1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.584 2009/10/30 15:08:25 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 Bugzilla 437: Prevent Maildix aux files being created with mode 000
54 NM/05 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
146 TK/01 Add preliminary DKIM support. Currently requires a forked version of
147 ALT-N's libdkim that I have put here:
148 http://duncanthrax.net/exim-experimental/
150 Note to Michael Haardt: I had to rename some vars in sieve.c. They
151 were called 'true' and it seems that C99 defines that as a reserved
152 keyword to be used with 'bool' variable types. That means you could
153 not include C99-style headers which use bools without triggering
154 build errors in sieve.c.
156 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
157 as mailq or other aliases. Changed the --help handling significantly
158 to do whats expected. exim_usage() emits usage/help information.
160 SC/01 Added the -bylocaldomain option to eximstats.
162 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
164 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
166 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
172 PH/01 Another patch from the Sieve maintainer.
174 PH/02 When an IPv6 address is converted to a string for single-key lookup
175 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
176 dots are used instead of colons so that keys in lsearch files need not
177 contain colons. This was done some time before quoting was made available
178 in lsearch files. However, iplsearch files do require colons in IPv6 keys
179 (notated using the quote facility) so as to distinguish them from IPv4
180 keys. This meant that lookups for IP addresses in host lists did not work
181 for iplsearch lookups.
183 This has been fixed by arranging for IPv6 addresses to be expressed with
184 colons if the lookup type is iplsearch. This is not incompatible, because
185 previously such lookups could never work.
187 The situation is now rather anomolous, since one *can* have colons in
188 ordinary lsearch keys. However, making the change in all cases is
189 incompatible and would probably break a number of configurations.
191 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
194 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
195 conversion specification without a maximum field width, thereby enabling
196 a rogue spamd server to cause a buffer overflow. While nobody in their
197 right mind would setup Exim to query an untrusted spamd server, an
198 attacker that gains access to a server running spamd could potentially
199 exploit this vulnerability to run arbitrary code as the Exim user.
201 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
202 $primary_hostname instead of what libspf2 thinks the hosts name is.
204 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
205 a directory entry by the name of the lookup key. Previously, if a
206 symlink pointed to a non-existing file or a file in a directory that
207 Exim lacked permissions to read, a lookup for a key matching that
208 symlink would fail. Now it is enough that a matching directory entry
209 exists, symlink or not. (Bugzilla 503.)
211 PH/03 The body_linecount and body_zerocount variables are now exported in the
214 PH/04 Added the $dnslist_matched variable.
216 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
217 This means they are set thereafter only if the connection becomes
220 PH/06 Added the client_condition to authenticators so that some can be skipped
221 by clients under certain conditions.
223 PH/07 The error message for a badly-placed control=no_multiline_responses left
224 "_responses" off the end of the name.
226 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
228 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
229 (without spaces) instead of just copying the configuration text.
231 PH/10 Added the /noupdate option to the ratelimit ACL condition.
233 PH/11 Added $max_received_linelength.
235 PH/12 Added +ignore_defer and +include_defer to host lists.
237 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
238 way in which PCRE > 7.0 is built.
240 PH/14 Implemented queue_only_load_latch.
242 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
243 MAIL command. The effect was to mangle the value on 64-bit systems.
245 PH/16 Another patch from the Sieve maintainer.
247 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
249 PH/18 If a system quota error occurred while trying to create the file for
250 a maildir delivery, the message "Mailbox is full" was not appended to the
251 bounce if the delivery eventually timed out. Change 4.67/27 below applied
252 only to a quota excession during the actual writing of the file.
254 PH/19 It seems that peer DN values may contain newlines (and other non-printing
255 characters?) which causes problems in log lines. The DN values are now
256 passed through string_printing() before being added to log lines.
258 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
259 and InterBase are left for another time.)
261 PH/21 Added message_body_newlines option.
263 PH/22 Guard against possible overflow in moan_check_errorcopy().
265 PH/23 POSIX allows open() to be a macro; guard against that.
267 PH/24 If the recipient of an error message contained an @ in the local part
268 (suitably quoted, of course), incorrect values were put in $domain and
269 $local_part during the evaluation of errors_copy.
275 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
276 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
279 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
280 issue a MAIL command.
282 PH/02 In an ACL statement such as
284 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
286 if a client was not listed at all, or was listed with a value other than
287 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
288 the condition was not true (as it should be), so access was not denied.
289 The bug was that the ! inversion was incorrectly passed on to the second
290 item. This has been fixed.
292 PH/03 Added additional dnslists conditions == and =& which are different from
293 = and & when the dns lookup returns more than one IP address.
295 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
296 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
298 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
299 FSYNC, which compiles an option called disable_fsync that allows for
300 bypassing fsync(). The documentation is heavily laced with warnings.
302 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
304 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
305 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
306 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
307 including adding "make clean"; (3) Added -fPIC when compiling the test
308 dynamically loaded module, to get rid of a warning.
310 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
311 message fails, move_frozen_messages = true and ignore_bounce_errors_after
312 = 0s. The bug is otherwise harmless.
314 PH/07 There was a bug in the dovecot authenticator such that the value of
315 $auth1 could be overwritten, and so not correctly preserved, after a
316 successful authentication. This usually meant that the value preserved by
317 the server_setid option was incorrect.
319 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
321 PH/09 Installed PCRE release 7.0.
323 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
324 run for batched SMTP input. It is now run at the start of every message
325 in the batch. While fixing this I discovered that the process information
326 (output by running exiwhat) was not always getting set for -bs and -bS
327 input. This is fixed, and it now also says "batched" for BSMTP.
329 PH/11 Added control=no_pipelining.
331 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
332 patch, slightly modified), and move the expansion of helo_data till after
333 the connection is made in the smtp transport (so it can use these
336 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
338 PH/14 Added log_selector = +pid.
340 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
342 PH/16 Add ${if forany and ${if forall.
344 PH/17 Added dsn_from option to vary the From: line in DSNs.
346 PH/18 Flush SMTP output before performing a callout, unless control =
347 no_callout_flush is set.
349 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
350 was true (the default) a successful delivery failed to delete the retry
351 item, thus causing premature timeout of the address. The bug is now
354 PH/20 Added hosts_avoid_pipelining to the smtp transport.
356 PH/21 Long custom messages for fakedefer and fakereject are now split up
357 into multiline reponses in the same way that messages for "deny" and
358 other ACL rejections are.
360 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
361 with slight modification.
363 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
364 draft, changing the syntax and factoring some duplicate code".
366 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
367 for deliveries of the second and subsequent messages over the same SMTP
370 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
371 ${reduce, with only minor "tidies".
373 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
375 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
376 expansion side effects.
378 PH/27 When a message times out after an over-quota error from an Exim-imposed
379 quota, the bounce message says "mailbox is full". This message was not
380 being given when it was a system quota that was exceeded. It now should
383 MH/03 Made $recipients available in local_scan(). local_scan() already has
384 better access to the recipient list through recipients_list[], but
385 $recipients can be useful in postmaster-provided expansion strings.
387 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
388 in the case of a MAIL command with additional options following the
389 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
390 were accidentally chopped off.
392 PH/29 SMTP synchronization checks are implemented when a command is read -
393 there is a check that no more input is waiting when there shouldn't be
394 any. However, for some commands, a delay in an ACL can mean that it is
395 some time before the response is written. In this time, more input might
396 arrive, invalidly. So now there are extra checks after an ACL has run for
397 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
398 pipelining has not been advertised.
400 PH/30 MH's patch to allow iscntrl() characters to be list separators.
402 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
403 returned in the SMTP response when smtp_return_error_details was false.
406 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
407 instead of the C I/O that was originally supplied, because problems were
410 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
411 Exim which did not show up earlier: it was assuming that a call to
412 SSL_CTX_set_info_callback() might give an error value. In fact, there is
413 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
414 was a macro that became an assignment, so it seemed to work. This has
415 changed to a proper function call with a void return, hence the compile
416 error. Exim's code has been fixed.
418 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
421 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
423 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
425 PH/37 If a message is not accepted after it has had an id assigned (e.g.
426 because it turns out to be too big or there is a timeout) there is no
427 "Completed" line in the log. When some messages of this type were
428 selected by exigrep, they were listed as "not completed". Others were
429 picked up by some special patterns. I have improved the selection
430 criteria to be more general.
432 PH/38 The host_find_failed option in the manualroute router can now be set
433 to "ignore", to completely ignore a host whose IP address cannot be
434 found. If all hosts are ignored, the behaviour is controlled by the new
435 host_all_ignored option.
437 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
438 homing or because of multiple MX records with /mx) generated more than
439 one IP address, and the following item turned out to be the local host,
440 all the secondary addresses of the first item were incorrectly removed
441 from the list, along with the local host and any following hosts (which
442 is what is supposed to happen).
444 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
445 whoever called Exim into the -H file. In the case of the daemon it was
446 behaving confusingly. When first started, it used values for whoever
447 started the daemon, but after a SIGHUP it used the Exim user (because it
448 calls itself on a restart). I have changed the code so that it now always
451 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
452 message are rejected with the same error (e.g. no authentication or bad
453 sender address), and a DATA command is nevertheless sent (as can happen
454 with PIPELINING or a stupid MUA), the error message that was given to the
455 RCPT commands is included in the rejection of the DATA command. This is
456 intended to be helpful for MUAs that show only the final error to their
459 PH/42 Another patch from the Sieve maintainer.
461 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
462 Eximstats - Fixed some broken HTML links and added missing column headers
464 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
465 columns for Rejects, Temp Rejects, Ham, and Spam rows.
467 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
469 PH/43 Yet another patch from the Sieve maintainer.
471 PH/44 I found a way to check for a TCP/IP connection going away before sending
472 the response to the final '.' that terminates a message, but only in the
473 case where the client has not sent further data following the '.'
474 (unfortunately, this is allowed). However, in many cases there won't be
475 any further data because there won't be any more messages to send. A call
476 to select() can be used: if it shows that the input is "ready", there is
477 either input waiting, or the socket has been closed. An attempt to read
478 the next input character can distinguish the two cases. Previously, Exim
479 would have sent an OK response which the client would never have see.
480 This could lead to message repetition. This fix should cure that, at
481 least in a lot of common cases.
483 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
484 advertised in response to EHLO.
490 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
491 fixed by 4.65/MH/01 (is this a record?) are fixed:
493 (i) An empty string was always treated as zero by the numeric comparison
494 operators. This behaviour has been restored.
496 (ii) It is documented that the numeric comparison operators always treat
497 their arguments as decimal numbers. This was broken in that numbers
498 starting with 0 were being interpreted as octal.
500 While fixing these problems I realized that there was another issue that
501 hadn't been noticed. Values of message_size_limit (both the global option
502 and the transport option) were treated as octal if they started with 0.
503 The documentation was vague. These values are now always treated as
504 decimal, and I will make that clear in the documentation.
510 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
511 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
514 MH/01 Don't check that the operands of numeric comparison operators are
515 integers when their expansion is in "skipping" mode (fixes bug
516 introduced by 4.64-PH/07).
518 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
519 child addresses, Exim now panics and dies. Previously, because the count
520 is held in a short int, deliveries were likely to be lost. As such a
521 large number of recipients for a single message is ridiculous
522 (performance will be very, very poor), I have chosen to impose a limit
523 rather than extend the field.
529 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
530 leftover -K file (the existence of which was triggered by #402).
531 While we were at it, introduced process PID as part of the -K
532 filename. This should rule out race conditions when creating
535 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
536 processing considerably. Previous code took too long for large mails,
537 triggering a timeout which in turn triggers #401.
539 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
540 in the DK code in transports.c. sendfile() is not really portable,
541 hence the _LINUX specificness.
543 TF/01 In the add_headers option to the mail command in an Exim filter,
544 there was a bug that Exim would claim a syntax error in any
545 header after the first one which had an odd number of characters
548 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
549 callout verification, Exim cached a "reject" for the entire domain. This
550 is correct for most verifications, but it is not correct for a recipient
551 verification with use_sender or use_postmaster set, because in that case
552 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
553 case of MAIL FROM:<> rejection from other early rejections (e.g.
554 rejection of HELO). When verifying a recipient using a non-null MAIL
555 address, the cache is ignored if it shows MAIL FROM:<> rejection.
556 Whatever the result of the callout, the value of the domain cache is
557 left unchanged (for any other kind of callout, getting as far as trying
558 RCPT means that the domain itself is ok).
560 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
563 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
564 manifest itself as EPIPE rather than ECONNECT. When tidying away a
565 session, the daemon ignores ECONNECT errors and logs others; it now
566 ignores EPIPE as well.
568 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
569 (quoted-printable decoding).
571 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
572 later the small subsequent patch to fix an introduced bug.
574 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
576 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
578 PH/08 An error is now given if message_size_limit is specified negative.
580 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
581 to be given (somewhat) arbitrary names.
583 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
586 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
587 miscellaneous code fixes
589 PH/10 Added the log_reject_target ACL modifier to specify where to log
592 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
593 hostname. This is wrong, because it relates to the incoming message (and
594 probably the interface on which it is arriving) and not to the outgoing
595 callout (which could be using a different interface). This has been
596 changed to use the value of the helo_data option from the smtp transport
597 instead - this is what is used when a message is actually being sent. If
598 there is no remote transport (possible with a router that sets up host
599 addresses), $smtp_active_hostname is used.
601 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
602 tweaks were necessary in order to get it to work (see also 21 below):
603 (a) The code assumed that strncpy() returns a negative number on buffer
604 overflow, which isn't the case. Replaced with Exim's string_format()
606 (b) There were several signed/unsigned issues. I just did the minimum
607 hacking in of casts. There is scope for a larger refactoring.
608 (c) The code used strcasecmp() which is not a standard C function.
609 Replaced with Exim's strcmpic() function.
610 (d) The code set only $1; it now sets $auth1 as well.
611 (e) A simple test gave the error "authentication client didn't specify
612 service in request". It would seem that Dovecot has changed its
613 interface. Fortunately there's a specification; I followed it and
614 changed what the client sends and it appears to be working now.
616 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
619 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
620 address A is aliased to B and C, where B exists and C does not. Without
621 -v the output is "A verified" because verification stops after a
622 successful redirection if more than one address is generated. However,
623 with -v the child addresses are also verified. Exim was outputting "A
624 failed to verify" and then showing the successful verification for C,
625 with its parentage. It now outputs "B failed to verify", showing B's
626 parentage before showing the successful verification of C.
628 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
629 look up a TXT record in a specific list after matching in a combined
632 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
633 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
634 they consult the DNS. I had assumed they would set it the way they
635 wanted; and indeed my experiments on Linux seem to show that in some
636 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
637 To be on the safe side, however, I have now made the interface to
638 host_find_byname() similar to host_find_bydns(), with an argument
639 containing the DNS resolver options. The host_find_byname() function now
640 sets these options at its start, just as host_find_bydns() does. The smtp
641 transport options dns_qualify_single and dns_search_parents are passed to
642 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
643 of host_find_byname() use the default settings of RES_DEFNAMES
644 (qualify_single) but not RES_DNSRCH (search_parents).
646 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
647 spool_read_header() do less string testing, by means of a preliminary
648 switch on the second character of optional "-foo" lines. (This is
649 overdue, caused by the large number of possibilities that now exist.
650 Originally there were few.) While I was there, I also converted the
651 str(n)cmp tests so they don't re-test the leading "-" and the first
652 character, in the hope this might squeeze out yet more improvement.
654 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
655 flag allowing group syntax was set by the header_syntax check but not
656 turned off, possible causing trouble later; (2) The flag was not being
657 set at all for the header_verify test, causing "group"-style headers to
658 be rejected. I have now set it in this case, and also caused header_
659 verify to ignore an empty address taken from a group. While doing this, I
660 came across some other cases where the code for allowing group syntax
661 while scanning a header line wasn't quite right (mostly, not resetting
662 the flag correctly in the right place). These bugs could have caused
663 trouble for malformed header lines. I hope it is now all correct.
665 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
666 with the "reply" argument non-NULL. The code, however (which originally
667 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
668 but it didn't always do it. This confused somebody who was copying the
669 code for some other use. I have removed all the tests.
671 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
672 feature that was used to support insecure browsers during the U.S. crypto
673 embargo. It requires special client support, and Exim is probably the
674 only MTA that supported it -- and would never use it because real RSA is
675 always available. This code has been removed, because it had the bad
676 effect of slowing Exim down by computing (never used) parameters for the
677 RSA_EXPORT functionality.
679 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
680 authenticator to fail if there's a tab character in the incoming data
681 (there should never be unless someone is messing about, as it's supposed
682 to be base64-encoded). Also added, on Timo's advice, the "secured" option
683 if the connection is using TLS or if the remote IP is the same as the
684 local IP, and the "valid-client-cert option" if a client certificate has
687 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
688 authenticators. This can be used for authorization after authentication
689 succeeds. (In the case of plaintext, it servers for both authentication
692 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
693 if any retry times were supplied.
695 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
696 connection, where there is no client IP address to check. In this
697 situation, the verify now always succeeds.
699 PH/25 Applied John Jetmore's -Mset patch.
701 PH/26 Added -bem to be like -Mset, but loading a message from a file.
703 PH/27 In a string expansion for a processed (not raw) header when multiple
704 headers of the same name were present, leading whitespace was being
705 removed from all of them, but trailing whitespace was being removed only
706 from the last one. Now trailing whitespace is removed from each header
707 before concatenation. Completely empty headers in a concatenation (as
710 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
711 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
713 PH/29 [Removed. This was a change that I later backed out, and forgot to
714 correct the ChangeLog entry (that I had efficiently created) before
715 committing the later change.]
717 PH/30 Exim was sometimes attempting to deliver messages that had suffered
718 address errors (4xx response to RCPT) over the same connection as other
719 messages routed to the same hosts. Such deliveries are always "forced",
720 so retry times are not inspected. This resulted in far too many retries
721 for the affected addresses. The effect occurred only when there were more
722 hosts than the hosts_max_try setting in the smtp transport when it had
723 the 4xx errors. Those hosts that it had tried were not added to the list
724 of hosts for which the message was waiting, so if all were tried, there
725 was no problem. Two fixes have been applied:
727 (i) If there are any address or message errors in an SMTP delivery, none
728 of the hosts (tried or untried) are now added to the list of hosts
729 for which the message is waiting, so the message should not be a
730 candidate for sending over the same connection that was used for a
731 successful delivery of some other message. This seems entirely
732 reasonable: after all the message is NOT "waiting for some host".
733 This is so "obvious" that I'm not sure why it wasn't done
734 previously. Hope I haven't missed anything, but it can't do any
735 harm, as the worst effect is to miss an optimization.
737 (ii) If, despite (i), such a delivery is accidentally attempted, the
738 routing retry time is respected, so at least it doesn't keep
739 hammering the server.
741 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
742 in ${readsocket because some servers need this prod.
744 PH/32 Added some extra debug output when updating a wait-xxx database.
746 PH/33 The hint "could be header name not terminated by colon", which has been
747 given for certain expansion errors for a long time, was not being given
748 for the ${if def:h_colon_omitted{... case.
750 PH/34 The spec says: "With one important exception, whenever a domain list is
751 being scanned, $domain contains the subject domain." There was at least
752 one case where this was not true.
754 PH/35 The error "getsockname() failed: connection reset by peer" was being
755 written to the panic log as well as the main log, but it isn't really
756 panic-worthy as it just means the connection died rather early on. I have
757 removed the panic log writing for the ECONNRESET error when getsockname()
760 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
761 runs only) independently of the message's sender address. This meant
762 that, if the 4xx error was in fact related to the sender, a different
763 message to the same recipient with a different sender could confuse
764 things. In particualar, this can happen when sending to a greylisting
765 server, but other circumstances could also provoke similar problems.
766 I have changed the default so that the retry time for these errors is now
767 based a combination of the sender and recipient addresses. This change
768 can be overridden by setting address_retry_include_sender=false in the
771 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
772 remote server are returned as part of bounce messages. This was not
773 happening for LMTP over a pipe (the lmtp transport), but now it is the
774 same for both kinds of LMTP.
776 PH/38 Despite being documented as not happening, Exim was rewriting addresses
777 in header lines that were in fact CNAMEs. This is no longer the case.
779 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
780 and queue runs started by the daemon processed all messages. This has
781 been fixed so that -R and -S can now usefully be given with -q<time>.
783 PH/40 Import PCRE release 6.7 (fixes some bugs).
785 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
787 PH/42 Give an error if -q is specified more than once.
789 PH/43 Renamed the variables $interface_address and $interface_port as
790 $received_ip_address and $received_port, to make it clear that these
791 values apply to message reception, and not to the outgoing interface when
792 a message is delivered. (The old names remain recognized, of course.)
794 PH/44 There was no timeout on the connect() call when using a Unix domain
795 socket in the ${readsocket expansion. There now is.
797 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
798 be meaningful with "accept".
800 SC/01 Eximstats V1.43
801 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
803 SC/02 Eximstats V1.44
804 Use a glob alias rather than an array ref in the generated
805 parser. This improves both readability and performance.
807 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
808 Collect SpamAssassin and rejection statistics.
809 Don't display local sender or destination tables unless
810 there is data to show.
811 Added average volumes into the top table text output.
813 SC/04 Eximstats V1.46
814 Collect data on the number of addresses (recipients)
815 as well as the number of messages.
817 SC/05 Eximstats V1.47
818 Added 'Message too big' to the list of mail rejection
819 reasons (thanks to Marco Gaiarin).
821 SC/06 Eximstats V1.48
822 Mainlog lines which have GMT offsets and are too short to
823 have a flag are now skipped.
825 SC/07 Eximstats V1.49 (Alain Williams)
826 Added the -emptyok flag.
828 SC/08 Eximstats V1.50
829 Fixes for obtaining the IP address from reject messages.
831 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
832 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
833 whitesspace changes from 4.64-PH/27
835 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
838 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
839 are found, allow negative numbers in numeric criteria)
841 JJ/06 exipick.20061117.2, added new $message_body_missing variable
843 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
844 to match changes made in 4.64-PH/43
846 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
848 PH/47 Put in an explicit test for a DNS lookup of an address record where the
849 "domain" is actually an IP address, and force a failure. This locks out
850 those revolvers/nameservers that support "A-for-A" lookups, in
851 contravention of the specifications.
853 PH/48 When a host name was looked up from an IP address, and the subsequent
854 forward lookup of the name timed out, the host name was left in
855 $sender_host_name, contrary to the specification.
857 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
858 restricted to single-key lookups, Exim was not diagnosing an error if
859 * or *@ was used with a query-style lookup.
861 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
863 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
864 long ago, but noone interested enough thought of it. Let's just say that
865 the "1.1" means that there are some new functions that weren't there at
866 some point in the past.
868 PH/51 Error processing for expansion failure of helo_data from an smtp
869 transport during callout processing was broken.
871 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
872 tested/used via the -bh/-bhc/-bs options.
874 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
875 bug, fixed in subsequent PCRE releases).
877 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
878 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
880 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
886 SC/01 Use a glob alias rather than an array ref in eximstats generated
887 parser. This improves both readability and performance.
889 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
890 Don't display local sender or destination tables in eximstats unless
891 there is data to show.
892 Added average volumes into the eximstats top table text output.
894 SC/03 Collect data on the number of addresses (recipients) as well
895 as the number of messages in eximstats.
897 TF/01 Correct an error in the documentation for the redirect router. Exim
898 does (usually) call initgroups() when daemonizing.
900 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
901 with consistent privilege compared to when running as a daemon.
903 TF/03 Note in the spec that $authenticated_id is not set for local
904 submissions from trusted users.
906 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
907 Thanks to Dean Brooks <dean@iglou.com> for the patch.
909 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
910 by adding some example configuration directives to the default
911 configuration file. A little bit of work is required to uncomment the
912 directives and define how usernames and passwords are checked, but
913 there is now a framework to start from.
915 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
916 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
917 without this. I don't know how relevant this is to other LDAP libraries.
919 PH/02 Add the verb name to the "unknown ACL verb" error.
921 PH/03 Magnus Holmgren's patch for filter_prepend_home.
923 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
925 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
926 directory not expanded when it should be if an expanded home directory
927 was set for the address (which is overridden by the transport).
929 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
932 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
933 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
934 because it is too late at that time, and has no effect.
936 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
937 security issue with \' (bugzilla #107). I could not use the
938 PQescapeStringConn() function, because it needs a PGconn value as one of
941 PH/08 When testing addresses using -bt, indicate those final addresses that
942 are duplicates that would not cause an additional delivery. At least one
943 person was confused, thinking that -bt output corresponded to deliveries.
944 (Suppressing duplicates isn't a good idea as you lose the information
945 about possibly different redirections that led to the duplicates.)
947 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
948 systems where poll() doesn't work, in particular OS X.
950 PH/10 Added more information to debugging output for retry time not reached.
952 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
953 operations in malware.c.
955 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
958 PH/13 If write_rejectlog was set false when logging was sent to syslog with
959 syslog_duplication set false, log lines that would normally be written
960 both the the main log and to the reject log were not written to syslog at
963 PH/14 In the default configuration, change the use of "message" in ACL warn
964 statements to "add_header".
966 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
967 not followed by a command (e.g. "seen endif").
969 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
970 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
973 PH/17 Added extra conditions to the default value of delay_warning_condition
977 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
978 { match{$h_precedence:}{(?i)bulk|list|junk} } \
979 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
982 The Auto-Submitted: and various List- headers are standardised, whereas I
983 don't think Precedence: ever was.
985 PH/18 Refactored debugging code in route_finduser() to show more information,
986 in particular, the error code if getpwnam() issues one.
988 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
989 This is apparently needed in addition to the PH/07 change above to avoid
990 any possible encoding problems.
992 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
993 but not after initializing Perl.
995 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
996 output them only if debugging. By default they are written stderr,
997 apparently, which is not desirable.
999 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1002 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1005 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1007 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1008 authenticated or an ident call has been made. Suppress the default
1009 values for $authenticated_id and $authenticated_sender (but permit -oMai
1010 and -oMas) when testing with -bh.
1012 PH/24 Re-jigged the order of the tests in the default configuration so that the
1013 tests for valid domains and recipients precede the DNS black list and CSA
1014 tests, on the grounds that those ones are more expensive.
1016 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1017 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1018 This bug exists in every version of Exim that I still have, right back to
1021 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1022 However, an attempt to turn on case-sensitivity in a regex key by
1023 including (?-i) didn't work because the subject string was already
1024 lowercased, and the effects were non-intuitive. It turns out that a
1025 one-line patch can be used to allow (?-i) to work as expected.
1031 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1032 other effects) broke the use of negated acl sub-conditions.
1034 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1037 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1038 "Deny" causes Exim to reject the incoming connection with a 554 error.
1039 Unfortunately, if there is a major crisis, such as a disk failure,
1040 tcp-wrappers gives "deny", whereas what one would like would be some
1041 kind of temporary error. A kludge has been added to help with this.
1042 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1043 554 error is used if errno is still zero or contains ENOENT (which occurs
1044 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1047 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1049 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1050 errors. Otherwise a message that provokes a temporary error (when other
1051 messages do not) can cause a whole host to time out.
1053 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1054 addresses were routed directly to files or pipes from a redirect router.
1055 File deliveries just didn't batch; pipe deliveries might have suffered
1058 PH/06 A failure to get a lock for a hints database would erroneously always say
1059 "Failed to get write lock", even when it was really a read lock.
1061 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1062 of 0600. This has been changed to use the value of the lockfile_mode
1063 option (which defaults to 0600).
1065 PH/08 Applied small patch from the Sieve maintainer.
1067 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1068 folder from quota calculations, a direct delivery into this folder messed
1069 up the contents of the maildirsize file. This was because the regex was
1070 used only to exclude .Trash (or whatever) when the size of the mailbox
1071 was calculated. There was no check that a delivery was happening into an
1072 excluded directory. This bug has been fixed by ignoring all quota
1073 processing for deliveries into excluded directories.
1075 PH/10 Added the maildirfolder_create_regex option to appendfile.
1081 PH/01 The code for finding all the local interface addresses on a FreeBSD
1082 system running IPv6 was broken. This may well have applied to all BSD
1083 systems, as well as to others that have similar system calls. The broken
1084 code found IPv4 interfaces correctly, but gave incorrect values for the
1085 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1086 that it would not match correctly against @[] and not recognize the IPv6
1089 PH/02 The ipliteral router was not recognizing addresses of the form user@
1090 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1092 PH/03 Added disable_ipv6.
1094 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1095 decoded form, because it is most often used to construct To: headers
1096 lines in autoreplies, and the decoded form may well be syntactically
1097 invalid. However, $reply_address has leading white space removed, and all
1098 newlines turned into spaces so that the autoreply transport does not
1101 PH/05 If group was specified without a user on a router, and no group or user
1102 was specified on a transport, the group from the router was ignored.
1104 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1105 for visible compile-time settings that can be used to change these
1106 numbers, for those that want even more. Backwards compatibility with old
1107 spool files has been maintained. However, going back to a previous Exim
1108 release will lost any variables that are in spool files.
1110 PH/07 Two small changes when running in the test harness: increase delay when
1111 passing a TCP/IP connection to a new process, in case the original
1112 process has to generate a bounce, and remove special handling of
1113 127.0.0.2 (sic), which is no longer necessary.
1115 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1116 be the same on different OS.
1118 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1121 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1122 whether --show-vars was specified or not
1124 JJ/02 exipick: Added support for new ACL variable spool format introduced
1127 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1128 syntactically invalid From: or Reply-to: line, and a filter used this to
1129 generate an autoreply, and therefore failed to obtain an address for the
1130 autoreply, Exim could try to deliver to a non-existent relative file
1131 name, causing unrelated and misleading errors. What now happens is that
1132 it logs this as a hard delivery error, but does not attempt to create a
1135 PH/11 The exinext utility has a -C option for testing purposes, but although
1136 the given file was scanned by exinext itself; it wasn't being passed on
1137 when Exim was called.
1139 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1140 an end-of-file indication when reading a command response.
1142 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1143 compiled. In many other places in Exim, IPv6 addresses are always
1144 recognized, so I have changed this. It also means that IPv4 domain
1145 literals of the form [IPV4:n.n.n.n] are now always recognized.
1147 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1148 used if the router is not running as root, for example, when verifying at
1149 ACL time, or when using -bh. The debugging output from this situation was
1150 non-existent - all you got was a failure to exec. I have made two
1153 (a) Failures to set uid/gid, the current directory, or a process leader
1154 in a subprocess such as that created by queryprogram now generate
1155 suitable debugging ouput when -d is set.
1157 (b) The queryprogram router detects when it is not running as root,
1158 outputs suitable debugging information if -d is set, and then runs
1159 the subprocess without attempting to change uid/gid.
1161 PH/15 Minor change to Makefile for building test_host (undocumented testing
1164 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1165 additional section of a DNS packet that returns MX or SRV records.
1166 Instead, it always explicitly searches for A/AAAA records. This avoids
1167 major problems that occur when a DNS server includes only records of one
1168 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1169 fixed another bug: if SRV records were looked up and the corresponding
1170 address records were *not* found in the additional section, the port
1171 values from the SRV records were lost.
1173 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1174 using the correct key (the original address) when searching the retry
1175 rules in order to find which one to use for generating the retry hint.
1177 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1178 adding the default one. Similarly, if it contains a Reply-To: header, the
1179 errors_reply_to option, if set, is not used.
1181 PH/19 When calculating a retry time, Exim used to measure the "time since
1182 failure" by looking at the "first failed" field in the retry record. Now
1183 it does not use this if it is later than than the arrival time of the
1184 message. Instead it uses the arrival time. This makes for better
1185 behaviour in cases where some deliveries succeed, thus re-setting the
1186 "first failed" field. An example is a quota failure for a huge message
1187 when small messages continue to be delivered. Without this change, the
1188 "time since failure" will always be short, possible causing more frequent
1189 delivery attempts for the huge message than are intended.
1190 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1192 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1193 $1, $2, $3) because the numerical variables can be reset during some
1194 expansion items (e.g. "match"), thereby losing the authentication data.
1196 PH/21 Make -bV show the size of off_t variables so that the test suite can
1197 decide whether to run tests for quotas > 2G.
1199 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1200 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1201 filecount value is greater than 2G or if a quota value is greater than 2G
1202 on a system where the size of off_t is not greater than 4, a panic error
1205 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1206 never match. The debug and -bh output now contains an explicit error
1207 message indicating a malformed IPv4 address or mask.
1209 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1210 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1211 PH/23 above applies.
1213 PH/25 Do not write to syslog when running in the test harness. The only
1214 occasion when this arises is a failure to open the main or panic logs
1215 (for which there is an explicit test).
1217 PH/26 Added the /no_tell option to "control=freeze".
1219 PH/27 If a host name lookup failed very early in a connection, for example, if
1220 the IP address matched host_lookup and the reverse lookup yielded a name
1221 that did not have a forward lookup, an error message of the form "no IP
1222 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1223 could be logged. Now it outputs the IP address instead of "NULL".
1225 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1226 allows the sender and the authenticated sender to be set when
1227 submitting a message from within Exim. Since child_open_exim() is
1228 documented for local_scan(), the new function should be too.
1230 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1231 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1232 results in an empty string is now treated as unset.
1234 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1236 PH/31 Added sender_verify_fail logging option.
1238 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1239 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1240 tidied the source and removed it altogether.
1242 PH/33 When a queue run was abandoned because the load average was too high, a
1243 log line was always written; now it is written only if the queue_run log
1244 selector is set. In addition, the log line for abandonment now contains
1245 information about the queue run such as the pid. This is always present
1246 in "start" and "stop" lines but was omitted from the "abandon" line.
1248 PH/34 Omit spaces between a header name and the colon in the error message that
1249 is given when verify = headers_syntax fails (if there are lots of them,
1250 the message gets confusing).
1252 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1253 names, as there are now some PTR records that contain slashes. This check
1254 is only to protect against broken name servers that fall over on strange
1255 characters, so the fact that it applies to all lookups doesn't matter.
1257 PH/36 Now that the new test suite is complete, we can remove some of the
1258 special code in Exim that was needed for the old test suite. For example,
1259 sorting DNS records because real resolvers return them in an arbitrary
1260 order. The new test suite's fake resolver always returns records in the
1263 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1264 bounces) except when queue_only is set, to avoid logging races between
1265 the different processes.
1267 PH/38 Panic-die if .include specifies a non-absolute path.
1269 PH/39 A tweak to the "H" retry rule from its user.
1271 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1272 a label. They prevented compilation on older perls.
1274 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1275 a warning to be raised on newish perls.
1277 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1278 on queue. Changes to match documented behaviour of showing count of
1279 messages matching specified criteria.
1281 PH/40 Changed the default ident timeout from 30s to 5s.
1283 PH/41 Added support for the use of login_cap features, on those BSD systems
1284 that have them, for controlling the resources used by pipe deliveries.
1286 PH/42 The content-scanning code uses fopen() to create files in which to put
1287 message data. Previously it was not paying any attention to the mode of
1288 the files. Exim runs with umask(0) because the rest of the code creates
1289 files with open(), and sets the required mode explicitly. Thus, these
1290 files were ending up world-writeable. This was not a big issue, because,
1291 being within the spool directory, they were not world-accessible. I have
1292 created a function called modefopen, which takes an additional mode
1293 argument. It sets umask(777), creates the file, chmods it to the required
1294 mode, then resets the umask. All the relevant calls to fopen() in the
1295 content scanning code have been changed to use this function.
1297 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1298 to 24 hours. This avoids potential overflow problems when processing G
1299 and H retry rules. I suspect nobody ever tinkers with this value.
1301 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1303 PH/45 When the plaintext authenticator is running as a client, the server's
1304 challenges are checked to ensure they are valid base64 strings. By
1305 default, the authentication attempt is cancelled if an invalid string is
1306 received. Setting client_ignore_invalid_base64 true ignores these errors.
1307 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1308 they are received. Thus, the responses can be made to depend on the
1309 challenges. If an invalid string is ignored, an empty string is placed in
1312 PH/46 Messages that are created by the autoreply transport now contains a
1313 References: header, in accordance with RFCs 2822 and 3834.
1315 PH/47 Added authenticated_sender_force to the smtp transport.
1317 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1319 PH/49 Installed latest patch from the Sieve maintainer.
1321 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1322 was also set, the appendfile transport was unnecessarily scanning a
1323 directory of message files (e.g. for maildir delivery) to find the count
1324 of files (along with the size), even though it did not need this
1325 information. It now does the scan only if it needs to find either the
1326 size of the count of files.
1328 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1330 PH/52 Two bugs concerned with error handling when the smtp transport is
1333 (i) Exim was not creating retry information for temporary errors given
1334 for individual recipients after the DATA command when the smtp transport
1335 was used in LMTP mode. This meant that they could be retried too
1336 frequently, and not timed out correctly.
1338 (ii) Exim was setting the flag that allows error details to be returned
1339 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1340 recipients that were returned after the DATA command.
1342 PH/53 This is related to PH/52, but is more general: for any failing address,
1343 when detailed error information was permitted to be returned to the
1344 sender, but the error was temporary, then after the final timeout, only
1345 "retry timeout exceeded" was returned. Now it returns the full error as
1346 well as "retry timeout exceeded".
1348 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1349 do this, and (what is worse) MTAs that accept it.
1351 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1352 will now be deprecated.
1354 PH/56 New os.c-cygwin from the Cygwin maintainer.
1356 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1357 formats (previously only available in exim formats via -bpr, -bpru,
1358 and -bpra. Now also available in native and exiqgrep formats)
1360 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1361 with very large, slow to parse queues
1363 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1365 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1367 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1368 responses to authentication challenges, though it was showing the
1369 challenges; (ii) I've removed the CR characters from the debug output for
1372 PH/58 Allow for the insertion of a newline as well as a space when a string
1373 is turned into more than one encoded-word during RFC 2047 encoding. The
1374 Sieve code now uses this.
1376 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1377 data_4xx, lost_connection, tls_required.
1379 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1380 message was being sent as an SMTP response.
1382 PH/61 Add -l and -k options to exicyclog.
1384 PH/62 When verifying, if an address was redirected to one new address, so that
1385 verification continued, and the new address failed or deferred after
1386 having set something in $address_data, the value of $address_data was not
1387 passed back to the ACL. This was different to the case when no
1388 redirection occurred. The value is now passed back in both cases.
1390 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1391 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1392 use login_cap.h, so on its own it isn't the distinguishing feature. The
1393 new name refers directly to the setclassresources() function.
1395 PH/65 Added configuration files for NetBSD3.
1397 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1399 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1400 is preferred over IPv4.
1402 PH/68 The bounce_return_message and bounce_return_body options were not being
1403 honoured for bounces generated during the reception of non-SMTP messages.
1404 In particular, this applied to messages rejected by the ACL. This bug has
1405 been fixed. However, if bounce_return_message is true and bounce_return_
1406 body is false, the headers that are returned for a non-SMTP message
1407 include only those that have been read before the error was detected.
1408 (In the case of an ACL rejection, they have all been read.)
1410 PH/69 The HTML version of the specification is now built in a directory called
1411 spec_html instead of spec.html, because the latter looks like a path with
1412 a MIME-type, and this confuses some software.
1414 PH/70 Catch two compiler warnings in sieve.c.
1416 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1417 function verify_get_ident() calls ip_connect() to connect a socket, but
1418 if the "connect()" function timed out, ip_connect() used to close the
1419 socket. However, verify_get_ident() also closes the socket later, and in
1420 between Exim writes to the log, which may get opened at this point. When
1421 the socket was closed in ip_connect(), the log could get the same file
1422 descriptor number as the socket. This naturally causes chaos. The fix is
1423 not to close the socket in ip_connect(); the socket should be closed by
1424 the function that creates it. There was only one place in the code where
1425 this was missing, in the iplookup router, which I don't think anybody now
1426 uses, but I've fixed it anyway.
1428 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1429 well as to direct DNS lookups. Otherwise the handling of names in host
1430 lists is inconsistent and therefore confusing.
1436 PH/01 Two changes to the default runtime configuration:
1438 (1) Move the checks for relay_from_hosts and authenticated clients from
1439 after to before the (commented out) DNS black list checks.
1441 (2) Add control=submission to the relay_from_hosts and authenticated
1442 clients checks, on the grounds that messages accepted by these
1443 statements are most likely to be submissions.
1445 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1447 (1) Generate an error if the third argument for the ${prvs expansion is
1450 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1453 (3) Reset the variables that are obtained from the first argument of
1454 ${prvscheck and used in the second argument before leaving the code,
1455 because their memory is reclaimed, so using them afterwards may do
1458 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1459 one (it's much easier than Tom thought :-).
1461 (5) Because of (4), we can now allow for the use of $prvscheck_result
1462 inside the third argument.
1464 PH/03 For some reason, the default setting of PATH when running a command from
1465 a pipe transport was just "/usr/bin". I have changed it to
1468 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1469 anything to be listed in the output from -bV.
1471 PH/05 When a filter generated an autoreply, the entire To: header line was
1472 quoted in the delivery log line, like this:
1474 => >A.N.Other <ano@some.domain> <original@ddress> ...
1476 This has been changed so that it extracts the operative address. There
1477 may be more than one such address. If so, they are comma-separated, like
1480 => >ano@some.domain,ona@other.domain <original@ddress> ...
1482 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1483 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1484 not being looked up in the rDNS to get a host name, Exim was showing the
1485 IP address twice in Received: lines, even though the IP addresses were
1486 identical. For example:
1488 Received: from [1.2.3.4] (helo=[1.2.3.4])
1490 However, if the real host name was known, it was omitting the HELO data
1491 if it matched the actual IP address. This has been tidied up so that it
1492 doesn't show the same IP address twice.
1494 PH/07 When both +timestamp and +memory debugging was on, the value given by
1495 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1496 called by the debug printing, thereby overwriting the timestamp buffer.
1497 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1499 PH/08 When the original message was included in an autoreply transport, it
1500 always said "this is a copy of the message, including all the headers",
1501 even if body_only or headers_only was set. It now gives an appropriate
1504 PH/09 Applied a patch from the Sieve maintainer which:
1506 o fixes some comments
1507 o adds the (disabled) notify extension core
1508 o adds some debug output for the result of if/elsif tests
1509 o points to the current vacation draft in the documentation
1510 and documents the missing references header update
1514 o fixes a bug in processing the envelope test (when testing
1515 multiple envelope elements, the last element determinted the
1518 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1519 Electronic Mail") by including:
1521 Auto-submitted: auto-generated
1523 in the messages that it generates (bounce messages and others, such as
1524 warnings). In the case of bounce messages for non-SMTP mesages, there was
1525 also a typo: it was using "Auto_submitted" (underscore instead of
1526 hyphen). Since every message generated by Exim is necessarily in response
1527 to another message, thes have all been changed to:
1529 Auto-Submitted: auto-replied
1531 in accordance with these statements in the RFC:
1533 The auto-replied keyword:
1535 - SHOULD be used on messages sent in direct response to another
1536 message by an automatic process,
1538 - MUST NOT be used on manually-generated messages,
1540 - MAY be used on Delivery Status Notifications (DSNs) and Message
1541 Disposition Notifications (MDNs),
1543 - MUST NOT be used on messages generated by automatic or periodic
1544 processes, except for messages which are automatic responses to
1547 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1548 to the default Received: header definition.
1550 PH/12 Added log selector acl_warn_skipped (default on).
1552 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1553 variables because (a) they are in the wrong storage pool and (b) even if
1554 they were copied, it wouldn't work properly because of the caching.
1556 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1557 checking when decoding. Apparently there are clients that generate
1558 overlong encoded strings. Why am I not surprised?
1560 PH/15 If the first argument of "${if match_address" was not empty, but did not
1561 contain an "@" character, Exim crashed. Now it writes a panic log message
1562 and treats the condition as false.
1564 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1566 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1567 extension "envelope-auth". The code is finished and in agreement with
1568 other implementations, but there is no documentation so far and in fact,
1569 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1570 not changing the active code.
1572 Print executed "if" and "elsif" statements when debugging is used. This
1573 helps a great deal to understand what a filter does.
1575 Document more things not specified clearly in RFC3028. I had all this
1576 sorted out, when out of a sudden new issues came to my mind. Oops."
1578 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1581 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1582 canonical form (as documented). However, after a host name lookup from
1583 the IP address, check_host() was doing a simple string comparison with
1584 addresses acquired from the DNS when checking that the found name did
1585 have the original IP as one of its addresses. Since any found IPv6
1586 addresses are likely to be in abbreviated form, the comparison could
1587 fail. Luckily, there already exists a function for doing the comparison
1588 by converting both addresses to binary, so now that is used instead of
1589 the text comparison.
1591 PH/20 There was another similar case to PH/19, when a complete host name was
1592 given in a host list; looking up its IP address could give an abbreviated
1593 form, whereas the current host's name might or might not be abbreviated.
1594 The same fix has been applied.
1600 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1601 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1604 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1605 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1607 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1609 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1610 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1611 Also, for "Auto-Submitted", treat anything other than "no" as
1612 non-personal, in accordance with RFC 3834. (Previously it treated
1613 anything starting "auto-" as non-personal.)
1615 TF/01 The control=submission/name=... option had a problem with syntax
1616 errors if the name included a slash character. The /name= option
1617 now slurps the rest of the string, so it can include any characters
1618 but it must come last in the list of options (after /sender_retain
1621 PH/05 Some modifications to the interface to the fake nameserver for the new
1629 TK/01 Added the "success_on_redirect" address verification option. See
1630 NewStuff for rationale and an example.
1632 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1634 PH/02 Patch to exigrep to allow it to work on syslog lines.
1636 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1637 fread() to skip over the body file's header line, because in Cygwin the
1638 header line is locked and is inaccessible.
1640 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1641 co-exist for some time) to make it clear that it is the Exim ID that is
1642 referenced, not the Message-ID: header line.
1644 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1645 string_format() function, because snprintf() does not exist on all
1648 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1649 ${stat: expansion item.
1651 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1652 protocol synchronization error", to keep the pedants happy.
1654 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1655 well as for IRIX systems, when gcc is being used. See the host.c source
1658 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1660 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1663 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1664 transport and to the smtp transport in LMTP mode.
1666 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1668 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1669 run a filter in a subprocess. This could lead to confusion in subsequent
1670 lookups in the parent process. There should also be a search_tidyup() at
1671 the end of the subprocess.
1673 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1674 only if the host matched helo_try_verify_hosts, which caused the
1675 verification to occur when the EHLO/HELO command was issued. The ACL just
1676 tested the remembered result. Now, if a previous verification attempt has
1677 not happened, "verify = helo" does it there and then.
1679 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1681 TK/03 Fix log output including CR from clamd.
1683 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1684 exist provoked a memory error which could cause a segfault.
1686 PH/15 Installed PCRE 6.2
1688 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1690 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1691 of the problem. Specifically, suggested +O2 rather than +O1 for the
1694 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1696 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1697 record, the port was not correctly logged when the outgoing_port log
1698 selector was set (it logged the transort's default port).
1700 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1701 fallback_hosts, and "hosts" in the smtp transport.
1703 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1704 host errors such as "Connection refused".
1706 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1707 authentication with radiusclient 0.4.9:
1709 - Error returned from rc_read_config was caught wrongly
1710 - Username/password not passed on to radius server due to wrong length.
1712 The presumption is that some radiusclient API changes for 4.51/PH/17
1713 were not taken care of correctly. The code is still untested by me (my
1714 Linux distribution still has 0.3.2 of radiusclient), but it was
1715 contributed by a Radius user.
1717 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1718 expanding the "port" option of the smtp transport.
1720 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1721 while reading a MIME header. Thanks to Tom Hughes for a patch.
1723 PH/24 Include config.h inside local_scan.h so that configuration settings are
1726 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1727 that in an ACL for RCPT (for example), you can examine exactly what was
1730 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1731 commands, but it was not correctly comparing the address with the actual
1732 client host address. Thus, it would show the EHLO address in Received:
1733 header lines when this was not necessary.
1735 PH/27 Added the % operator to ${eval:}.
1737 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1738 it should be ignoring failures (because with -C, for example, it has lost
1739 privilege). It wasn't ignoring creation failures other than "already
1742 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1745 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1746 because an input line was too long, either on its own, or by virtue of
1747 too many continuations, the temporary file was not being removed, and the
1748 return code was incorrect.
1750 PH/31 Missing "BOOL" in function definition in filtertest.c.
1752 PH/32 Applied Sieve patches from the maintainer.
1754 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1756 PH/33 Added "verify = not_blind".
1758 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1759 Local/Makefile (with some defaults set). These are used in built scripts
1760 such as exicyclog, but they have never been used in the exim_install
1761 script (though there are many overriding facilities there). I have
1762 arranged that the exim_install script now takes note of these two
1765 PH/35 Installed configuration files for Dragonfly.
1767 PH/36 When a locally submitted message by a trusted user did not contain a
1768 From: header, and the sender address was obtained from -f or from an SMTP
1769 MAIL command, and the trusted user did not use -F to supply a sender
1770 name, $originator_name was incorrectly used when constructing a From:
1771 header. Furthermore, $originator_name was used for submission mode
1772 messages from external hosts without From: headers in a similar way,
1773 which is clearly wrong.
1775 PH/37 Added control=suppress_local_fixups.
1777 PH/38 When log_selector = +received_sender was set, and the addition of the
1778 sender made the log line's construction buffer exactly full, or one byte
1779 less than full, an overflow happened when the terminating "\n" was
1782 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1783 when the result of a list match is failure because a DNS lookup failed.
1785 PH/40 RM_COMMAND is now used in the building process.
1787 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1788 the "build-* directories that it finds.
1790 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1791 address in a domain literal was a prefix of an interface address.
1793 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1794 when verifying a sender address, unless rewrite_headers is false.
1796 PH/44 Wrote a long comment about why errors_to addresses are verified as
1797 recipients, not senders.
1799 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1800 the ratelimit ACL was added.
1802 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1804 PH/46 Added extra information about PostgreSQL errors to the error string.
1806 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1807 avoiding the need to install special zones in a real server. This is
1808 backwards compatible; if it can't find the fake resolver, it drops back.
1809 Thus, both old and new test suites can be run.
1811 TF/02 Added util/ratelimit.pl
1813 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1814 clock is set back in time.
1816 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1817 Candler <B.Candler@pobox.com>.
1819 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1820 OK for addresses that are the result of redirections.
1822 PH/48 A number of further additions for the benefit of the new test suite,
1823 including a fake gethostbyname() that interfaces to the fake DNS resolver
1826 TF/06 The fix for widen_domains has also been applied to qualify_single and
1827 search_parents which are the other dnslookup options that can cause
1830 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1833 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1835 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1836 service" if it encountered an MX record with an empty target hostname.
1837 The message is now "an MX or SRV record indicated no SMTP service".
1839 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1840 if the DNS of the sending site is misconfigured. This is quite a
1841 common situation. This change restores the behaviour of treating a
1842 helo verification defer as a failure.
1844 PH/51 If self=fail was set on a router, the bounce message did not include the
1845 actual error message.
1851 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1853 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1854 command itself ended in error, the underlying message about the transport
1855 filter timeout was being overwritten with the pipe command error. Now the
1856 underlying error message should be appended to the second error message.
1858 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1860 PH/02 Reduce the amount of output that "make" produces by default. Full output
1861 can still be requested.
1863 PH/03 The warning log line about a condition test deferring for a "warn" verb
1864 was being output only once per connection, rather than after each
1865 occurrence (because it was using the same function as for successful
1866 "warn" verbs). This seems wrong, so I have changed it.
1868 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1869 it should not have, which might have caused a crash in the right
1870 circumstances, but probably never did.
1872 PH/04 Installed a modified version of Tony Finch's patch to make submission
1873 mode fix the return path as well as the Sender: header line, and to
1874 add a /name= option so that you can make the user's friendly name appear
1877 TF/03 Added the control = fakedefer ACL modifier.
1879 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1880 Mark Lowes for thorough testing.
1882 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1884 TK/03 Merged latest SRS patch from Miles Wilton.
1886 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1887 with the definition in sysexits.h (which is #included earlier).
1888 Fortunately, Exim does not actually use EX_OK. The code used to try to
1889 preserve the sysexits.h value, by assumimg that macro definitions were
1890 scanned for macro replacements. I have been disabused of this notion,
1891 so now the code just undefines EX_OK before #including unistd.h.
1893 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1894 in the smtp transport. When a block could not be written in a single
1895 write() function, the timeout was being re-applied to each part-write.
1896 This seems wrong - if the receiver was accepting one byte at a time it
1897 would take for ever. The timeout is now adjusted when this happens. It
1898 doesn't have to be particularly precise.
1900 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1901 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1903 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1904 without a domain if the check to <postmaster@domain> fails.
1906 SC/01 Eximstats: added -xls and the ability to specify output files
1907 (patch written by Frank Heydlauf).
1909 SC/02 Eximstats: use FileHandles for outputing results.
1911 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1913 SC/04 Eximstats: fixed display of large numbers with -nvr option
1915 SC/05 Eximstats: fixed merging of reports with empty tables.
1917 SC/06 Eximstats: added the -include_original_destination flag
1919 SC/07 Eximstats: removed tabs and trailing whitespace.
1921 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1923 TK/06 MBOX spool code: Add real "From " MBOX separator line
1924 so the .eml file is really in mbox format (even though
1925 most programs do not really care). Patch from Alex Miller.
1927 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1928 The latter is generated from $received_to and is only set if the
1929 message has one envelope recipient. SA can use these headers,
1930 obviously out-of-the-box. Patch from Alex Miller.
1932 PH/08 The ${def test on a variable was returning false if the variable's
1933 value was "0", contrary to what the specification has always said!
1934 The result should be true unless the variable is empty.
1936 PH/09 The syntax error of a character other than { following "${if
1937 def:variable_name" (after optional whitespace) was not being diagnosed.
1938 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1939 accidental colon was present, for example, could give incorrect results.
1941 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1942 result is used (not including appendfile, where other changes are about
1945 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1946 This involved changing a lot of size variables from int to off_t. It
1947 should work with maildirs and everything.
1949 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1950 spamd dying while we are connected to it.
1952 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1955 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1956 function for BDB 4.3.
1958 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1960 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1961 the sources for PCRE have been re-arranged and more files are now
1964 PH/15 The code I had for printing potentially long long variables in PH/11
1965 above was not the best (it lost precision). The length of off_t variables
1966 is now inspected at build time, and an appropriate printing format (%ld
1967 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1968 to be "long long int" or "long int". This is needed for the internal
1969 formatting function string_vformat().
1971 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1972 the configuration file to be ":syslog", then the script "guesses" where
1973 the logs files are, rather than using the compiled in default. In our
1974 case the guess is not the same as the compiled default, so the script
1975 suddenly stopped working when I started to use syslog. The patch checks
1976 to see if log_file_path is "". If so, it attempts to read it from exim
1977 with no configuration file to get the compiled in version, before it
1978 falls back to the previous guessing code."
1980 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1981 implementing BATV in an Exim configuration. See NewStuff for the gory
1984 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1985 Makefile that are specific to HP-UX.
1987 PH/18 If the "use_postmaster" option was set for a recipient callout together
1988 with the "random" option, the postmaster address was used as the MAIL
1989 FROM address for the random test, but not for the subsequent recipient
1990 test. It is now used for both.
1992 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1993 patch removes a few documentation additions to RFC 3028, because the
1994 latest draft now contains them. It adds the new en;ascii-case comparator
1995 and a new error check for 8bit text in MIME parts. Comparator and
1996 require names are now matched exactly. I enabled the subaddress
1997 extension, but it is not well tested yet (read: it works for me)."
1999 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2000 rework some of the code of TK/09 above to avoid the hardwired use of
2001 "%lld" and "long long". Replaced the call to snprintf() with a call to
2004 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2005 records point to non-existent hosts", "retry timeout exceeded", and
2006 "retry time not reached for any host after a long failure period".
2008 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2009 experimental DomainKeys support:
2011 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2012 (2) On an error such as an illegally used "control", the wrong name for
2013 the control was given.
2015 These problems did NOT occur unless DomainKeys support was compiled.
2017 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2019 PH/24 Added ${if match_ip condition.
2021 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2022 databases so that it will be absolutely obvious if a crash occurs in the
2023 DB library. This is a regular occurrence (often caused by mis-matched
2026 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2027 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2028 picked up on a user's system that detects such things. There doesn't seem
2029 to be a gcc warning option for this - only an attribute that has to be
2030 put on the function's prototype. It seems that in Fedora Core 4 they have
2031 set this on a number of new functions. No doubt there will be more in due
2034 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2035 specify a transport. However, if an address that was verified by such a
2036 router was the subject of a callout, Exim crashed because it tried to
2037 read the rcpt_include_affixes from the non-existent transport. Now it
2038 just assumes that the setting of that option is false. This bug was
2039 introduced by 4.51/PH/31.
2041 PH/28 Changed -d+all to exclude +memory, because that information is very
2042 rarely of interest, but it makes the output a lot bigger. People tend to
2043 do -d+all out of habit.
2045 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2046 code in os-type was giving problems when libc.so lives in lib64, like on
2049 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2050 aren't the modern standard, and it seems that some systems' include files
2051 don't always have them. Exim was already checking for some of the newer
2052 ones like T_AAAA, and defining it itself. I've added checks for all the
2053 record types that Exim uses.
2055 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2056 not automatically generating a new one, as it is supposed to. This
2057 prevented TLS from working. If the file did exist, but contained invalid
2058 data, a new version was generated, as expected. It was only the case of a
2059 non-existent file that was broken.
2061 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2062 with a change in libdomainkeys > 0.64.
2064 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2065 from DNS. If the selector record carries the flag, it now has
2066 precedence over the domain-wide flag.
2068 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2070 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2071 the use of a port name that isn't defined in /etc/services) occurred, the
2072 message was deferred as in a normal delivery, and thus remained on the
2073 spool, instead of being failed because of the mua_wrapper setting. This
2074 is now fixed, and I tidied up some of the mua_wrapper messages at the
2077 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2078 the messages in a hash of arrays rather than using individual hashes.
2079 This is a bit cleaner and results in dramatic memory savings, albeit
2080 at a slight CPU cost.
2082 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2083 as requested by Marc Sherman.
2085 SC/10 Eximstats: added histograms for user specified patterns as requested
2088 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2090 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2091 fopen() in the content-scanning modules that did not already have it.
2097 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2098 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2100 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2102 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2104 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2105 to test Sieve filters that use "vacation".
2107 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2108 that changes the way the GnuTLS parameters are stored in the cache file.
2109 The new format can be generated externally. For backward compatibility,
2110 if the data in the cache doesn't make sense, Exim assumes it has read an
2111 old-format file, and it generates new data and writes a new file. This
2112 means that you can't go back to an older release without removing the
2115 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2116 work if there are any delivery delays because "one_time" forces the
2117 parent to be marked "delivered", so its unseen clone is never tried
2118 again. For this reason, Exim now forbids the simultaneous setting of
2121 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2122 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2123 entry if you want to know the details. The fix, however, neglected to
2124 consider the case when local delivery batching is involved. The test for
2125 "previously delivered" was not happening when checking to see if an
2126 address could be batched with a previous (undelivered) one; under
2127 certain circumstances this could lead to multiple deliveries to the same
2130 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2131 in its include files, and this causes problems building Exim.
2133 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2134 header_syntax) but Exim was just ignoring anything given after a slash.
2135 In particular, this caused confusion with an attempt to use "verify =
2136 reverse_host_lookup/defer_ok". An error is now given when options are
2137 supplied for verify items that do not have them. (Maybe reverse_host_
2138 lookup should have a defer_ok option, but that's a different point.)
2140 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2141 defined by RFC 821) to 2048, because there were problems with some AUTH
2142 commands, and RFC 1869 says the size should be increased for extended
2143 SMTP commands that take arguments.
2145 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2148 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2149 "unknown" error; now it says that the functionality isn't in the binary.
2151 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2152 an address' error message when a string expansion fails (syntax or
2153 whatever). Otherwise the password may appear in the log. Following change
2154 PH/42 below, there is no longer a chance of it appearing in a bounce
2157 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2159 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2160 of its addresses was ever tried. (Bugzilla bug #2.)
2162 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2163 the result incorrectly in the debug output. (It correctly added a newline
2164 to what was transported.)
2166 TF/01 Added $received_time.
2168 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2169 commented out examples of how to interface to a virus scanner and to
2170 SpamAssassin. Also added commented examples of av_scanner and
2171 spamd_address settings.
2173 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2174 and controls are allowed in which ACLs. There were a couple of minor
2175 errors. Some of the entries in the conditions table (which is a table of
2176 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2177 negation of where the condition IS allowed.
2179 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2181 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2182 header file does not have a version number, so I've had to invent a new
2183 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2184 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2185 radiusclient), but it was contributed by a Radius user.
2187 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2188 files or directories, for OpenSSL.
2190 PH/19 When an Exim process that is running as root has to create an Exim log
2191 file, it does so in a subprocess that runs as exim:exim so as to get the
2192 ownership right at creation (otherwise, other Exim processes might see
2193 the file with the wrong ownership). There was no test for failure of this
2194 fork() call, which would lead to the process getting stuck as it waited
2195 for a non-existent subprocess. Forks do occasionally fail when resources
2196 run out. I reviewed all the other calls to fork(); they all seem to check
2199 PH/20 When checking for unexpected SMTP input at connect time (before writing
2200 the banner), Exim was not dealing correctly with a non-positive return
2201 from the read() function. If the client had disconnected by this time,
2202 the result was a log entry for a synchronization error with an empty
2203 string after "input=" when read() returned zero. If read() returned -1
2204 (an event I could not check), uninitialized data bytes were printed.
2205 There were reports of junk text (parts of files, etc) appearing after
2208 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2210 PH/22 Added support for macro redefinition, and (re)definition in between
2211 driver and ACL definitions.
2213 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2214 forgetting to use the resulting value; it was using the unexpanded value.
2216 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2217 hadn't been configured. The fix is from Juergen Kreileder, who
2218 understands it better than I do:
2220 "Here's what I see happening with three configured cyrus_sasl
2221 authenticators configured (plain, login, cram-md5):
2223 On startup auth_cyrus_sasl_init() gets called for each of these.
2224 This means three calls to sasl_listmech() without a specified mech_list.
2225 => SASL tests which mechs of all available mechs actually work
2226 => three warnings about OTP not working
2227 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2229 With the patch, sasl_listmech() also gets called three times. But now
2230 SASL's mech_list option is set to the server_mech specified in the the
2231 authenticator. Or in other words, the answer from sasl_listmech()
2232 gets limited to just the mech you're testing for (which is different
2234 => the return list contains just 'plain' or 'login', 'cram-md5' or
2235 nothing depending on the value of ob->server_mech.
2237 I've just tested the patch: Authentication still works fine,
2238 unavailable mechs specified in the exim configuration are still
2239 caught, and the auth.log warnings about OTP are gone."
2241 PH/25 When debugging is enabled, the contents of the command line are added
2242 to the debugging output, even when log_selector=+arguments is not
2245 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2246 answer is "GNU", and only if the return is "GNU/something" is the answer
2249 PH/27 $acl_verify_message is now set immediately after the failure of a
2250 verification in an ACL, and so is available in subsequent modifiers. In
2251 particular, the message can be preserved by coding like this:
2253 warn !verify = sender
2254 set acl_m0 = $acl_verify_message
2256 Previously, $acl_verify_message was set only while expanding "message"
2257 and "log_message" when a very denied access.
2259 PH/28 Modified OS/os.c-Linux with
2261 -#ifndef OS_LOAD_AVERAGE
2262 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2264 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2265 nomenclature these days.)
2267 PH/29 Installed patch from the Sieve maintainer that adds the options
2268 sieve_useraddress and sieve_subaddress to the redirect router.
2270 PH/30 In these circumstances:
2271 . Two addresses routed to the same list of hosts;
2272 . First host does not offer TLS;
2273 . First host accepts first address;
2274 . First host gives temporary error to second address;
2275 . Second host offers TLS and a TLS session is established;
2276 . Second host accepts second address.
2277 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2278 and peerdn, if requested) that were in fact used only for the second
2281 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2282 attention to any local part prefix or suffix that was matched by the
2283 router that accepted the address. It now behaves in the same way as it
2284 does for delivery: the affixes are removed from the local part unless
2285 rcpt_include_affixes is set on the transport.
2287 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2288 timeout during the DATA phase of an incoming message.
2290 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2291 applied a patch sanctioned by the Sieve maintainer.
2293 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2294 the uid or gid is negative. A case of a negative gid caused this to be
2295 noticed. The fix allows for either to be negative.
2297 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2298 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2301 PH/36 The change PH/12 above was broken. Fixed it.
2303 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2304 the grounds that routing the same address twice would always produce the
2305 same answer. This might have been true once, but it is certainly no
2306 longer true now. Routing a child address may depend on the previous
2307 routing that produced that child. Some complicated redirection strategies
2308 went wrong when messages had multiple recipients, and made Exim's
2309 behaviour dependent on the order in which the addresses were given.
2311 I have moved the duplicate checking until after the routing is complete.
2312 Exim scans the addresses that are assigned to local and remote
2313 transports, and removes any duplicates. This means that more work will be
2314 done, as duplicates will always all be routed, but duplicates are
2315 presumably rare, so I don't expect this is of any significance.
2317 For deliveries to pipes, files, and autoreplies, the duplicate checking
2318 still happens during the routing process, since they are not going to be
2321 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2322 It corrects a timeout issue with spamd. This is Ian's comment: "The
2323 background is that sometimes spamd either never reads data from a
2324 connection it has accepted, or it never writes response data. The exiscan
2325 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2326 blindly assumes that writes won't block so it may never time out."
2328 PH/39 Allow G after quota size as well as K and M.
2330 PH/40 The value set for $authenticated_id in an authenticator may not contain
2331 binary zeroes or newlines because the value is written to log lines and
2332 to spool files. There was no check on this. Now the value is run through
2333 the string_printing() function so that such characters are converted to
2334 printable escape sequences.
2336 PH/41 $message_linecount is a new variable that contains the total number of
2337 lines in the message. Compare $body_linecount, which is the count for the
2340 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2341 bounce and delay warning messages, except in certain special cases, which
2344 (a) An SMTP error message from a remote host;
2345 (b) A message specified in a :fail: redirection;
2346 (c) A message specified in a "fail" command in a system filter;
2347 (d) A message specified in a FAIL return from the queryprogram router;
2348 (e) A message specified by the cannot_route_message router option.
2350 In these cases only, Exim does include the error details in bounce and
2351 warning messages. There are also a few cases where bland messages such
2352 as "unrouteable address" or "local delivery error" are given.
2354 PH/43 $value is now also set for the "else" part of a ${run expansion.
2356 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2357 being worked on, but at least Exim now implements the latest version to
2360 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2361 process to complete was treated as a delivery failure, a timeout while
2362 writing the message to the pipe was logged, but erroneously treated as a
2363 successful delivery. Such timeouts include transport filter timeouts. For
2364 consistency with the overall process timeout, these timeouts are now
2365 treated as errors, giving rise to delivery failures by default. However,
2366 there is now a new Boolean option for the pipe transport called
2367 timeout_defer, which, if set TRUE, converts the failures into defers for
2368 both kinds of timeout. A transport filter timeout is now identified in
2371 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2372 systems where "make" and "gmake" are different, calling "gmake" at top
2373 level broke things. I've arranged for the value of $(MAKE) to be passed
2374 from the Makefile to this script so that it can call the same version of
2378 A note about Exim versions 4.44 and 4.50
2379 ----------------------------------------
2381 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2382 changes of various kinds. As a consequence, a big documentation update was
2383 needed. This delayed the release for rather longer than seemed good, especially
2384 in the light of a couple of (minor) security issues. Therefore, the changes
2385 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2386 release. So 4.44 and 4.50 are in effect two different branches that both start
2389 I have left the 4.50 change log unchanged; it contains all the changes since
2390 4.43. The change log for 4.44 is below; many of its items are identical to
2391 those for 4.50. This seems to be the most sensible way to preserve the
2392 historical information.
2398 1. Minor wording change to the doc/README.SIEVE file.
2400 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2401 computation of the current number of files was incorrect.
2403 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2404 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2407 4. Give more explanation in the error message when the command for a transport
2408 filter fails to execute.
2410 5. There are several places where Exim runs a non-Exim command in a
2411 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2412 was being done only for the command run by the queryprogram router. It is
2413 now done for all such subprocesses. The other cases are: ${run, transport
2414 filters, and the commands run by the lmtp and pipe transports.
2416 6. Added CONFIGURE_GROUP build-time option.
2418 7. Some older OS have a limit of 256 on the maximum number of file
2419 descriptors. Exim was using setrlimit() to set 1000 as a large value
2420 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2421 systems. I've change it so that if it can't get 1000, it tries for 256.
2423 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2424 was an oversight, and furthermore, ever since the addition of extra
2425 controls (e.g. 4.43/32), the checks on when to allow different forms of
2426 "control" were broken. There should now be diagnostics for all cases when a
2427 control that does not make sense is encountered.
2429 9. Added the /retain_sender option to "control=submission".
2431 10. $recipients is now available in the predata ACL (oversight).
2433 11. Tidy the search cache before the fork to do a delivery from a message
2434 received from the command line. Otherwise the child will trigger a lookup
2435 failure and thereby defer the delivery if it tries to use (for example) a
2436 cached ldap connection that the parent has called unbind on.
2438 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2439 of $address_data from the recipient verification was clobbered by the
2440 sender verification.
2442 13. The value of address_data from a sender verification is now available in
2443 $sender_address_data in subsequent conditions in the ACL statement.
2445 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2447 15. Added a new option "connect=<time>" to callout options, to set a different
2450 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2451 was its contents. (It was OK if the option was not defined at all.)
2453 17. A "Completed" log line is now written for messages that are removed from
2454 the spool by the -Mrm option.
2456 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2457 information about exactly what failed.
2459 19. Added -dd to debug only the daemon process.
2461 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2462 handles timeouts, both on the server side and network timeouts. Renamed the
2463 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2465 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2466 It is now set to "smtps".
2468 22. $host_address is now set to the target address during the checking of
2469 ignore_target_hosts.
2471 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2472 being passed; this would have caused $sender_host_name to have been used if
2473 matching the list had actually called for a host name (not very likely,
2474 since this list is usually IP addresses). A host name is now passed as
2477 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2478 code that specifies a non-restarting handler (typically sigaction() in
2479 modern systems) in an attempt to fix a rare and obscure crash bug.
2481 25. Narrowed the window for a race in the daemon that could cause it to ignore
2482 SIGCHLD signals. This is not a major problem, because they are used only to
2483 wake it up if nothing else does.
2485 26. A malformed maildirsize file could cause Exim to calculate negative values
2486 for the mailbox size or file count. Odd effects could occur as a result.
2487 The maildirsize information is now recalculated if the size or filecount
2490 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2491 support for a long time. Removed HAVE_SYS_VFS_H.
2493 28. Installed the latest version of exipick from John Jetmore.
2495 29. In an address list, if the pattern was not a regular expression, an empty
2496 subject address (from a bounce message) matched only if the pattern was an
2497 empty string. Non-empty patterns were not even tested. This was the wrong
2498 because it is perfectly reasonable to use an empty address as part of a
2499 database query. An empty address is now tested by patterns that are
2500 lookups. However, all the other forms of pattern expect the subject to
2501 contain a local part and a domain, and therefore, for them, an empty
2502 address still always fails if the pattern is not itself empty.
2504 30. Exim went into a mad DNS loop when attempting to do a callout where the
2505 host was specified on an smtp transport, and looking it up yielded more
2506 than one IP address.
2508 31. Re-factored the code for checking spool and log partition space into a
2509 function that finds that data and another that does the check. The former
2510 is then used to implement four new variables: $spool_space, $log_space,
2511 $spool_inodes, and $log_inodes.
2513 32. The RFC2047 encoding function was originally intended for short strings
2514 such as real names; it was not keeping to the 75-character limit for
2515 encoded words that the RFC imposes. It now respects the limit, and
2516 generates multiple encoded words if necessary. To be on the safe side, I
2517 have increased the buffer size for the ${rfc2047: expansion operator from
2520 33. It is now permitted to omit both strings after an "if" condition; if the
2521 condition is true, the result is "true". As before, when the second string
2522 is omitted, a false condition yields an empty string. This makes it less
2523 cumbersome to write custom ACL and router conditions.
2525 34. Failure to deliver a bounce message always caused it to be frozen, even if
2526 there was an errors_to setting on the router. The errors_to setting is now
2529 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2530 canonical form (fully expanded) before being placed in
2531 $sender_host_address.
2533 36. The table in the code that translates DNS record types into text (T_A to
2534 "A" for instance) was missing entries for NS and CNAME. It is just possible
2535 that this could have caused confusion if both these types were looked up
2536 for the same domain, because the text type is used as part of Exim's
2537 per-process caching. But the chance of anyone hitting this buglet seems
2540 37. The dnsdb lookup has been extended in a number of ways.
2542 (1) There is a new type, "zns", which walks up the domain tree until it
2543 finds some nameserver records. It should be used with care.
2545 (2) There is a new type, "mxh", which is like "mx" except that it returns
2546 just the host names, not the priorities.
2548 (3) It is now possible to give a list of domains (or IP addresses) to be
2549 looked up. The behaviour when one of the lookups defers can be
2550 controlled by a keyword.
2552 (4) It is now possible to specify the separator character for use when
2553 multiple records are returned.
2555 38. The dnslists ACL condition has been extended: it is now possible to supply
2556 a list of IP addresses and/or domains to be looked up in a particular DNS
2559 39. Added log_selector=+queue_time_overall.
2561 40. When running the queue in the test harness, wait just a tad after forking a
2562 delivery process, to get repeatability of debugging output.
2564 41. Include certificate and key file names in error message when GnuTLS fails
2565 to set them up, because the GnuTLS error message doesn't include the name
2566 of the failing file when there is a problem reading it.
2568 42. Allow both -bf and -bF in the same test run.
2570 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2572 44. The "Exiscan patch" is now merged into the mainline Exim source.
2574 45. Sometimes the final signoff response after QUIT could fail to get
2575 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2576 < 0 before doing a fflush(). This bug looks as though it goes back to the
2577 introduction of TLS in release 3.20, but "sometimes" must have been rare
2578 because the tests only now provoked it.
2580 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2581 (this can affect the format of dates).
2583 47. exim_tidydb, when checking for the continued existence of a message for
2584 which it has found a message-specific retry record, was not finding
2585 messages that were in split spool directories. Consequently, it was
2586 deleting retry records that should have stayed in existence.
2588 48. Steve fixed some bugs in eximstats.
2590 49. The SPA authentication driver was not abandoning authentication and moving
2591 on to the next authenticator when an expansion was forced to fail,
2592 contradicting the general specification for all authenticators. Instead it
2593 was generating a temporary error. It now behaves as specified.
2595 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2596 (the order specifies the preference for clients). The order is now AES256,
2597 AES128, 3DES, ARCFOUR128.
2599 51. Small patch to Sieve code - explicitly set From: when generating an
2602 52. Exim crashed if a remote delivery caused a very long error message to be
2603 recorded - for instance if somebody sent an entire SpamAssassin report back
2604 as a large number of 550 error lines. This bug was coincidentally fixed by
2605 increasing the size of one of Exim's internal buffers (big_buffer) that
2606 happened as part of the Exiscan merge. However, to be on the safe side, I
2607 have made the code more robust (and fixed the comments that describe what
2610 53. Now that there can be additional text after "Completed" in log lines (if
2611 the queue_time_overall log selector is set), a one-byte patch to exigrep
2612 was needed to allow it to recognize "Completed" as not the last thing in
2615 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2616 patch that reportedly fixes this has been added. I am not expert enough to
2617 create a test for it. This is what the patch creator wrote:
2619 "I found a little strange behaviour of ldap code when working with
2620 Windows 2003 AD Domain, where users was placed in more than one
2621 Organization Units. When I tried to give exim partial DN, the exit code
2622 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2623 But simultaneously result of request was absolutely normal ldap result,
2624 so I produce this patch..."
2626 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2627 REFERENCE, so I have modified the code to exclude the patch when that macro
2630 55. Some experimental protocols are using DNS PTR records for new purposes. The
2631 keys for these records are domain names, not reversed IP addresses. The
2632 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2633 leaves it alone. Component reversal etc. now happens only for IP addresses.
2636 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2638 57. Double the size of the debug message buffer (to 2048) so that more of very
2639 long debug lines gets shown.
2641 58. The exicyclog utility now does better if the number of log files to keep
2642 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2644 59. Two changes related to the smtp_active_hostname option:
2646 (1) $smtp_active_hostname is now available as a variable.
2647 (2) The default for smtp_banner uses $smtp_active_hostname instead
2648 of $primary_hostname.
2650 60. The host_aton() function is supposed to be passed a string that is known
2651 to be a valid IP address. However, in the case of IPv6 addresses, it was
2652 not checking this. This is a hostage to fortune. Exim now panics and dies
2653 if the condition is not met. A case was found where this could be provoked
2654 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2655 components; fortuitously, this particular loophole had already been fixed
2656 by change 4.50/55 above.
2658 If there are any other similar loopholes, the new check in host_aton()
2659 itself should stop them being exploited. The report I received stated that
2660 data on the command line could provoke the exploit when Exim was running as
2661 exim, but did not say which command line option was involved. All I could
2662 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2663 running as the user.
2666 61. There was a buffer overflow vulnerability in the SPA authentication code
2667 (which came originally from the Samba project). I have added a test to the
2668 spa_base64_to_bits() function which I hope fixes it.
2671 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2672 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2674 63. The daemon start-up calls getloadavg() while still root for those OS that
2675 need the first call to be done as root, but it missed one case: when
2676 deliver_queue_load_max is set with deliver_drop_privilege. This is
2677 necessary for the benefit of the queue runner, because there is no re-exec
2678 when deliver_drop_privilege is set.
2680 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2681 This has been fixed.
2683 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2684 was in use, was not putting the data itself into the right store pool;
2685 consequently, it could be overwritten for a subsequent message in the same
2686 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2689 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2691 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2692 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2693 the function were treating the return as a boolean value, which happened to
2694 work because 0=false and not-0=true, but is not correct code.
2696 68. The host_aton() function was not handling scoped IPv6 addresses (those
2697 with, for example, "%eth0" on the end) correctly.
2699 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2700 negated items (that is, ~something) in unsigned ints. Some compilers
2701 apparently mutter when there is no cast.
2703 70. If an address verification called from an ACL failed, and did not produce a
2704 user-specific message (i.e. there was only a "system" message), nothing was
2705 put in $acl_verify_message. In this situation, it now puts the system
2708 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2709 session; change 4.31/43 added the unwanted input to the log line - except
2710 that it did not do this in the start of session case. It now does.
2712 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2713 This is wrong and can cause the other end to generate a synchronization
2714 error if it is another Exim or anything else that does the synchronization
2715 check. A QUIT command is no longer sent after a timeout.
2717 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2718 during host lookups.
2720 74. The defer_ok option of callout verification was not working if it was used
2721 when verifying addresses in header lines, that is, for this case:
2723 verify = header_sender/callout=defer_ok
2725 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2726 those file descriptors could be used for SMTP connections. If anything
2727 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2728 could be sent to the SMTP client, causing chaos. The daemon now opens
2729 stdin, stdout, and stderr to /dev/null when it puts itself into the
2732 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2733 log by default. The user can override this with suitable Perl magic.
2735 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2736 the log message when discard triggers, was not working for the DATA ACL or
2737 for the non-SMTP ACL.
2739 78. Error message wording change in sieve.c.
2741 79. If smtp_accept_max_per_host was set, the number of connections could be
2742 restricted to fewer than expected, because the daemon was trying to set up
2743 a new connection before checking whether the processes handling previous
2744 connections had finished. The check for completed processes is now done
2745 earlier. On busy systems, this bug wouldn't be noticed because something
2746 else would have woken the daemon, and it would have reaped the completed
2749 80. If a message was submitted locally by a user whose login name contained one
2750 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2751 It caused a spool format error. I have fixed the spool reading code. A
2752 related problem was that the "from" clause in the Received: line became
2753 illegal because of the space(s). It is now covered by ${quote_local_part.
2755 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2758 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2760 83. Patch from Sieve maintainer to fix unterminated string problem in
2761 "vacation" handling.
2763 84. Some minor changes to the Linux configuration files to help with other
2764 OS variants using glibc.
2766 85. One more patch for Sieve to update vacation handling to latest spec.
2769 ----------------------------------------------------
2770 See the note above about the 4.44 and 4.50 releases.
2771 ----------------------------------------------------
2777 1. Change 4.43/35 introduced a bug that caused file counts to be
2778 incorrectly computed when quota_filecount was set in an appendfile
2781 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2782 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2785 3. Give more explanation in the error message when the command for a transport
2786 filter fails to execute.
2788 4. There are several places where Exim runs a non-Exim command in a
2789 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2790 was being done only for the command run by the queryprogram router. It is
2791 now done for all such subprocesses. The other cases are: ${run, transport
2792 filters, and the commands run by the lmtp and pipe transports.
2794 5. Some older OS have a limit of 256 on the maximum number of file
2795 descriptors. Exim was using setrlimit() to set 1000 as a large value
2796 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2797 systems. I've change it so that if it can't get 1000, it tries for 256.
2799 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2800 was an oversight, and furthermore, ever since the addition of extra
2801 controls (e.g. 4.43/32), the checks on when to allow different forms of
2802 "control" were broken. There should now be diagnostics for all cases when a
2803 control that does not make sense is encountered.
2805 7. $recipients is now available in the predata ACL (oversight).
2807 8. Tidy the search cache before the fork to do a delivery from a message
2808 received from the command line. Otherwise the child will trigger a lookup
2809 failure and thereby defer the delivery if it tries to use (for example) a
2810 cached ldap connection that the parent has called unbind on.
2812 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2813 of $address_data from the recipient verification was clobbered by the
2814 sender verification.
2816 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2817 was its contents. (It was OK if the option was not defined at all.)
2819 11. A "Completed" log line is now written for messages that are removed from
2820 the spool by the -Mrm option.
2822 12. $host_address is now set to the target address during the checking of
2823 ignore_target_hosts.
2825 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2826 being passed; this would have caused $sender_host_name to have been used if
2827 matching the list had actually called for a host name (not very likely,
2828 since this list is usually IP addresses). A host name is now passed as
2831 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2832 code that specifies a non-restarting handler (typically sigaction() in
2833 modern systems) in an attempt to fix a rare and obscure crash bug.
2835 15. Narrowed the window for a race in the daemon that could cause it to ignore
2836 SIGCHLD signals. This is not a major problem, because they are used only to
2837 wake it up if nothing else does.
2839 16. A malformed maildirsize file could cause Exim to calculate negative values
2840 for the mailbox size or file count. Odd effects could occur as a result.
2841 The maildirsize information is now recalculated if the size or filecount
2844 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2845 support for a long time. Removed HAVE_SYS_VFS_H.
2847 18. Updated exipick to current release from John Jetmore.
2849 19. Allow an empty sender to be matched against a lookup in an address list.
2850 Previously the only cases considered were a regular expression, or an
2853 20. Exim went into a mad DNS lookup loop when doing a callout where the
2854 host was specified on the transport, if the DNS lookup yielded more than
2857 21. The RFC2047 encoding function was originally intended for short strings
2858 such as real names; it was not keeping to the 75-character limit for
2859 encoded words that the RFC imposes. It now respects the limit, and
2860 generates multiple encoded words if necessary. To be on the safe side, I
2861 have increased the buffer size for the ${rfc2047: expansion operator from
2864 22. Failure to deliver a bounce message always caused it to be frozen, even if
2865 there was an errors_to setting on the router. The errors_to setting is now
2868 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2869 canonical form (fully expanded) before being placed in
2870 $sender_host_address.
2872 24. Updated eximstats to version 1.33
2874 25. Include certificate and key file names in error message when GnuTLS fails
2875 to set them up, because the GnuTLS error message doesn't include the name
2876 of the failing file when there is a problem reading it.
2878 26. Expand error message when OpenSSL has problems setting up cert/key files.
2881 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2882 (this can affect the format of dates).
2884 28. exim_tidydb, when checking for the continued existence of a message for
2885 which it has found a message-specific retry record, was not finding
2886 messages that were in split spool directories. Consequently, it was
2887 deleting retry records that should have stayed in existence.
2889 29. eximstats updated to version 1.35
2890 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2891 1.35 - bugfix such that pie charts by volume are generated correctly
2893 30. The SPA authentication driver was not abandoning authentication and moving
2894 on to the next authenticator when an expansion was forced to fail,
2895 contradicting the general specification for all authenticators. Instead it
2896 was generating a temporary error. It now behaves as specified.
2898 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2899 (the order specifies the preference for clients). The order is now AES256,
2900 AES128, 3DES, ARCFOUR128.
2902 31. Small patch to Sieve code - explicitly set From: when generating an
2905 32. Exim crashed if a remote delivery caused a very long error message to be
2906 recorded - for instance if somebody sent an entire SpamAssassin report back
2907 as a large number of 550 error lines. This bug was coincidentally fixed by
2908 increasing the size of one of Exim's internal buffers (big_buffer) that
2909 happened as part of the Exiscan merge. However, to be on the safe side, I
2910 have made the code more robust (and fixed the comments that describe what
2913 33. Some experimental protocols are using DNS PTR records for new purposes. The
2914 keys for these records are domain names, not reversed IP addresses. The
2915 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2916 leaves it alone. Component reversal etc. now happens only for IP addresses.
2919 34. The host_aton() function is supposed to be passed a string that is known
2920 to be a valid IP address. However, in the case of IPv6 addresses, it was
2921 not checking this. This is a hostage to fortune. Exim now panics and dies
2922 if the condition is not met. A case was found where this could be provoked
2923 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2924 components; fortuitously, this particular loophole had already been fixed
2925 by change 4.50/55 or 4.44/33 above.
2927 If there are any other similar loopholes, the new check in host_aton()
2928 itself should stop them being exploited. The report I received stated that
2929 data on the command line could provoke the exploit when Exim was running as
2930 exim, but did not say which command line option was involved. All I could
2931 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2932 running as the user.
2935 35. There was a buffer overflow vulnerability in the SPA authentication code
2936 (which came originally from the Samba project). I have added a test to the
2937 spa_base64_to_bits() function which I hope fixes it.
2940 36. The daemon start-up calls getloadavg() while still root for those OS that
2941 need the first call to be done as root, but it missed one case: when
2942 deliver_queue_load_max is set with deliver_drop_privilege. This is
2943 necessary for the benefit of the queue runner, because there is no re-exec
2944 when deliver_drop_privilege is set.
2946 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2947 was in use, was not putting the data itself into the right store pool;
2948 consequently, it could be overwritten for a subsequent message in the same
2949 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2952 38. Sometimes the final signoff response after QUIT could fail to get
2953 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2954 < 0 before doing a fflush(). This bug looks as though it goes back to the
2955 introduction of TLS in release 3.20, but "sometimes" must have been rare
2956 because the tests only now provoked it.
2962 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2963 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2964 Now it returns an integer. A number of calls were still expecting a T/F
2965 return. Fortuitously, in all cases, the tests worked in OK situations,
2966 which is the norm. However, things would have gone wrong on any write
2967 failures on the smtp file descriptor. This function is used when sending
2968 messages over SMTP and also when doing verify callouts.
2970 2. When Exim is called to do synchronous delivery of a locally submitted
2971 message (the -odf or -odi options), it no longer closes stderr before doing
2974 3. Implemented the mua_wrapper option.
2976 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2978 5. Implemented the functions header_remove(), header_testname(),
2979 header_add_at_position(), and receive_remove_recipient(), and exported them
2982 6. If an ACL "warn" statement specified the addition of headers, Exim already
2983 inserted X-ACL-Warn: at the start if there was no header name. However, it
2984 was not making this test for the second and subsequent header lines if
2985 there were newlines in the string. This meant that an invalid header could
2986 be inserted if Exim was badly configured.
2988 7. Allow an ACL "warn" statement to add header lines at the start or after all
2989 the Received: headers, as well as at the end.
2991 8. Added the rcpt_4xx retry error code.
2993 9. Added postmaster_mailfrom=xxx to callout verification option.
2995 10. Added mailfrom=xxxx to the callout verification option, for verify=
2998 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2999 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3001 12. Inserted some casts to stop certain compilers warning when using pointer
3002 differences as field lengths or precisions in printf-type calls (mostly
3003 affecting debugging statements).
3005 13. Added optional readline() support for -be (dynamically loaded).
3007 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3008 same clock tick as a message's arrival, so that its received time was the
3009 same as the "first fail" time on the retry record, and that message
3010 remained on the queue past the ultimate address timeout, every queue runner
3011 would try a delivery (because it was past the ultimate address timeout) but
3012 after another failure, the ultimate address timeout, which should have then
3013 bounced the address, did not kick in. This was a "< instead of <=" error;
3014 in most cases the first failure would have been in the next clock tick
3015 after the received time, and all would be well.
3017 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3018 being recognized when the domain list was tested by the match_domain
3019 condition in an expansion string.
3021 16. Added the ${str2b64: operator.
3023 17. Exim was always calling setrlimit() to set a large limit for the number of
3024 processes, without checking whether the existing limit was already
3025 adequate. (It did check for the limit on file descriptors.) Furthermore,
3026 errors from getrlimit() and setrlimit() were being ignored. Now they are
3027 logged to the main and panic logs, but Exim does carry on, to try to do its
3028 job under whatever limits there are.
3030 18. Imported PCRE 5.0.
3032 19. Trivial typo in log message " temporarily refused connection" (the leading
3035 20. If the log selector return_path_on_delivery was set and an address was
3036 redirected to /dev/null, the delivery process crashed because it assumed
3037 that a return path would always be set for a "successful" delivery. In this
3038 case, the whole delivery is bypassed as an optimization, and therefore no
3041 21. Internal re-arrangement: the function for sending a challenge and reading
3042 a response while authentication was assuming a zero-terminated challenge
3043 string. It's now changed to take a pointer and a length, to allow for
3044 binary data in such strings.
3046 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3048 23. Exim was not respecting finduser_retries when seeking the login of the
3049 uid under which it was called; it was always trying 10 times. (The default
3050 setting of finduser_retries is zero.) Also, it was sleeping after the final
3051 failure, which is pointless.
3053 24. Implemented tls_on_connect_ports.
3055 25. Implemented acl_smtp_predata.
3057 26. If the domain in control=submission is set empty, Exim assumes that the
3058 authenticated id is a complete email address when it generates From: or
3059 Sender: header lines.
3061 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3062 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3063 chown and chgrp in /bin and hostname in /usr/bin.
3065 28. Exim was keeping the "process log" file open after each use, just as it
3066 does for the main log. This opens the possibility of it remaining open for
3067 long periods when the USR1 signal hits a daemon. Occasional processlog
3068 errors were reported, that could have been caused by this. Anyway, it seems
3069 much more sensible not to leave this file open at all, so that is what now
3072 29. The long-running daemon process does not normally write to the log once it
3073 has entered its main loop, and it closes the log before doing so. This is
3074 so that log files can straightforwardly be renamed and moved. However,
3075 there are a couple of unusual error situations where the daemon does write
3076 log entries, and I had neglected to close the log afterwards.
3078 30. The text of an SMTP error response that was received during a remote
3079 delivery was being truncated at 512 bytes. This is too short for some of
3080 the long messages that one sometimes sees. I've increased the limit to
3083 31. It is now possible to make retry rules that apply only when a message has a
3084 specific sender, in particular, an empty sender.
3086 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3087 it possible to be selective about when SMTP synchronization is enforced.
3089 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3091 32. Implemented hosts_connection_nolog.
3093 33. Added an ACL for QUIT.
3095 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3098 35. Added mailbox_size and mailbox_filecount to appendfile.
3100 36. Added control = no_multiline_responses to ACLs.
3102 37. There was a bug in the logic of the code that waits for the clock to tick
3103 in the case where the clock went backwards by a substantial amount such
3104 that the microsecond fraction of "now" was more than the microsecond
3105 fraction of "then" (but the whole seconds number was less).
3107 38. Added support for the libradius Radius client library this is found on
3108 FreeBSD (previously only the radiusclient library was supported).
3114 1. When certain lookups returned multiple values in the form name=value, the
3115 quoting of the values was not always being done properly. Specifically:
3116 (a) If the value started with a double quote, but contained no whitespace,
3118 (b) If the value contained whitespace other than a space character (i.e.
3119 tabs or newlines or carriage returns) it was not quoted.
3120 This fix has been applied to the mysql and pgsql lookups by writing a
3121 separate quoting function and calling it from the lookup code. The fix
3122 should probably also be applied to nisplus, ibase and oracle lookups, but
3123 since I cannot test any of those, I have not disturbed their existing code.
3125 2. A hit in the callout cache for a specific address caused a log line with no
3126 reason for rejecting RCPT. Now it says "Previous (cached) callout
3127 verification failure".
3129 3. There was an off-by-one bug in the queryprogram router. An over-long
3130 return line was truncated at 256 instead of 255 characters, thereby
3131 overflowing its buffer with the terminating zero. As well as fixing this, I
3132 have increased the buffer size to 1024 (and made a note to document this).
3134 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3135 when Exim is waiting for an SMTP response from a remote server, Exim
3136 restarts its select() call on the socket, thereby resetting its timeout.
3137 This is not a problem when such interrupts are rare. Somebody set up a cron
3138 job to run exiwhat every 2 minutes, which is less than the normal select()
3139 timeout (5 or 10 minutes). This meant that the select() timeout never
3140 kicked in because it was always reset. I have fixed this by comparing the
3141 time when an interrupt arrives with the time at the start of the first call
3142 to select(). If more time than the timeout has elapsed, the interrupt is
3143 treated as a timeout.
3145 5. Some internal re-factoring in preparation for the addition of Sieve
3146 extensions (by MH). In particular, the "personal" test is moved to a
3147 separate function, and given an option for scanning Cc: and Bcc: (which is
3148 not set for Exim filters).
3150 6. When Exim created an email address using the login of the caller as the
3151 local part (e.g. when creating a From: or Sender: header line), it was not
3152 quoting the local part when it contained special characters such as @.
3154 7. Installed new OpenBSD configuration files.
3156 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3157 try to make them clearer.
3159 9. Callout options, other than the timeout value, were being ignored when
3160 verifying sender addresses in header lines. For example, when using
3162 verify = header_sender/callout=no_cache
3164 the cache was (incorrectly) being used.
3166 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3167 only the Cygwin environment.
3169 11. When return_path_on_delivery was set as a log selector, if different remote
3170 addresses in the same message used different return paths and parallel
3171 remote delivery occurred, the wrong values would sometimes be logged.
3172 (Whenever a remote delivery process finished, the return path value from
3173 the most recently started remote delivery process was logged.)
3175 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3176 lines when AUTH and/or TLS are in use. This is the "received protocol"
3177 field. Exim used to use "asmtp" for authenticated SMTP, without any
3178 indication (in the protocol name) for TLS use. Now it follows the RFC and
3179 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3180 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3181 names appear in log lines as well as in Received: header lines.
3183 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3184 extensions, and comparison tests, and to fix some bugs.
3186 14. Changes to the "personal" filter test:
3188 (1) The test was buggy in that it was just doing the equivalent of
3189 "contains" tests on header lines. For example, if a user's address was
3190 anne@some.where, the "personal" test would incorrectly be true for
3192 To: susanne@some.where
3194 This test is now done by extracting each address from the header in turn,
3195 and checking the entire address. Other tests that are part of "personal"
3196 are now done using regular expressions (for example, to check local parts
3197 of addresses in From: header lines).
3199 (2) The list of non-personal local parts in From: addresses has been
3200 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3201 taken from the Sieve specification recommendations.
3203 (3) If the message contains any header line starting with "List-" it is
3204 treated as non-personal.
3206 (4) The test for "circular" in the Subject: header line has been removed
3207 because it now seems ill-conceived.
3209 15. Minor typos in src/EDITME comments corrected.
3211 16. Installed latest exipick from John Jetmore.
3213 17. If headers_add on a router specified a text string that was too long for
3214 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3215 of string_sprintf() is now avoided.
3217 18. $message_body_size was not set (it was always zero) when running the DATA
3218 ACL and the local_scan() function.
3220 19. For the "mail" command in an Exim filter, no default was being set for
3221 the once_repeat time, causing a random time value to be used if "once" was
3222 specified. (If the value happened to be <= 0, no repeat happened.) The
3223 default is now 0s, meaning "never repeat". The "vacation" command was OK
3224 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3225 (I found it when inspecting the code).
3227 20. There is now an overall timeout for performing a callout verification. It
3228 defaults to 4 times the callout timeout, which applies to individual SMTP
3229 commands during the callout. The overall timeout applies when there is more
3230 than one host that can be tried. The timeout is checked before trying the
3231 next host. This prevents very long delays if there are a large number of
3232 hosts and all are timing out (e.g. when the network connections are timing
3233 out). The value of the overall timeout can be changed by specifying an
3234 additional sub-option for "callout", called "maxwait". For example:
3236 verify = sender/callout=5s,maxwait=20s
3238 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3239 to the end before writing, but this should make it even safer).
3241 22. Exim was forgetting that it had advertised PIPELINING for the second and
3242 subsequent messages on an SMTP connection. It was also not resetting its
3243 memory on STARTTLS and an internal HELO.
3245 23. When Exim logs an SMTP synchronization error within a session, it now
3246 records whether PIPELINING has been advertised or not.
3248 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3249 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3250 rather than long int.
3252 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3254 26. Added the never_mail option to autoreply.
3260 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3261 crash if the getsockname() call failed; this can happen if a connection is
3262 closed very soon after it is established. The problem was simply in the
3263 order in which certain operations were done, causing Exim to try to write
3264 to the SMTP stream before it had set up the file descriptor. The bug has
3265 been fixed by making things happen in the correct order.
3271 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3272 before the connection was closed, thus losing the rejection response.
3274 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3275 some early Solaris releases, but causes trouble in current releases where
3276 socklen_t is defined.
3278 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3281 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3284 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3285 that releasing the top of it at the end releases what was used for sub-
3286 expansions (unless the block got too big). However, discard this block if
3287 the first thing is a variable or header, so that we can use its block when
3288 it is dynamic (useful for very large $message_headers, for example).
3290 6. Lookups now cache *every* query, not just the most recent. A new, separate
3291 store pool is used for this. It can be recovered when all lookup caches are
3292 flushed. Lookups now release memory at the end of their result strings.
3293 This has involved some general refactoring of the lookup sources.
3295 7. Some code has been added to the store_xxx() functions to reduce the amount
3296 of flapping under certain conditions.
3298 8. log_incoming_interface used to affect only the <= reception log lines. Now
3299 it causes the local interface and port to be added to several more SMTP log
3300 lines, for example "SMTP connection from", and rejection lines.
3302 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3304 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3306 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3307 could be overwritten at the end of the current message (or the start of a
3308 new message if it was set in a HELO ACL). The value is now preserved for
3309 the duration of the SMTP connection.
3311 12. If a transport had a headers_rewrite setting, and a matching header line
3312 contained an unqualified address, that address was qualified, even if it
3313 did not match any rewriting rules. The underlying bug was that the values
3314 of the flags that permit the existence of unqualified sender and recipient
3315 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3316 non-local messages, and by -bnq for local messages) were not being
3317 preserved with the message after it was received.
3319 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3320 "next input=" as part of the text comprising the host identity instead of
3321 the correct text. The code was using the same buffer for two different
3322 strings. However, depending on which order the printing function evaluated
3323 its arguments, the bug did not always show up. Under Linux, for example, my
3324 test suite worked just fine.
3326 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3327 doesn't work with some older versions of Perl. It has been changed to "my",
3328 which in any case is probably the better facility to use.
3330 15. A really picky compiler found some instances of statements for creating
3331 error messages that either had too many or two few arguments for the format
3334 16. The size of the buffer for calls to the DNS resolver has been increased
3335 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3336 for addresses that have a lot of PTR records. This alleviates a problem; it
3337 does not fully solve it.
3339 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3340 buffer now truncates the list and logs the incident, which is the same
3341 action as happens when Exim is looking up a host name and its aliases.
3342 Previously in this situation something unpredictable would happen;
3343 sometimes it was "internal error: store_reset failed".
3345 18. If a server dropped the connection unexpectedly when an Exim client was
3346 using GnuTLS and trying to read a response, the client delivery process
3347 crashed while trying to generate an error log message.
3349 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3350 string, for example:
3352 warn message = H1: something\nH2: something
3354 the text was added as a single header line from Exim's point of view
3355 though it ended up OK in the delivered message. However, searching for the
3356 second and subsequent header lines using $h_h2: did not work. This has been
3357 fixed. Similarly, if a system filter added multiple headers in this way,
3358 the routers could not see them.
3360 20. Expanded the error message when iplsearch is called with an invalid key to
3361 suggest using net-iplsearch in a host list.
3363 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3364 ACLs are no longer actually imposed (and a message to that effect is
3367 22. If a "gecos" field in a passwd entry contained escaped characters, in
3368 particular, if it contained a \" sequence, Exim got it wrong when building
3369 a From: or a Sender: header from that name. A second bug also caused
3370 incorrect handling when an unquoted " was present following a character
3371 that needed quoting.
3373 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3374 was not being matched caselessly.
3376 24. Arranged for all hyphens in the exim.8 source to be escaped with
3379 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3380 properly. Recipient callout cache records were still being keyed to include
3381 the sender, even when use_sender was set false. This led to far more
3382 callouts that were necessary. The sender is no longer included in the key
3383 when use_sender is false.
3385 26. Added "control = submission" modifier to ACLs.
3387 27. Added the ${base62d: operator to decode base 62 numbers.
3389 28. dnsdb lookups can now access SRV records.
3391 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3392 the configuration file.
3394 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3395 (-v) mode. This makes the output for a verbose queue run more intelligible.
3397 31. Added a use_postmaster feature to recipient callouts.
3399 32. Added the $body_zerocount variable, containing the number of binary zero
3400 bytes in the message body.
3402 33. The time of last modification of the "new" subdirectory is now used as the
3403 "mailbox time last read" when there is a quota error for a maildir
3406 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3408 35. Added +ignore_unknown as a special item in host lists.
3410 36. Code for decoding IPv6 addresses in host lists is now included, even if
3411 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3412 address was recognized as an IP address, but was then not correctly decoded
3413 into binary, causing unexpected and incorrect effects when compared with
3420 1. Very minor rewording of debugging text in manualroute to say "list of
3421 hosts" instead of "hostlist".
3423 2. If verify=header_syntax was set, and a header line with an unqualified
3424 address (no domain) and a large number of spaces between the end of the
3425 name and the colon was received, the reception process suffered a buffer
3426 overflow, and (when I tested it) crashed. This was caused by some obsolete
3427 code that should have been removed. The fix is to remove it!
3429 3. When running in the test harness, delay a bit after writing a bounce
3430 message to get a bit more predictability in the log output.
3432 4. Added a call to search_tidyup() just before forking a reception process. In
3433 theory, someone could use a lookup in the expansion of smtp_accept_max_
3434 per_host which, without the tidyup, could leave open a database connection.
3436 5. Added the variables $recipient_data and $sender_data which get set from a
3437 lookup success in an ACL "recipients" or "senders" condition, or a router
3438 "senders" option, similar to $domain_data and $local_part_data.
3440 6. Moved the writing of debug_print from before to after the "senders" test
3443 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3444 problems for message scanning, either using a data ACL, or using
3445 local_scan() because the Received: header was not generated till after they
3446 were called (in order to set the time as the time of reception completion).
3447 I have revised the way this works. The header is now generated after the
3448 body is received, but before the ACL or local_scan() are called. After they
3449 are run, the timestamp in the header is updated.
3455 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3456 before starting a queue runner without re-exec. This happened only when
3457 deliver_drop_privilege was set or when the Exim user was set to root. The
3458 effect of the bug was that timeouts during subsequent deliveries caused
3459 crashes instead of being properly handled. The handler is now left at its
3460 default (and expected) setting.
3462 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3463 message, again when deliver_drop_privilege is set or Exim is run as root.
3464 The bug described in (1) was not present in this case, but the tidying up
3465 of the other signals was missing. I have made the two cases consistent.
3467 3. The ignore_target_hosts setting on a manualroute router was being ignored
3468 for hosts that were looked up using the /MX notation.
3470 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3473 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3474 operated on the sender address. After changing the $sender_address to <>
3475 for the sender address verify, Exim was re-instated it as the original
3476 (before rewriting) address, but remembering that it had rewritten it, so it
3477 wasn't rewriting it again. This bug also had the effect of breaking the
3478 sender address verification caching when the sender address was rewritten.
3480 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3481 This has been changed so that if the ip literal address matches
3482 ignore_target_hosts, the router declines.
3484 7. Added expansion conditions match_domain, match_address, and match_local_
3485 part (NOT match_host).
3487 8. The placeholder for the Received: header didn't have a length field set.
3489 9. Added code to Exim itself and to exim_lock to test for a specific race
3490 condition that could lead to file corruption when using MBX delivery. The
3491 issue is with the lockfile that is created in /tmp. If this file is removed
3492 after a process has opened it but before that process has acquired a lock,
3493 there is the potential for a second process to recreate the file and also
3494 acquire a lock. This could lead to two Exim processes writing to the file
3495 at the same time. The added code performs the same test as UW imapd; it
3496 checks after acquiring the lock that its file descriptor still refers to
3497 the same named file.
3499 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3500 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3503 11. Added the smtp_active_hostname option. If used, this will typically be made
3504 to depend on the incoming interface address. Because $interface_address is
3505 not set up until the daemon has forked a reception process, error responses
3506 that can happen earlier (such as "too many connections") no longer contain
3509 12. If an expansion in a condition on a "warn" statement fails because a lookup
3510 defers, the "warn" statement is abandoned, and the next ACL statement is
3511 processed. Previously this caused the whole ACL to be aborted.
3513 13. Added the iplsearch lookup type.
3515 14. Added ident_timeout as a log selector.
3517 15. Added tls_certificate_verified as a log selector.
3519 16. Added a global option tls_require_ciphers (compare the smtp transport
3520 option of the same name). This controls incoming TLS connections.
3522 17. I finally figured out how to make tls_require_ciphers do a similar thing
3523 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3524 before starting the TLS session.
3526 18. Tabs are now shown as \t in -bP output.
3528 19. If the log selector return_path_on_delivery was set, Exim crashed when
3529 bouncing a message because it had too many Received: header lines.
3531 20. If two routers both had headers_remove settings, and the first one included
3532 a superfluous trailing colon, the final name in the first list and the
3533 first name in the second list were incorrectly joined into one item (with a
3534 colon in the middle).
3540 1. Added -C and -D options to the exinext utility, mainly to make it easier
3541 to include in the automated testing, but these could be helpful when
3542 multiple configurations are in use.
3544 2. The exinext utility was not formatting the output nicely when there was
3545 an alternate port involved in the retry record key, nor when there was a
3546 message id as well (for retries that were specific to a specific message
3547 and a specific host). It was also confused by IPv6 addresses, because of
3548 the additional colons they contain. I have fixed the IPv4 problem, and
3549 patched it up to do a reasonable job for IPv6.
3551 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3552 delivery, the log line now contains "pipelined" if PIPELINING was used.
3554 4. An SMTP transport process used to panic and die if the bind() call to set
3555 an explicit outgoing interface failed. This has been changed; it is now
3556 treated in the same way as a connect() failure.
3558 5. A reference to $sender_host_name in the part of a conditional expansion
3559 that was being skipped was still causing a DNS lookup. This no longer
3562 6. The def: expansion condition was not recognizing references to header lines
3563 that used bh_ and bheader_.
3565 7. Added the _cache feature to named lists.
3567 8. The code for checking quota_filecount in the appendfile transport was
3568 allowing one more file than it should have been.
3570 9. For compatibility with Sendmail, the command line option
3578 and sets the incoming protocol and host name (for trusted callers). The
3579 host name and its colon can be omitted when only the protocol is to be set.
3580 Note the Exim already has two private options, -pd and -ps, that refer to
3581 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3582 "s", but I don't think that's a major issue.
3584 10. A number of refactoring changes to the code, none of which should affect
3587 (a) The number of logging options was getting close to filling up the
3588 32-bit word that was used as a bit map. I have split them into two classes:
3589 those that are passed in the argument to log_write(), and those that are
3590 only ever tested independently outside of that function. These are now in
3591 separate 32-bit words, so there is plenty of room for expansion again.
3592 There is no change in the user interface or the logging behaviour.
3594 (b) When building, for example, log lines, the code previously used a
3595 macro that called string_cat() twice, in order to add two strings. This is
3596 not really sufficiently general. Furthermore, there was one instance where
3597 it was actually wrong because one of the argument was used twice, and in
3598 one call a function was used. (As it happened, calling the function twice
3599 did not affect the overall behaviour.) The macro has been replaced by a
3600 function that can join an arbitrary number of extra strings onto a growing
3603 (c) The code for expansion conditions now uses a table and a binary chop
3604 instead of a serial search (which was left over from when there were very
3605 few conditions). Also, it now recognizes conditions like "pam" even when
3606 the relevant support is not compiled in: a suitably worded error message is
3607 given if an attempt is made to use such a condition.
3609 11. Added ${time_interval:xxxxx}.
3611 12. A bug was causing one of the ddress fields not to be passed back correctly
3612 from remote delivery subprocesses. The field in question was not being
3613 subsequently used, so this caused to problems in practice.
3615 13. Added new log selectors queue_time and deliver_time.
3617 14. Might have fixed a bug in maildirsizefile handling that threw up
3618 "unexpected character" debug warnings, and recalculated the data
3619 unnecessarily. In any case, I expanded the warning message to give more
3622 15. Added the message "Restricted characters in address" to the statements in
3623 the default ACL that block characters like @ and % in local parts.
3625 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3626 Three changes have been made:
3628 (a) There was a serious bug; a negative response to MAIL caused the whole
3629 recipient domain to be cached as invalid, thereby blocking all messages
3630 to all local parts at the same domain, from all senders. This bug has
3631 been fixed. The domain is no longer cached after a negative response to
3632 MAIL if the sender used is not empty.
3634 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3637 (c) A new callout option, "use_sender" has been added for people who want
3638 the modified behaviour.
3644 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3647 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3648 indeed breaks things for older releases.
3650 3. Added additional logging to the case where there is a problem reading data
3651 from a filter that is running in a subprocess using a pipe, in order to
3652 try to track down a specific problem.
3654 4. Testing facility fudge: when running in the test harness and attempting
3655 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3656 getting "No route to host". Convert this to a timeout.
3658 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3661 6. Some OS don't have socklen_t but use size_t instead. This affects the
3662 fifth argument of getsockopt() amongst other things. This is now
3663 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3664 can be set for individual OS. I have set it for SunOS5, OSF1, and
3665 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3666 some earlier ones do not.
3668 7. Change 4.30/15 was not doing the test caselessly.
3670 8. The standard form for an IPv6 address literal was being rejected by address
3671 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3672 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3673 this, as well as the form without the "IPv6" on the front (but only when
3674 address literals are enabled, of course).
3676 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3678 10. Exim crashed if a message with an empty sender address specified by -f
3679 encountered a router with an errors_to setting. This could be provoked only
3680 by a command such as
3684 where an empty string was supplied; "<>" did not hit this bug.
3686 11. Installed PCRE release 4.5.
3688 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3689 remained set. It is now erased.
3691 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3692 times from message ids (which are base 36 rather than the normal 62).
3694 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3695 were being counted as actual protocol errors, and logged if the log
3696 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3697 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3698 and DATA following a set of rejected RCPTs do not count as protocol errors.
3699 In other words, Exim assumes they were pipelined, though this may not
3700 actually be the case. Of course, in all cases the client gets an
3701 appropriate error code.
3703 15. If a lookup fails in an ACL condition, a message about the failure may
3704 be available; it is used if testing the ACL cannot continue, because most
3705 such messages specify what the cause of the deferral is. However, some
3706 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3707 that caused an old message to be retained and used if a later statement
3708 caused a defer, replacing the real cause of the deferral.
3710 16. If an IP address had so many PTR records that the DNS lookup buffer
3711 was not large enough to hold them, Exim could crash while trying to process
3712 the truncated data. It now detects and logs this case.
3714 17. Further to 4.21/58, another change has been made: if (and only if) the
3715 first line of a message (the first header line) ends with CRLF, a bare LF
3716 in a subsequent header line has a space inserted after it, so as not to
3717 terminate the header.
3719 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3720 unnecessarily, used atoi() instead of strtol(), and didn't check the
3721 termination when getting file sizes from file names by regex.
3723 19. Completely re-implemented the support for maildirsize files, in the light
3724 of a number of problems with the previous contributed implementation
3725 (4.30/29). In particular:
3727 . If the quota is zero, the maildirsize file is maintained, but no quota is
3730 . If the maildir directory does not exist, it is created before any attempt
3731 to write a maildirsize file.
3733 . The quota value in the file is just a cache; if the quota is changed in
3734 the transport, the new value overrides.
3736 . A regular expression is available for excluding directories from the
3739 20. The autoreply transport checks the characters in options that define the
3740 message's headers; it allows continued headers, but it was checking with
3741 isspace() after an embedded newline instead of explicitly looking for a
3744 21. If all the "regular" hosts to which an address was routed had passed their
3745 expiry times, and had not reached their retry times, the address was
3746 bounced, even if fallback hosts were defined. Now Exim should go on to try
3749 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3750 equivalent code in the SMTP transport. Some hosts send humungous responses
3751 to HELO/EHLO, more than 1024 it seems.
3753 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3754 gives compiler warnings in some environments. I've now done it "properly",
3757 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3758 (because of problems with the built-in one) was declared to return uschar *
3759 instead of char *, causing compiler failure.
3761 25. Fixed a file descriptor leak when processing alias/forward files.
3763 26. Fixed a minor format string issue in dbfn.c.
3765 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3767 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3768 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3771 29. When a sender address is verified, it is cached, to save repeating the test
3772 when there is more than one recipient in a message. However, when the
3773 verification involves a callout, it is possible for different callout
3774 options to be set for different recipients. It is too complicated to keep
3775 track of this in the cache, so now Exim always runs a verification when a
3776 callout is required, relying on the callout cache for the optimization.
3777 The overhead is duplication of the address routing, but this should not be
3780 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3781 to be verified with callout=postmaster, and the main callout worked but the
3782 postmaster check failed, the verification correctly failed. However, if a
3783 subsequent RCPT command asked for sender verification *without* the
3784 postmaster check, incorrect caching caused this verification also to fail,
3787 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3788 it was not caching the DNS options (qualify_single, search_parents) that
3789 were used when the lookup failed. A subsequent lookup with different
3790 options therefore always gave the same answer, though there were cases
3791 where it should not have. (Example: a "domains = !$mx_any" option on a
3792 dnslookup router: the "domains" option is always processed without any
3793 widening, but the router might have qualify_single set.) Now Exim uses the
3794 cached value only when the same options are set.
3796 32. Added John Jetmore's "exipick" utility to the distribution.
3798 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3799 than a timeout (e.g. a certificate is required, and is not provided), an
3800 Exim server now closes the connection immediately. Previously it waited for
3801 the client to close - but if the client is SSL, it seems that they each
3802 wait for each other, leading to a delay before one of them times out.
3804 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3805 maintained 0.8.x compatibility because I don't think many are using it, and
3806 it is clearly obsolete.
3808 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3811 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3812 client certificate was expired. A simple patch fixes this, though I don't
3813 understand the full logic of why the verify callback is called multiple
3816 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3817 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3818 which causes problems with some clients (such as the Certicom SSL Plus
3819 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3820 disables the coutermeasure allowing Eudora to connect."
3822 38. Exim was not checking that a write() to a log file succeeded. This could
3823 lead to Bad Things if a log got too big, in particular if it hit a file
3824 size limit. Exim now panics and dies if it cannot write to a log file, just
3825 as it does if it cannot open a log file.
3827 39. Modified OS/Makefile-Linux so that it now contains
3829 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3831 The two -D definitions ensure that Exim is compiled with large file
3832 support, which makes it possible to handle log files that are bigger than
3835 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3836 instance) a domain was checked against a named list that involved a lookup,
3837 causing $domain_data to be set, then another domain was checked against the
3838 same list, then the first domain was re-checked, the value of $domain_data
3839 after the final check could be wrong. In particular, if the second check
3840 failed, it could be set empty. This bug probably also applied to
3843 41. The strip_trailing_dot option was not being applied to the address given
3844 with the -f command-line option.
3846 42. The code for reading a message's header from the spool was incrementing
3847 $received_count, but never initializing it. This meant that the value was
3848 incorrect (doubled) while delivering a message in the same process in which
3849 it was received. In the most common configuration of Exim, this never
3850 happens - a fresh exec is done - but it can happen when
3851 deliver_drop_privilege is set.
3853 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3854 it now includes up to 150 characters of the unexpected data in the log
3857 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3858 and building data strings. The size of both of these buffers was 10 000
3859 bytes - far larger than anybody would *ever* want, thought I. Needless to
3860 say, somebody hit the limit. I have increased the maximum line length to
3861 20 000 and the maximum data length of concatenated lines to 100 000. I have
3862 also fixed two bugs, because there was no checking on these buffers. Tsk,
3863 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3864 buffer is too small.
3866 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3867 lsearch lookups. Now it does.
3869 46. When parsing a route_list item in a manualroute router, a fixed-length
3870 buffer was used for the list of hosts. I made this 1024 bytes long,
3871 thinking that nobody would ever have a list of hosts that long. Wrong.
3872 Somebody had a whole pile of complicated expansion conditions, and the
3873 string was silently truncated, leading to an expansion error. It turns out
3874 that it is easier to change to an unlimited length (owing to other changes
3875 that have happened since this code was originally written) than to build
3876 structure for giving a limitation error. The length of the item that
3877 expands into the list of hosts is now unlimited.
3879 47. The lsearch lookup could not handle data where the length of text line was
3880 more than 4095 characters. Such lines were truncated, leading to shortened
3881 data being returned. It should now handle lines of any length.
3883 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3884 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3887 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3888 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3889 no longer happens in comment lines. A list of replacements is now placed
3890 at the head of all of the source files, except those whose only change is
3891 to replace PERL_COMMAND in the very first #! line.
3893 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3894 messages on the queue, with a bottom-up merge sort, using code contributed
3895 by Michael Haardt. This should make operations like -bp somewhat faster on
3896 large queues. It won't affect queue runners, except when queue_run_in_order
3899 51. Installed eximstats 1.31 in the distribution.
3901 52. Added support for SRV lookups to the dnslookup router.
3903 53. If an ACL referred to $message_body or $message_body_end, the value was not
3904 reset for any messages that followed in the same SMTP session.
3906 54. The store-handling optimization for building very long strings was not
3907 differentiating between the different store pools. I don't think this
3908 actually made any difference in practice, but I've tidied it.
3910 55. While running the routers to verify a sender address, $sender_address
3911 was still set to the sender address. This is wrong, because when routing to
3912 send a bounce to the sender, it would be empty. Therefore, I have changed
3913 it so that, while verifying a sender address, $sender_address is set to <>.
3914 (There is no change to what happens when verifying a recipient address.)
3916 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3917 target A or AAAA records (if not already returned) without resetting the
3918 qualify_single or search_parents options of the DNS resolver. These are
3919 inappropriate in this case because the targets of MX and SRV records must
3920 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3921 target that, when qualified, matched something in the local domain. These
3922 two options are now turned off when doing these lookups.
3924 57. It seems that at least some releases of Reiserfs (which does not have the
3925 concept of a fixed number of inodes) returns zero and not -1 for the
3926 number of available inodes. This interacted badly with check_spool_inodes,
3927 which assumed that -1 was the "no such thing" setting. What I have done is
3928 to check that the total number of inodes is greater than zero before doing
3929 the test of how many are available.
3931 58. When a "warn" ACL statement has a log_message modifier, the message is
3932 remembered, and not repeated. This is to avoid a lot of repetition when a
3933 message has many recipients that cause the same warning to be written.
3934 Howewer, Exim was preserving the list of already written lines for an
3935 entire SMTP session, which doesn't seem right. The memory is now reset if a
3936 new message is started.
3938 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3939 debugging output unless log_rewrite was also set.
3941 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3942 of (int)(handle) when we know that handle contains (void *)(-1).
3944 61. The Exim daemon panic-logs an error return when it closes the incoming
3945 connection. However "connection reset by peer" seems to be common, and
3946 isn't really an error worthy of noting specially, so that particular error
3949 62. When Exim is trying to find all the local interfaces, it used to panic and
3950 die if the ioctl to get the interface flags failed. However, it seems that
3951 on at least one OS (Solaris 9) it is possible to have an interface that is
3952 included in the list of interfaces, but for which you get a failure error
3953 for this call. This happens when the interface is not "plumbed" into a
3954 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3955 failure of the "get flags" call assumes that the interface is down.
3957 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3958 makes life easier for people who are doing arithmetic on fields extracted
3959 from dates, where you often get leading zeros that should not be
3960 interpreted as octal.
3962 64. Added qualify_domain to the redirect router, to override the global
3965 65. If a pathologically long header line contained very many addresses (the
3966 report of this problem mentioned 10 000) and each of them was rewritten,
3967 Exim could use up a very large amount of memory. (It kept on making new
3968 copies of the header line as it rewrote, and never released the old ones.)
3969 At the expense of a bit more processing, the header rewriting function has
3970 been changed so that it no longer eats memory in this way.
3972 66. The generation of the Received: header has been moved from the time that a
3973 message starts to be received, to the time that it finishes. The timestamp
3974 in the Received: header should now be very close to that of the <= log
3975 line. There are two side-effects of this change:
3977 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3978 logged header lines no longer include the local Received: line, because
3979 it has not yet been created. The same applies to a copy of the message
3980 that is returned to a non-SMTP sender when a message is rejected.
3982 (b) When a filter file is tested using -bf, no additional Received: header
3983 is added to the test message. After some thought, I decided that this
3986 This change does not affect the value of $received_for. It is still set
3987 after address rewriting, but before local_scan() is called.
3989 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3991 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3992 gave an unhelpful panic error message, and a defer error. I have managed to
3993 change this behaviour so that it now rejects any supplied certificate,
3994 which seems right, as the list of acceptable certificates is empty.
3996 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3997 gave an unhelpful defer error. I have not managed to make this reject any
3998 supplied certificates, but the error message it gives is "no certificate
3999 supplied", which is not helpful.
4001 70. exigrep's output now also includes lines that are not associated with any
4002 message, but which match the given pattern. Implemented by a patch from
4003 Martin Sluka, which also tidied up the Perl a bit.
4005 71. Recipient callout verification, like sender verification, was using <> in
4006 the MAIL FROM command. This isn't really the right thing, since the actual
4007 sender may affect whether the remote host accepts the recipient or not. I
4008 have changed it to use the actual sender in the callout; this means that
4009 the cache record is now keyed on a recipient/sender pair, not just the
4010 recipient address. There doesn't seem to be a real danger of callout loops,
4011 since a callout by the remote host to check the sender would use <>.
4012 [SEE ABOVE: changed after hitting problems.]
4014 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4015 temporary errors. However, in the case of such a code being given after
4016 the end of a data transmission (i.e. after ".") Exim was failing to write
4017 a retry record for the message. (Yes, there was some broken host that was
4018 actually sending 8xx at this point.)
4020 73. An unknown lookup type in a host list could cause Exim to panic-die when
4021 the list was checked. (An example that provoked this was putting <; in the
4022 middle of a list instead of at the start.) If this happened during a DATA
4023 ACL check, a -D file could be left lying around. This kind of configuration
4024 error no longer causes Exim to die; instead it causes a defer errror. The
4025 incident is still logged to the main and panic logs.
4027 74. Buglet left over from Exim 3 conversion. The message "too many messages
4028 in one connection" was written to the rejectlog but not the mainlog, except
4029 when address rewriting (yes!) was being logged.
4031 75. Added write_rejectlog option.
4033 76. When a system filter was run not as root (that is, when system_filter_user
4034 was set), the values of the $n variables were not being returned to the
4035 main process; thus, they were not subsequently available in the $sn
4038 77. Added +return_path_on_delivery log selector.
4040 78. A connection timeout was being treated differently from recipients deferred
4041 when testing hosts_max_try with a message that was older than the host's
4042 retry timeout. (The host should not be counted, thus allowing all hosts to
4043 be tried at least once before bouncing.) This may have been the cause of an
4044 occasionally reported bug whereby a message would remain on the queue
4045 longer than the retry timeout, but would be bounced if a delivery was
4046 forced. I say "may" because I never totally pinned down the problem;
4047 setting up timeout/retry tests is difficult. See also the next item.
4049 79. The ultimate address timeout was not being applied to errors that involved
4050 a combination of host plus message (for example, a timeout on a MAIL
4051 command). When an address resolved to a number of possible hosts, and they
4052 were not all tried for each delivery (e.g. because of hosts_max_try), a
4053 message could remain on the queue longer than the retry timeout.
4055 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4058 81. Fixed an obscure SMTP outgoing bug which required at least the following
4059 conditions: (a) there was another message waiting for the same server;
4060 (b) the server returned 5xx to all RCPT commands in the first message so
4061 that the message was not completed; (c) the server dropped the connection
4062 or gave a negative response to the RSET that Exim sends to abort the
4063 transaction. The observed case was a dropped connection after DATA that had
4064 been sent in pipelining mode. That is, the server had advertised PIPELINING
4065 but was not implementing it correctly. The effect of the bug was incorrect
4066 behaviour, such as trying another host, and this could lead to a crash.
4072 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4073 and daemon.c were passed as pointers to ints; they should have been
4074 pointers to socklen_t variables (which are typically unsigned ints).
4076 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4079 3. Fixed a really odd bug that affected only the testing scheme; patching a
4080 certain fixed string in the binary changed the value of another string that
4081 happened to be identical to the end of the original first string.
4083 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4084 name", it returns that address as the IP address. On some operating
4085 systems (e.g. Solaris), it also passes back the IP address string as the
4086 "host name". However, on others (e.g. Linux), it passes back an empty
4087 string. Exim wasn't checking for this, and was changing the host name to an
4088 empty string, assuming it had been canonicized.
4090 5. Although rare, it is permitted to have more than one PTR record for a given
4091 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4092 all the names associated with an address, because they do in Solaris.
4093 However, it seems that they do not in Linux for data that comes from the
4094 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4095 I found this out when I moved to a new Linux workstation and tried to run
4096 the Exim test suite.
4098 To get round this problem I have changed the code so that it now does its
4099 own call to the DNS to look up PTR records when searching for a host name.
4100 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4101 addresses that are only in /etc/hosts are still found.
4103 This behaviour is, however, controlled by an option called host_lookup_
4104 order, which defaults to "bydns:byaddr". If people want to use the other
4105 order, or indeed, just use one or the other means of lookup, they can
4106 specify it in this variable.
4108 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4109 some operating systems, this comes back from gethostbyaddr() as an empty
4110 string, and this is what Exim used to test for. However, it seems that in
4111 other systems, "." is yielded. Exim now tests for this case too.
4113 7. The values of check_spool_space and check_log_space are now held internally
4114 as a number of kilobytes instead of an absolute number of bytes. If a
4115 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4116 kilobyte. This means that much larger values can be stored.
4118 8. Exim monitor: an attempt to get the action menu when not actually pointing
4119 at a message produces an empty menu entitled "No message selected". This
4120 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4121 no entries in it ("Shell widget menu has zero width and/or height"). So I
4122 have added a single, blank menu entry in this case.
4124 9. Added ${quote_local_part.
4126 10. MIME decoding is now applied to the contents of Subject: header lines when
4129 11. Now that a reference to $sender_host_address automatically causes a reverse
4130 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4131 host lookup before query-style lookups in lists that might use this
4132 variable. This has therefore been abolished, and the "net-" prefix is no
4133 longer necessary for query-style lookups.
4135 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4136 have been a typo for LFLAGS, so it has been changed.
4138 13. The install script calls Exim with "-C /dev/null" in order to find the
4139 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4140 to be output. Howeve, since Exim outputs its version number before the
4141 error, it didn't break the script. It just looked ugly. I fixed this by
4142 always allowing "-C /dev/null" if the caller is root.
4144 14. Ignore overlarge ACL variable number when reading spool file - insurance
4145 against a later release with more variables having written the file.
4147 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4148 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4149 this, as well as the form without the "IPv6" on the front.
4151 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4152 OS/Makefile-Darwin file.
4154 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4155 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4157 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4158 variable. However, some OS use ints for inodes. I've added cast to long int
4159 to get rid of the compiler warning.
4161 19. I had forgotten to lock out "/../" in configuration file names when
4162 ALT_CONFIG_PREFIX was set.
4164 20. Routers used for verification do not need to specify transports. However,
4165 if such a router generated a host list, and callout was configured, Exim
4166 crashed, because it could not find a port number from the (non-existent)
4167 transport. It now assumes port 25 in this circumstance.
4169 21. Added the -t option to exigrep.
4171 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4172 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4173 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4174 them is set, LOOKUP_LSEARCH is forced.
4176 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4178 24. Added sender and host information to the "rejected by local_scan()" log
4179 line; previously there was no indication of these.
4181 25. Added .include_if_exists.
4183 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4184 Linux. It turns out that not all file systems support this. Apparently some
4185 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4186 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4187 not supported on the file descriptor, is now ignored when Exim is trying to
4188 sync a directory. This applies only to Linux.
4190 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4192 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4193 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4196 29. Added support for maildirsize files from supplied patch (modified a bit).
4198 30. The use of :fail: followed by an empty string could lead Exim to respond to
4199 sender verification failures with (e.g.):
4201 550 Verification failed for <xxx>
4202 550 Sender verify failed
4204 where the first response line was missing the '-' that indicates it is not
4205 the final line of the response.
4207 31. The loop for finding the name of the user that called Exim had a hardwired
4208 limit of 10; it now uses the value of finduser_retries, which is used for
4209 all other user lookups.
4211 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4214 33. Exim was neglecting to zero errno before one call of strtol() when
4215 expanding a string and expecting an integer value. On some systems this
4216 resulted in spurious "integer overflow" errors. Also, it was casting the
4217 result into an int without checking.
4219 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4220 did not work. The code looks as if it has *never* worked, though it appears
4221 to have been documented since at least releast 1.62. I have made it work.
4223 35. The "timeout_DNS" error in retry rules, also documented since at least
4224 1.62, also never worked. As it isn't clear exactly what this means, and
4225 clearly it isn't a major issue, I have abolished the feature by treating it
4226 as "timeout", and writing a warning to the main and panic logs.
4228 36. The display of retry rules for -brt wasn't always showing the error code
4231 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4232 timeout_connect_A, timeout_connect_MX.
4234 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4235 to the empty sender.
4237 39. The daemon was not analysing the content of -oX till after it had closed
4238 stderr and disconnected from the controlling terminal. This meant that any
4239 syntax errors were only noted on the panic log, and the return code from
4240 the command was 0. By re-arranging the code a little, I've made the
4241 decoding happen first, so such errors now appear on stderr, and the return
4242 code is 1. However, the actual setting up of the sockets still happens in
4243 the disconnected process, so errors there are still only recorded on the
4246 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4247 connections (as happens on some IP stacks) was logged at start up time as
4248 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4249 it from "IPv6 and IPv4", which means that two separate sockets are being
4252 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4253 says whether AF_INET or AF_INET6 was passed as an argument.
4255 42. Exiwhat output was messed up when time zones were included in log
4258 43. Exiwhat now gives more information about the daemon's listening ports,
4259 and whether -tls-on-connect was used.
4261 44. The "port" option of the smtp transport is now expanded.
4263 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4264 silently ignored. Now an error message is written to the main and panic
4267 46. There's a new ACL modifier called "logwrite" which writes to a log file
4268 as soon as it is encountered.
4270 47. Added $local_user_uid and $local_user_gid at routing time.
4272 48. Exim crashed when trying to verify a sender address that was being
4275 49. Exim was recognizing only a space character after ".include". It now also
4276 recognizes a tab character.
4278 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4279 extracting the relevant information from the specification. The man page no
4280 longer contains scrambled data for the -d option, and I've added a section
4281 at the front about calling Exim under different names.
4283 51. Added "extra_headers" argument to the "mail" command in filter files.
4285 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4288 53. Installed eximstats 1.29.
4290 54. Added transport_filter_timeout as a generic transport option.
4292 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4293 Cc: header lines. This was required by RFC 822, but it not required by RFC
4296 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4297 incoming messages that did not have them. Now it does so only if the
4298 message originates locally, that is, if there is no associated remote host
4299 address. When Resent- header lines are present, this applies to the Resent-
4300 lines rather than the non-Resent- lines.
4302 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4303 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4305 58. Messages for configuration errors now include the name of the main
4306 configuration file - useful now that there may be more than one file in a
4307 list (.included file names were always shown).
4309 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4310 for those rare installations that do not start the daemon as root or run it
4311 setuid root. I've cut out the call to initgroups() if the daemon is not
4314 60. The Exim user and group can now be bound into the binary as text strings
4315 that are looked up at the start of Exim's processing.
4317 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4319 62. Added $mailstore_basename variable.
4321 63. Installed patch to sieve.c from Michael Haardt.
4323 64. When Exim failed to open the panic log after failing to open the main log,
4324 the original message it was trying to log was written to stderr and debug
4325 output, but if they were not available (the usual case in production), it
4326 was lost. Now it is written to syslog before the two lines that record the
4327 failures to open the logs.
4329 65. Users' Exim filters run in subprocesses under the user's uid. It is
4330 possible for a "deliver" command or an alias in a "personal" command to
4331 provoke an address rewrite. If logging of address rewriting is configured,
4332 this fails because the process is not running as root or exim. There may be
4333 a better way of dealing with this, but for the moment (because 4.30 needs
4334 to be released), I have disabled address rewrite logging when running a
4335 filter in a non-root, non-exim process.
4341 1. The buildconfig auxiliary program wasn't quoting the value set for
4342 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4343 not defined. This bug was masked in 4.22 by the effect that was fixed in
4346 2. Some messages that were rejected after a message id was allocated were
4347 shown as "incomplete" by exigrep. It no longer does this for messages that
4348 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4350 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4351 have allow_domain_literals set, the ID did not get logged in the <= line.
4352 Domain literals are now always recognized in Message-ID: header lines.
4354 4. The first argument for a ${extract expansion item is the key name or field
4355 number. Leading and trailing spaces in this item were not being ignored,
4356 causing some misleading effects.
4358 5. When deliver_drop_privilege was set, single queue runner processes started
4359 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4360 same command in the process it spins off) were not dropping privilege.
4362 6. When the daemon running as "exim" started a queue runner, it always
4363 re-executed Exim in the spun-off process. This is a waste of effort when
4364 deliver_drop_privilege is set. The new process now just calls the
4365 queue-runner function directly.
4371 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4374 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4375 ignored. Though the use of -oP was forcing the writing of a pid file, it
4376 was always written to the default place.
4378 3. If the message "no IP address found for host xxxx" is generated during
4379 incoming verification, it is now followed by identification of the incoming
4380 connection (so you can more easily find what provoked it).
4382 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4384 5. Added some features to "harden" Exim a bit more against certain attacks:
4386 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4387 be put in Local/Makefile. This is like the never_users runtime option,
4388 but it cannot be overridden. The default setting is "root".
4390 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4391 prefix string with which any file named in a -C command line option
4394 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4395 is retained for -C and -D only if the caller of Exim is root. Without
4396 it, the exim user may also use -C and -D and retain privilege.
4398 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4399 command line option is disabled.
4401 6. Macro names set by the -D option must start with an upper case letter, just
4402 like macro names defined in the configuration file.
4404 7. Added "dereference=" facility to LDAP.
4406 8. Two instances of the typo "uknown" in the source files are fixed.
4408 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4409 the Configure-Makefile script screwed up while processing it.
4411 10. Incorporated PCRE 4.4.
4413 11. The SMTP synchronization check was not operating right at the start of an
4414 SMTP session. For example, it could not catch a HELO sent before the client
4415 waited for the greeting. There is now a check for outstanding input at the
4416 point when the greeting is written. Because of the duplex, asynchronous
4417 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4418 way, but not yet received, when the check is performed.
4420 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4421 on TCP/IP sockets, because this apparently causes some broken clients to
4424 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4425 unchanged) from the Cygwin maintainer.
4427 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4428 was supported instead of testing for mbx. Effectively a typo.
4430 15. The spa authenticator server code was not checking that the input it
4431 received was valid base64.
4433 16. The debug output line for the "set" modifier in ACLs was not showing the
4434 name of the variable that was being set.
4436 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4438 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4439 lookup on its own. Something else was needed to trigger the lookup. For
4440 example, a match in host_lookup or the need for a host name in a host list.
4441 Now, if $sender_host_name is referenced and the host name has not yet been
4442 looked up, a lookup is performed. If the lookup fails, the variable remains
4443 empty, and $host_lookup_failed is set to "1".
4445 19. Added "eqi" as a case-independent comparison operator.
4447 20. The saslauthd authentication condition could segfault if neither service
4448 nor realm was specified.
4450 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4451 error message that was logged was misleading, and incoming SMTP
4452 connections were dropped. The message is now more accurate, and temporary
4453 errors are given to SMTP connections.
4455 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4456 (except RSET) with a 421 error, until QUIT is received. However, it was
4457 failing to send a response to QUIT.
4459 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4460 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4461 ACL is now run after the helo_try_verify_hosts code.
4463 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4464 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4465 case-independent, so other case variants are also recognized). Apparently
4466 some systems use these upper case variants.
4468 25. If more than two messages were waiting for the same host, and a transport
4469 filter was specified for the transport, Exim sent two messages over the
4470 same TCP/IP connection, and then failed with "socket operation on non-
4471 socket" when it tried to send the third.
4473 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4475 27. The extern definition of crypt16() in expand.c was not being excluded when
4476 the OS had its own crypt16() function.
4478 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4479 as a preferred synonym for return_size_limit, both as an option and as an
4482 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4484 30. Changed the default configuration ACL to relax the local part checking rule
4485 for addresses that are not in any local domains. For these addresses,
4486 slashes and pipe symbols are allowed within local parts, but the sequence
4487 /../ is explicitly forbidden.
4489 31. SPA server authentication was not clearing the challenge buffer before
4492 32. log_message in a "warn" ACL statement was writing to the reject log as
4493 well as to the main log, which contradicts the documentation and doesn't
4494 seem right (because no rejection is happening). So I have stopped it.
4496 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4497 However, I am unable to do any testing of this.
4499 34. Fixed an infelicity in the appendfile transport. When checking directories
4500 for a mailbox, to see if any needed to be created, it was accidentally
4501 using path names with one or more superfluous leading slashes; tracing
4502 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4504 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4505 contents are added to the log line that is written for every discarded
4506 recipient. (Previously a log_message setting was ignored.)
4508 36. The ${quote: operator now quotes the string if it is empty.
4510 37. The install script runs exim in order to find its version number. If for
4511 some reason other than non-existence or emptiness, which it checks, it
4512 could not run './exim', it was installing it with an empty version number,
4513 i.e. as "exim-". This error state is now caught, and the installation is
4516 38. An argument was missing from the function that creates an error message
4517 when Exim fails to connect to the socket for saslauthd authentication.
4518 This could cause Exim to crash, or give a corrupted message.
4520 39. Added isip, isip4, and isip6 to ${if conditions.
4522 40. The ACL variables $acl_xx are now saved with the message, and can be
4523 accessed later in routers, transports, and filters.
4525 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4526 strings in the file are not string-expanded.
4528 42. If a MAIL command specified a SIZE value that was too large to fit into an
4529 int variable, the check against message_size_limit failed. Such values are
4530 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4531 day this will have to be increased, but I don't think I want to be around
4532 when emails are that large.
4539 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4540 iconv() is not standard in FreeBSD.
4542 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4543 IPv6 enabled. The observed symptom was a segmentation fault on return from
4544 the function os_common_find_running_interfaces() in src/os.c.
4546 3. In the check_special_case() function in daemon.c I had used "errno" as an
4547 argument name, which causes warnings on some systems. This was basically a
4548 typo, since it was named "eno" in the comments!
4550 4. The code that waits for the clock to tick (at a resolution of some fraction
4551 of a second) so as to ensure message-id uniqueness was always waiting for
4552 at least one whole tick, when it could have waited for less. [This is
4553 almost certainly not relevant at current processor speeds, where it is
4554 unlikely to ever wait at all. But we try to future-proof.]
4556 5. The function that sleeps for a time interval that includes fractions of a
4557 second contained a race. It did not block SIGALRM between setting the
4558 timer, and suspending (a couple of lines later). If the interval was short
4559 and the sigsuspend() was delayed until after it had expired, the suspension
4560 never ended. On busy systems this could lead to processes getting stuck for
4563 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4564 process, before it forks any delivery processes. The open lookup caching
4565 mechanism meant that the open file or database connection was passed into
4566 the delivery process. The problem was that delivery processes always tidy
4567 up cached lookup data. This could cause a problem for the next delivery
4568 process started by the queue runner, because the external queue runner
4569 process does not know about the closure. So the next delivery process
4570 still has data in the lookup cache. In the case of a file lookup, there was
4571 no problem because closing a file descriptor in a subprocess doesn't affect
4572 the parent. However, if the lookup was caching a connection to a database,
4573 the connection was closed, and the second delivery process was likely to
4574 see errors such as "PGSQL: query failed: server closed the connection
4575 unexpectedly". The problem has been fixed by closing all cached lookups
4576 in a queue runner before running a delivery process.
4578 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4579 seem to have the "const" qualifier which it has on other OS. I've
4582 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4583 *of the same type* (client or server) with the same public name that an
4584 error should be diagnosed.
4586 9. When Exim looked up a host name for an IP address, but failed to find the
4587 original IP address when looking up the host name (a safety check), it
4588 output the message "<ip address> does not match any IP for NULL", which was
4589 confusing, to say the least. The bug was that the host name should have
4590 appeared instead of "NULL".
4592 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4593 user that is built into the binary, and the -C or -D options is used, root
4594 privilege is dropped before the configuration file is read. In addition,
4595 logging is switched to stderr instead of the normal log files. If the
4596 configuration then re-defines the Exim user, the unprivileged environment
4597 is probably not what is expected, so Exim logs a panic warning message (but
4600 However, if deliver_drop_privilege is set, the unprivileged state may well
4601 be exactly what is intended, so the warning has been cut out in that case,
4602 and Exim is allowed to try to write to its normal log files.
4608 1. smtp_return_error_details was not giving details for temporary sender
4609 or receiver verification errors.
4611 2. Diagnose a configuration error if two authenticators have the same public
4614 3. Exim used not to create the message log file for a message until the first
4615 delivery attempt. This could be confusing when incoming messages were held
4616 for policy or load reasons. The message log file is now created at the time
4617 the message is received, and an initial "Received" line is written to it.
4619 4. The automatically generated man page for command line options had a minor
4620 bug that caused no ill effects; however, a more serious problem was that
4621 the procedure for building the man page automatically didn't always
4622 operate. Consequently, release 4.20 contains an out-of-date version. This
4623 shouldn't happen again.
4625 5. When building Exim with embedded Perl support, the script that builds the
4626 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4627 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4629 6. The freeze_tell option was not being used for messages that were frozen on
4630 arrival, either by an ACL or by local_scan().
4632 7. Added the smtp_incomplete_transaction log selector.
4634 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4635 was accepting AUTH without a new EHLO.
4637 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4638 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4641 10. Logging of TCP/IP connections (when configured) now happens in the main
4642 daemon process instead of the child process, so that the TCP/IP connection
4643 count is more accurate (but it can never be perfect).
4645 11. The use of "drop" in a nested ACL was not being handled correctly in the
4646 outer ACL. Now, if condition failure induced by the nested "drop" causes
4647 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4648 or "require"), the connection is dropped.
4650 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4651 that yield "discard" can now be used with an "accept" or a "discard" verb,
4652 but an error is generated for any others (because I can't see a useful way
4653 to define what should happen).
4655 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4656 are now processed in the same way as lines in the Exim configuration file.
4657 In particular, continuation lines are supported.
4659 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4661 15. Added -ti meaning -t -i.
4663 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4664 domains, and warn by logging if others are found.
4666 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4667 returned from GIFCONF when Exim is trying to find the list of interfaces on
4668 a host. The code in os.c has been modified to copy each ifreq to an aligned
4669 structure in all cases.
4671 Also, in some cases, the returned ifreq's were being copied to a 'struct
4672 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4673 means the last couple of bytes of an IPv6 address could be chopped if the
4674 ifreq contained only a normal sockaddr (14 bytes storage).
4676 18. Named domain lists were not supported in the hosts_treat_as_local option.
4677 An entry such as +xxxx was not recognized, and was treated as a literal
4680 19. Ensure that header lines added by a DATA ACL are included in the reject log
4681 if the ACL subsequently rejects the message.
4683 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4684 MD5 (which is deprecated).
4686 21. When testing a filter file using -bf, Exim was writing a message when it
4687 took the sender from a "From " line in the message, but it was not doing so
4688 when it took $return_path from a Return-Path: header line. It now does.
4690 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4691 with a valid header line field name (a series of printing characters
4692 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4694 23. Changed "disc" in the source to "disk" to conform to the documentation and
4695 the book and for uniformity.
4697 24. Ignore Sendmail's -Ooption=value command line item.
4699 25. When execve() failed while trying to run a command in a pipe transport,
4700 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4701 could be confused with a return value of 69 from the command itself. This
4702 has been changed to 127, the value the shell returns if it is asked to run
4703 a non-existent command. The wording for the related log line suggests a
4704 non-existent command as the problem.
4706 26. If received_header_text expands to an empty string, do not add a Received:
4707 header line to the message. (Well, it adds a token one on the spool, but
4708 marks it "old" so that it doesn't get used or transmitted.)
4710 27. Installed eximstats 1.28 (addition of -nt option).
4712 28. There was no check for failure on the call to getsockname() in the daemon
4713 code. This can fail if there is a shortage of resources on the system, with
4714 ENOMEM, for example. A temporary error is now given on failure.
4716 29. Contrary to the C standard, it seems that in some environments, the
4717 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4718 program. Exim now does this explicitly; it affects the formatting of
4719 timestamps using strftime().
4721 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4722 complaints. I've now initialized all the variables, to avoid this.
4724 32. Header lines added by a system filter were not being "seen" during
4725 transport-time rewrites.
4727 33. The info_callback() function passed to OpenSSL is set up with type void
4728 (*)(SSL *, int, int), as described somewhere. However, when calling the
4729 function (actually a macro) that sets it up, the type void(*)() is
4730 expected. I've put in a cast to prevent warnings from picky compilers.
4732 34. If a DNS black list lookup found a CNAME record, but there were no A
4733 records associated with the domain it pointed at, Exim crashed.
4735 35. If a DNS black list lookup returned more than one A record, Exim ignored
4736 all but the first. It now scans all returned addresses if a particular IP
4737 value is being sought. In this situation, the contents of the
4738 $dnslist_value variable are a list of all the addresses, separated by a
4741 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4742 to accept a host name and all its aliases if the forward lookup for any of
4743 them yielded the IP address of the incoming connection. Now it accepts only
4744 those names whose forward lookup yields the correct IP address. Any other
4745 names are discarded. This closes a loophole whereby a rogue DNS
4746 administrator could create reverse DNS records to break through a
4747 wildcarded host restriction in an ACL.
4749 37. If a user filter or a system filter that ran in a subprocess used any of
4750 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4751 the wrong values were passed to the pipe command ($thisaddress had the
4752 value of $0, $0 had the value of $1, etc). This bug was introduced by
4753 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4755 38. Improved the line breaking for long SMTP error messages from ACLs.
4756 Previously, if there was no break point between 40 and 75 characters, Exim
4757 left the rest of the message alone. Two changes have been made: (a) I've
4758 reduced the minimum length to 35 characters; (b) if it can't find a break
4759 point between 35 and 75 characters, it looks ahead and uses the first one
4760 that it finds. This may give the occasional overlong line, but at least the
4761 remaining text gets split now.
4763 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4764 file descriptors might be low, and that setting 1000 would always raise it.
4765 It turns out that in some environments, the limit is already over 1000 and
4766 that lowering it causes trouble. So now Exim takes care not to decrease it.
4768 40. When delivering a message, the value of $return_path is set to $sender_
4769 address at the start of routing (routers may change the value). By an
4770 oversight, this default was not being set up when an address was tested by
4771 -bt or -bv, which affected the outcome if any router or filter referred to
4774 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4775 log only when "message" or "log_message" are set. However, if one of the
4776 conditions was an address verification, or a call to a nested ACL, the
4777 messages generated by the underlying test were being passed through. This
4778 no longer happens. The underlying message is available in $acl_verify_
4779 message for both "message" and "log_message" expansions, so it can be
4780 passed through if needed.
4782 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4783 new expansion $bh_ to give the encoded byte string without charset
4784 translation. Translation happens only if iconv() is available; HAVE_ICONV
4785 indicates this at build time. HEADERS_CHARSET gives the charset to
4786 translate to; headers_charset can change it in the configuration, and
4787 "headers charset" can change it in an individual filter file.
4789 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4790 that creates RFC 2047 encoded "words" labels them as that charset instead
4791 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4792 expansion operator; (ii) when Exim creates a From: line for a local
4793 message; (iii) when a header line is rewritten to include a "phrase" part.
4795 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4796 buggy, causing it to skip the first lines of messages whose message ID
4797 ended in 'D'. This would not have bitten before Exim release 4.14, because
4798 message IDs were unlikely to end in 'D' before then. The effect was to have
4799 incorrect size information for certain domains.
4801 45. #include "config.h" was missing at the start of the crypt16.c module. This
4802 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4805 46. If there was a timeout during a "random" callout check, Exim treated it as
4806 a failure of the random address, and carried on sending RSET and the real
4807 address. If the delay was just some slowness somewhere, the response to the
4808 original RCPT would be taken as a response to RSET and so on, causing
4809 mayhem of various kinds.
4811 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4812 when I implemented it. It didn't allow for the fact that some option values
4813 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4814 the right test for positive values.
4816 48. Domain names in DNS records are case-independent. Exim always looks them up
4817 in lower case. Some resolvers return domain names in exactly the case they
4818 appear in the zone file, that is, they may contain uppercase letters. Not
4819 all resolvers do this - some return always lower case. Exim was treating a
4820 change of case by a resolver as a change of domain, similar to a widening
4821 of a domain abbreviation. This triggered its re-routing code and so it was
4822 trying to route what was effectively the same domain again. This normally
4823 caused routing to fail (because the router wouldn't handle the domain
4824 twice). Now Exim checks for this case specially, and just changes the
4825 casing of the domain that it ultimately uses when it transmits the message
4828 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4831 50. If a filter generated a file delivery with a non-absolute name (possible if
4832 no home directory exists for the router), the forbid_file option was not
4835 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4836 the existing equality matching.
4838 52. Exim was using ints instead of ino_t variables in some places where it was
4839 dealing with inode numbers.
4841 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4842 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4843 called TMPDIR, and if it finds it is different, it changes its value.
4845 54. The smtp_printf() function is now made available to local_scan() so
4846 additional output lines can be written before returning. There is also an
4847 smtp_fflush() function to enable the detection of a dropped connection.
4848 The variables smtp_input and smtp_batched_input are exported to
4851 55. Changed the default runtime configuration: the message "Unknown user"
4852 has been removed from the ACL, and instead placed on the localuser router,
4853 using the cannot_route_message feature. This means that any verification
4854 failures that generate their own messages won't get overridden. Similarly,
4855 the "Unrouteable address" message that was in the ACL for unverifiable
4856 relay addresses has also been removed.
4858 56. Added hosts_avoid_esmtp to the smtp transport.
4860 57. The exicyclog script was not checking for the esoteric option
4861 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4862 will work only if exicyclog is run under the appropriate euid.
4864 58. Following a discussion on the list, the rules by which Exim recognises line
4865 endings on incoming messages have been changed. The -dropcr and drop_cr
4866 options are now no-ops, retained only for backwards compatibility. The
4867 following line terminators are recognized: LF CRLF CR. However, special
4868 processing applies to CR:
4870 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4871 nor a local message in the state where . is a terminator.
4873 (ii) If a bare CR is encountered in a header line, an extra space is added
4874 after the line terminator so as not to end the header. The reasoning
4875 behind this is that bare CRs in header lines are most likely either
4876 to be mistakes, or people trying to play silly games.
4878 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4879 was being incorrectly given as 18 bytes larger than it should have been.
4880 This is a VOB (very old bug).
4882 60. This may never have affected anything current, but just in case it has:
4883 When the local host is found other than at the start of a list of hosts,
4884 the local host, those with the same MX, and any that follow, are discarded.
4885 When the list in question was part of a longer list of hosts, the following
4886 hosts (not currently being processed) were also being discarded. This no
4887 longer happens. I'm not sure if this situation could ever has previously
4890 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4893 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4894 header. This is something that is recommended in a new Internet Draft, and
4895 is something that is documented as being done by Sendmail. There are two
4896 possible values. For messages generated by the autoreply transport, Exim
4899 Auto-Submitted: auto-replied
4901 whereas for all other generated messages (e.g. bounces) it adds
4903 Auto-Submitted: auto-generated
4905 63. The "personal" condition in filters now includes a test for the
4906 Auto-Submitted: header. If it contains the string "auto-" the message it
4907 not considered personal.
4909 64. Added rcpt_include_affixes as a generic transport option.
4911 65. Added queue_only_override (default true).
4913 66. Added the syslog_duplication option.
4915 67. If what should have been the first header line of a message consisted of
4916 a space followed by a colon, Exim was mis-interpreting it as a header line.
4917 It isn't of course - it is syntactically invalid and should therefore be
4918 treated as the start of the message body. The misbehaviour could have
4919 caused a number of strange effects, including loss of data in subsequent
4920 header lines, and spool format errors.
4922 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4923 client host had authenticated. This control can now be exercised by an ACL
4924 for more flexibility.
4926 69. By default, callouts do not happen when testing with -bh. There is now a
4927 variant, -bhc, which does actually run the callout code, including
4928 consulting and updating the callout cache.
4930 70. Added support for saslauthd authentication, courtesy of Alexander
4933 71. If statvfs() failed on the spool or log directories while checking their
4934 size for availability, Exim confusingly gave the error "space shortage".
4935 Furthermore, in debugging mode it crashed with a floating point exception.
4936 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4937 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4938 serious problem, Exim now writes to the main and panic logs when this
4939 happens, with details of the failure. It then refuses to accept the
4940 incoming message, giving the message "spool directory problem" or "log
4941 directory problem" with a 421 code for SMTP messages.
4943 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4944 0, 1, and 2 exist, because some OS complain for execs without them (see
4945 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4946 descriptors. However, the code omitted to check that the open succeeded,
4947 causing mysterious errors if for some reason the permissions on /dev/null
4948 got screwed. Now Exim writes a message to the main and panic logs, and
4949 bombs out if it can't open /dev/null.
4951 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4952 interact so that it is all more flexible. It is supposed to remain
4953 backwards compatible. Also added extra_local_interfaces.
4955 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4956 to bomb out with an assertion failure - to the client this appears as a
4957 connection drop. This problem occurs in the part of the code that was taken
4958 from the Samba project. Fortunately, the assertion is in a very simple
4959 function, so I have fixed this by reproducing the function inline in the
4960 one place where it is called, and arranging for authentication to fail
4961 instead of killing the process with assert().
4963 75. The SPA client code was not working when the server requested OEM rather
4964 than Unicode encoding.
4966 76. Added code to make require_files with a specific uid setting more usable in
4967 the case where statting the file as root fails - usually a non-root-mounted
4968 NFS file system. When this happens and the failure is EACCES, Exim now
4969 forks a subprocess and does the per-uid checking as the relevant uid.
4971 77. Added process_log_path.
4973 78. If log_file_path was not explicitly set, a setting of check_log_space or
4974 check_log_inodes was ignored.
4976 79. If a space check for the spool or log partitions fails, the incident is now
4977 logged. Of course, in the latter case the data may get lost...
4979 80. Added the %p formatting code to string_format() so that it can be used to
4980 print addresses in debug_print(). Adjusted all the address printing in the
4981 debugging in store.c to use %p rather than %d.
4983 81. There was a concern that a line of code in smtp_in.c could overflow a
4984 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4985 initially expressed, the concern was not well-founded, because trailing
4986 spaces are removed early. However, if the trailing spaces were followed by
4987 a NULL, they did not get removed, so the overflow was possible. Two fixes
4990 (a) I re-wrote the offending code in a cleaner fashion.
4991 (b) If an incoming SMTP command contains a NULL character, it is rejected
4994 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4995 runs initgroups(), so that if the Exim user is in any additional groups,
4996 they will be used during message reception.
5002 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob