't think anybody now
1440 uses, but I've fixed it anyway.
1442 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1443 well as to direct DNS lookups. Otherwise the handling of names in host
1444 lists is inconsistent and therefore confusing.
1450 PH/01 Two changes to the default runtime configuration:
1452 (1) Move the checks for relay_from_hosts and authenticated clients from
1453 after to before the (commented out) DNS black list checks.
1455 (2) Add control=submission to the relay_from_hosts and authenticated
1456 clients checks, on the grounds that messages accepted by these
1457 statements are most likely to be submissions.
1459 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1461 (1) Generate an error if the third argument for the ${prvs expansion is
1464 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1467 (3) Reset the variables that are obtained from the first argument of
1468 ${prvscheck and used in the second argument before leaving the code,
1469 because their memory is reclaimed, so using them afterwards may do
1472 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1473 one (it's much easier than Tom thought :-).
1475 (5) Because of (4), we can now allow for the use of $prvscheck_result
1476 inside the third argument.
1478 PH/03 For some reason, the default setting of PATH when running a command from
1479 a pipe transport was just "/usr/bin". I have changed it to
1482 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1483 anything to be listed in the output from -bV.
1485 PH/05 When a filter generated an autoreply, the entire To: header line was
1486 quoted in the delivery log line, like this:
1488 => >A.N.Other <ano@some.domain> <original@ddress> ...
1490 This has been changed so that it extracts the operative address. There
1491 may be more than one such address. If so, they are comma-separated, like
1494 => >ano@some.domain,ona@other.domain <original@ddress> ...
1496 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1497 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1498 not being looked up in the rDNS to get a host name, Exim was showing the
1499 IP address twice in Received: lines, even though the IP addresses were
1500 identical. For example:
1502 Received: from [1.2.3.4] (helo=[1.2.3.4])
1504 However, if the real host name was known, it was omitting the HELO data
1505 if it matched the actual IP address. This has been tidied up so that it
1506 doesn't show the same IP address twice.
1508 PH/07 When both +timestamp and +memory debugging was on, the value given by
1509 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1510 called by the debug printing, thereby overwriting the timestamp buffer.
1511 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1513 PH/08 When the original message was included in an autoreply transport, it
1514 always said "this is a copy of the message, including all the headers",
1515 even if body_only or headers_only was set. It now gives an appropriate
1518 PH/09 Applied a patch from the Sieve maintainer which:
1520 o fixes some comments
1521 o adds the (disabled) notify extension core
1522 o adds some debug output for the result of if/elsif tests
1523 o points to the current vacation draft in the documentation
1524 and documents the missing references header update
1528 o fixes a bug in processing the envelope test (when testing
1529 multiple envelope elements, the last element determinted the
1532 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1533 Electronic Mail") by including:
1535 Auto-submitted: auto-generated
1537 in the messages that it generates (bounce messages and others, such as
1538 warnings). In the case of bounce messages for non-SMTP mesages, there was
1539 also a typo: it was using "Auto_submitted" (underscore instead of
1540 hyphen). Since every message generated by Exim is necessarily in response
1541 to another message, thes have all been changed to:
1543 Auto-Submitted: auto-replied
1545 in accordance with these statements in the RFC:
1547 The auto-replied keyword:
1549 - SHOULD be used on messages sent in direct response to another
1550 message by an automatic process,
1552 - MUST NOT be used on manually-generated messages,
1554 - MAY be used on Delivery Status Notifications (DSNs) and Message
1555 Disposition Notifications (MDNs),
1557 - MUST NOT be used on messages generated by automatic or periodic
1558 processes, except for messages which are automatic responses to
1561 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1562 to the default Received: header definition.
1564 PH/12 Added log selector acl_warn_skipped (default on).
1566 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1567 variables because (a) they are in the wrong storage pool and (b) even if
1568 they were copied, it wouldn't work properly because of the caching.
1570 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1571 checking when decoding. Apparently there are clients that generate
1572 overlong encoded strings. Why am I not surprised?
1574 PH/15 If the first argument of "${if match_address" was not empty, but did not
1575 contain an "@" character, Exim crashed. Now it writes a panic log message
1576 and treats the condition as false.
1578 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1580 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1581 extension "envelope-auth". The code is finished and in agreement with
1582 other implementations, but there is no documentation so far and in fact,
1583 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1584 not changing the active code.
1586 Print executed "if" and "elsif" statements when debugging is used. This
1587 helps a great deal to understand what a filter does.
1589 Document more things not specified clearly in RFC3028. I had all this
1590 sorted out, when out of a sudden new issues came to my mind. Oops."
1592 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1595 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1596 canonical form (as documented). However, after a host name lookup from
1597 the IP address, check_host() was doing a simple string comparison with
1598 addresses acquired from the DNS when checking that the found name did
1599 have the original IP as one of its addresses. Since any found IPv6
1600 addresses are likely to be in abbreviated form, the comparison could
1601 fail. Luckily, there already exists a function for doing the comparison
1602 by converting both addresses to binary, so now that is used instead of
1603 the text comparison.
1605 PH/20 There was another similar case to PH/19, when a complete host name was
1606 given in a host list; looking up its IP address could give an abbreviated
1607 form, whereas the current host's name might or might not be abbreviated.
1608 The same fix has been applied.
1614 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1615 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1618 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1619 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1621 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1623 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1624 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1625 Also, for "Auto-Submitted", treat anything other than "no" as
1626 non-personal, in accordance with RFC 3834. (Previously it treated
1627 anything starting "auto-" as non-personal.)
1629 TF/01 The control=submission/name=... option had a problem with syntax
1630 errors if the name included a slash character. The /name= option
1631 now slurps the rest of the string, so it can include any characters
1632 but it must come last in the list of options (after /sender_retain
1635 PH/05 Some modifications to the interface to the fake nameserver for the new
1643 TK/01 Added the "success_on_redirect" address verification option. See
1644 NewStuff for rationale and an example.
1646 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1648 PH/02 Patch to exigrep to allow it to work on syslog lines.
1650 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1651 fread() to skip over the body file's header line, because in Cygwin the
1652 header line is locked and is inaccessible.
1654 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1655 co-exist for some time) to make it clear that it is the Exim ID that is
1656 referenced, not the Message-ID: header line.
1658 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1659 string_format() function, because snprintf() does not exist on all
1662 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1663 ${stat: expansion item.
1665 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1666 protocol synchronization error", to keep the pedants happy.
1668 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1669 well as for IRIX systems, when gcc is being used. See the host.c source
1672 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1674 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1677 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1678 transport and to the smtp transport in LMTP mode.
1680 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1682 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1683 run a filter in a subprocess. This could lead to confusion in subsequent
1684 lookups in the parent process. There should also be a search_tidyup() at
1685 the end of the subprocess.
1687 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1688 only if the host matched helo_try_verify_hosts, which caused the
1689 verification to occur when the EHLO/HELO command was issued. The ACL just
1690 tested the remembered result. Now, if a previous verification attempt has
1691 not happened, "verify = helo" does it there and then.
1693 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1695 TK/03 Fix log output including CR from clamd.
1697 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1698 exist provoked a memory error which could cause a segfault.
1700 PH/15 Installed PCRE 6.2
1702 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1704 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1705 of the problem. Specifically, suggested +O2 rather than +O1 for the
1708 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1710 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1711 record, the port was not correctly logged when the outgoing_port log
1712 selector was set (it logged the transort's default port).
1714 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1715 fallback_hosts, and "hosts" in the smtp transport.
1717 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1718 host errors such as "Connection refused".
1720 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1721 authentication with radiusclient 0.4.9:
1723 - Error returned from rc_read_config was caught wrongly
1724 - Username/password not passed on to radius server due to wrong length.
1726 The presumption is that some radiusclient API changes for 4.51/PH/17
1727 were not taken care of correctly. The code is still untested by me (my
1728 Linux distribution still has 0.3.2 of radiusclient), but it was
1729 contributed by a Radius user.
1731 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1732 expanding the "port" option of the smtp transport.
1734 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1735 while reading a MIME header. Thanks to Tom Hughes for a patch.
1737 PH/24 Include config.h inside local_scan.h so that configuration settings are
1740 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1741 that in an ACL for RCPT (for example), you can examine exactly what was
1744 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1745 commands, but it was not correctly comparing the address with the actual
1746 client host address. Thus, it would show the EHLO address in Received:
1747 header lines when this was not necessary.
1749 PH/27 Added the % operator to ${eval:}.
1751 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1752 it should be ignoring failures (because with -C, for example, it has lost
1753 privilege). It wasn't ignoring creation failures other than "already
1756 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1759 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1760 because an input line was too long, either on its own, or by virtue of
1761 too many continuations, the temporary file was not being removed, and the
1762 return code was incorrect.
1764 PH/31 Missing "BOOL" in function definition in filtertest.c.
1766 PH/32 Applied Sieve patches from the maintainer.
1768 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1770 PH/33 Added "verify = not_blind".
1772 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1773 Local/Makefile (with some defaults set). These are used in built scripts
1774 such as exicyclog, but they have never been used in the exim_install
1775 script (though there are many overriding facilities there). I have
1776 arranged that the exim_install script now takes note of these two
1779 PH/35 Installed configuration files for Dragonfly.
1781 PH/36 When a locally submitted message by a trusted user did not contain a
1782 From: header, and the sender address was obtained from -f or from an SMTP
1783 MAIL command, and the trusted user did not use -F to supply a sender
1784 name, $originator_name was incorrectly used when constructing a From:
1785 header. Furthermore, $originator_name was used for submission mode
1786 messages from external hosts without From: headers in a similar way,
1787 which is clearly wrong.
1789 PH/37 Added control=suppress_local_fixups.
1791 PH/38 When log_selector = +received_sender was set, and the addition of the
1792 sender made the log line's construction buffer exactly full, or one byte
1793 less than full, an overflow happened when the terminating "\n" was
1796 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1797 when the result of a list match is failure because a DNS lookup failed.
1799 PH/40 RM_COMMAND is now used in the building process.
1801 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1802 the "build-* directories that it finds.
1804 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1805 address in a domain literal was a prefix of an interface address.
1807 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1808 when verifying a sender address, unless rewrite_headers is false.
1810 PH/44 Wrote a long comment about why errors_to addresses are verified as
1811 recipients, not senders.
1813 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1814 the ratelimit ACL was added.
1816 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1818 PH/46 Added extra information about PostgreSQL errors to the error string.
1820 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1821 avoiding the need to install special zones in a real server. This is
1822 backwards compatible; if it can't find the fake resolver, it drops back.
1823 Thus, both old and new test suites can be run.
1825 TF/02 Added util/ratelimit.pl
1827 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1828 clock is set back in time.
1830 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1831 Candler <B.Candler@pobox.com>.
1833 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1834 OK for addresses that are the result of redirections.
1836 PH/48 A number of further additions for the benefit of the new test suite,
1837 including a fake gethostbyname() that interfaces to the fake DNS resolver
1840 TF/06 The fix for widen_domains has also been applied to qualify_single and
1841 search_parents which are the other dnslookup options that can cause
1844 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1847 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1849 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1850 service" if it encountered an MX record with an empty target hostname.
1851 The message is now "an MX or SRV record indicated no SMTP service".
1853 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1854 if the DNS of the sending site is misconfigured. This is quite a
1855 common situation. This change restores the behaviour of treating a
1856 helo verification defer as a failure.
1858 PH/51 If self=fail was set on a router, the bounce message did not include the
1859 actual error message.
1865 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1867 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1868 command itself ended in error, the underlying message about the transport
1869 filter timeout was being overwritten with the pipe command error. Now the
1870 underlying error message should be appended to the second error message.
1872 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1874 PH/02 Reduce the amount of output that "make" produces by default. Full output
1875 can still be requested.
1877 PH/03 The warning log line about a condition test deferring for a "warn" verb
1878 was being output only once per connection, rather than after each
1879 occurrence (because it was using the same function as for successful
1880 "warn" verbs). This seems wrong, so I have changed it.
1882 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1883 it should not have, which might have caused a crash in the right
1884 circumstances, but probably never did.
1886 PH/04 Installed a modified version of Tony Finch's patch to make submission
1887 mode fix the return path as well as the Sender: header line, and to
1888 add a /name= option so that you can make the user's friendly name appear
1891 TF/03 Added the control = fakedefer ACL modifier.
1893 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1894 Mark Lowes for thorough testing.
1896 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1898 TK/03 Merged latest SRS patch from Miles Wilton.
1900 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1901 with the definition in sysexits.h (which is #included earlier).
1902 Fortunately, Exim does not actually use EX_OK. The code used to try to
1903 preserve the sysexits.h value, by assumimg that macro definitions were
1904 scanned for macro replacements. I have been disabused of this notion,
1905 so now the code just undefines EX_OK before #including unistd.h.
1907 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1908 in the smtp transport. When a block could not be written in a single
1909 write() function, the timeout was being re-applied to each part-write.
1910 This seems wrong - if the receiver was accepting one byte at a time it
1911 would take for ever. The timeout is now adjusted when this happens. It
1912 doesn't have to be particularly precise.
1914 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1915 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1917 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1918 without a domain if the check to <postmaster@domain> fails.
1920 SC/01 Eximstats: added -xls and the ability to specify output files
1921 (patch written by Frank Heydlauf).
1923 SC/02 Eximstats: use FileHandles for outputing results.
1925 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1927 SC/04 Eximstats: fixed display of large numbers with -nvr option
1929 SC/05 Eximstats: fixed merging of reports with empty tables.
1931 SC/06 Eximstats: added the -include_original_destination flag
1933 SC/07 Eximstats: removed tabs and trailing whitespace.
1935 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1937 TK/06 MBOX spool code: Add real "From " MBOX separator line
1938 so the .eml file is really in mbox format (even though
1939 most programs do not really care). Patch from Alex Miller.
1941 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1942 The latter is generated from $received_to and is only set if the
1943 message has one envelope recipient. SA can use these headers,
1944 obviously out-of-the-box. Patch from Alex Miller.
1946 PH/08 The ${def test on a variable was returning false if the variable's
1947 value was "0", contrary to what the specification has always said!
1948 The result should be true unless the variable is empty.
1950 PH/09 The syntax error of a character other than { following "${if
1951 def:variable_name" (after optional whitespace) was not being diagnosed.
1952 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1953 accidental colon was present, for example, could give incorrect results.
1955 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1956 result is used (not including appendfile, where other changes are about
1959 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1960 This involved changing a lot of size variables from int to off_t. It
1961 should work with maildirs and everything.
1963 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1964 spamd dying while we are connected to it.
1966 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1969 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1970 function for BDB 4.3.
1972 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1974 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1975 the sources for PCRE have been re-arranged and more files are now
1978 PH/15 The code I had for printing potentially long long variables in PH/11
1979 above was not the best (it lost precision). The length of off_t variables
1980 is now inspected at build time, and an appropriate printing format (%ld
1981 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1982 to be "long long int" or "long int". This is needed for the internal
1983 formatting function string_vformat().
1985 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1986 the configuration file to be ":syslog", then the script "guesses" where
1987 the logs files are, rather than using the compiled in default. In our
1988 case the guess is not the same as the compiled default, so the script
1989 suddenly stopped working when I started to use syslog. The patch checks
1990 to see if log_file_path is "". If so, it attempts to read it from exim
1991 with no configuration file to get the compiled in version, before it
1992 falls back to the previous guessing code."
1994 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1995 implementing BATV in an Exim configuration. See NewStuff for the gory
1998 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1999 Makefile that are specific to HP-UX.
2001 PH/18 If the "use_postmaster" option was set for a recipient callout together
2002 with the "random" option, the postmaster address was used as the MAIL
2003 FROM address for the random test, but not for the subsequent recipient
2004 test. It is now used for both.
2006 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2007 patch removes a few documentation additions to RFC 3028, because the
2008 latest draft now contains them. It adds the new en;ascii-case comparator
2009 and a new error check for 8bit text in MIME parts. Comparator and
2010 require names are now matched exactly. I enabled the subaddress
2011 extension, but it is not well tested yet (read: it works for me)."
2013 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2014 rework some of the code of TK/09 above to avoid the hardwired use of
2015 "%lld" and "long long". Replaced the call to snprintf() with a call to
2018 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2019 records point to non-existent hosts", "retry timeout exceeded", and
2020 "retry time not reached for any host after a long failure period".
2022 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2023 experimental DomainKeys support:
2025 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2026 (2) On an error such as an illegally used "control", the wrong name for
2027 the control was given.
2029 These problems did NOT occur unless DomainKeys support was compiled.
2031 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2033 PH/24 Added ${if match_ip condition.
2035 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2036 databases so that it will be absolutely obvious if a crash occurs in the
2037 DB library. This is a regular occurrence (often caused by mis-matched
2040 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2041 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2042 picked up on a user's system that detects such things. There doesn't seem
2043 to be a gcc warning option for this - only an attribute that has to be
2044 put on the function's prototype. It seems that in Fedora Core 4 they have
2045 set this on a number of new functions. No doubt there will be more in due
2048 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2049 specify a transport. However, if an address that was verified by such a
2050 router was the subject of a callout, Exim crashed because it tried to
2051 read the rcpt_include_affixes from the non-existent transport. Now it
2052 just assumes that the setting of that option is false. This bug was
2053 introduced by 4.51/PH/31.
2055 PH/28 Changed -d+all to exclude +memory, because that information is very
2056 rarely of interest, but it makes the output a lot bigger. People tend to
2057 do -d+all out of habit.
2059 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2060 code in os-type was giving problems when libc.so lives in lib64, like on
2063 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2064 aren't the modern standard, and it seems that some systems' include files
2065 don't always have them. Exim was already checking for some of the newer
2066 ones like T_AAAA, and defining it itself. I've added checks for all the
2067 record types that Exim uses.
2069 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2070 not automatically generating a new one, as it is supposed to. This
2071 prevented TLS from working. If the file did exist, but contained invalid
2072 data, a new version was generated, as expected. It was only the case of a
2073 non-existent file that was broken.
2075 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2076 with a change in libdomainkeys > 0.64.
2078 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2079 from DNS. If the selector record carries the flag, it now has
2080 precedence over the domain-wide flag.
2082 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2084 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2085 the use of a port name that isn't defined in /etc/services) occurred, the
2086 message was deferred as in a normal delivery, and thus remained on the
2087 spool, instead of being failed because of the mua_wrapper setting. This
2088 is now fixed, and I tidied up some of the mua_wrapper messages at the
2091 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2092 the messages in a hash of arrays rather than using individual hashes.
2093 This is a bit cleaner and results in dramatic memory savings, albeit
2094 at a slight CPU cost.
2096 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2097 as requested by Marc Sherman.
2099 SC/10 Eximstats: added histograms for user specified patterns as requested
2102 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2104 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2105 fopen() in the content-scanning modules that did not already have it.
2111 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2112 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2114 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2116 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2118 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2119 to test Sieve filters that use "vacation".
2121 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2122 that changes the way the GnuTLS parameters are stored in the cache file.
2123 The new format can be generated externally. For backward compatibility,
2124 if the data in the cache doesn't make sense, Exim assumes it has read an
2125 old-format file, and it generates new data and writes a new file. This
2126 means that you can't go back to an older release without removing the
2129 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2130 work if there are any delivery delays because "one_time" forces the
2131 parent to be marked "delivered", so its unseen clone is never tried
2132 again. For this reason, Exim now forbids the simultaneous setting of
2135 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2136 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2137 entry if you want to know the details. The fix, however, neglected to
2138 consider the case when local delivery batching is involved. The test for
2139 "previously delivered" was not happening when checking to see if an
2140 address could be batched with a previous (undelivered) one; under
2141 certain circumstances this could lead to multiple deliveries to the same
2144 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2145 in its include files, and this causes problems building Exim.
2147 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2148 header_syntax) but Exim was just ignoring anything given after a slash.
2149 In particular, this caused confusion with an attempt to use "verify =
2150 reverse_host_lookup/defer_ok". An error is now given when options are
2151 supplied for verify items that do not have them. (Maybe reverse_host_
2152 lookup should have a defer_ok option, but that's a different point.)
2154 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2155 defined by RFC 821) to 2048, because there were problems with some AUTH
2156 commands, and RFC 1869 says the size should be increased for extended
2157 SMTP commands that take arguments.
2159 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2162 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2163 "unknown" error; now it says that the functionality isn't in the binary.
2165 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2166 an address' error message when a string expansion fails (syntax or
2167 whatever). Otherwise the password may appear in the log. Following change
2168 PH/42 below, there is no longer a chance of it appearing in a bounce
2171 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2173 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2174 of its addresses was ever tried. (Bugzilla bug #2.)
2176 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2177 the result incorrectly in the debug output. (It correctly added a newline
2178 to what was transported.)
2180 TF/01 Added $received_time.
2182 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2183 commented out examples of how to interface to a virus scanner and to
2184 SpamAssassin. Also added commented examples of av_scanner and
2185 spamd_address settings.
2187 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2188 and controls are allowed in which ACLs. There were a couple of minor
2189 errors. Some of the entries in the conditions table (which is a table of
2190 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2191 negation of where the condition IS allowed.
2193 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2195 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2196 header file does not have a version number, so I've had to invent a new
2197 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2198 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2199 radiusclient), but it was contributed by a Radius user.
2201 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2202 files or directories, for OpenSSL.
2204 PH/19 When an Exim process that is running as root has to create an Exim log
2205 file, it does so in a subprocess that runs as exim:exim so as to get the
2206 ownership right at creation (otherwise, other Exim processes might see
2207 the file with the wrong ownership). There was no test for failure of this
2208 fork() call, which would lead to the process getting stuck as it waited
2209 for a non-existent subprocess. Forks do occasionally fail when resources
2210 run out. I reviewed all the other calls to fork(); they all seem to check
2213 PH/20 When checking for unexpected SMTP input at connect time (before writing
2214 the banner), Exim was not dealing correctly with a non-positive return
2215 from the read() function. If the client had disconnected by this time,
2216 the result was a log entry for a synchronization error with an empty
2217 string after "input=" when read() returned zero. If read() returned -1
2218 (an event I could not check), uninitialized data bytes were printed.
2219 There were reports of junk text (parts of files, etc) appearing after
2222 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2224 PH/22 Added support for macro redefinition, and (re)definition in between
2225 driver and ACL definitions.
2227 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2228 forgetting to use the resulting value; it was using the unexpanded value.
2230 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2231 hadn't been configured. The fix is from Juergen Kreileder, who
2232 understands it better than I do:
2234 "Here's what I see happening with three configured cyrus_sasl
2235 authenticators configured (plain, login, cram-md5):
2237 On startup auth_cyrus_sasl_init() gets called for each of these.
2238 This means three calls to sasl_listmech() without a specified mech_list.
2239 => SASL tests which mechs of all available mechs actually work
2240 => three warnings about OTP not working
2241 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2243 With the patch, sasl_listmech() also gets called three times. But now
2244 SASL's mech_list option is set to the server_mech specified in the the
2245 authenticator. Or in other words, the answer from sasl_listmech()
2246 gets limited to just the mech you're testing for (which is different
2248 => the return list contains just 'plain' or 'login', 'cram-md5' or
2249 nothing depending on the value of ob->server_mech.
2251 I've just tested the patch: Authentication still works fine,
2252 unavailable mechs specified in the exim configuration are still
2253 caught, and the auth.log warnings about OTP are gone."
2255 PH/25 When debugging is enabled, the contents of the command line are added
2256 to the debugging output, even when log_selector=+arguments is not
2259 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2260 answer is "GNU", and only if the return is "GNU/something" is the answer
2263 PH/27 $acl_verify_message is now set immediately after the failure of a
2264 verification in an ACL, and so is available in subsequent modifiers. In
2265 particular, the message can be preserved by coding like this:
2267 warn !verify = sender
2268 set acl_m0 = $acl_verify_message
2270 Previously, $acl_verify_message was set only while expanding "message"
2271 and "log_message" when a very denied access.
2273 PH/28 Modified OS/os.c-Linux with
2275 -#ifndef OS_LOAD_AVERAGE
2276 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2278 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2279 nomenclature these days.)
2281 PH/29 Installed patch from the Sieve maintainer that adds the options
2282 sieve_useraddress and sieve_subaddress to the redirect router.
2284 PH/30 In these circumstances:
2285 . Two addresses routed to the same list of hosts;
2286 . First host does not offer TLS;
2287 . First host accepts first address;
2288 . First host gives temporary error to second address;
2289 . Second host offers TLS and a TLS session is established;
2290 . Second host accepts second address.
2291 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2292 and peerdn, if requested) that were in fact used only for the second
2295 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2296 attention to any local part prefix or suffix that was matched by the
2297 router that accepted the address. It now behaves in the same way as it
2298 does for delivery: the affixes are removed from the local part unless
2299 rcpt_include_affixes is set on the transport.
2301 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2302 timeout during the DATA phase of an incoming message.
2304 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2305 applied a patch sanctioned by the Sieve maintainer.
2307 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2308 the uid or gid is negative. A case of a negative gid caused this to be
2309 noticed. The fix allows for either to be negative.
2311 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2312 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2315 PH/36 The change PH/12 above was broken. Fixed it.
2317 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2318 the grounds that routing the same address twice would always produce the
2319 same answer. This might have been true once, but it is certainly no
2320 longer true now. Routing a child address may depend on the previous
2321 routing that produced that child. Some complicated redirection strategies
2322 went wrong when messages had multiple recipients, and made Exim's
2323 behaviour dependent on the order in which the addresses were given.
2325 I have moved the duplicate checking until after the routing is complete.
2326 Exim scans the addresses that are assigned to local and remote
2327 transports, and removes any duplicates. This means that more work will be
2328 done, as duplicates will always all be routed, but duplicates are
2329 presumably rare, so I don't expect this is of any significance.
2331 For deliveries to pipes, files, and autoreplies, the duplicate checking
2332 still happens during the routing process, since they are not going to be
2335 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2336 It corrects a timeout issue with spamd. This is Ian's comment: "The
2337 background is that sometimes spamd either never reads data from a
2338 connection it has accepted, or it never writes response data. The exiscan
2339 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2340 blindly assumes that writes won't block so it may never time out."
2342 PH/39 Allow G after quota size as well as K and M.
2344 PH/40 The value set for $authenticated_id in an authenticator may not contain
2345 binary zeroes or newlines because the value is written to log lines and
2346 to spool files. There was no check on this. Now the value is run through
2347 the string_printing() function so that such characters are converted to
2348 printable escape sequences.
2350 PH/41 $message_linecount is a new variable that contains the total number of
2351 lines in the message. Compare $body_linecount, which is the count for the
2354 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2355 bounce and delay warning messages, except in certain special cases, which
2358 (a) An SMTP error message from a remote host;
2359 (b) A message specified in a :fail: redirection;
2360 (c) A message specified in a "fail" command in a system filter;
2361 (d) A message specified in a FAIL return from the queryprogram router;
2362 (e) A message specified by the cannot_route_message router option.
2364 In these cases only, Exim does include the error details in bounce and
2365 warning messages. There are also a few cases where bland messages such
2366 as "unrouteable address" or "local delivery error" are given.
2368 PH/43 $value is now also set for the "else" part of a ${run expansion.
2370 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2371 being worked on, but at least Exim now implements the latest version to
2374 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2375 process to complete was treated as a delivery failure, a timeout while
2376 writing the message to the pipe was logged, but erroneously treated as a
2377 successful delivery. Such timeouts include transport filter timeouts. For
2378 consistency with the overall process timeout, these timeouts are now
2379 treated as errors, giving rise to delivery failures by default. However,
2380 there is now a new Boolean option for the pipe transport called
2381 timeout_defer, which, if set TRUE, converts the failures into defers for
2382 both kinds of timeout. A transport filter timeout is now identified in
2385 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2386 systems where "make" and "gmake" are different, calling "gmake" at top
2387 level broke things. I've arranged for the value of $(MAKE) to be passed
2388 from the Makefile to this script so that it can call the same version of
2392 A note about Exim versions 4.44 and 4.50
2393 ----------------------------------------
2395 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2396 changes of various kinds. As a consequence, a big documentation update was
2397 needed. This delayed the release for rather longer than seemed good, especially
2398 in the light of a couple of (minor) security issues. Therefore, the changes
2399 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2400 release. So 4.44 and 4.50 are in effect two different branches that both start
2403 I have left the 4.50 change log unchanged; it contains all the changes since
2404 4.43. The change log for 4.44 is below; many of its items are identical to
2405 those for 4.50. This seems to be the most sensible way to preserve the
2406 historical information.
2412 1. Minor wording change to the doc/README.SIEVE file.
2414 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2415 computation of the current number of files was incorrect.
2417 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2418 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2421 4. Give more explanation in the error message when the command for a transport
2422 filter fails to execute.
2424 5. There are several places where Exim runs a non-Exim command in a
2425 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2426 was being done only for the command run by the queryprogram router. It is
2427 now done for all such subprocesses. The other cases are: ${run, transport
2428 filters, and the commands run by the lmtp and pipe transports.
2430 6. Added CONFIGURE_GROUP build-time option.
2432 7. Some older OS have a limit of 256 on the maximum number of file
2433 descriptors. Exim was using setrlimit() to set 1000 as a large value
2434 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2435 systems. I've change it so that if it can't get 1000, it tries for 256.
2437 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2438 was an oversight, and furthermore, ever since the addition of extra
2439 controls (e.g. 4.43/32), the checks on when to allow different forms of
2440 "control" were broken. There should now be diagnostics for all cases when a
2441 control that does not make sense is encountered.
2443 9. Added the /retain_sender option to "control=submission".
2445 10. $recipients is now available in the predata ACL (oversight).
2447 11. Tidy the search cache before the fork to do a delivery from a message
2448 received from the command line. Otherwise the child will trigger a lookup
2449 failure and thereby defer the delivery if it tries to use (for example) a
2450 cached ldap connection that the parent has called unbind on.
2452 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2453 of $address_data from the recipient verification was clobbered by the
2454 sender verification.
2456 13. The value of address_data from a sender verification is now available in
2457 $sender_address_data in subsequent conditions in the ACL statement.
2459 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2461 15. Added a new option "connect=<time>" to callout options, to set a different
2464 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2465 was its contents. (It was OK if the option was not defined at all.)
2467 17. A "Completed" log line is now written for messages that are removed from
2468 the spool by the -Mrm option.
2470 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2471 information about exactly what failed.
2473 19. Added -dd to debug only the daemon process.
2475 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2476 handles timeouts, both on the server side and network timeouts. Renamed the
2477 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2479 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2480 It is now set to "smtps".
2482 22. $host_address is now set to the target address during the checking of
2483 ignore_target_hosts.
2485 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2486 being passed; this would have caused $sender_host_name to have been used if
2487 matching the list had actually called for a host name (not very likely,
2488 since this list is usually IP addresses). A host name is now passed as
2491 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2492 code that specifies a non-restarting handler (typically sigaction() in
2493 modern systems) in an attempt to fix a rare and obscure crash bug.
2495 25. Narrowed the window for a race in the daemon that could cause it to ignore
2496 SIGCHLD signals. This is not a major problem, because they are used only to
2497 wake it up if nothing else does.
2499 26. A malformed maildirsize file could cause Exim to calculate negative values
2500 for the mailbox size or file count. Odd effects could occur as a result.
2501 The maildirsize information is now recalculated if the size or filecount
2504 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2505 support for a long time. Removed HAVE_SYS_VFS_H.
2507 28. Installed the latest version of exipick from John Jetmore.
2509 29. In an address list, if the pattern was not a regular expression, an empty
2510 subject address (from a bounce message) matched only if the pattern was an
2511 empty string. Non-empty patterns were not even tested. This was the wrong
2512 because it is perfectly reasonable to use an empty address as part of a
2513 database query. An empty address is now tested by patterns that are
2514 lookups. However, all the other forms of pattern expect the subject to
2515 contain a local part and a domain, and therefore, for them, an empty
2516 address still always fails if the pattern is not itself empty.
2518 30. Exim went into a mad DNS loop when attempting to do a callout where the
2519 host was specified on an smtp transport, and looking it up yielded more
2520 than one IP address.
2522 31. Re-factored the code for checking spool and log partition space into a
2523 function that finds that data and another that does the check. The former
2524 is then used to implement four new variables: $spool_space, $log_space,
2525 $spool_inodes, and $log_inodes.
2527 32. The RFC2047 encoding function was originally intended for short strings
2528 such as real names; it was not keeping to the 75-character limit for
2529 encoded words that the RFC imposes. It now respects the limit, and
2530 generates multiple encoded words if necessary. To be on the safe side, I
2531 have increased the buffer size for the ${rfc2047: expansion operator from
2534 33. It is now permitted to omit both strings after an "if" condition; if the
2535 condition is true, the result is "true". As before, when the second string
2536 is omitted, a false condition yields an empty string. This makes it less
2537 cumbersome to write custom ACL and router conditions.
2539 34. Failure to deliver a bounce message always caused it to be frozen, even if
2540 there was an errors_to setting on the router. The errors_to setting is now
2543 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2544 canonical form (fully expanded) before being placed in
2545 $sender_host_address.
2547 36. The table in the code that translates DNS record types into text (T_A to
2548 "A" for instance) was missing entries for NS and CNAME. It is just possible
2549 that this could have caused confusion if both these types were looked up
2550 for the same domain, because the text type is used as part of Exim's
2551 per-process caching. But the chance of anyone hitting this buglet seems
2554 37. The dnsdb lookup has been extended in a number of ways.
2556 (1) There is a new type, "zns", which walks up the domain tree until it
2557 finds some nameserver records. It should be used with care.
2559 (2) There is a new type, "mxh", which is like "mx" except that it returns
2560 just the host names, not the priorities.
2562 (3) It is now possible to give a list of domains (or IP addresses) to be
2563 looked up. The behaviour when one of the lookups defers can be
2564 controlled by a keyword.
2566 (4) It is now possible to specify the separator character for use when
2567 multiple records are returned.
2569 38. The dnslists ACL condition has been extended: it is now possible to supply
2570 a list of IP addresses and/or domains to be looked up in a particular DNS
2573 39. Added log_selector=+queue_time_overall.
2575 40. When running the queue in the test harness, wait just a tad after forking a
2576 delivery process, to get repeatability of debugging output.
2578 41. Include certificate and key file names in error message when GnuTLS fails
2579 to set them up, because the GnuTLS error message doesn't include the name
2580 of the failing file when there is a problem reading it.
2582 42. Allow both -bf and -bF in the same test run.
2584 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2586 44. The "Exiscan patch" is now merged into the mainline Exim source.
2588 45. Sometimes the final signoff response after QUIT could fail to get
2589 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2590 < 0 before doing a fflush(). This bug looks as though it goes back to the
2591 introduction of TLS in release 3.20, but "sometimes" must have been rare
2592 because the tests only now provoked it.
2594 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2595 (this can affect the format of dates).
2597 47. exim_tidydb, when checking for the continued existence of a message for
2598 which it has found a message-specific retry record, was not finding
2599 messages that were in split spool directories. Consequently, it was
2600 deleting retry records that should have stayed in existence.
2602 48. Steve fixed some bugs in eximstats.
2604 49. The SPA authentication driver was not abandoning authentication and moving
2605 on to the next authenticator when an expansion was forced to fail,
2606 contradicting the general specification for all authenticators. Instead it
2607 was generating a temporary error. It now behaves as specified.
2609 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2610 (the order specifies the preference for clients). The order is now AES256,
2611 AES128, 3DES, ARCFOUR128.
2613 51. Small patch to Sieve code - explicitly set From: when generating an
2616 52. Exim crashed if a remote delivery caused a very long error message to be
2617 recorded - for instance if somebody sent an entire SpamAssassin report back
2618 as a large number of 550 error lines. This bug was coincidentally fixed by
2619 increasing the size of one of Exim's internal buffers (big_buffer) that
2620 happened as part of the Exiscan merge. However, to be on the safe side, I
2621 have made the code more robust (and fixed the comments that describe what
2624 53. Now that there can be additional text after "Completed" in log lines (if
2625 the queue_time_overall log selector is set), a one-byte patch to exigrep
2626 was needed to allow it to recognize "Completed" as not the last thing in
2629 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2630 patch that reportedly fixes this has been added. I am not expert enough to
2631 create a test for it. This is what the patch creator wrote:
2633 "I found a little strange behaviour of ldap code when working with
2634 Windows 2003 AD Domain, where users was placed in more than one
2635 Organization Units. When I tried to give exim partial DN, the exit code
2636 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2637 But simultaneously result of request was absolutely normal ldap result,
2638 so I produce this patch..."
2640 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2641 REFERENCE, so I have modified the code to exclude the patch when that macro
2644 55. Some experimental protocols are using DNS PTR records for new purposes. The
2645 keys for these records are domain names, not reversed IP addresses. The
2646 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2647 leaves it alone. Component reversal etc. now happens only for IP addresses.
2650 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2652 57. Double the size of the debug message buffer (to 2048) so that more of very
2653 long debug lines gets shown.
2655 58. The exicyclog utility now does better if the number of log files to keep
2656 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2658 59. Two changes related to the smtp_active_hostname option:
2660 (1) $smtp_active_hostname is now available as a variable.
2661 (2) The default for smtp_banner uses $smtp_active_hostname instead
2662 of $primary_hostname.
2664 60. The host_aton() function is supposed to be passed a string that is known
2665 to be a valid IP address. However, in the case of IPv6 addresses, it was
2666 not checking this. This is a hostage to fortune. Exim now panics and dies
2667 if the condition is not met. A case was found where this could be provoked
2668 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2669 components; fortuitously, this particular loophole had already been fixed
2670 by change 4.50/55 above.
2672 If there are any other similar loopholes, the new check in host_aton()
2673 itself should stop them being exploited. The report I received stated that
2674 data on the command line could provoke the exploit when Exim was running as
2675 exim, but did not say which command line option was involved. All I could
2676 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2677 running as the user.
2680 61. There was a buffer overflow vulnerability in the SPA authentication code
2681 (which came originally from the Samba project). I have added a test to the
2682 spa_base64_to_bits() function which I hope fixes it.
2685 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2686 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2688 63. The daemon start-up calls getloadavg() while still root for those OS that
2689 need the first call to be done as root, but it missed one case: when
2690 deliver_queue_load_max is set with deliver_drop_privilege. This is
2691 necessary for the benefit of the queue runner, because there is no re-exec
2692 when deliver_drop_privilege is set.
2694 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2695 This has been fixed.
2697 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2698 was in use, was not putting the data itself into the right store pool;
2699 consequently, it could be overwritten for a subsequent message in the same
2700 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2703 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2705 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2706 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2707 the function were treating the return as a boolean value, which happened to
2708 work because 0=false and not-0=true, but is not correct code.
2710 68. The host_aton() function was not handling scoped IPv6 addresses (those
2711 with, for example, "%eth0" on the end) correctly.
2713 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2714 negated items (that is, ~something) in unsigned ints. Some compilers
2715 apparently mutter when there is no cast.
2717 70. If an address verification called from an ACL failed, and did not produce a
2718 user-specific message (i.e. there was only a "system" message), nothing was
2719 put in $acl_verify_message. In this situation, it now puts the system
2722 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2723 session; change 4.31/43 added the unwanted input to the log line - except
2724 that it did not do this in the start of session case. It now does.
2726 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2727 This is wrong and can cause the other end to generate a synchronization
2728 error if it is another Exim or anything else that does the synchronization
2729 check. A QUIT command is no longer sent after a timeout.
2731 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2732 during host lookups.
2734 74. The defer_ok option of callout verification was not working if it was used
2735 when verifying addresses in header lines, that is, for this case:
2737 verify = header_sender/callout=defer_ok
2739 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2740 those file descriptors could be used for SMTP connections. If anything
2741 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2742 could be sent to the SMTP client, causing chaos. The daemon now opens
2743 stdin, stdout, and stderr to /dev/null when it puts itself into the
2746 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2747 log by default. The user can override this with suitable Perl magic.
2749 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2750 the log message when discard triggers, was not working for the DATA ACL or
2751 for the non-SMTP ACL.
2753 78. Error message wording change in sieve.c.
2755 79. If smtp_accept_max_per_host was set, the number of connections could be
2756 restricted to fewer than expected, because the daemon was trying to set up
2757 a new connection before checking whether the processes handling previous
2758 connections had finished. The check for completed processes is now done
2759 earlier. On busy systems, this bug wouldn't be noticed because something
2760 else would have woken the daemon, and it would have reaped the completed
2763 80. If a message was submitted locally by a user whose login name contained one
2764 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2765 It caused a spool format error. I have fixed the spool reading code. A
2766 related problem was that the "from" clause in the Received: line became
2767 illegal because of the space(s). It is now covered by ${quote_local_part.
2769 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2772 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2774 83. Patch from Sieve maintainer to fix unterminated string problem in
2775 "vacation" handling.
2777 84. Some minor changes to the Linux configuration files to help with other
2778 OS variants using glibc.
2780 85. One more patch for Sieve to update vacation handling to latest spec.
2783 ----------------------------------------------------
2784 See the note above about the 4.44 and 4.50 releases.
2785 ----------------------------------------------------
2791 1. Change 4.43/35 introduced a bug that caused file counts to be
2792 incorrectly computed when quota_filecount was set in an appendfile
2795 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2796 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2799 3. Give more explanation in the error message when the command for a transport
2800 filter fails to execute.
2802 4. There are several places where Exim runs a non-Exim command in a
2803 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2804 was being done only for the command run by the queryprogram router. It is
2805 now done for all such subprocesses. The other cases are: ${run, transport
2806 filters, and the commands run by the lmtp and pipe transports.
2808 5. Some older OS have a limit of 256 on the maximum number of file
2809 descriptors. Exim was using setrlimit() to set 1000 as a large value
2810 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2811 systems. I've change it so that if it can't get 1000, it tries for 256.
2813 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2814 was an oversight, and furthermore, ever since the addition of extra
2815 controls (e.g. 4.43/32), the checks on when to allow different forms of
2816 "control" were broken. There should now be diagnostics for all cases when a
2817 control that does not make sense is encountered.
2819 7. $recipients is now available in the predata ACL (oversight).
2821 8. Tidy the search cache before the fork to do a delivery from a message
2822 received from the command line. Otherwise the child will trigger a lookup
2823 failure and thereby defer the delivery if it tries to use (for example) a
2824 cached ldap connection that the parent has called unbind on.
2826 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2827 of $address_data from the recipient verification was clobbered by the
2828 sender verification.
2830 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2831 was its contents. (It was OK if the option was not defined at all.)
2833 11. A "Completed" log line is now written for messages that are removed from
2834 the spool by the -Mrm option.
2836 12. $host_address is now set to the target address during the checking of
2837 ignore_target_hosts.
2839 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2840 being passed; this would have caused $sender_host_name to have been used if
2841 matching the list had actually called for a host name (not very likely,
2842 since this list is usually IP addresses). A host name is now passed as
2845 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2846 code that specifies a non-restarting handler (typically sigaction() in
2847 modern systems) in an attempt to fix a rare and obscure crash bug.
2849 15. Narrowed the window for a race in the daemon that could cause it to ignore
2850 SIGCHLD signals. This is not a major problem, because they are used only to
2851 wake it up if nothing else does.
2853 16. A malformed maildirsize file could cause Exim to calculate negative values
2854 for the mailbox size or file count. Odd effects could occur as a result.
2855 The maildirsize information is now recalculated if the size or filecount
2858 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2859 support for a long time. Removed HAVE_SYS_VFS_H.
2861 18. Updated exipick to current release from John Jetmore.
2863 19. Allow an empty sender to be matched against a lookup in an address list.
2864 Previously the only cases considered were a regular expression, or an
2867 20. Exim went into a mad DNS lookup loop when doing a callout where the
2868 host was specified on the transport, if the DNS lookup yielded more than
2871 21. The RFC2047 encoding function was originally intended for short strings
2872 such as real names; it was not keeping to the 75-character limit for
2873 encoded words that the RFC imposes. It now respects the limit, and
2874 generates multiple encoded words if necessary. To be on the safe side, I
2875 have increased the buffer size for the ${rfc2047: expansion operator from
2878 22. Failure to deliver a bounce message always caused it to be frozen, even if
2879 there was an errors_to setting on the router. The errors_to setting is now
2882 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2883 canonical form (fully expanded) before being placed in
2884 $sender_host_address.
2886 24. Updated eximstats to version 1.33
2888 25. Include certificate and key file names in error message when GnuTLS fails
2889 to set them up, because the GnuTLS error message doesn't include the name
2890 of the failing file when there is a problem reading it.
2892 26. Expand error message when OpenSSL has problems setting up cert/key files.
2895 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2896 (this can affect the format of dates).
2898 28. exim_tidydb, when checking for the continued existence of a message for
2899 which it has found a message-specific retry record, was not finding
2900 messages that were in split spool directories. Consequently, it was
2901 deleting retry records that should have stayed in existence.
2903 29. eximstats updated to version 1.35
2904 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2905 1.35 - bugfix such that pie charts by volume are generated correctly
2907 30. The SPA authentication driver was not abandoning authentication and moving
2908 on to the next authenticator when an expansion was forced to fail,
2909 contradicting the general specification for all authenticators. Instead it
2910 was generating a temporary error. It now behaves as specified.
2912 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2913 (the order specifies the preference for clients). The order is now AES256,
2914 AES128, 3DES, ARCFOUR128.
2916 31. Small patch to Sieve code - explicitly set From: when generating an
2919 32. Exim crashed if a remote delivery caused a very long error message to be
2920 recorded - for instance if somebody sent an entire SpamAssassin report back
2921 as a large number of 550 error lines. This bug was coincidentally fixed by
2922 increasing the size of one of Exim's internal buffers (big_buffer) that
2923 happened as part of the Exiscan merge. However, to be on the safe side, I
2924 have made the code more robust (and fixed the comments that describe what
2927 33. Some experimental protocols are using DNS PTR records for new purposes. The
2928 keys for these records are domain names, not reversed IP addresses. The
2929 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2930 leaves it alone. Component reversal etc. now happens only for IP addresses.
2933 34. The host_aton() function is supposed to be passed a string that is known
2934 to be a valid IP address. However, in the case of IPv6 addresses, it was
2935 not checking this. This is a hostage to fortune. Exim now panics and dies
2936 if the condition is not met. A case was found where this could be provoked
2937 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2938 components; fortuitously, this particular loophole had already been fixed
2939 by change 4.50/55 or 4.44/33 above.
2941 If there are any other similar loopholes, the new check in host_aton()
2942 itself should stop them being exploited. The report I received stated that
2943 data on the command line could provoke the exploit when Exim was running as
2944 exim, but did not say which command line option was involved. All I could
2945 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2946 running as the user.
2949 35. There was a buffer overflow vulnerability in the SPA authentication code
2950 (which came originally from the Samba project). I have added a test to the
2951 spa_base64_to_bits() function which I hope fixes it.
2954 36. The daemon start-up calls getloadavg() while still root for those OS that
2955 need the first call to be done as root, but it missed one case: when
2956 deliver_queue_load_max is set with deliver_drop_privilege. This is
2957 necessary for the benefit of the queue runner, because there is no re-exec
2958 when deliver_drop_privilege is set.
2960 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2961 was in use, was not putting the data itself into the right store pool;
2962 consequently, it could be overwritten for a subsequent message in the same
2963 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2966 38. Sometimes the final signoff response after QUIT could fail to get
2967 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2968 < 0 before doing a fflush(). This bug looks as though it goes back to the
2969 introduction of TLS in release 3.20, but "sometimes" must have been rare
2970 because the tests only now provoked it.
2976 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2977 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2978 Now it returns an integer. A number of calls were still expecting a T/F
2979 return. Fortuitously, in all cases, the tests worked in OK situations,
2980 which is the norm. However, things would have gone wrong on any write
2981 failures on the smtp file descriptor. This function is used when sending
2982 messages over SMTP and also when doing verify callouts.
2984 2. When Exim is called to do synchronous delivery of a locally submitted
2985 message (the -odf or -odi options), it no longer closes stderr before doing
2988 3. Implemented the mua_wrapper option.
2990 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2992 5. Implemented the functions header_remove(), header_testname(),
2993 header_add_at_position(), and receive_remove_recipient(), and exported them
2996 6. If an ACL "warn" statement specified the addition of headers, Exim already
2997 inserted X-ACL-Warn: at the start if there was no header name. However, it
2998 was not making this test for the second and subsequent header lines if
2999 there were newlines in the string. This meant that an invalid header could
3000 be inserted if Exim was badly configured.
3002 7. Allow an ACL "warn" statement to add header lines at the start or after all
3003 the Received: headers, as well as at the end.
3005 8. Added the rcpt_4xx retry error code.
3007 9. Added postmaster_mailfrom=xxx to callout verification option.
3009 10. Added mailfrom=xxxx to the callout verification option, for verify=
3012 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3013 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3015 12. Inserted some casts to stop certain compilers warning when using pointer
3016 differences as field lengths or precisions in printf-type calls (mostly
3017 affecting debugging statements).
3019 13. Added optional readline() support for -be (dynamically loaded).
3021 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3022 same clock tick as a message's arrival, so that its received time was the
3023 same as the "first fail" time on the retry record, and that message
3024 remained on the queue past the ultimate address timeout, every queue runner
3025 would try a delivery (because it was past the ultimate address timeout) but
3026 after another failure, the ultimate address timeout, which should have then
3027 bounced the address, did not kick in. This was a "< instead of <=" error;
3028 in most cases the first failure would have been in the next clock tick
3029 after the received time, and all would be well.
3031 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3032 being recognized when the domain list was tested by the match_domain
3033 condition in an expansion string.
3035 16. Added the ${str2b64: operator.
3037 17. Exim was always calling setrlimit() to set a large limit for the number of
3038 processes, without checking whether the existing limit was already
3039 adequate. (It did check for the limit on file descriptors.) Furthermore,
3040 errors from getrlimit() and setrlimit() were being ignored. Now they are
3041 logged to the main and panic logs, but Exim does carry on, to try to do its
3042 job under whatever limits there are.
3044 18. Imported PCRE 5.0.
3046 19. Trivial typo in log message " temporarily refused connection" (the leading
3049 20. If the log selector return_path_on_delivery was set and an address was
3050 redirected to /dev/null, the delivery process crashed because it assumed
3051 that a return path would always be set for a "successful" delivery. In this
3052 case, the whole delivery is bypassed as an optimization, and therefore no
3055 21. Internal re-arrangement: the function for sending a challenge and reading
3056 a response while authentication was assuming a zero-terminated challenge
3057 string. It's now changed to take a pointer and a length, to allow for
3058 binary data in such strings.
3060 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3062 23. Exim was not respecting finduser_retries when seeking the login of the
3063 uid under which it was called; it was always trying 10 times. (The default
3064 setting of finduser_retries is zero.) Also, it was sleeping after the final
3065 failure, which is pointless.
3067 24. Implemented tls_on_connect_ports.
3069 25. Implemented acl_smtp_predata.
3071 26. If the domain in control=submission is set empty, Exim assumes that the
3072 authenticated id is a complete email address when it generates From: or
3073 Sender: header lines.
3075 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3076 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3077 chown and chgrp in /bin and hostname in /usr/bin.
3079 28. Exim was keeping the "process log" file open after each use, just as it
3080 does for the main log. This opens the possibility of it remaining open for
3081 long periods when the USR1 signal hits a daemon. Occasional processlog
3082 errors were reported, that could have been caused by this. Anyway, it seems
3083 much more sensible not to leave this file open at all, so that is what now
3086 29. The long-running daemon process does not normally write to the log once it
3087 has entered its main loop, and it closes the log before doing so. This is
3088 so that log files can straightforwardly be renamed and moved. However,
3089 there are a couple of unusual error situations where the daemon does write
3090 log entries, and I had neglected to close the log afterwards.
3092 30. The text of an SMTP error response that was received during a remote
3093 delivery was being truncated at 512 bytes. This is too short for some of
3094 the long messages that one sometimes sees. I've increased the limit to
3097 31. It is now possible to make retry rules that apply only when a message has a
3098 specific sender, in particular, an empty sender.
3100 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3101 it possible to be selective about when SMTP synchronization is enforced.
3103 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3105 32. Implemented hosts_connection_nolog.
3107 33. Added an ACL for QUIT.
3109 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3112 35. Added mailbox_size and mailbox_filecount to appendfile.
3114 36. Added control = no_multiline_responses to ACLs.
3116 37. There was a bug in the logic of the code that waits for the clock to tick
3117 in the case where the clock went backwards by a substantial amount such
3118 that the microsecond fraction of "now" was more than the microsecond
3119 fraction of "then" (but the whole seconds number was less).
3121 38. Added support for the libradius Radius client library this is found on
3122 FreeBSD (previously only the radiusclient library was supported).
3128 1. When certain lookups returned multiple values in the form name=value, the
3129 quoting of the values was not always being done properly. Specifically:
3130 (a) If the value started with a double quote, but contained no whitespace,
3132 (b) If the value contained whitespace other than a space character (i.e.
3133 tabs or newlines or carriage returns) it was not quoted.
3134 This fix has been applied to the mysql and pgsql lookups by writing a
3135 separate quoting function and calling it from the lookup code. The fix
3136 should probably also be applied to nisplus, ibase and oracle lookups, but
3137 since I cannot test any of those, I have not disturbed their existing code.
3139 2. A hit in the callout cache for a specific address caused a log line with no
3140 reason for rejecting RCPT. Now it says "Previous (cached) callout
3141 verification failure".
3143 3. There was an off-by-one bug in the queryprogram router. An over-long
3144 return line was truncated at 256 instead of 255 characters, thereby
3145 overflowing its buffer with the terminating zero. As well as fixing this, I
3146 have increased the buffer size to 1024 (and made a note to document this).
3148 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3149 when Exim is waiting for an SMTP response from a remote server, Exim
3150 restarts its select() call on the socket, thereby resetting its timeout.
3151 This is not a problem when such interrupts are rare. Somebody set up a cron
3152 job to run exiwhat every 2 minutes, which is less than the normal select()
3153 timeout (5 or 10 minutes). This meant that the select() timeout never
3154 kicked in because it was always reset. I have fixed this by comparing the
3155 time when an interrupt arrives with the time at the start of the first call
3156 to select(). If more time than the timeout has elapsed, the interrupt is
3157 treated as a timeout.
3159 5. Some internal re-factoring in preparation for the addition of Sieve
3160 extensions (by MH). In particular, the "personal" test is moved to a
3161 separate function, and given an option for scanning Cc: and Bcc: (which is
3162 not set for Exim filters).
3164 6. When Exim created an email address using the login of the caller as the
3165 local part (e.g. when creating a From: or Sender: header line), it was not
3166 quoting the local part when it contained special characters such as @.
3168 7. Installed new OpenBSD configuration files.
3170 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3171 try to make them clearer.
3173 9. Callout options, other than the timeout value, were being ignored when
3174 verifying sender addresses in header lines. For example, when using
3176 verify = header_sender/callout=no_cache
3178 the cache was (incorrectly) being used.
3180 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3181 only the Cygwin environment.
3183 11. When return_path_on_delivery was set as a log selector, if different remote
3184 addresses in the same message used different return paths and parallel
3185 remote delivery occurred, the wrong values would sometimes be logged.
3186 (Whenever a remote delivery process finished, the return path value from
3187 the most recently started remote delivery process was logged.)
3189 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3190 lines when AUTH and/or TLS are in use. This is the "received protocol"
3191 field. Exim used to use "asmtp" for authenticated SMTP, without any
3192 indication (in the protocol name) for TLS use. Now it follows the RFC and
3193 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3194 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3195 names appear in log lines as well as in Received: header lines.
3197 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3198 extensions, and comparison tests, and to fix some bugs.
3200 14. Changes to the "personal" filter test:
3202 (1) The test was buggy in that it was just doing the equivalent of
3203 "contains" tests on header lines. For example, if a user's address was
3204 anne@some.where, the "personal" test would incorrectly be true for
3206 To: susanne@some.where
3208 This test is now done by extracting each address from the header in turn,
3209 and checking the entire address. Other tests that are part of "personal"
3210 are now done using regular expressions (for example, to check local parts
3211 of addresses in From: header lines).
3213 (2) The list of non-personal local parts in From: addresses has been
3214 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3215 taken from the Sieve specification recommendations.
3217 (3) If the message contains any header line starting with "List-" it is
3218 treated as non-personal.
3220 (4) The test for "circular" in the Subject: header line has been removed
3221 because it now seems ill-conceived.
3223 15. Minor typos in src/EDITME comments corrected.
3225 16. Installed latest exipick from John Jetmore.
3227 17. If headers_add on a router specified a text string that was too long for
3228 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3229 of string_sprintf() is now avoided.
3231 18. $message_body_size was not set (it was always zero) when running the DATA
3232 ACL and the local_scan() function.
3234 19. For the "mail" command in an Exim filter, no default was being set for
3235 the once_repeat time, causing a random time value to be used if "once" was
3236 specified. (If the value happened to be <= 0, no repeat happened.) The
3237 default is now 0s, meaning "never repeat". The "vacation" command was OK
3238 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3239 (I found it when inspecting the code).
3241 20. There is now an overall timeout for performing a callout verification. It
3242 defaults to 4 times the callout timeout, which applies to individual SMTP
3243 commands during the callout. The overall timeout applies when there is more
3244 than one host that can be tried. The timeout is checked before trying the
3245 next host. This prevents very long delays if there are a large number of
3246 hosts and all are timing out (e.g. when the network connections are timing
3247 out). The value of the overall timeout can be changed by specifying an
3248 additional sub-option for "callout", called "maxwait". For example:
3250 verify = sender/callout=5s,maxwait=20s
3252 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3253 to the end before writing, but this should make it even safer).
3255 22. Exim was forgetting that it had advertised PIPELINING for the second and
3256 subsequent messages on an SMTP connection. It was also not resetting its
3257 memory on STARTTLS and an internal HELO.
3259 23. When Exim logs an SMTP synchronization error within a session, it now
3260 records whether PIPELINING has been advertised or not.
3262 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3263 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3264 rather than long int.
3266 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3268 26. Added the never_mail option to autoreply.
3274 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3275 crash if the getsockname() call failed; this can happen if a connection is
3276 closed very soon after it is established. The problem was simply in the
3277 order in which certain operations were done, causing Exim to try to write
3278 to the SMTP stream before it had set up the file descriptor. The bug has
3279 been fixed by making things happen in the correct order.
3285 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3286 before the connection was closed, thus losing the rejection response.
3288 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3289 some early Solaris releases, but causes trouble in current releases where
3290 socklen_t is defined.
3292 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3295 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3298 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3299 that releasing the top of it at the end releases what was used for sub-
3300 expansions (unless the block got too big). However, discard this block if
3301 the first thing is a variable or header, so that we can use its block when
3302 it is dynamic (useful for very large $message_headers, for example).
3304 6. Lookups now cache *every* query, not just the most recent. A new, separate
3305 store pool is used for this. It can be recovered when all lookup caches are
3306 flushed. Lookups now release memory at the end of their result strings.
3307 This has involved some general refactoring of the lookup sources.
3309 7. Some code has been added to the store_xxx() functions to reduce the amount
3310 of flapping under certain conditions.
3312 8. log_incoming_interface used to affect only the <= reception log lines. Now
3313 it causes the local interface and port to be added to several more SMTP log
3314 lines, for example "SMTP connection from", and rejection lines.
3316 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3318 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3320 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3321 could be overwritten at the end of the current message (or the start of a
3322 new message if it was set in a HELO ACL). The value is now preserved for
3323 the duration of the SMTP connection.
3325 12. If a transport had a headers_rewrite setting, and a matching header line
3326 contained an unqualified address, that address was qualified, even if it
3327 did not match any rewriting rules. The underlying bug was that the values
3328 of the flags that permit the existence of unqualified sender and recipient
3329 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3330 non-local messages, and by -bnq for local messages) were not being
3331 preserved with the message after it was received.
3333 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3334 "next input=" as part of the text comprising the host identity instead of
3335 the correct text. The code was using the same buffer for two different
3336 strings. However, depending on which order the printing function evaluated
3337 its arguments, the bug did not always show up. Under Linux, for example, my
3338 test suite worked just fine.
3340 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3341 doesn't work with some older versions of Perl. It has been changed to "my",
3342 which in any case is probably the better facility to use.
3344 15. A really picky compiler found some instances of statements for creating
3345 error messages that either had too many or two few arguments for the format
3348 16. The size of the buffer for calls to the DNS resolver has been increased
3349 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3350 for addresses that have a lot of PTR records. This alleviates a problem; it
3351 does not fully solve it.
3353 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3354 buffer now truncates the list and logs the incident, which is the same
3355 action as happens when Exim is looking up a host name and its aliases.
3356 Previously in this situation something unpredictable would happen;
3357 sometimes it was "internal error: store_reset failed".
3359 18. If a server dropped the connection unexpectedly when an Exim client was
3360 using GnuTLS and trying to read a response, the client delivery process
3361 crashed while trying to generate an error log message.
3363 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3364 string, for example:
3366 warn message = H1: something\nH2: something
3368 the text was added as a single header line from Exim's point of view
3369 though it ended up OK in the delivered message. However, searching for the
3370 second and subsequent header lines using $h_h2: did not work. This has been
3371 fixed. Similarly, if a system filter added multiple headers in this way,
3372 the routers could not see them.
3374 20. Expanded the error message when iplsearch is called with an invalid key to
3375 suggest using net-iplsearch in a host list.
3377 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3378 ACLs are no longer actually imposed (and a message to that effect is
3381 22. If a "gecos" field in a passwd entry contained escaped characters, in
3382 particular, if it contained a \" sequence, Exim got it wrong when building
3383 a From: or a Sender: header from that name. A second bug also caused
3384 incorrect handling when an unquoted " was present following a character
3385 that needed quoting.
3387 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3388 was not being matched caselessly.
3390 24. Arranged for all hyphens in the exim.8 source to be escaped with
3393 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3394 properly. Recipient callout cache records were still being keyed to include
3395 the sender, even when use_sender was set false. This led to far more
3396 callouts that were necessary. The sender is no longer included in the key
3397 when use_sender is false.
3399 26. Added "control = submission" modifier to ACLs.
3401 27. Added the ${base62d: operator to decode base 62 numbers.
3403 28. dnsdb lookups can now access SRV records.
3405 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3406 the configuration file.
3408 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3409 (-v) mode. This makes the output for a verbose queue run more intelligible.
3411 31. Added a use_postmaster feature to recipient callouts.
3413 32. Added the $body_zerocount variable, containing the number of binary zero
3414 bytes in the message body.
3416 33. The time of last modification of the "new" subdirectory is now used as the
3417 "mailbox time last read" when there is a quota error for a maildir
3420 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3422 35. Added +ignore_unknown as a special item in host lists.
3424 36. Code for decoding IPv6 addresses in host lists is now included, even if
3425 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3426 address was recognized as an IP address, but was then not correctly decoded
3427 into binary, causing unexpected and incorrect effects when compared with
3434 1. Very minor rewording of debugging text in manualroute to say "list of
3435 hosts" instead of "hostlist".
3437 2. If verify=header_syntax was set, and a header line with an unqualified
3438 address (no domain) and a large number of spaces between the end of the
3439 name and the colon was received, the reception process suffered a buffer
3440 overflow, and (when I tested it) crashed. This was caused by some obsolete
3441 code that should have been removed. The fix is to remove it!
3443 3. When running in the test harness, delay a bit after writing a bounce
3444 message to get a bit more predictability in the log output.
3446 4. Added a call to search_tidyup() just before forking a reception process. In
3447 theory, someone could use a lookup in the expansion of smtp_accept_max_
3448 per_host which, without the tidyup, could leave open a database connection.
3450 5. Added the variables $recipient_data and $sender_data which get set from a
3451 lookup success in an ACL "recipients" or "senders" condition, or a router
3452 "senders" option, similar to $domain_data and $local_part_data.
3454 6. Moved the writing of debug_print from before to after the "senders" test
3457 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3458 problems for message scanning, either using a data ACL, or using
3459 local_scan() because the Received: header was not generated till after they
3460 were called (in order to set the time as the time of reception completion).
3461 I have revised the way this works. The header is now generated after the
3462 body is received, but before the ACL or local_scan() are called. After they
3463 are run, the timestamp in the header is updated.
3469 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3470 before starting a queue runner without re-exec. This happened only when
3471 deliver_drop_privilege was set or when the Exim user was set to root. The
3472 effect of the bug was that timeouts during subsequent deliveries caused
3473 crashes instead of being properly handled. The handler is now left at its
3474 default (and expected) setting.
3476 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3477 message, again when deliver_drop_privilege is set or Exim is run as root.
3478 The bug described in (1) was not present in this case, but the tidying up
3479 of the other signals was missing. I have made the two cases consistent.
3481 3. The ignore_target_hosts setting on a manualroute router was being ignored
3482 for hosts that were looked up using the /MX notation.
3484 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3487 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3488 operated on the sender address. After changing the $sender_address to <>
3489 for the sender address verify, Exim was re-instated it as the original
3490 (before rewriting) address, but remembering that it had rewritten it, so it
3491 wasn't rewriting it again. This bug also had the effect of breaking the
3492 sender address verification caching when the sender address was rewritten.
3494 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3495 This has been changed so that if the ip literal address matches
3496 ignore_target_hosts, the router declines.
3498 7. Added expansion conditions match_domain, match_address, and match_local_
3499 part (NOT match_host).
3501 8. The placeholder for the Received: header didn't have a length field set.
3503 9. Added code to Exim itself and to exim_lock to test for a specific race
3504 condition that could lead to file corruption when using MBX delivery. The
3505 issue is with the lockfile that is created in /tmp. If this file is removed
3506 after a process has opened it but before that process has acquired a lock,
3507 there is the potential for a second process to recreate the file and also
3508 acquire a lock. This could lead to two Exim processes writing to the file
3509 at the same time. The added code performs the same test as UW imapd; it
3510 checks after acquiring the lock that its file descriptor still refers to
3511 the same named file.
3513 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3514 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3517 11. Added the smtp_active_hostname option. If used, this will typically be made
3518 to depend on the incoming interface address. Because $interface_address is
3519 not set up until the daemon has forked a reception process, error responses
3520 that can happen earlier (such as "too many connections") no longer contain
3523 12. If an expansion in a condition on a "warn" statement fails because a lookup
3524 defers, the "warn" statement is abandoned, and the next ACL statement is
3525 processed. Previously this caused the whole ACL to be aborted.
3527 13. Added the iplsearch lookup type.
3529 14. Added ident_timeout as a log selector.
3531 15. Added tls_certificate_verified as a log selector.
3533 16. Added a global option tls_require_ciphers (compare the smtp transport
3534 option of the same name). This controls incoming TLS connections.
3536 17. I finally figured out how to make tls_require_ciphers do a similar thing
3537 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3538 before starting the TLS session.
3540 18. Tabs are now shown as \t in -bP output.
3542 19. If the log selector return_path_on_delivery was set, Exim crashed when
3543 bouncing a message because it had too many Received: header lines.
3545 20. If two routers both had headers_remove settings, and the first one included
3546 a superfluous trailing colon, the final name in the first list and the
3547 first name in the second list were incorrectly joined into one item (with a
3548 colon in the middle).
3554 1. Added -C and -D options to the exinext utility, mainly to make it easier
3555 to include in the automated testing, but these could be helpful when
3556 multiple configurations are in use.
3558 2. The exinext utility was not formatting the output nicely when there was
3559 an alternate port involved in the retry record key, nor when there was a
3560 message id as well (for retries that were specific to a specific message
3561 and a specific host). It was also confused by IPv6 addresses, because of
3562 the additional colons they contain. I have fixed the IPv4 problem, and
3563 patched it up to do a reasonable job for IPv6.
3565 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3566 delivery, the log line now contains "pipelined" if PIPELINING was used.
3568 4. An SMTP transport process used to panic and die if the bind() call to set
3569 an explicit outgoing interface failed. This has been changed; it is now
3570 treated in the same way as a connect() failure.
3572 5. A reference to $sender_host_name in the part of a conditional expansion
3573 that was being skipped was still causing a DNS lookup. This no longer
3576 6. The def: expansion condition was not recognizing references to header lines
3577 that used bh_ and bheader_.
3579 7. Added the _cache feature to named lists.
3581 8. The code for checking quota_filecount in the appendfile transport was
3582 allowing one more file than it should have been.
3584 9. For compatibility with Sendmail, the command line option
3592 and sets the incoming protocol and host name (for trusted callers). The
3593 host name and its colon can be omitted when only the protocol is to be set.
3594 Note the Exim already has two private options, -pd and -ps, that refer to
3595 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3596 "s", but I don't think that's a major issue.
3598 10. A number of refactoring changes to the code, none of which should affect
3601 (a) The number of logging options was getting close to filling up the
3602 32-bit word that was used as a bit map. I have split them into two classes:
3603 those that are passed in the argument to log_write(), and those that are
3604 only ever tested independently outside of that function. These are now in
3605 separate 32-bit words, so there is plenty of room for expansion again.
3606 There is no change in the user interface or the logging behaviour.
3608 (b) When building, for example, log lines, the code previously used a
3609 macro that called string_cat() twice, in order to add two strings. This is
3610 not really sufficiently general. Furthermore, there was one instance where
3611 it was actually wrong because one of the argument was used twice, and in
3612 one call a function was used. (As it happened, calling the function twice
3613 did not affect the overall behaviour.) The macro has been replaced by a
3614 function that can join an arbitrary number of extra strings onto a growing
3617 (c) The code for expansion conditions now uses a table and a binary chop
3618 instead of a serial search (which was left over from when there were very
3619 few conditions). Also, it now recognizes conditions like "pam" even when
3620 the relevant support is not compiled in: a suitably worded error message is
3621 given if an attempt is made to use such a condition.
3623 11. Added ${time_interval:xxxxx}.
3625 12. A bug was causing one of the ddress fields not to be passed back correctly
3626 from remote delivery subprocesses. The field in question was not being
3627 subsequently used, so this caused to problems in practice.
3629 13. Added new log selectors queue_time and deliver_time.
3631 14. Might have fixed a bug in maildirsizefile handling that threw up
3632 "unexpected character" debug warnings, and recalculated the data
3633 unnecessarily. In any case, I expanded the warning message to give more
3636 15. Added the message "Restricted characters in address" to the statements in
3637 the default ACL that block characters like @ and % in local parts.
3639 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3640 Three changes have been made:
3642 (a) There was a serious bug; a negative response to MAIL caused the whole
3643 recipient domain to be cached as invalid, thereby blocking all messages
3644 to all local parts at the same domain, from all senders. This bug has
3645 been fixed. The domain is no longer cached after a negative response to
3646 MAIL if the sender used is not empty.
3648 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3651 (c) A new callout option, "use_sender" has been added for people who want
3652 the modified behaviour.
3658 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3661 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3662 indeed breaks things for older releases.
3664 3. Added additional logging to the case where there is a problem reading data
3665 from a filter that is running in a subprocess using a pipe, in order to
3666 try to track down a specific problem.
3668 4. Testing facility fudge: when running in the test harness and attempting
3669 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3670 getting "No route to host". Convert this to a timeout.
3672 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3675 6. Some OS don't have socklen_t but use size_t instead. This affects the
3676 fifth argument of getsockopt() amongst other things. This is now
3677 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3678 can be set for individual OS. I have set it for SunOS5, OSF1, and
3679 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3680 some earlier ones do not.
3682 7. Change 4.30/15 was not doing the test caselessly.
3684 8. The standard form for an IPv6 address literal was being rejected by address
3685 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3686 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3687 this, as well as the form without the "IPv6" on the front (but only when
3688 address literals are enabled, of course).
3690 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3692 10. Exim crashed if a message with an empty sender address specified by -f
3693 encountered a router with an errors_to setting. This could be provoked only
3694 by a command such as
3698 where an empty string was supplied; "<>" did not hit this bug.
3700 11. Installed PCRE release 4.5.
3702 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3703 remained set. It is now erased.
3705 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3706 times from message ids (which are base 36 rather than the normal 62).
3708 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3709 were being counted as actual protocol errors, and logged if the log
3710 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3711 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3712 and DATA following a set of rejected RCPTs do not count as protocol errors.
3713 In other words, Exim assumes they were pipelined, though this may not
3714 actually be the case. Of course, in all cases the client gets an
3715 appropriate error code.
3717 15. If a lookup fails in an ACL condition, a message about the failure may
3718 be available; it is used if testing the ACL cannot continue, because most
3719 such messages specify what the cause of the deferral is. However, some
3720 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3721 that caused an old message to be retained and used if a later statement
3722 caused a defer, replacing the real cause of the deferral.
3724 16. If an IP address had so many PTR records that the DNS lookup buffer
3725 was not large enough to hold them, Exim could crash while trying to process
3726 the truncated data. It now detects and logs this case.
3728 17. Further to 4.21/58, another change has been made: if (and only if) the
3729 first line of a message (the first header line) ends with CRLF, a bare LF
3730 in a subsequent header line has a space inserted after it, so as not to
3731 terminate the header.
3733 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3734 unnecessarily, used atoi() instead of strtol(), and didn't check the
3735 termination when getting file sizes from file names by regex.
3737 19. Completely re-implemented the support for maildirsize files, in the light
3738 of a number of problems with the previous contributed implementation
3739 (4.30/29). In particular:
3741 . If the quota is zero, the maildirsize file is maintained, but no quota is
3744 . If the maildir directory does not exist, it is created before any attempt
3745 to write a maildirsize file.
3747 . The quota value in the file is just a cache; if the quota is changed in
3748 the transport, the new value overrides.
3750 . A regular expression is available for excluding directories from the
3753 20. The autoreply transport checks the characters in options that define the
3754 message's headers; it allows continued headers, but it was checking with
3755 isspace() after an embedded newline instead of explicitly looking for a
3758 21. If all the "regular" hosts to which an address was routed had passed their
3759 expiry times, and had not reached their retry times, the address was
3760 bounced, even if fallback hosts were defined. Now Exim should go on to try
3763 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3764 equivalent code in the SMTP transport. Some hosts send humungous responses
3765 to HELO/EHLO, more than 1024 it seems.
3767 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3768 gives compiler warnings in some environments. I've now done it "properly",
3771 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3772 (because of problems with the built-in one) was declared to return uschar *
3773 instead of char *, causing compiler failure.
3775 25. Fixed a file descriptor leak when processing alias/forward files.
3777 26. Fixed a minor format string issue in dbfn.c.
3779 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3781 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3782 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3785 29. When a sender address is verified, it is cached, to save repeating the test
3786 when there is more than one recipient in a message. However, when the
3787 verification involves a callout, it is possible for different callout
3788 options to be set for different recipients. It is too complicated to keep
3789 track of this in the cache, so now Exim always runs a verification when a
3790 callout is required, relying on the callout cache for the optimization.
3791 The overhead is duplication of the address routing, but this should not be
3794 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3795 to be verified with callout=postmaster, and the main callout worked but the
3796 postmaster check failed, the verification correctly failed. However, if a
3797 subsequent RCPT command asked for sender verification *without* the
3798 postmaster check, incorrect caching caused this verification also to fail,
3801 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3802 it was not caching the DNS options (qualify_single, search_parents) that
3803 were used when the lookup failed. A subsequent lookup with different
3804 options therefore always gave the same answer, though there were cases
3805 where it should not have. (Example: a "domains = !$mx_any" option on a
3806 dnslookup router: the "domains" option is always processed without any
3807 widening, but the router might have qualify_single set.) Now Exim uses the
3808 cached value only when the same options are set.
3810 32. Added John Jetmore's "exipick" utility to the distribution.
3812 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3813 than a timeout (e.g. a certificate is required, and is not provided), an
3814 Exim server now closes the connection immediately. Previously it waited for
3815 the client to close - but if the client is SSL, it seems that they each
3816 wait for each other, leading to a delay before one of them times out.
3818 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3819 maintained 0.8.x compatibility because I don't think many are using it, and
3820 it is clearly obsolete.
3822 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3825 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3826 client certificate was expired. A simple patch fixes this, though I don't
3827 understand the full logic of why the verify callback is called multiple
3830 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3831 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3832 which causes problems with some clients (such as the Certicom SSL Plus
3833 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3834 disables the coutermeasure allowing Eudora to connect."
3836 38. Exim was not checking that a write() to a log file succeeded. This could
3837 lead to Bad Things if a log got too big, in particular if it hit a file
3838 size limit. Exim now panics and dies if it cannot write to a log file, just
3839 as it does if it cannot open a log file.
3841 39. Modified OS/Makefile-Linux so that it now contains
3843 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3845 The two -D definitions ensure that Exim is compiled with large file
3846 support, which makes it possible to handle log files that are bigger than
3849 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3850 instance) a domain was checked against a named list that involved a lookup,
3851 causing $domain_data to be set, then another domain was checked against the
3852 same list, then the first domain was re-checked, the value of $domain_data
3853 after the final check could be wrong. In particular, if the second check
3854 failed, it could be set empty. This bug probably also applied to
3857 41. The strip_trailing_dot option was not being applied to the address given
3858 with the -f command-line option.
3860 42. The code for reading a message's header from the spool was incrementing
3861 $received_count, but never initializing it. This meant that the value was
3862 incorrect (doubled) while delivering a message in the same process in which
3863 it was received. In the most common configuration of Exim, this never
3864 happens - a fresh exec is done - but it can happen when
3865 deliver_drop_privilege is set.
3867 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3868 it now includes up to 150 characters of the unexpected data in the log
3871 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3872 and building data strings. The size of both of these buffers was 10 000
3873 bytes - far larger than anybody would *ever* want, thought I. Needless to
3874 say, somebody hit the limit. I have increased the maximum line length to
3875 20 000 and the maximum data length of concatenated lines to 100 000. I have
3876 also fixed two bugs, because there was no checking on these buffers. Tsk,
3877 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3878 buffer is too small.
3880 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3881 lsearch lookups. Now it does.
3883 46. When parsing a route_list item in a manualroute router, a fixed-length
3884 buffer was used for the list of hosts. I made this 1024 bytes long,
3885 thinking that nobody would ever have a list of hosts that long. Wrong.
3886 Somebody had a whole pile of complicated expansion conditions, and the
3887 string was silently truncated, leading to an expansion error. It turns out
3888 that it is easier to change to an unlimited length (owing to other changes
3889 that have happened since this code was originally written) than to build
3890 structure for giving a limitation error. The length of the item that
3891 expands into the list of hosts is now unlimited.
3893 47. The lsearch lookup could not handle data where the length of text line was
3894 more than 4095 characters. Such lines were truncated, leading to shortened
3895 data being returned. It should now handle lines of any length.
3897 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3898 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3901 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3902 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3903 no longer happens in comment lines. A list of replacements is now placed
3904 at the head of all of the source files, except those whose only change is
3905 to replace PERL_COMMAND in the very first #! line.
3907 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3908 messages on the queue, with a bottom-up merge sort, using code contributed
3909 by Michael Haardt. This should make operations like -bp somewhat faster on
3910 large queues. It won't affect queue runners, except when queue_run_in_order
3913 51. Installed eximstats 1.31 in the distribution.
3915 52. Added support for SRV lookups to the dnslookup router.
3917 53. If an ACL referred to $message_body or $message_body_end, the value was not
3918 reset for any messages that followed in the same SMTP session.
3920 54. The store-handling optimization for building very long strings was not
3921 differentiating between the different store pools. I don't think this
3922 actually made any difference in practice, but I've tidied it.
3924 55. While running the routers to verify a sender address, $sender_address
3925 was still set to the sender address. This is wrong, because when routing to
3926 send a bounce to the sender, it would be empty. Therefore, I have changed
3927 it so that, while verifying a sender address, $sender_address is set to <>.
3928 (There is no change to what happens when verifying a recipient address.)
3930 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3931 target A or AAAA records (if not already returned) without resetting the
3932 qualify_single or search_parents options of the DNS resolver. These are
3933 inappropriate in this case because the targets of MX and SRV records must
3934 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3935 target that, when qualified, matched something in the local domain. These
3936 two options are now turned off when doing these lookups.
3938 57. It seems that at least some releases of Reiserfs (which does not have the
3939 concept of a fixed number of inodes) returns zero and not -1 for the
3940 number of available inodes. This interacted badly with check_spool_inodes,
3941 which assumed that -1 was the "no such thing" setting. What I have done is
3942 to check that the total number of inodes is greater than zero before doing
3943 the test of how many are available.
3945 58. When a "warn" ACL statement has a log_message modifier, the message is
3946 remembered, and not repeated. This is to avoid a lot of repetition when a
3947 message has many recipients that cause the same warning to be written.
3948 Howewer, Exim was preserving the list of already written lines for an
3949 entire SMTP session, which doesn't seem right. The memory is now reset if a
3950 new message is started.
3952 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3953 debugging output unless log_rewrite was also set.
3955 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3956 of (int)(handle) when we know that handle contains (void *)(-1).
3958 61. The Exim daemon panic-logs an error return when it closes the incoming
3959 connection. However "connection reset by peer" seems to be common, and
3960 isn't really an error worthy of noting specially, so that particular error
3963 62. When Exim is trying to find all the local interfaces, it used to panic and
3964 die if the ioctl to get the interface flags failed. However, it seems that
3965 on at least one OS (Solaris 9) it is possible to have an interface that is
3966 included in the list of interfaces, but for which you get a failure error
3967 for this call. This happens when the interface is not "plumbed" into a
3968 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3969 failure of the "get flags" call assumes that the interface is down.
3971 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3972 makes life easier for people who are doing arithmetic on fields extracted
3973 from dates, where you often get leading zeros that should not be
3974 interpreted as octal.
3976 64. Added qualify_domain to the redirect router, to override the global
3979 65. If a pathologically long header line contained very many addresses (the
3980 report of this problem mentioned 10 000) and each of them was rewritten,
3981 Exim could use up a very large amount of memory. (It kept on making new
3982 copies of the header line as it rewrote, and never released the old ones.)
3983 At the expense of a bit more processing, the header rewriting function has
3984 been changed so that it no longer eats memory in this way.
3986 66. The generation of the Received: header has been moved from the time that a
3987 message starts to be received, to the time that it finishes. The timestamp
3988 in the Received: header should now be very close to that of the <= log
3989 line. There are two side-effects of this change:
3991 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3992 logged header lines no longer include the local Received: line, because
3993 it has not yet been created. The same applies to a copy of the message
3994 that is returned to a non-SMTP sender when a message is rejected.
3996 (b) When a filter file is tested using -bf, no additional Received: header
3997 is added to the test message. After some thought, I decided that this
4000 This change does not affect the value of $received_for. It is still set
4001 after address rewriting, but before local_scan() is called.
4003 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4005 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4006 gave an unhelpful panic error message, and a defer error. I have managed to
4007 change this behaviour so that it now rejects any supplied certificate,
4008 which seems right, as the list of acceptable certificates is empty.
4010 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4011 gave an unhelpful defer error. I have not managed to make this reject any
4012 supplied certificates, but the error message it gives is "no certificate
4013 supplied", which is not helpful.
4015 70. exigrep's output now also includes lines that are not associated with any
4016 message, but which match the given pattern. Implemented by a patch from
4017 Martin Sluka, which also tidied up the Perl a bit.
4019 71. Recipient callout verification, like sender verification, was using <> in
4020 the MAIL FROM command. This isn't really the right thing, since the actual
4021 sender may affect whether the remote host accepts the recipient or not. I
4022 have changed it to use the actual sender in the callout; this means that
4023 the cache record is now keyed on a recipient/sender pair, not just the
4024 recipient address. There doesn't seem to be a real danger of callout loops,
4025 since a callout by the remote host to check the sender would use <>.
4026 [SEE ABOVE: changed after hitting problems.]
4028 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4029 temporary errors. However, in the case of such a code being given after
4030 the end of a data transmission (i.e. after ".") Exim was failing to write
4031 a retry record for the message. (Yes, there was some broken host that was
4032 actually sending 8xx at this point.)
4034 73. An unknown lookup type in a host list could cause Exim to panic-die when
4035 the list was checked. (An example that provoked this was putting <; in the
4036 middle of a list instead of at the start.) If this happened during a DATA
4037 ACL check, a -D file could be left lying around. This kind of configuration
4038 error no longer causes Exim to die; instead it causes a defer errror. The
4039 incident is still logged to the main and panic logs.
4041 74. Buglet left over from Exim 3 conversion. The message "too many messages
4042 in one connection" was written to the rejectlog but not the mainlog, except
4043 when address rewriting (yes!) was being logged.
4045 75. Added write_rejectlog option.
4047 76. When a system filter was run not as root (that is, when system_filter_user
4048 was set), the values of the $n variables were not being returned to the
4049 main process; thus, they were not subsequently available in the $sn
4052 77. Added +return_path_on_delivery log selector.
4054 78. A connection timeout was being treated differently from recipients deferred
4055 when testing hosts_max_try with a message that was older than the host's
4056 retry timeout. (The host should not be counted, thus allowing all hosts to
4057 be tried at least once before bouncing.) This may have been the cause of an
4058 occasionally reported bug whereby a message would remain on the queue
4059 longer than the retry timeout, but would be bounced if a delivery was
4060 forced. I say "may" because I never totally pinned down the problem;
4061 setting up timeout/retry tests is difficult. See also the next item.
4063 79. The ultimate address timeout was not being applied to errors that involved
4064 a combination of host plus message (for example, a timeout on a MAIL
4065 command). When an address resolved to a number of possible hosts, and they
4066 were not all tried for each delivery (e.g. because of hosts_max_try), a
4067 message could remain on the queue longer than the retry timeout.
4069 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4072 81. Fixed an obscure SMTP outgoing bug which required at least the following
4073 conditions: (a) there was another message waiting for the same server;
4074 (b) the server returned 5xx to all RCPT commands in the first message so
4075 that the message was not completed; (c) the server dropped the connection
4076 or gave a negative response to the RSET that Exim sends to abort the
4077 transaction. The observed case was a dropped connection after DATA that had
4078 been sent in pipelining mode. That is, the server had advertised PIPELINING
4079 but was not implementing it correctly. The effect of the bug was incorrect
4080 behaviour, such as trying another host, and this could lead to a crash.
4086 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4087 and daemon.c were passed as pointers to ints; they should have been
4088 pointers to socklen_t variables (which are typically unsigned ints).
4090 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4093 3. Fixed a really odd bug that affected only the testing scheme; patching a
4094 certain fixed string in the binary changed the value of another string that
4095 happened to be identical to the end of the original first string.
4097 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4098 name", it returns that address as the IP address. On some operating
4099 systems (e.g. Solaris), it also passes back the IP address string as the
4100 "host name". However, on others (e.g. Linux), it passes back an empty
4101 string. Exim wasn't checking for this, and was changing the host name to an
4102 empty string, assuming it had been canonicized.
4104 5. Although rare, it is permitted to have more than one PTR record for a given
4105 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4106 all the names associated with an address, because they do in Solaris.
4107 However, it seems that they do not in Linux for data that comes from the
4108 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4109 I found this out when I moved to a new Linux workstation and tried to run
4110 the Exim test suite.
4112 To get round this problem I have changed the code so that it now does its
4113 own call to the DNS to look up PTR records when searching for a host name.
4114 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4115 addresses that are only in /etc/hosts are still found.
4117 This behaviour is, however, controlled by an option called host_lookup_
4118 order, which defaults to "bydns:byaddr". If people want to use the other
4119 order, or indeed, just use one or the other means of lookup, they can
4120 specify it in this variable.
4122 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4123 some operating systems, this comes back from gethostbyaddr() as an empty
4124 string, and this is what Exim used to test for. However, it seems that in
4125 other systems, "." is yielded. Exim now tests for this case too.
4127 7. The values of check_spool_space and check_log_space are now held internally
4128 as a number of kilobytes instead of an absolute number of bytes. If a
4129 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4130 kilobyte. This means that much larger values can be stored.
4132 8. Exim monitor: an attempt to get the action menu when not actually pointing
4133 at a message produces an empty menu entitled "No message selected". This
4134 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4135 no entries in it ("Shell widget menu has zero width and/or height"). So I
4136 have added a single, blank menu entry in this case.
4138 9. Added ${quote_local_part.
4140 10. MIME decoding is now applied to the contents of Subject: header lines when
4143 11. Now that a reference to $sender_host_address automatically causes a reverse
4144 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4145 host lookup before query-style lookups in lists that might use this
4146 variable. This has therefore been abolished, and the "net-" prefix is no
4147 longer necessary for query-style lookups.
4149 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4150 have been a typo for LFLAGS, so it has been changed.
4152 13. The install script calls Exim with "-C /dev/null" in order to find the
4153 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4154 to be output. Howeve, since Exim outputs its version number before the
4155 error, it didn't break the script. It just looked ugly. I fixed this by
4156 always allowing "-C /dev/null" if the caller is root.
4158 14. Ignore overlarge ACL variable number when reading spool file - insurance
4159 against a later release with more variables having written the file.
4161 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4162 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4163 this, as well as the form without the "IPv6" on the front.
4165 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4166 OS/Makefile-Darwin file.
4168 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4169 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4171 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4172 variable. However, some OS use ints for inodes. I've added cast to long int
4173 to get rid of the compiler warning.
4175 19. I had forgotten to lock out "/../" in configuration file names when
4176 ALT_CONFIG_PREFIX was set.
4178 20. Routers used for verification do not need to specify transports. However,
4179 if such a router generated a host list, and callout was configured, Exim
4180 crashed, because it could not find a port number from the (non-existent)
4181 transport. It now assumes port 25 in this circumstance.
4183 21. Added the -t option to exigrep.
4185 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4186 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4187 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4188 them is set, LOOKUP_LSEARCH is forced.
4190 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4192 24. Added sender and host information to the "rejected by local_scan()" log
4193 line; previously there was no indication of these.
4195 25. Added .include_if_exists.
4197 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4198 Linux. It turns out that not all file systems support this. Apparently some
4199 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4200 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4201 not supported on the file descriptor, is now ignored when Exim is trying to
4202 sync a directory. This applies only to Linux.
4204 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4206 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4207 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4210 29. Added support for maildirsize files from supplied patch (modified a bit).
4212 30. The use of :fail: followed by an empty string could lead Exim to respond to
4213 sender verification failures with (e.g.):
4215 550 Verification failed for <xxx>
4216 550 Sender verify failed
4218 where the first response line was missing the '-' that indicates it is not
4219 the final line of the response.
4221 31. The loop for finding the name of the user that called Exim had a hardwired
4222 limit of 10; it now uses the value of finduser_retries, which is used for
4223 all other user lookups.
4225 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4228 33. Exim was neglecting to zero errno before one call of strtol() when
4229 expanding a string and expecting an integer value. On some systems this
4230 resulted in spurious "integer overflow" errors. Also, it was casting the
4231 result into an int without checking.
4233 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4234 did not work. The code looks as if it has *never* worked, though it appears
4235 to have been documented since at least releast 1.62. I have made it work.
4237 35. The "timeout_DNS" error in retry rules, also documented since at least
4238 1.62, also never worked. As it isn't clear exactly what this means, and
4239 clearly it isn't a major issue, I have abolished the feature by treating it
4240 as "timeout", and writing a warning to the main and panic logs.
4242 36. The display of retry rules for -brt wasn't always showing the error code
4245 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4246 timeout_connect_A, timeout_connect_MX.
4248 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4249 to the empty sender.
4251 39. The daemon was not analysing the content of -oX till after it had closed
4252 stderr and disconnected from the controlling terminal. This meant that any
4253 syntax errors were only noted on the panic log, and the return code from
4254 the command was 0. By re-arranging the code a little, I've made the
4255 decoding happen first, so such errors now appear on stderr, and the return
4256 code is 1. However, the actual setting up of the sockets still happens in
4257 the disconnected process, so errors there are still only recorded on the
4260 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4261 connections (as happens on some IP stacks) was logged at start up time as
4262 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4263 it from "IPv6 and IPv4", which means that two separate sockets are being
4266 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4267 says whether AF_INET or AF_INET6 was passed as an argument.
4269 42. Exiwhat output was messed up when time zones were included in log
4272 43. Exiwhat now gives more information about the daemon's listening ports,
4273 and whether -tls-on-connect was used.
4275 44. The "port" option of the smtp transport is now expanded.
4277 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4278 silently ignored. Now an error message is written to the main and panic
4281 46. There's a new ACL modifier called "logwrite" which writes to a log file
4282 as soon as it is encountered.
4284 47. Added $local_user_uid and $local_user_gid at routing time.
4286 48. Exim crashed when trying to verify a sender address that was being
4289 49. Exim was recognizing only a space character after ".include". It now also
4290 recognizes a tab character.
4292 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4293 extracting the relevant information from the specification. The man page no
4294 longer contains scrambled data for the -d option, and I've added a section
4295 at the front about calling Exim under different names.
4297 51. Added "extra_headers" argument to the "mail" command in filter files.
4299 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4302 53. Installed eximstats 1.29.
4304 54. Added transport_filter_timeout as a generic transport option.
4306 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4307 Cc: header lines. This was required by RFC 822, but it not required by RFC
4310 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4311 incoming messages that did not have them. Now it does so only if the
4312 message originates locally, that is, if there is no associated remote host
4313 address. When Resent- header lines are present, this applies to the Resent-
4314 lines rather than the non-Resent- lines.
4316 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4317 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4319 58. Messages for configuration errors now include the name of the main
4320 configuration file - useful now that there may be more than one file in a
4321 list (.included file names were always shown).
4323 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4324 for those rare installations that do not start the daemon as root or run it
4325 setuid root. I've cut out the call to initgroups() if the daemon is not
4328 60. The Exim user and group can now be bound into the binary as text strings
4329 that are looked up at the start of Exim's processing.
4331 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4333 62. Added $mailstore_basename variable.
4335 63. Installed patch to sieve.c from Michael Haardt.
4337 64. When Exim failed to open the panic log after failing to open the main log,
4338 the original message it was trying to log was written to stderr and debug
4339 output, but if they were not available (the usual case in production), it
4340 was lost. Now it is written to syslog before the two lines that record the
4341 failures to open the logs.
4343 65. Users' Exim filters run in subprocesses under the user's uid. It is
4344 possible for a "deliver" command or an alias in a "personal" command to
4345 provoke an address rewrite. If logging of address rewriting is configured,
4346 this fails because the process is not running as root or exim. There may be
4347 a better way of dealing with this, but for the moment (because 4.30 needs
4348 to be released), I have disabled address rewrite logging when running a
4349 filter in a non-root, non-exim process.
4355 1. The buildconfig auxiliary program wasn't quoting the value set for
4356 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4357 not defined. This bug was masked in 4.22 by the effect that was fixed in
4360 2. Some messages that were rejected after a message id was allocated were
4361 shown as "incomplete" by exigrep. It no longer does this for messages that
4362 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4364 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4365 have allow_domain_literals set, the ID did not get logged in the <= line.
4366 Domain literals are now always recognized in Message-ID: header lines.
4368 4. The first argument for a ${extract expansion item is the key name or field
4369 number. Leading and trailing spaces in this item were not being ignored,
4370 causing some misleading effects.
4372 5. When deliver_drop_privilege was set, single queue runner processes started
4373 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4374 same command in the process it spins off) were not dropping privilege.
4376 6. When the daemon running as "exim" started a queue runner, it always
4377 re-executed Exim in the spun-off process. This is a waste of effort when
4378 deliver_drop_privilege is set. The new process now just calls the
4379 queue-runner function directly.
4385 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4388 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4389 ignored. Though the use of -oP was forcing the writing of a pid file, it
4390 was always written to the default place.
4392 3. If the message "no IP address found for host xxxx" is generated during
4393 incoming verification, it is now followed by identification of the incoming
4394 connection (so you can more easily find what provoked it).
4396 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4398 5. Added some features to "harden" Exim a bit more against certain attacks:
4400 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4401 be put in Local/Makefile. This is like the never_users runtime option,
4402 but it cannot be overridden. The default setting is "root".
4404 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4405 prefix string with which any file named in a -C command line option
4408 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4409 is retained for -C and -D only if the caller of Exim is root. Without
4410 it, the exim user may also use -C and -D and retain privilege.
4412 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4413 command line option is disabled.
4415 6. Macro names set by the -D option must start with an upper case letter, just
4416 like macro names defined in the configuration file.
4418 7. Added "dereference=" facility to LDAP.
4420 8. Two instances of the typo "uknown" in the source files are fixed.
4422 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4423 the Configure-Makefile script screwed up while processing it.
4425 10. Incorporated PCRE 4.4.
4427 11. The SMTP synchronization check was not operating right at the start of an
4428 SMTP session. For example, it could not catch a HELO sent before the client
4429 waited for the greeting. There is now a check for outstanding input at the
4430 point when the greeting is written. Because of the duplex, asynchronous
4431 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4432 way, but not yet received, when the check is performed.
4434 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4435 on TCP/IP sockets, because this apparently causes some broken clients to
4438 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4439 unchanged) from the Cygwin maintainer.
4441 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4442 was supported instead of testing for mbx. Effectively a typo.
4444 15. The spa authenticator server code was not checking that the input it
4445 received was valid base64.
4447 16. The debug output line for the "set" modifier in ACLs was not showing the
4448 name of the variable that was being set.
4450 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4452 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4453 lookup on its own. Something else was needed to trigger the lookup. For
4454 example, a match in host_lookup or the need for a host name in a host list.
4455 Now, if $sender_host_name is referenced and the host name has not yet been
4456 looked up, a lookup is performed. If the lookup fails, the variable remains
4457 empty, and $host_lookup_failed is set to "1".
4459 19. Added "eqi" as a case-independent comparison operator.
4461 20. The saslauthd authentication condition could segfault if neither service
4462 nor realm was specified.
4464 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4465 error message that was logged was misleading, and incoming SMTP
4466 connections were dropped. The message is now more accurate, and temporary
4467 errors are given to SMTP connections.
4469 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4470 (except RSET) with a 421 error, until QUIT is received. However, it was
4471 failing to send a response to QUIT.
4473 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4474 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4475 ACL is now run after the helo_try_verify_hosts code.
4477 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4478 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4479 case-independent, so other case variants are also recognized). Apparently
4480 some systems use these upper case variants.
4482 25. If more than two messages were waiting for the same host, and a transport
4483 filter was specified for the transport, Exim sent two messages over the
4484 same TCP/IP connection, and then failed with "socket operation on non-
4485 socket" when it tried to send the third.
4487 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4489 27. The extern definition of crypt16() in expand.c was not being excluded when
4490 the OS had its own crypt16() function.
4492 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4493 as a preferred synonym for return_size_limit, both as an option and as an
4496 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4498 30. Changed the default configuration ACL to relax the local part checking rule
4499 for addresses that are not in any local domains. For these addresses,
4500 slashes and pipe symbols are allowed within local parts, but the sequence
4501 /../ is explicitly forbidden.
4503 31. SPA server authentication was not clearing the challenge buffer before
4506 32. log_message in a "warn" ACL statement was writing to the reject log as
4507 well as to the main log, which contradicts the documentation and doesn't
4508 seem right (because no rejection is happening). So I have stopped it.
4510 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4511 However, I am unable to do any testing of this.
4513 34. Fixed an infelicity in the appendfile transport. When checking directories
4514 for a mailbox, to see if any needed to be created, it was accidentally
4515 using path names with one or more superfluous leading slashes; tracing
4516 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4518 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4519 contents are added to the log line that is written for every discarded
4520 recipient. (Previously a log_message setting was ignored.)
4522 36. The ${quote: operator now quotes the string if it is empty.
4524 37. The install script runs exim in order to find its version number. If for
4525 some reason other than non-existence or emptiness, which it checks, it
4526 could not run './exim', it was installing it with an empty version number,
4527 i.e. as "exim-". This error state is now caught, and the installation is
4530 38. An argument was missing from the function that creates an error message
4531 when Exim fails to connect to the socket for saslauthd authentication.
4532 This could cause Exim to crash, or give a corrupted message.
4534 39. Added isip, isip4, and isip6 to ${if conditions.
4536 40. The ACL variables $acl_xx are now saved with the message, and can be
4537 accessed later in routers, transports, and filters.
4539 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4540 strings in the file are not string-expanded.
4542 42. If a MAIL command specified a SIZE value that was too large to fit into an
4543 int variable, the check against message_size_limit failed. Such values are
4544 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4545 day this will have to be increased, but I don't think I want to be around
4546 when emails are that large.
4553 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4554 iconv() is not standard in FreeBSD.
4556 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4557 IPv6 enabled. The observed symptom was a segmentation fault on return from
4558 the function os_common_find_running_interfaces() in src/os.c.
4560 3. In the check_special_case() function in daemon.c I had used "errno" as an
4561 argument name, which causes warnings on some systems. This was basically a
4562 typo, since it was named "eno" in the comments!
4564 4. The code that waits for the clock to tick (at a resolution of some fraction
4565 of a second) so as to ensure message-id uniqueness was always waiting for
4566 at least one whole tick, when it could have waited for less. [This is
4567 almost certainly not relevant at current processor speeds, where it is
4568 unlikely to ever wait at all. But we try to future-proof.]
4570 5. The function that sleeps for a time interval that includes fractions of a
4571 second contained a race. It did not block SIGALRM between setting the
4572 timer, and suspending (a couple of lines later). If the interval was short
4573 and the sigsuspend() was delayed until after it had expired, the suspension
4574 never ended. On busy systems this could lead to processes getting stuck for
4577 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4578 process, before it forks any delivery processes. The open lookup caching
4579 mechanism meant that the open file or database connection was passed into
4580 the delivery process. The problem was that delivery processes always tidy
4581 up cached lookup data. This could cause a problem for the next delivery
4582 process started by the queue runner, because the external queue runner
4583 process does not know about the closure. So the next delivery process
4584 still has data in the lookup cache. In the case of a file lookup, there was
4585 no problem because closing a file descriptor in a subprocess doesn't affect
4586 the parent. However, if the lookup was caching a connection to a database,
4587 the connection was closed, and the second delivery process was likely to
4588 see errors such as "PGSQL: query failed: server closed the connection
4589 unexpectedly". The problem has been fixed by closing all cached lookups
4590 in a queue runner before running a delivery process.
4592 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4593 seem to have the "const" qualifier which it has on other OS. I've
4596 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4597 *of the same type* (client or server) with the same public name that an
4598 error should be diagnosed.
4600 9. When Exim looked up a host name for an IP address, but failed to find the
4601 original IP address when looking up the host name (a safety check), it
4602 output the message "<ip address> does not match any IP for NULL", which was
4603 confusing, to say the least. The bug was that the host name should have
4604 appeared instead of "NULL".
4606 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4607 user that is built into the binary, and the -C or -D options is used, root
4608 privilege is dropped before the configuration file is read. In addition,
4609 logging is switched to stderr instead of the normal log files. If the
4610 configuration then re-defines the Exim user, the unprivileged environment
4611 is probably not what is expected, so Exim logs a panic warning message (but
4614 However, if deliver_drop_privilege is set, the unprivileged state may well
4615 be exactly what is intended, so the warning has been cut out in that case,
4616 and Exim is allowed to try to write to its normal log files.
4622 1. smtp_return_error_details was not giving details for temporary sender
4623 or receiver verification errors.
4625 2. Diagnose a configuration error if two authenticators have the same public
4628 3. Exim used not to create the message log file for a message until the first
4629 delivery attempt. This could be confusing when incoming messages were held
4630 for policy or load reasons. The message log file is now created at the time
4631 the message is received, and an initial "Received" line is written to it.
4633 4. The automatically generated man page for command line options had a minor
4634 bug that caused no ill effects; however, a more serious problem was that
4635 the procedure for building the man page automatically didn't always
4636 operate. Consequently, release 4.20 contains an out-of-date version. This
4637 shouldn't happen again.
4639 5. When building Exim with embedded Perl support, the script that builds the
4640 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4641 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4643 6. The freeze_tell option was not being used for messages that were frozen on
4644 arrival, either by an ACL or by local_scan().
4646 7. Added the smtp_incomplete_transaction log selector.
4648 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4649 was accepting AUTH without a new EHLO.
4651 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4652 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4655 10. Logging of TCP/IP connections (when configured) now happens in the main
4656 daemon process instead of the child process, so that the TCP/IP connection
4657 count is more accurate (but it can never be perfect).
4659 11. The use of "drop" in a nested ACL was not being handled correctly in the
4660 outer ACL. Now, if condition failure induced by the nested "drop" causes
4661 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4662 or "require"), the connection is dropped.
4664 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4665 that yield "discard" can now be used with an "accept" or a "discard" verb,
4666 but an error is generated for any others (because I can't see a useful way
4667 to define what should happen).
4669 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4670 are now processed in the same way as lines in the Exim configuration file.
4671 In particular, continuation lines are supported.
4673 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4675 15. Added -ti meaning -t -i.
4677 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4678 domains, and warn by logging if others are found.
4680 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4681 returned from GIFCONF when Exim is trying to find the list of interfaces on
4682 a host. The code in os.c has been modified to copy each ifreq to an aligned
4683 structure in all cases.
4685 Also, in some cases, the returned ifreq's were being copied to a 'struct
4686 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4687 means the last couple of bytes of an IPv6 address could be chopped if the
4688 ifreq contained only a normal sockaddr (14 bytes storage).
4690 18. Named domain lists were not supported in the hosts_treat_as_local option.
4691 An entry such as +xxxx was not recognized, and was treated as a literal
4694 19. Ensure that header lines added by a DATA ACL are included in the reject log
4695 if the ACL subsequently rejects the message.
4697 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4698 MD5 (which is deprecated).
4700 21. When testing a filter file using -bf, Exim was writing a message when it
4701 took the sender from a "From " line in the message, but it was not doing so
4702 when it took $return_path from a Return-Path: header line. It now does.
4704 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4705 with a valid header line field name (a series of printing characters
4706 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4708 23. Changed "disc" in the source to "disk" to conform to the documentation and
4709 the book and for uniformity.
4711 24. Ignore Sendmail's -Ooption=value command line item.
4713 25. When execve() failed while trying to run a command in a pipe transport,
4714 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4715 could be confused with a return value of 69 from the command itself. This
4716 has been changed to 127, the value the shell returns if it is asked to run
4717 a non-existent command. The wording for the related log line suggests a
4718 non-existent command as the problem.
4720 26. If received_header_text expands to an empty string, do not add a Received:
4721 header line to the message. (Well, it adds a token one on the spool, but
4722 marks it "old" so that it doesn't get used or transmitted.)
4724 27. Installed eximstats 1.28 (addition of -nt option).
4726 28. There was no check for failure on the call to getsockname() in the daemon
4727 code. This can fail if there is a shortage of resources on the system, with
4728 ENOMEM, for example. A temporary error is now given on failure.
4730 29. Contrary to the C standard, it seems that in some environments, the
4731 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4732 program. Exim now does this explicitly; it affects the formatting of
4733 timestamps using strftime().
4735 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4736 complaints. I've now initialized all the variables, to avoid this.
4738 32. Header lines added by a system filter were not being "seen" during
4739 transport-time rewrites.
4741 33. The info_callback() function passed to OpenSSL is set up with type void
4742 (*)(SSL *, int, int), as described somewhere. However, when calling the
4743 function (actually a macro) that sets it up, the type void(*)() is
4744 expected. I've put in a cast to prevent warnings from picky compilers.
4746 34. If a DNS black list lookup found a CNAME record, but there were no A
4747 records associated with the domain it pointed at, Exim crashed.
4749 35. If a DNS black list lookup returned more than one A record, Exim ignored
4750 all but the first. It now scans all returned addresses if a particular IP
4751 value is being sought. In this situation, the contents of the
4752 $dnslist_value variable are a list of all the addresses, separated by a
4755 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4756 to accept a host name and all its aliases if the forward lookup for any of
4757 them yielded the IP address of the incoming connection. Now it accepts only
4758 those names whose forward lookup yields the correct IP address. Any other
4759 names are discarded. This closes a loophole whereby a rogue DNS
4760 administrator could create reverse DNS records to break through a
4761 wildcarded host restriction in an ACL.
4763 37. If a user filter or a system filter that ran in a subprocess used any of
4764 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4765 the wrong values were passed to the pipe command ($thisaddress had the
4766 value of $0, $0 had the value of $1, etc). This bug was introduced by
4767 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4769 38. Improved the line breaking for long SMTP error messages from ACLs.
4770 Previously, if there was no break point between 40 and 75 characters, Exim
4771 left the rest of the message alone. Two changes have been made: (a) I've
4772 reduced the minimum length to 35 characters; (b) if it can't find a break
4773 point between 35 and 75 characters, it looks ahead and uses the first one
4774 that it finds. This may give the occasional overlong line, but at least the
4775 remaining text gets split now.
4777 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4778 file descriptors might be low, and that setting 1000 would always raise it.
4779 It turns out that in some environments, the limit is already over 1000 and
4780 that lowering it causes trouble. So now Exim takes care not to decrease it.
4782 40. When delivering a message, the value of $return_path is set to $sender_
4783 address at the start of routing (routers may change the value). By an
4784 oversight, this default was not being set up when an address was tested by
4785 -bt or -bv, which affected the outcome if any router or filter referred to
4788 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4789 log only when "message" or "log_message" are set. However, if one of the
4790 conditions was an address verification, or a call to a nested ACL, the
4791 messages generated by the underlying test were being passed through. This
4792 no longer happens. The underlying message is available in $acl_verify_
4793 message for both "message" and "log_message" expansions, so it can be
4794 passed through if needed.
4796 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4797 new expansion $bh_ to give the encoded byte string without charset
4798 translation. Translation happens only if iconv() is available; HAVE_ICONV
4799 indicates this at build time. HEADERS_CHARSET gives the charset to
4800 translate to; headers_charset can change it in the configuration, and
4801 "headers charset" can change it in an individual filter file.
4803 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4804 that creates RFC 2047 encoded "words" labels them as that charset instead
4805 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4806 expansion operator; (ii) when Exim creates a From: line for a local
4807 message; (iii) when a header line is rewritten to include a "phrase" part.
4809 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4810 buggy, causing it to skip the first lines of messages whose message ID
4811 ended in 'D'. This would not have bitten before Exim release 4.14, because
4812 message IDs were unlikely to end in 'D' before then. The effect was to have
4813 incorrect size information for certain domains.
4815 45. #include "config.h" was missing at the start of the crypt16.c module. This
4816 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4819 46. If there was a timeout during a "random" callout check, Exim treated it as
4820 a failure of the random address, and carried on sending RSET and the real
4821 address. If the delay was just some slowness somewhere, the response to the
4822 original RCPT would be taken as a response to RSET and so on, causing
4823 mayhem of various kinds.
4825 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4826 when I implemented it. It didn't allow for the fact that some option values
4827 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4828 the right test for positive values.
4830 48. Domain names in DNS records are case-independent. Exim always looks them up
4831 in lower case. Some resolvers return domain names in exactly the case they
4832 appear in the zone file, that is, they may contain uppercase letters. Not
4833 all resolvers do this - some return always lower case. Exim was treating a
4834 change of case by a resolver as a change of domain, similar to a widening
4835 of a domain abbreviation. This triggered its re-routing code and so it was
4836 trying to route what was effectively the same domain again. This normally
4837 caused routing to fail (because the router wouldn't handle the domain
4838 twice). Now Exim checks for this case specially, and just changes the
4839 casing of the domain that it ultimately uses when it transmits the message
4842 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4845 50. If a filter generated a file delivery with a non-absolute name (possible if
4846 no home directory exists for the router), the forbid_file option was not
4849 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4850 the existing equality matching.
4852 52. Exim was using ints instead of ino_t variables in some places where it was
4853 dealing with inode numbers.
4855 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4856 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4857 called TMPDIR, and if it finds it is different, it changes its value.
4859 54. The smtp_printf() function is now made available to local_scan() so
4860 additional output lines can be written before returning. There is also an
4861 smtp_fflush() function to enable the detection of a dropped connection.
4862 The variables smtp_input and smtp_batched_input are exported to
4865 55. Changed the default runtime configuration: the message "Unknown user"
4866 has been removed from the ACL, and instead placed on the localuser router,
4867 using the cannot_route_message feature. This means that any verification
4868 failures that generate their own messages won't get overridden. Similarly,
4869 the "Unrouteable address" message that was in the ACL for unverifiable
4870 relay addresses has also been removed.
4872 56. Added hosts_avoid_esmtp to the smtp transport.
4874 57. The exicyclog script was not checking for the esoteric option
4875 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4876 will work only if exicyclog is run under the appropriate euid.
4878 58. Following a discussion on the list, the rules by which Exim recognises line
4879 endings on incoming messages have been changed. The -dropcr and drop_cr
4880 options are now no-ops, retained only for backwards compatibility. The
4881 following line terminators are recognized: LF CRLF CR. However, special
4882 processing applies to CR:
4884 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4885 nor a local message in the state where . is a terminator.
4887 (ii) If a bare CR is encountered in a header line, an extra space is added
4888 after the line terminator so as not to end the header. The reasoning
4889 behind this is that bare CRs in header lines are most likely either
4890 to be mistakes, or people trying to play silly games.
4892 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4893 was being incorrectly given as 18 bytes larger than it should have been.
4894 This is a VOB (very old bug).
4896 60. This may never have affected anything current, but just in case it has:
4897 When the local host is found other than at the start of a list of hosts,
4898 the local host, those with the same MX, and any that follow, are discarded.
4899 When the list in question was part of a longer list of hosts, the following
4900 hosts (not currently being processed) were also being discarded. This no
4901 longer happens. I'm not sure if this situation could ever has previously
4904 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4907 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4908 header. This is something that is recommended in a new Internet Draft, and
4909 is something that is documented as being done by Sendmail. There are two
4910 possible values. For messages generated by the autoreply transport, Exim
4913 Auto-Submitted: auto-replied
4915 whereas for all other generated messages (e.g. bounces) it adds
4917 Auto-Submitted: auto-generated
4919 63. The "personal" condition in filters now includes a test for the
4920 Auto-Submitted: header. If it contains the string "auto-" the message it
4921 not considered personal.
4923 64. Added rcpt_include_affixes as a generic transport option.
4925 65. Added queue_only_override (default true).
4927 66. Added the syslog_duplication option.
4929 67. If what should have been the first header line of a message consisted of
4930 a space followed by a colon, Exim was mis-interpreting it as a header line.
4931 It isn't of course - it is syntactically invalid and should therefore be
4932 treated as the start of the message body. The misbehaviour could have
4933 caused a number of strange effects, including loss of data in subsequent
4934 header lines, and spool format errors.
4936 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4937 client host had authenticated. This control can now be exercised by an ACL
4938 for more flexibility.
4940 69. By default, callouts do not happen when testing with -bh. There is now a
4941 variant, -bhc, which does actually run the callout code, including
4942 consulting and updating the callout cache.
4944 70. Added support for saslauthd authentication, courtesy of Alexander
4947 71. If statvfs() failed on the spool or log directories while checking their
4948 size for availability, Exim confusingly gave the error "space shortage".
4949 Furthermore, in debugging mode it crashed with a floating point exception.
4950 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4951 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4952 serious problem, Exim now writes to the main and panic logs when this
4953 happens, with details of the failure. It then refuses to accept the
4954 incoming message, giving the message "spool directory problem" or "log
4955 directory problem" with a 421 code for SMTP messages.
4957 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4958 0, 1, and 2 exist, because some OS complain for execs without them (see
4959 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4960 descriptors. However, the code omitted to check that the open succeeded,
4961 causing mysterious errors if for some reason the permissions on /dev/null
4962 got screwed. Now Exim writes a message to the main and panic logs, and
4963 bombs out if it can't open /dev/null.
4965 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4966 interact so that it is all more flexible. It is supposed to remain
4967 backwards compatible. Also added extra_local_interfaces.
4969 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4970 to bomb out with an assertion failure - to the client this appears as a
4971 connection drop. This problem occurs in the part of the code that was taken
4972 from the Samba project. Fortunately, the assertion is in a very simple
4973 function, so I have fixed this by reproducing the function inline in the
4974 one place where it is called, and arranging for authentication to fail
4975 instead of killing the process with assert().
4977 75. The SPA client code was not working when the server requested OEM rather
4978 than Unicode encoding.
4980 76. Added code to make require_files with a specific uid setting more usable in
4981 the case where statting the file as root fails - usually a non-root-mounted
4982 NFS file system. When this happens and the failure is EACCES, Exim now
4983 forks a subprocess and does the per-uid checking as the relevant uid.
4985 77. Added process_log_path.
4987 78. If log_file_path was not explicitly set, a setting of check_log_space or
4988 check_log_inodes was ignored.
4990 79. If a space check for the spool or log partitions fails, the incident is now
4991 logged. Of course, in the latter case the data may get lost...
4993 80. Added the %p formatting code to string_format() so that it can be used to
4994 print addresses in debug_print(). Adjusted all the address printing in the
4995 debugging in store.c to use %p rather than %d.
4997 81. There was a concern that a line of code in smtp_in.c could overflow a
4998 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4999 initially expressed, the concern was not well-founded, because trailing
5000 spaces are removed early. However, if the trailing spaces were followed by
5001 a NULL, they did not get removed, so the overflow was possible. Two fixes
5004 (a) I re-wrote the offending code in a cleaner fashion.
5005 (b) If an incoming SMTP command contains a NULL character, it is rejected
5008 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5009 runs initgroups(), so that if the Exim user is in any additional groups,
5010 they will be used during message reception.
5016 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob