1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.580 2009/10/26 13:10:23 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05 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
128 NM/26 Bugzilla 799: Documentation correction for ratelimit
130 NM/27 Bugzilla 802: Improvements to local interface IP addr detection
131 Patch provided by David Brownlee
133 NM/28 Bugzilla 807: Improvements to LMTP delivery logging
135 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes
141 TK/01 Add preliminary DKIM support. Currently requires a forked version of
142 ALT-N's libdkim that I have put here:
143 http://duncanthrax.net/exim-experimental/
145 Note to Michael Haardt: I had to rename some vars in sieve.c. They
146 were called 'true' and it seems that C99 defines that as a reserved
147 keyword to be used with 'bool' variable types. That means you could
148 not include C99-style headers which use bools without triggering
149 build errors in sieve.c.
151 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
152 as mailq or other aliases. Changed the --help handling significantly
153 to do whats expected. exim_usage() emits usage/help information.
155 SC/01 Added the -bylocaldomain option to eximstats.
157 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
159 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
161 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
167 PH/01 Another patch from the Sieve maintainer.
169 PH/02 When an IPv6 address is converted to a string for single-key lookup
170 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
171 dots are used instead of colons so that keys in lsearch files need not
172 contain colons. This was done some time before quoting was made available
173 in lsearch files. However, iplsearch files do require colons in IPv6 keys
174 (notated using the quote facility) so as to distinguish them from IPv4
175 keys. This meant that lookups for IP addresses in host lists did not work
176 for iplsearch lookups.
178 This has been fixed by arranging for IPv6 addresses to be expressed with
179 colons if the lookup type is iplsearch. This is not incompatible, because
180 previously such lookups could never work.
182 The situation is now rather anomolous, since one *can* have colons in
183 ordinary lsearch keys. However, making the change in all cases is
184 incompatible and would probably break a number of configurations.
186 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
189 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
190 conversion specification without a maximum field width, thereby enabling
191 a rogue spamd server to cause a buffer overflow. While nobody in their
192 right mind would setup Exim to query an untrusted spamd server, an
193 attacker that gains access to a server running spamd could potentially
194 exploit this vulnerability to run arbitrary code as the Exim user.
196 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
197 $primary_hostname instead of what libspf2 thinks the hosts name is.
199 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
200 a directory entry by the name of the lookup key. Previously, if a
201 symlink pointed to a non-existing file or a file in a directory that
202 Exim lacked permissions to read, a lookup for a key matching that
203 symlink would fail. Now it is enough that a matching directory entry
204 exists, symlink or not. (Bugzilla 503.)
206 PH/03 The body_linecount and body_zerocount variables are now exported in the
209 PH/04 Added the $dnslist_matched variable.
211 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
212 This means they are set thereafter only if the connection becomes
215 PH/06 Added the client_condition to authenticators so that some can be skipped
216 by clients under certain conditions.
218 PH/07 The error message for a badly-placed control=no_multiline_responses left
219 "_responses" off the end of the name.
221 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
223 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
224 (without spaces) instead of just copying the configuration text.
226 PH/10 Added the /noupdate option to the ratelimit ACL condition.
228 PH/11 Added $max_received_linelength.
230 PH/12 Added +ignore_defer and +include_defer to host lists.
232 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
233 way in which PCRE > 7.0 is built.
235 PH/14 Implemented queue_only_load_latch.
237 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
238 MAIL command. The effect was to mangle the value on 64-bit systems.
240 PH/16 Another patch from the Sieve maintainer.
242 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
244 PH/18 If a system quota error occurred while trying to create the file for
245 a maildir delivery, the message "Mailbox is full" was not appended to the
246 bounce if the delivery eventually timed out. Change 4.67/27 below applied
247 only to a quota excession during the actual writing of the file.
249 PH/19 It seems that peer DN values may contain newlines (and other non-printing
250 characters?) which causes problems in log lines. The DN values are now
251 passed through string_printing() before being added to log lines.
253 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
254 and InterBase are left for another time.)
256 PH/21 Added message_body_newlines option.
258 PH/22 Guard against possible overflow in moan_check_errorcopy().
260 PH/23 POSIX allows open() to be a macro; guard against that.
262 PH/24 If the recipient of an error message contained an @ in the local part
263 (suitably quoted, of course), incorrect values were put in $domain and
264 $local_part during the evaluation of errors_copy.
270 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
271 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
274 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
275 issue a MAIL command.
277 PH/02 In an ACL statement such as
279 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
281 if a client was not listed at all, or was listed with a value other than
282 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
283 the condition was not true (as it should be), so access was not denied.
284 The bug was that the ! inversion was incorrectly passed on to the second
285 item. This has been fixed.
287 PH/03 Added additional dnslists conditions == and =& which are different from
288 = and & when the dns lookup returns more than one IP address.
290 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
291 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
293 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
294 FSYNC, which compiles an option called disable_fsync that allows for
295 bypassing fsync(). The documentation is heavily laced with warnings.
297 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
299 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
300 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
301 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
302 including adding "make clean"; (3) Added -fPIC when compiling the test
303 dynamically loaded module, to get rid of a warning.
305 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
306 message fails, move_frozen_messages = true and ignore_bounce_errors_after
307 = 0s. The bug is otherwise harmless.
309 PH/07 There was a bug in the dovecot authenticator such that the value of
310 $auth1 could be overwritten, and so not correctly preserved, after a
311 successful authentication. This usually meant that the value preserved by
312 the server_setid option was incorrect.
314 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
316 PH/09 Installed PCRE release 7.0.
318 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
319 run for batched SMTP input. It is now run at the start of every message
320 in the batch. While fixing this I discovered that the process information
321 (output by running exiwhat) was not always getting set for -bs and -bS
322 input. This is fixed, and it now also says "batched" for BSMTP.
324 PH/11 Added control=no_pipelining.
326 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
327 patch, slightly modified), and move the expansion of helo_data till after
328 the connection is made in the smtp transport (so it can use these
331 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
333 PH/14 Added log_selector = +pid.
335 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
337 PH/16 Add ${if forany and ${if forall.
339 PH/17 Added dsn_from option to vary the From: line in DSNs.
341 PH/18 Flush SMTP output before performing a callout, unless control =
342 no_callout_flush is set.
344 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
345 was true (the default) a successful delivery failed to delete the retry
346 item, thus causing premature timeout of the address. The bug is now
349 PH/20 Added hosts_avoid_pipelining to the smtp transport.
351 PH/21 Long custom messages for fakedefer and fakereject are now split up
352 into multiline reponses in the same way that messages for "deny" and
353 other ACL rejections are.
355 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
356 with slight modification.
358 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
359 draft, changing the syntax and factoring some duplicate code".
361 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
362 for deliveries of the second and subsequent messages over the same SMTP
365 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
366 ${reduce, with only minor "tidies".
368 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
370 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
371 expansion side effects.
373 PH/27 When a message times out after an over-quota error from an Exim-imposed
374 quota, the bounce message says "mailbox is full". This message was not
375 being given when it was a system quota that was exceeded. It now should
378 MH/03 Made $recipients available in local_scan(). local_scan() already has
379 better access to the recipient list through recipients_list[], but
380 $recipients can be useful in postmaster-provided expansion strings.
382 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
383 in the case of a MAIL command with additional options following the
384 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
385 were accidentally chopped off.
387 PH/29 SMTP synchronization checks are implemented when a command is read -
388 there is a check that no more input is waiting when there shouldn't be
389 any. However, for some commands, a delay in an ACL can mean that it is
390 some time before the response is written. In this time, more input might
391 arrive, invalidly. So now there are extra checks after an ACL has run for
392 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
393 pipelining has not been advertised.
395 PH/30 MH's patch to allow iscntrl() characters to be list separators.
397 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
398 returned in the SMTP response when smtp_return_error_details was false.
401 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
402 instead of the C I/O that was originally supplied, because problems were
405 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
406 Exim which did not show up earlier: it was assuming that a call to
407 SSL_CTX_set_info_callback() might give an error value. In fact, there is
408 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
409 was a macro that became an assignment, so it seemed to work. This has
410 changed to a proper function call with a void return, hence the compile
411 error. Exim's code has been fixed.
413 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
416 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
418 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
420 PH/37 If a message is not accepted after it has had an id assigned (e.g.
421 because it turns out to be too big or there is a timeout) there is no
422 "Completed" line in the log. When some messages of this type were
423 selected by exigrep, they were listed as "not completed". Others were
424 picked up by some special patterns. I have improved the selection
425 criteria to be more general.
427 PH/38 The host_find_failed option in the manualroute router can now be set
428 to "ignore", to completely ignore a host whose IP address cannot be
429 found. If all hosts are ignored, the behaviour is controlled by the new
430 host_all_ignored option.
432 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
433 homing or because of multiple MX records with /mx) generated more than
434 one IP address, and the following item turned out to be the local host,
435 all the secondary addresses of the first item were incorrectly removed
436 from the list, along with the local host and any following hosts (which
437 is what is supposed to happen).
439 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
440 whoever called Exim into the -H file. In the case of the daemon it was
441 behaving confusingly. When first started, it used values for whoever
442 started the daemon, but after a SIGHUP it used the Exim user (because it
443 calls itself on a restart). I have changed the code so that it now always
446 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
447 message are rejected with the same error (e.g. no authentication or bad
448 sender address), and a DATA command is nevertheless sent (as can happen
449 with PIPELINING or a stupid MUA), the error message that was given to the
450 RCPT commands is included in the rejection of the DATA command. This is
451 intended to be helpful for MUAs that show only the final error to their
454 PH/42 Another patch from the Sieve maintainer.
456 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
457 Eximstats - Fixed some broken HTML links and added missing column headers
459 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
460 columns for Rejects, Temp Rejects, Ham, and Spam rows.
462 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
464 PH/43 Yet another patch from the Sieve maintainer.
466 PH/44 I found a way to check for a TCP/IP connection going away before sending
467 the response to the final '.' that terminates a message, but only in the
468 case where the client has not sent further data following the '.'
469 (unfortunately, this is allowed). However, in many cases there won't be
470 any further data because there won't be any more messages to send. A call
471 to select() can be used: if it shows that the input is "ready", there is
472 either input waiting, or the socket has been closed. An attempt to read
473 the next input character can distinguish the two cases. Previously, Exim
474 would have sent an OK response which the client would never have see.
475 This could lead to message repetition. This fix should cure that, at
476 least in a lot of common cases.
478 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
479 advertised in response to EHLO.
485 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
486 fixed by 4.65/MH/01 (is this a record?) are fixed:
488 (i) An empty string was always treated as zero by the numeric comparison
489 operators. This behaviour has been restored.
491 (ii) It is documented that the numeric comparison operators always treat
492 their arguments as decimal numbers. This was broken in that numbers
493 starting with 0 were being interpreted as octal.
495 While fixing these problems I realized that there was another issue that
496 hadn't been noticed. Values of message_size_limit (both the global option
497 and the transport option) were treated as octal if they started with 0.
498 The documentation was vague. These values are now always treated as
499 decimal, and I will make that clear in the documentation.
505 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
506 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
509 MH/01 Don't check that the operands of numeric comparison operators are
510 integers when their expansion is in "skipping" mode (fixes bug
511 introduced by 4.64-PH/07).
513 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
514 child addresses, Exim now panics and dies. Previously, because the count
515 is held in a short int, deliveries were likely to be lost. As such a
516 large number of recipients for a single message is ridiculous
517 (performance will be very, very poor), I have chosen to impose a limit
518 rather than extend the field.
524 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
525 leftover -K file (the existence of which was triggered by #402).
526 While we were at it, introduced process PID as part of the -K
527 filename. This should rule out race conditions when creating
530 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
531 processing considerably. Previous code took too long for large mails,
532 triggering a timeout which in turn triggers #401.
534 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
535 in the DK code in transports.c. sendfile() is not really portable,
536 hence the _LINUX specificness.
538 TF/01 In the add_headers option to the mail command in an Exim filter,
539 there was a bug that Exim would claim a syntax error in any
540 header after the first one which had an odd number of characters
543 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
544 callout verification, Exim cached a "reject" for the entire domain. This
545 is correct for most verifications, but it is not correct for a recipient
546 verification with use_sender or use_postmaster set, because in that case
547 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
548 case of MAIL FROM:<> rejection from other early rejections (e.g.
549 rejection of HELO). When verifying a recipient using a non-null MAIL
550 address, the cache is ignored if it shows MAIL FROM:<> rejection.
551 Whatever the result of the callout, the value of the domain cache is
552 left unchanged (for any other kind of callout, getting as far as trying
553 RCPT means that the domain itself is ok).
555 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
558 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
559 manifest itself as EPIPE rather than ECONNECT. When tidying away a
560 session, the daemon ignores ECONNECT errors and logs others; it now
561 ignores EPIPE as well.
563 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
564 (quoted-printable decoding).
566 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
567 later the small subsequent patch to fix an introduced bug.
569 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
571 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
573 PH/08 An error is now given if message_size_limit is specified negative.
575 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
576 to be given (somewhat) arbitrary names.
578 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
581 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
582 miscellaneous code fixes
584 PH/10 Added the log_reject_target ACL modifier to specify where to log
587 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
588 hostname. This is wrong, because it relates to the incoming message (and
589 probably the interface on which it is arriving) and not to the outgoing
590 callout (which could be using a different interface). This has been
591 changed to use the value of the helo_data option from the smtp transport
592 instead - this is what is used when a message is actually being sent. If
593 there is no remote transport (possible with a router that sets up host
594 addresses), $smtp_active_hostname is used.
596 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
597 tweaks were necessary in order to get it to work (see also 21 below):
598 (a) The code assumed that strncpy() returns a negative number on buffer
599 overflow, which isn't the case. Replaced with Exim's string_format()
601 (b) There were several signed/unsigned issues. I just did the minimum
602 hacking in of casts. There is scope for a larger refactoring.
603 (c) The code used strcasecmp() which is not a standard C function.
604 Replaced with Exim's strcmpic() function.
605 (d) The code set only $1; it now sets $auth1 as well.
606 (e) A simple test gave the error "authentication client didn't specify
607 service in request". It would seem that Dovecot has changed its
608 interface. Fortunately there's a specification; I followed it and
609 changed what the client sends and it appears to be working now.
611 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
614 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
615 address A is aliased to B and C, where B exists and C does not. Without
616 -v the output is "A verified" because verification stops after a
617 successful redirection if more than one address is generated. However,
618 with -v the child addresses are also verified. Exim was outputting "A
619 failed to verify" and then showing the successful verification for C,
620 with its parentage. It now outputs "B failed to verify", showing B's
621 parentage before showing the successful verification of C.
623 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
624 look up a TXT record in a specific list after matching in a combined
627 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
628 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
629 they consult the DNS. I had assumed they would set it the way they
630 wanted; and indeed my experiments on Linux seem to show that in some
631 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
632 To be on the safe side, however, I have now made the interface to
633 host_find_byname() similar to host_find_bydns(), with an argument
634 containing the DNS resolver options. The host_find_byname() function now
635 sets these options at its start, just as host_find_bydns() does. The smtp
636 transport options dns_qualify_single and dns_search_parents are passed to
637 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
638 of host_find_byname() use the default settings of RES_DEFNAMES
639 (qualify_single) but not RES_DNSRCH (search_parents).
641 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
642 spool_read_header() do less string testing, by means of a preliminary
643 switch on the second character of optional "-foo" lines. (This is
644 overdue, caused by the large number of possibilities that now exist.
645 Originally there were few.) While I was there, I also converted the
646 str(n)cmp tests so they don't re-test the leading "-" and the first
647 character, in the hope this might squeeze out yet more improvement.
649 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
650 flag allowing group syntax was set by the header_syntax check but not
651 turned off, possible causing trouble later; (2) The flag was not being
652 set at all for the header_verify test, causing "group"-style headers to
653 be rejected. I have now set it in this case, and also caused header_
654 verify to ignore an empty address taken from a group. While doing this, I
655 came across some other cases where the code for allowing group syntax
656 while scanning a header line wasn't quite right (mostly, not resetting
657 the flag correctly in the right place). These bugs could have caused
658 trouble for malformed header lines. I hope it is now all correct.
660 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
661 with the "reply" argument non-NULL. The code, however (which originally
662 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
663 but it didn't always do it. This confused somebody who was copying the
664 code for some other use. I have removed all the tests.
666 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
667 feature that was used to support insecure browsers during the U.S. crypto
668 embargo. It requires special client support, and Exim is probably the
669 only MTA that supported it -- and would never use it because real RSA is
670 always available. This code has been removed, because it had the bad
671 effect of slowing Exim down by computing (never used) parameters for the
672 RSA_EXPORT functionality.
674 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
675 authenticator to fail if there's a tab character in the incoming data
676 (there should never be unless someone is messing about, as it's supposed
677 to be base64-encoded). Also added, on Timo's advice, the "secured" option
678 if the connection is using TLS or if the remote IP is the same as the
679 local IP, and the "valid-client-cert option" if a client certificate has
682 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
683 authenticators. This can be used for authorization after authentication
684 succeeds. (In the case of plaintext, it servers for both authentication
687 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
688 if any retry times were supplied.
690 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
691 connection, where there is no client IP address to check. In this
692 situation, the verify now always succeeds.
694 PH/25 Applied John Jetmore's -Mset patch.
696 PH/26 Added -bem to be like -Mset, but loading a message from a file.
698 PH/27 In a string expansion for a processed (not raw) header when multiple
699 headers of the same name were present, leading whitespace was being
700 removed from all of them, but trailing whitespace was being removed only
701 from the last one. Now trailing whitespace is removed from each header
702 before concatenation. Completely empty headers in a concatenation (as
705 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
706 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
708 PH/29 [Removed. This was a change that I later backed out, and forgot to
709 correct the ChangeLog entry (that I had efficiently created) before
710 committing the later change.]
712 PH/30 Exim was sometimes attempting to deliver messages that had suffered
713 address errors (4xx response to RCPT) over the same connection as other
714 messages routed to the same hosts. Such deliveries are always "forced",
715 so retry times are not inspected. This resulted in far too many retries
716 for the affected addresses. The effect occurred only when there were more
717 hosts than the hosts_max_try setting in the smtp transport when it had
718 the 4xx errors. Those hosts that it had tried were not added to the list
719 of hosts for which the message was waiting, so if all were tried, there
720 was no problem. Two fixes have been applied:
722 (i) If there are any address or message errors in an SMTP delivery, none
723 of the hosts (tried or untried) are now added to the list of hosts
724 for which the message is waiting, so the message should not be a
725 candidate for sending over the same connection that was used for a
726 successful delivery of some other message. This seems entirely
727 reasonable: after all the message is NOT "waiting for some host".
728 This is so "obvious" that I'm not sure why it wasn't done
729 previously. Hope I haven't missed anything, but it can't do any
730 harm, as the worst effect is to miss an optimization.
732 (ii) If, despite (i), such a delivery is accidentally attempted, the
733 routing retry time is respected, so at least it doesn't keep
734 hammering the server.
736 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
737 in ${readsocket because some servers need this prod.
739 PH/32 Added some extra debug output when updating a wait-xxx database.
741 PH/33 The hint "could be header name not terminated by colon", which has been
742 given for certain expansion errors for a long time, was not being given
743 for the ${if def:h_colon_omitted{... case.
745 PH/34 The spec says: "With one important exception, whenever a domain list is
746 being scanned, $domain contains the subject domain." There was at least
747 one case where this was not true.
749 PH/35 The error "getsockname() failed: connection reset by peer" was being
750 written to the panic log as well as the main log, but it isn't really
751 panic-worthy as it just means the connection died rather early on. I have
752 removed the panic log writing for the ECONNRESET error when getsockname()
755 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
756 runs only) independently of the message's sender address. This meant
757 that, if the 4xx error was in fact related to the sender, a different
758 message to the same recipient with a different sender could confuse
759 things. In particualar, this can happen when sending to a greylisting
760 server, but other circumstances could also provoke similar problems.
761 I have changed the default so that the retry time for these errors is now
762 based a combination of the sender and recipient addresses. This change
763 can be overridden by setting address_retry_include_sender=false in the
766 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
767 remote server are returned as part of bounce messages. This was not
768 happening for LMTP over a pipe (the lmtp transport), but now it is the
769 same for both kinds of LMTP.
771 PH/38 Despite being documented as not happening, Exim was rewriting addresses
772 in header lines that were in fact CNAMEs. This is no longer the case.
774 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
775 and queue runs started by the daemon processed all messages. This has
776 been fixed so that -R and -S can now usefully be given with -q<time>.
778 PH/40 Import PCRE release 6.7 (fixes some bugs).
780 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
782 PH/42 Give an error if -q is specified more than once.
784 PH/43 Renamed the variables $interface_address and $interface_port as
785 $received_ip_address and $received_port, to make it clear that these
786 values apply to message reception, and not to the outgoing interface when
787 a message is delivered. (The old names remain recognized, of course.)
789 PH/44 There was no timeout on the connect() call when using a Unix domain
790 socket in the ${readsocket expansion. There now is.
792 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
793 be meaningful with "accept".
795 SC/01 Eximstats V1.43
796 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
798 SC/02 Eximstats V1.44
799 Use a glob alias rather than an array ref in the generated
800 parser. This improves both readability and performance.
802 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
803 Collect SpamAssassin and rejection statistics.
804 Don't display local sender or destination tables unless
805 there is data to show.
806 Added average volumes into the top table text output.
808 SC/04 Eximstats V1.46
809 Collect data on the number of addresses (recipients)
810 as well as the number of messages.
812 SC/05 Eximstats V1.47
813 Added 'Message too big' to the list of mail rejection
814 reasons (thanks to Marco Gaiarin).
816 SC/06 Eximstats V1.48
817 Mainlog lines which have GMT offsets and are too short to
818 have a flag are now skipped.
820 SC/07 Eximstats V1.49 (Alain Williams)
821 Added the -emptyok flag.
823 SC/08 Eximstats V1.50
824 Fixes for obtaining the IP address from reject messages.
826 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
827 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
828 whitesspace changes from 4.64-PH/27
830 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
833 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
834 are found, allow negative numbers in numeric criteria)
836 JJ/06 exipick.20061117.2, added new $message_body_missing variable
838 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
839 to match changes made in 4.64-PH/43
841 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
843 PH/47 Put in an explicit test for a DNS lookup of an address record where the
844 "domain" is actually an IP address, and force a failure. This locks out
845 those revolvers/nameservers that support "A-for-A" lookups, in
846 contravention of the specifications.
848 PH/48 When a host name was looked up from an IP address, and the subsequent
849 forward lookup of the name timed out, the host name was left in
850 $sender_host_name, contrary to the specification.
852 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
853 restricted to single-key lookups, Exim was not diagnosing an error if
854 * or *@ was used with a query-style lookup.
856 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
858 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
859 long ago, but noone interested enough thought of it. Let's just say that
860 the "1.1" means that there are some new functions that weren't there at
861 some point in the past.
863 PH/51 Error processing for expansion failure of helo_data from an smtp
864 transport during callout processing was broken.
866 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
867 tested/used via the -bh/-bhc/-bs options.
869 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
870 bug, fixed in subsequent PCRE releases).
872 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
873 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
875 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
881 SC/01 Use a glob alias rather than an array ref in eximstats generated
882 parser. This improves both readability and performance.
884 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
885 Don't display local sender or destination tables in eximstats unless
886 there is data to show.
887 Added average volumes into the eximstats top table text output.
889 SC/03 Collect data on the number of addresses (recipients) as well
890 as the number of messages in eximstats.
892 TF/01 Correct an error in the documentation for the redirect router. Exim
893 does (usually) call initgroups() when daemonizing.
895 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
896 with consistent privilege compared to when running as a daemon.
898 TF/03 Note in the spec that $authenticated_id is not set for local
899 submissions from trusted users.
901 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
902 Thanks to Dean Brooks <dean@iglou.com> for the patch.
904 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
905 by adding some example configuration directives to the default
906 configuration file. A little bit of work is required to uncomment the
907 directives and define how usernames and passwords are checked, but
908 there is now a framework to start from.
910 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
911 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
912 without this. I don't know how relevant this is to other LDAP libraries.
914 PH/02 Add the verb name to the "unknown ACL verb" error.
916 PH/03 Magnus Holmgren's patch for filter_prepend_home.
918 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
920 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
921 directory not expanded when it should be if an expanded home directory
922 was set for the address (which is overridden by the transport).
924 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
927 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
928 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
929 because it is too late at that time, and has no effect.
931 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
932 security issue with \' (bugzilla #107). I could not use the
933 PQescapeStringConn() function, because it needs a PGconn value as one of
936 PH/08 When testing addresses using -bt, indicate those final addresses that
937 are duplicates that would not cause an additional delivery. At least one
938 person was confused, thinking that -bt output corresponded to deliveries.
939 (Suppressing duplicates isn't a good idea as you lose the information
940 about possibly different redirections that led to the duplicates.)
942 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
943 systems where poll() doesn't work, in particular OS X.
945 PH/10 Added more information to debugging output for retry time not reached.
947 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
948 operations in malware.c.
950 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
953 PH/13 If write_rejectlog was set false when logging was sent to syslog with
954 syslog_duplication set false, log lines that would normally be written
955 both the the main log and to the reject log were not written to syslog at
958 PH/14 In the default configuration, change the use of "message" in ACL warn
959 statements to "add_header".
961 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
962 not followed by a command (e.g. "seen endif").
964 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
965 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
968 PH/17 Added extra conditions to the default value of delay_warning_condition
972 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
973 { match{$h_precedence:}{(?i)bulk|list|junk} } \
974 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
977 The Auto-Submitted: and various List- headers are standardised, whereas I
978 don't think Precedence: ever was.
980 PH/18 Refactored debugging code in route_finduser() to show more information,
981 in particular, the error code if getpwnam() issues one.
983 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
984 This is apparently needed in addition to the PH/07 change above to avoid
985 any possible encoding problems.
987 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
988 but not after initializing Perl.
990 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
991 output them only if debugging. By default they are written stderr,
992 apparently, which is not desirable.
994 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
997 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1000 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1002 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1003 authenticated or an ident call has been made. Suppress the default
1004 values for $authenticated_id and $authenticated_sender (but permit -oMai
1005 and -oMas) when testing with -bh.
1007 PH/24 Re-jigged the order of the tests in the default configuration so that the
1008 tests for valid domains and recipients precede the DNS black list and CSA
1009 tests, on the grounds that those ones are more expensive.
1011 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1012 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1013 This bug exists in every version of Exim that I still have, right back to
1016 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1017 However, an attempt to turn on case-sensitivity in a regex key by
1018 including (?-i) didn't work because the subject string was already
1019 lowercased, and the effects were non-intuitive. It turns out that a
1020 one-line patch can be used to allow (?-i) to work as expected.
1026 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1027 other effects) broke the use of negated acl sub-conditions.
1029 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1032 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1033 "Deny" causes Exim to reject the incoming connection with a 554 error.
1034 Unfortunately, if there is a major crisis, such as a disk failure,
1035 tcp-wrappers gives "deny", whereas what one would like would be some
1036 kind of temporary error. A kludge has been added to help with this.
1037 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1038 554 error is used if errno is still zero or contains ENOENT (which occurs
1039 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1042 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1044 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1045 errors. Otherwise a message that provokes a temporary error (when other
1046 messages do not) can cause a whole host to time out.
1048 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1049 addresses were routed directly to files or pipes from a redirect router.
1050 File deliveries just didn't batch; pipe deliveries might have suffered
1053 PH/06 A failure to get a lock for a hints database would erroneously always say
1054 "Failed to get write lock", even when it was really a read lock.
1056 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1057 of 0600. This has been changed to use the value of the lockfile_mode
1058 option (which defaults to 0600).
1060 PH/08 Applied small patch from the Sieve maintainer.
1062 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1063 folder from quota calculations, a direct delivery into this folder messed
1064 up the contents of the maildirsize file. This was because the regex was
1065 used only to exclude .Trash (or whatever) when the size of the mailbox
1066 was calculated. There was no check that a delivery was happening into an
1067 excluded directory. This bug has been fixed by ignoring all quota
1068 processing for deliveries into excluded directories.
1070 PH/10 Added the maildirfolder_create_regex option to appendfile.
1076 PH/01 The code for finding all the local interface addresses on a FreeBSD
1077 system running IPv6 was broken. This may well have applied to all BSD
1078 systems, as well as to others that have similar system calls. The broken
1079 code found IPv4 interfaces correctly, but gave incorrect values for the
1080 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1081 that it would not match correctly against @[] and not recognize the IPv6
1084 PH/02 The ipliteral router was not recognizing addresses of the form user@
1085 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1087 PH/03 Added disable_ipv6.
1089 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1090 decoded form, because it is most often used to construct To: headers
1091 lines in autoreplies, and the decoded form may well be syntactically
1092 invalid. However, $reply_address has leading white space removed, and all
1093 newlines turned into spaces so that the autoreply transport does not
1096 PH/05 If group was specified without a user on a router, and no group or user
1097 was specified on a transport, the group from the router was ignored.
1099 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1100 for visible compile-time settings that can be used to change these
1101 numbers, for those that want even more. Backwards compatibility with old
1102 spool files has been maintained. However, going back to a previous Exim
1103 release will lost any variables that are in spool files.
1105 PH/07 Two small changes when running in the test harness: increase delay when
1106 passing a TCP/IP connection to a new process, in case the original
1107 process has to generate a bounce, and remove special handling of
1108 127.0.0.2 (sic), which is no longer necessary.
1110 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1111 be the same on different OS.
1113 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1116 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1117 whether --show-vars was specified or not
1119 JJ/02 exipick: Added support for new ACL variable spool format introduced
1122 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1123 syntactically invalid From: or Reply-to: line, and a filter used this to
1124 generate an autoreply, and therefore failed to obtain an address for the
1125 autoreply, Exim could try to deliver to a non-existent relative file
1126 name, causing unrelated and misleading errors. What now happens is that
1127 it logs this as a hard delivery error, but does not attempt to create a
1130 PH/11 The exinext utility has a -C option for testing purposes, but although
1131 the given file was scanned by exinext itself; it wasn't being passed on
1132 when Exim was called.
1134 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1135 an end-of-file indication when reading a command response.
1137 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1138 compiled. In many other places in Exim, IPv6 addresses are always
1139 recognized, so I have changed this. It also means that IPv4 domain
1140 literals of the form [IPV4:n.n.n.n] are now always recognized.
1142 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1143 used if the router is not running as root, for example, when verifying at
1144 ACL time, or when using -bh. The debugging output from this situation was
1145 non-existent - all you got was a failure to exec. I have made two
1148 (a) Failures to set uid/gid, the current directory, or a process leader
1149 in a subprocess such as that created by queryprogram now generate
1150 suitable debugging ouput when -d is set.
1152 (b) The queryprogram router detects when it is not running as root,
1153 outputs suitable debugging information if -d is set, and then runs
1154 the subprocess without attempting to change uid/gid.
1156 PH/15 Minor change to Makefile for building test_host (undocumented testing
1159 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1160 additional section of a DNS packet that returns MX or SRV records.
1161 Instead, it always explicitly searches for A/AAAA records. This avoids
1162 major problems that occur when a DNS server includes only records of one
1163 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1164 fixed another bug: if SRV records were looked up and the corresponding
1165 address records were *not* found in the additional section, the port
1166 values from the SRV records were lost.
1168 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1169 using the correct key (the original address) when searching the retry
1170 rules in order to find which one to use for generating the retry hint.
1172 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1173 adding the default one. Similarly, if it contains a Reply-To: header, the
1174 errors_reply_to option, if set, is not used.
1176 PH/19 When calculating a retry time, Exim used to measure the "time since
1177 failure" by looking at the "first failed" field in the retry record. Now
1178 it does not use this if it is later than than the arrival time of the
1179 message. Instead it uses the arrival time. This makes for better
1180 behaviour in cases where some deliveries succeed, thus re-setting the
1181 "first failed" field. An example is a quota failure for a huge message
1182 when small messages continue to be delivered. Without this change, the
1183 "time since failure" will always be short, possible causing more frequent
1184 delivery attempts for the huge message than are intended.
1185 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1187 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1188 $1, $2, $3) because the numerical variables can be reset during some
1189 expansion items (e.g. "match"), thereby losing the authentication data.
1191 PH/21 Make -bV show the size of off_t variables so that the test suite can
1192 decide whether to run tests for quotas > 2G.
1194 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1195 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1196 filecount value is greater than 2G or if a quota value is greater than 2G
1197 on a system where the size of off_t is not greater than 4, a panic error
1200 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1201 never match. The debug and -bh output now contains an explicit error
1202 message indicating a malformed IPv4 address or mask.
1204 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1205 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1206 PH/23 above applies.
1208 PH/25 Do not write to syslog when running in the test harness. The only
1209 occasion when this arises is a failure to open the main or panic logs
1210 (for which there is an explicit test).
1212 PH/26 Added the /no_tell option to "control=freeze".
1214 PH/27 If a host name lookup failed very early in a connection, for example, if
1215 the IP address matched host_lookup and the reverse lookup yielded a name
1216 that did not have a forward lookup, an error message of the form "no IP
1217 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1218 could be logged. Now it outputs the IP address instead of "NULL".
1220 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1221 allows the sender and the authenticated sender to be set when
1222 submitting a message from within Exim. Since child_open_exim() is
1223 documented for local_scan(), the new function should be too.
1225 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1226 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1227 results in an empty string is now treated as unset.
1229 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1231 PH/31 Added sender_verify_fail logging option.
1233 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1234 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1235 tidied the source and removed it altogether.
1237 PH/33 When a queue run was abandoned because the load average was too high, a
1238 log line was always written; now it is written only if the queue_run log
1239 selector is set. In addition, the log line for abandonment now contains
1240 information about the queue run such as the pid. This is always present
1241 in "start" and "stop" lines but was omitted from the "abandon" line.
1243 PH/34 Omit spaces between a header name and the colon in the error message that
1244 is given when verify = headers_syntax fails (if there are lots of them,
1245 the message gets confusing).
1247 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1248 names, as there are now some PTR records that contain slashes. This check
1249 is only to protect against broken name servers that fall over on strange
1250 characters, so the fact that it applies to all lookups doesn't matter.
1252 PH/36 Now that the new test suite is complete, we can remove some of the
1253 special code in Exim that was needed for the old test suite. For example,
1254 sorting DNS records because real resolvers return them in an arbitrary
1255 order. The new test suite's fake resolver always returns records in the
1258 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1259 bounces) except when queue_only is set, to avoid logging races between
1260 the different processes.
1262 PH/38 Panic-die if .include specifies a non-absolute path.
1264 PH/39 A tweak to the "H" retry rule from its user.
1266 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1267 a label. They prevented compilation on older perls.
1269 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1270 a warning to be raised on newish perls.
1272 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1273 on queue. Changes to match documented behaviour of showing count of
1274 messages matching specified criteria.
1276 PH/40 Changed the default ident timeout from 30s to 5s.
1278 PH/41 Added support for the use of login_cap features, on those BSD systems
1279 that have them, for controlling the resources used by pipe deliveries.
1281 PH/42 The content-scanning code uses fopen() to create files in which to put
1282 message data. Previously it was not paying any attention to the mode of
1283 the files. Exim runs with umask(0) because the rest of the code creates
1284 files with open(), and sets the required mode explicitly. Thus, these
1285 files were ending up world-writeable. This was not a big issue, because,
1286 being within the spool directory, they were not world-accessible. I have
1287 created a function called modefopen, which takes an additional mode
1288 argument. It sets umask(777), creates the file, chmods it to the required
1289 mode, then resets the umask. All the relevant calls to fopen() in the
1290 content scanning code have been changed to use this function.
1292 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1293 to 24 hours. This avoids potential overflow problems when processing G
1294 and H retry rules. I suspect nobody ever tinkers with this value.
1296 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1298 PH/45 When the plaintext authenticator is running as a client, the server's
1299 challenges are checked to ensure they are valid base64 strings. By
1300 default, the authentication attempt is cancelled if an invalid string is
1301 received. Setting client_ignore_invalid_base64 true ignores these errors.
1302 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1303 they are received. Thus, the responses can be made to depend on the
1304 challenges. If an invalid string is ignored, an empty string is placed in
1307 PH/46 Messages that are created by the autoreply transport now contains a
1308 References: header, in accordance with RFCs 2822 and 3834.
1310 PH/47 Added authenticated_sender_force to the smtp transport.
1312 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1314 PH/49 Installed latest patch from the Sieve maintainer.
1316 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1317 was also set, the appendfile transport was unnecessarily scanning a
1318 directory of message files (e.g. for maildir delivery) to find the count
1319 of files (along with the size), even though it did not need this
1320 information. It now does the scan only if it needs to find either the
1321 size of the count of files.
1323 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1325 PH/52 Two bugs concerned with error handling when the smtp transport is
1328 (i) Exim was not creating retry information for temporary errors given
1329 for individual recipients after the DATA command when the smtp transport
1330 was used in LMTP mode. This meant that they could be retried too
1331 frequently, and not timed out correctly.
1333 (ii) Exim was setting the flag that allows error details to be returned
1334 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1335 recipients that were returned after the DATA command.
1337 PH/53 This is related to PH/52, but is more general: for any failing address,
1338 when detailed error information was permitted to be returned to the
1339 sender, but the error was temporary, then after the final timeout, only
1340 "retry timeout exceeded" was returned. Now it returns the full error as
1341 well as "retry timeout exceeded".
1343 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1344 do this, and (what is worse) MTAs that accept it.
1346 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1347 will now be deprecated.
1349 PH/56 New os.c-cygwin from the Cygwin maintainer.
1351 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1352 formats (previously only available in exim formats via -bpr, -bpru,
1353 and -bpra. Now also available in native and exiqgrep formats)
1355 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1356 with very large, slow to parse queues
1358 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1360 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1362 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1363 responses to authentication challenges, though it was showing the
1364 challenges; (ii) I've removed the CR characters from the debug output for
1367 PH/58 Allow for the insertion of a newline as well as a space when a string
1368 is turned into more than one encoded-word during RFC 2047 encoding. The
1369 Sieve code now uses this.
1371 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1372 data_4xx, lost_connection, tls_required.
1374 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1375 message was being sent as an SMTP response.
1377 PH/61 Add -l and -k options to exicyclog.
1379 PH/62 When verifying, if an address was redirected to one new address, so that
1380 verification continued, and the new address failed or deferred after
1381 having set something in $address_data, the value of $address_data was not
1382 passed back to the ACL. This was different to the case when no
1383 redirection occurred. The value is now passed back in both cases.
1385 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1386 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1387 use login_cap.h, so on its own it isn't the distinguishing feature. The
1388 new name refers directly to the setclassresources() function.
1390 PH/65 Added configuration files for NetBSD3.
1392 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1394 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1395 is preferred over IPv4.
1397 PH/68 The bounce_return_message and bounce_return_body options were not being
1398 honoured for bounces generated during the reception of non-SMTP messages.
1399 In particular, this applied to messages rejected by the ACL. This bug has
1400 been fixed. However, if bounce_return_message is true and bounce_return_
1401 body is false, the headers that are returned for a non-SMTP message
1402 include only those that have been read before the error was detected.
1403 (In the case of an ACL rejection, they have all been read.)
1405 PH/69 The HTML version of the specification is now built in a directory called
1406 spec_html instead of spec.html, because the latter looks like a path with
1407 a MIME-type, and this confuses some software.
1409 PH/70 Catch two compiler warnings in sieve.c.
1411 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1412 function verify_get_ident() calls ip_connect() to connect a socket, but
1413 if the "connect()" function timed out, ip_connect() used to close the
1414 socket. However, verify_get_ident() also closes the socket later, and in
1415 between Exim writes to the log, which may get opened at this point. When
1416 the socket was closed in ip_connect(), the log could get the same file
1417 descriptor number as the socket. This naturally causes chaos. The fix is
1418 not to close the socket in ip_connect(); the socket should be closed by
1419 the function that creates it. There was only one place in the code where
1420 this was missing, in the iplookup router, which I don't think anybody now
1421 uses, but I've fixed it anyway.
1423 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1424 well as to direct DNS lookups. Otherwise the handling of names in host
1425 lists is inconsistent and therefore confusing.
1431 PH/01 Two changes to the default runtime configuration:
1433 (1) Move the checks for relay_from_hosts and authenticated clients from
1434 after to before the (commented out) DNS black list checks.
1436 (2) Add control=submission to the relay_from_hosts and authenticated
1437 clients checks, on the grounds that messages accepted by these
1438 statements are most likely to be submissions.
1440 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1442 (1) Generate an error if the third argument for the ${prvs expansion is
1445 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1448 (3) Reset the variables that are obtained from the first argument of
1449 ${prvscheck and used in the second argument before leaving the code,
1450 because their memory is reclaimed, so using them afterwards may do
1453 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1454 one (it's much easier than Tom thought :-).
1456 (5) Because of (4), we can now allow for the use of $prvscheck_result
1457 inside the third argument.
1459 PH/03 For some reason, the default setting of PATH when running a command from
1460 a pipe transport was just "/usr/bin". I have changed it to
1463 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1464 anything to be listed in the output from -bV.
1466 PH/05 When a filter generated an autoreply, the entire To: header line was
1467 quoted in the delivery log line, like this:
1469 => >A.N.Other <ano@some.domain> <original@ddress> ...
1471 This has been changed so that it extracts the operative address. There
1472 may be more than one such address. If so, they are comma-separated, like
1475 => >ano@some.domain,ona@other.domain <original@ddress> ...
1477 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1478 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1479 not being looked up in the rDNS to get a host name, Exim was showing the
1480 IP address twice in Received: lines, even though the IP addresses were
1481 identical. For example:
1483 Received: from [1.2.3.4] (helo=[1.2.3.4])
1485 However, if the real host name was known, it was omitting the HELO data
1486 if it matched the actual IP address. This has been tidied up so that it
1487 doesn't show the same IP address twice.
1489 PH/07 When both +timestamp and +memory debugging was on, the value given by
1490 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1491 called by the debug printing, thereby overwriting the timestamp buffer.
1492 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1494 PH/08 When the original message was included in an autoreply transport, it
1495 always said "this is a copy of the message, including all the headers",
1496 even if body_only or headers_only was set. It now gives an appropriate
1499 PH/09 Applied a patch from the Sieve maintainer which:
1501 o fixes some comments
1502 o adds the (disabled) notify extension core
1503 o adds some debug output for the result of if/elsif tests
1504 o points to the current vacation draft in the documentation
1505 and documents the missing references header update
1509 o fixes a bug in processing the envelope test (when testing
1510 multiple envelope elements, the last element determinted the
1513 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1514 Electronic Mail") by including:
1516 Auto-submitted: auto-generated
1518 in the messages that it generates (bounce messages and others, such as
1519 warnings). In the case of bounce messages for non-SMTP mesages, there was
1520 also a typo: it was using "Auto_submitted" (underscore instead of
1521 hyphen). Since every message generated by Exim is necessarily in response
1522 to another message, thes have all been changed to:
1524 Auto-Submitted: auto-replied
1526 in accordance with these statements in the RFC:
1528 The auto-replied keyword:
1530 - SHOULD be used on messages sent in direct response to another
1531 message by an automatic process,
1533 - MUST NOT be used on manually-generated messages,
1535 - MAY be used on Delivery Status Notifications (DSNs) and Message
1536 Disposition Notifications (MDNs),
1538 - MUST NOT be used on messages generated by automatic or periodic
1539 processes, except for messages which are automatic responses to
1542 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1543 to the default Received: header definition.
1545 PH/12 Added log selector acl_warn_skipped (default on).
1547 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1548 variables because (a) they are in the wrong storage pool and (b) even if
1549 they were copied, it wouldn't work properly because of the caching.
1551 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1552 checking when decoding. Apparently there are clients that generate
1553 overlong encoded strings. Why am I not surprised?
1555 PH/15 If the first argument of "${if match_address" was not empty, but did not
1556 contain an "@" character, Exim crashed. Now it writes a panic log message
1557 and treats the condition as false.
1559 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1561 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1562 extension "envelope-auth". The code is finished and in agreement with
1563 other implementations, but there is no documentation so far and in fact,
1564 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1565 not changing the active code.
1567 Print executed "if" and "elsif" statements when debugging is used. This
1568 helps a great deal to understand what a filter does.
1570 Document more things not specified clearly in RFC3028. I had all this
1571 sorted out, when out of a sudden new issues came to my mind. Oops."
1573 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1576 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1577 canonical form (as documented). However, after a host name lookup from
1578 the IP address, check_host() was doing a simple string comparison with
1579 addresses acquired from the DNS when checking that the found name did
1580 have the original IP as one of its addresses. Since any found IPv6
1581 addresses are likely to be in abbreviated form, the comparison could
1582 fail. Luckily, there already exists a function for doing the comparison
1583 by converting both addresses to binary, so now that is used instead of
1584 the text comparison.
1586 PH/20 There was another similar case to PH/19, when a complete host name was
1587 given in a host list; looking up its IP address could give an abbreviated
1588 form, whereas the current host's name might or might not be abbreviated.
1589 The same fix has been applied.
1595 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1596 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1599 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1600 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1602 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1604 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1605 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1606 Also, for "Auto-Submitted", treat anything other than "no" as
1607 non-personal, in accordance with RFC 3834. (Previously it treated
1608 anything starting "auto-" as non-personal.)
1610 TF/01 The control=submission/name=... option had a problem with syntax
1611 errors if the name included a slash character. The /name= option
1612 now slurps the rest of the string, so it can include any characters
1613 but it must come last in the list of options (after /sender_retain
1616 PH/05 Some modifications to the interface to the fake nameserver for the new
1624 TK/01 Added the "success_on_redirect" address verification option. See
1625 NewStuff for rationale and an example.
1627 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1629 PH/02 Patch to exigrep to allow it to work on syslog lines.
1631 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1632 fread() to skip over the body file's header line, because in Cygwin the
1633 header line is locked and is inaccessible.
1635 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1636 co-exist for some time) to make it clear that it is the Exim ID that is
1637 referenced, not the Message-ID: header line.
1639 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1640 string_format() function, because snprintf() does not exist on all
1643 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1644 ${stat: expansion item.
1646 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1647 protocol synchronization error", to keep the pedants happy.
1649 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1650 well as for IRIX systems, when gcc is being used. See the host.c source
1653 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1655 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1658 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1659 transport and to the smtp transport in LMTP mode.
1661 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1663 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1664 run a filter in a subprocess. This could lead to confusion in subsequent
1665 lookups in the parent process. There should also be a search_tidyup() at
1666 the end of the subprocess.
1668 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1669 only if the host matched helo_try_verify_hosts, which caused the
1670 verification to occur when the EHLO/HELO command was issued. The ACL just
1671 tested the remembered result. Now, if a previous verification attempt has
1672 not happened, "verify = helo" does it there and then.
1674 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1676 TK/03 Fix log output including CR from clamd.
1678 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1679 exist provoked a memory error which could cause a segfault.
1681 PH/15 Installed PCRE 6.2
1683 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1685 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1686 of the problem. Specifically, suggested +O2 rather than +O1 for the
1689 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1691 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1692 record, the port was not correctly logged when the outgoing_port log
1693 selector was set (it logged the transort's default port).
1695 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1696 fallback_hosts, and "hosts" in the smtp transport.
1698 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1699 host errors such as "Connection refused".
1701 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1702 authentication with radiusclient 0.4.9:
1704 - Error returned from rc_read_config was caught wrongly
1705 - Username/password not passed on to radius server due to wrong length.
1707 The presumption is that some radiusclient API changes for 4.51/PH/17
1708 were not taken care of correctly. The code is still untested by me (my
1709 Linux distribution still has 0.3.2 of radiusclient), but it was
1710 contributed by a Radius user.
1712 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1713 expanding the "port" option of the smtp transport.
1715 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1716 while reading a MIME header. Thanks to Tom Hughes for a patch.
1718 PH/24 Include config.h inside local_scan.h so that configuration settings are
1721 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1722 that in an ACL for RCPT (for example), you can examine exactly what was
1725 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1726 commands, but it was not correctly comparing the address with the actual
1727 client host address. Thus, it would show the EHLO address in Received:
1728 header lines when this was not necessary.
1730 PH/27 Added the % operator to ${eval:}.
1732 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1733 it should be ignoring failures (because with -C, for example, it has lost
1734 privilege). It wasn't ignoring creation failures other than "already
1737 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1740 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1741 because an input line was too long, either on its own, or by virtue of
1742 too many continuations, the temporary file was not being removed, and the
1743 return code was incorrect.
1745 PH/31 Missing "BOOL" in function definition in filtertest.c.
1747 PH/32 Applied Sieve patches from the maintainer.
1749 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1751 PH/33 Added "verify = not_blind".
1753 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1754 Local/Makefile (with some defaults set). These are used in built scripts
1755 such as exicyclog, but they have never been used in the exim_install
1756 script (though there are many overriding facilities there). I have
1757 arranged that the exim_install script now takes note of these two
1760 PH/35 Installed configuration files for Dragonfly.
1762 PH/36 When a locally submitted message by a trusted user did not contain a
1763 From: header, and the sender address was obtained from -f or from an SMTP
1764 MAIL command, and the trusted user did not use -F to supply a sender
1765 name, $originator_name was incorrectly used when constructing a From:
1766 header. Furthermore, $originator_name was used for submission mode
1767 messages from external hosts without From: headers in a similar way,
1768 which is clearly wrong.
1770 PH/37 Added control=suppress_local_fixups.
1772 PH/38 When log_selector = +received_sender was set, and the addition of the
1773 sender made the log line's construction buffer exactly full, or one byte
1774 less than full, an overflow happened when the terminating "\n" was
1777 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1778 when the result of a list match is failure because a DNS lookup failed.
1780 PH/40 RM_COMMAND is now used in the building process.
1782 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1783 the "build-* directories that it finds.
1785 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1786 address in a domain literal was a prefix of an interface address.
1788 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1789 when verifying a sender address, unless rewrite_headers is false.
1791 PH/44 Wrote a long comment about why errors_to addresses are verified as
1792 recipients, not senders.
1794 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1795 the ratelimit ACL was added.
1797 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1799 PH/46 Added extra information about PostgreSQL errors to the error string.
1801 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1802 avoiding the need to install special zones in a real server. This is
1803 backwards compatible; if it can't find the fake resolver, it drops back.
1804 Thus, both old and new test suites can be run.
1806 TF/02 Added util/ratelimit.pl
1808 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1809 clock is set back in time.
1811 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1812 Candler <B.Candler@pobox.com>.
1814 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1815 OK for addresses that are the result of redirections.
1817 PH/48 A number of further additions for the benefit of the new test suite,
1818 including a fake gethostbyname() that interfaces to the fake DNS resolver
1821 TF/06 The fix for widen_domains has also been applied to qualify_single and
1822 search_parents which are the other dnslookup options that can cause
1825 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1828 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1830 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1831 service" if it encountered an MX record with an empty target hostname.
1832 The message is now "an MX or SRV record indicated no SMTP service".
1834 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1835 if the DNS of the sending site is misconfigured. This is quite a
1836 common situation. This change restores the behaviour of treating a
1837 helo verification defer as a failure.
1839 PH/51 If self=fail was set on a router, the bounce message did not include the
1840 actual error message.
1846 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1848 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1849 command itself ended in error, the underlying message about the transport
1850 filter timeout was being overwritten with the pipe command error. Now the
1851 underlying error message should be appended to the second error message.
1853 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1855 PH/02 Reduce the amount of output that "make" produces by default. Full output
1856 can still be requested.
1858 PH/03 The warning log line about a condition test deferring for a "warn" verb
1859 was being output only once per connection, rather than after each
1860 occurrence (because it was using the same function as for successful
1861 "warn" verbs). This seems wrong, so I have changed it.
1863 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1864 it should not have, which might have caused a crash in the right
1865 circumstances, but probably never did.
1867 PH/04 Installed a modified version of Tony Finch's patch to make submission
1868 mode fix the return path as well as the Sender: header line, and to
1869 add a /name= option so that you can make the user's friendly name appear
1872 TF/03 Added the control = fakedefer ACL modifier.
1874 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1875 Mark Lowes for thorough testing.
1877 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1879 TK/03 Merged latest SRS patch from Miles Wilton.
1881 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1882 with the definition in sysexits.h (which is #included earlier).
1883 Fortunately, Exim does not actually use EX_OK. The code used to try to
1884 preserve the sysexits.h value, by assumimg that macro definitions were
1885 scanned for macro replacements. I have been disabused of this notion,
1886 so now the code just undefines EX_OK before #including unistd.h.
1888 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1889 in the smtp transport. When a block could not be written in a single
1890 write() function, the timeout was being re-applied to each part-write.
1891 This seems wrong - if the receiver was accepting one byte at a time it
1892 would take for ever. The timeout is now adjusted when this happens. It
1893 doesn't have to be particularly precise.
1895 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1896 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1898 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1899 without a domain if the check to <postmaster@domain> fails.
1901 SC/01 Eximstats: added -xls and the ability to specify output files
1902 (patch written by Frank Heydlauf).
1904 SC/02 Eximstats: use FileHandles for outputing results.
1906 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1908 SC/04 Eximstats: fixed display of large numbers with -nvr option
1910 SC/05 Eximstats: fixed merging of reports with empty tables.
1912 SC/06 Eximstats: added the -include_original_destination flag
1914 SC/07 Eximstats: removed tabs and trailing whitespace.
1916 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1918 TK/06 MBOX spool code: Add real "From " MBOX separator line
1919 so the .eml file is really in mbox format (even though
1920 most programs do not really care). Patch from Alex Miller.
1922 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1923 The latter is generated from $received_to and is only set if the
1924 message has one envelope recipient. SA can use these headers,
1925 obviously out-of-the-box. Patch from Alex Miller.
1927 PH/08 The ${def test on a variable was returning false if the variable's
1928 value was "0", contrary to what the specification has always said!
1929 The result should be true unless the variable is empty.
1931 PH/09 The syntax error of a character other than { following "${if
1932 def:variable_name" (after optional whitespace) was not being diagnosed.
1933 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1934 accidental colon was present, for example, could give incorrect results.
1936 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1937 result is used (not including appendfile, where other changes are about
1940 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1941 This involved changing a lot of size variables from int to off_t. It
1942 should work with maildirs and everything.
1944 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1945 spamd dying while we are connected to it.
1947 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1950 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1951 function for BDB 4.3.
1953 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1955 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1956 the sources for PCRE have been re-arranged and more files are now
1959 PH/15 The code I had for printing potentially long long variables in PH/11
1960 above was not the best (it lost precision). The length of off_t variables
1961 is now inspected at build time, and an appropriate printing format (%ld
1962 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1963 to be "long long int" or "long int". This is needed for the internal
1964 formatting function string_vformat().
1966 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1967 the configuration file to be ":syslog", then the script "guesses" where
1968 the logs files are, rather than using the compiled in default. In our
1969 case the guess is not the same as the compiled default, so the script
1970 suddenly stopped working when I started to use syslog. The patch checks
1971 to see if log_file_path is "". If so, it attempts to read it from exim
1972 with no configuration file to get the compiled in version, before it
1973 falls back to the previous guessing code."
1975 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1976 implementing BATV in an Exim configuration. See NewStuff for the gory
1979 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1980 Makefile that are specific to HP-UX.
1982 PH/18 If the "use_postmaster" option was set for a recipient callout together
1983 with the "random" option, the postmaster address was used as the MAIL
1984 FROM address for the random test, but not for the subsequent recipient
1985 test. It is now used for both.
1987 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1988 patch removes a few documentation additions to RFC 3028, because the
1989 latest draft now contains them. It adds the new en;ascii-case comparator
1990 and a new error check for 8bit text in MIME parts. Comparator and
1991 require names are now matched exactly. I enabled the subaddress
1992 extension, but it is not well tested yet (read: it works for me)."
1994 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1995 rework some of the code of TK/09 above to avoid the hardwired use of
1996 "%lld" and "long long". Replaced the call to snprintf() with a call to
1999 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2000 records point to non-existent hosts", "retry timeout exceeded", and
2001 "retry time not reached for any host after a long failure period".
2003 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2004 experimental DomainKeys support:
2006 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2007 (2) On an error such as an illegally used "control", the wrong name for
2008 the control was given.
2010 These problems did NOT occur unless DomainKeys support was compiled.
2012 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2014 PH/24 Added ${if match_ip condition.
2016 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2017 databases so that it will be absolutely obvious if a crash occurs in the
2018 DB library. This is a regular occurrence (often caused by mis-matched
2021 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2022 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2023 picked up on a user's system that detects such things. There doesn't seem
2024 to be a gcc warning option for this - only an attribute that has to be
2025 put on the function's prototype. It seems that in Fedora Core 4 they have
2026 set this on a number of new functions. No doubt there will be more in due
2029 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2030 specify a transport. However, if an address that was verified by such a
2031 router was the subject of a callout, Exim crashed because it tried to
2032 read the rcpt_include_affixes from the non-existent transport. Now it
2033 just assumes that the setting of that option is false. This bug was
2034 introduced by 4.51/PH/31.
2036 PH/28 Changed -d+all to exclude +memory, because that information is very
2037 rarely of interest, but it makes the output a lot bigger. People tend to
2038 do -d+all out of habit.
2040 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2041 code in os-type was giving problems when libc.so lives in lib64, like on
2044 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2045 aren't the modern standard, and it seems that some systems' include files
2046 don't always have them. Exim was already checking for some of the newer
2047 ones like T_AAAA, and defining it itself. I've added checks for all the
2048 record types that Exim uses.
2050 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2051 not automatically generating a new one, as it is supposed to. This
2052 prevented TLS from working. If the file did exist, but contained invalid
2053 data, a new version was generated, as expected. It was only the case of a
2054 non-existent file that was broken.
2056 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2057 with a change in libdomainkeys > 0.64.
2059 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2060 from DNS. If the selector record carries the flag, it now has
2061 precedence over the domain-wide flag.
2063 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2065 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2066 the use of a port name that isn't defined in /etc/services) occurred, the
2067 message was deferred as in a normal delivery, and thus remained on the
2068 spool, instead of being failed because of the mua_wrapper setting. This
2069 is now fixed, and I tidied up some of the mua_wrapper messages at the
2072 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2073 the messages in a hash of arrays rather than using individual hashes.
2074 This is a bit cleaner and results in dramatic memory savings, albeit
2075 at a slight CPU cost.
2077 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2078 as requested by Marc Sherman.
2080 SC/10 Eximstats: added histograms for user specified patterns as requested
2083 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2085 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2086 fopen() in the content-scanning modules that did not already have it.
2092 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2093 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2095 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2097 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2099 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2100 to test Sieve filters that use "vacation".
2102 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2103 that changes the way the GnuTLS parameters are stored in the cache file.
2104 The new format can be generated externally. For backward compatibility,
2105 if the data in the cache doesn't make sense, Exim assumes it has read an
2106 old-format file, and it generates new data and writes a new file. This
2107 means that you can't go back to an older release without removing the
2110 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2111 work if there are any delivery delays because "one_time" forces the
2112 parent to be marked "delivered", so its unseen clone is never tried
2113 again. For this reason, Exim now forbids the simultaneous setting of
2116 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2117 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2118 entry if you want to know the details. The fix, however, neglected to
2119 consider the case when local delivery batching is involved. The test for
2120 "previously delivered" was not happening when checking to see if an
2121 address could be batched with a previous (undelivered) one; under
2122 certain circumstances this could lead to multiple deliveries to the same
2125 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2126 in its include files, and this causes problems building Exim.
2128 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2129 header_syntax) but Exim was just ignoring anything given after a slash.
2130 In particular, this caused confusion with an attempt to use "verify =
2131 reverse_host_lookup/defer_ok". An error is now given when options are
2132 supplied for verify items that do not have them. (Maybe reverse_host_
2133 lookup should have a defer_ok option, but that's a different point.)
2135 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2136 defined by RFC 821) to 2048, because there were problems with some AUTH
2137 commands, and RFC 1869 says the size should be increased for extended
2138 SMTP commands that take arguments.
2140 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2143 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2144 "unknown" error; now it says that the functionality isn't in the binary.
2146 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2147 an address' error message when a string expansion fails (syntax or
2148 whatever). Otherwise the password may appear in the log. Following change
2149 PH/42 below, there is no longer a chance of it appearing in a bounce
2152 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2154 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2155 of its addresses was ever tried. (Bugzilla bug #2.)
2157 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2158 the result incorrectly in the debug output. (It correctly added a newline
2159 to what was transported.)
2161 TF/01 Added $received_time.
2163 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2164 commented out examples of how to interface to a virus scanner and to
2165 SpamAssassin. Also added commented examples of av_scanner and
2166 spamd_address settings.
2168 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2169 and controls are allowed in which ACLs. There were a couple of minor
2170 errors. Some of the entries in the conditions table (which is a table of
2171 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2172 negation of where the condition IS allowed.
2174 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2176 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2177 header file does not have a version number, so I've had to invent a new
2178 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2179 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2180 radiusclient), but it was contributed by a Radius user.
2182 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2183 files or directories, for OpenSSL.
2185 PH/19 When an Exim process that is running as root has to create an Exim log
2186 file, it does so in a subprocess that runs as exim:exim so as to get the
2187 ownership right at creation (otherwise, other Exim processes might see
2188 the file with the wrong ownership). There was no test for failure of this
2189 fork() call, which would lead to the process getting stuck as it waited
2190 for a non-existent subprocess. Forks do occasionally fail when resources
2191 run out. I reviewed all the other calls to fork(); they all seem to check
2194 PH/20 When checking for unexpected SMTP input at connect time (before writing
2195 the banner), Exim was not dealing correctly with a non-positive return
2196 from the read() function. If the client had disconnected by this time,
2197 the result was a log entry for a synchronization error with an empty
2198 string after "input=" when read() returned zero. If read() returned -1
2199 (an event I could not check), uninitialized data bytes were printed.
2200 There were reports of junk text (parts of files, etc) appearing after
2203 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2205 PH/22 Added support for macro redefinition, and (re)definition in between
2206 driver and ACL definitions.
2208 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2209 forgetting to use the resulting value; it was using the unexpanded value.
2211 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2212 hadn't been configured. The fix is from Juergen Kreileder, who
2213 understands it better than I do:
2215 "Here's what I see happening with three configured cyrus_sasl
2216 authenticators configured (plain, login, cram-md5):
2218 On startup auth_cyrus_sasl_init() gets called for each of these.
2219 This means three calls to sasl_listmech() without a specified mech_list.
2220 => SASL tests which mechs of all available mechs actually work
2221 => three warnings about OTP not working
2222 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2224 With the patch, sasl_listmech() also gets called three times. But now
2225 SASL's mech_list option is set to the server_mech specified in the the
2226 authenticator. Or in other words, the answer from sasl_listmech()
2227 gets limited to just the mech you're testing for (which is different
2229 => the return list contains just 'plain' or 'login', 'cram-md5' or
2230 nothing depending on the value of ob->server_mech.
2232 I've just tested the patch: Authentication still works fine,
2233 unavailable mechs specified in the exim configuration are still
2234 caught, and the auth.log warnings about OTP are gone."
2236 PH/25 When debugging is enabled, the contents of the command line are added
2237 to the debugging output, even when log_selector=+arguments is not
2240 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2241 answer is "GNU", and only if the return is "GNU/something" is the answer
2244 PH/27 $acl_verify_message is now set immediately after the failure of a
2245 verification in an ACL, and so is available in subsequent modifiers. In
2246 particular, the message can be preserved by coding like this:
2248 warn !verify = sender
2249 set acl_m0 = $acl_verify_message
2251 Previously, $acl_verify_message was set only while expanding "message"
2252 and "log_message" when a very denied access.
2254 PH/28 Modified OS/os.c-Linux with
2256 -#ifndef OS_LOAD_AVERAGE
2257 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2259 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2260 nomenclature these days.)
2262 PH/29 Installed patch from the Sieve maintainer that adds the options
2263 sieve_useraddress and sieve_subaddress to the redirect router.
2265 PH/30 In these circumstances:
2266 . Two addresses routed to the same list of hosts;
2267 . First host does not offer TLS;
2268 . First host accepts first address;
2269 . First host gives temporary error to second address;
2270 . Second host offers TLS and a TLS session is established;
2271 . Second host accepts second address.
2272 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2273 and peerdn, if requested) that were in fact used only for the second
2276 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2277 attention to any local part prefix or suffix that was matched by the
2278 router that accepted the address. It now behaves in the same way as it
2279 does for delivery: the affixes are removed from the local part unless
2280 rcpt_include_affixes is set on the transport.
2282 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2283 timeout during the DATA phase of an incoming message.
2285 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2286 applied a patch sanctioned by the Sieve maintainer.
2288 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2289 the uid or gid is negative. A case of a negative gid caused this to be
2290 noticed. The fix allows for either to be negative.
2292 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2293 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2296 PH/36 The change PH/12 above was broken. Fixed it.
2298 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2299 the grounds that routing the same address twice would always produce the
2300 same answer. This might have been true once, but it is certainly no
2301 longer true now. Routing a child address may depend on the previous
2302 routing that produced that child. Some complicated redirection strategies
2303 went wrong when messages had multiple recipients, and made Exim's
2304 behaviour dependent on the order in which the addresses were given.
2306 I have moved the duplicate checking until after the routing is complete.
2307 Exim scans the addresses that are assigned to local and remote
2308 transports, and removes any duplicates. This means that more work will be
2309 done, as duplicates will always all be routed, but duplicates are
2310 presumably rare, so I don't expect this is of any significance.
2312 For deliveries to pipes, files, and autoreplies, the duplicate checking
2313 still happens during the routing process, since they are not going to be
2316 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2317 It corrects a timeout issue with spamd. This is Ian's comment: "The
2318 background is that sometimes spamd either never reads data from a
2319 connection it has accepted, or it never writes response data. The exiscan
2320 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2321 blindly assumes that writes won't block so it may never time out."
2323 PH/39 Allow G after quota size as well as K and M.
2325 PH/40 The value set for $authenticated_id in an authenticator may not contain
2326 binary zeroes or newlines because the value is written to log lines and
2327 to spool files. There was no check on this. Now the value is run through
2328 the string_printing() function so that such characters are converted to
2329 printable escape sequences.
2331 PH/41 $message_linecount is a new variable that contains the total number of
2332 lines in the message. Compare $body_linecount, which is the count for the
2335 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2336 bounce and delay warning messages, except in certain special cases, which
2339 (a) An SMTP error message from a remote host;
2340 (b) A message specified in a :fail: redirection;
2341 (c) A message specified in a "fail" command in a system filter;
2342 (d) A message specified in a FAIL return from the queryprogram router;
2343 (e) A message specified by the cannot_route_message router option.
2345 In these cases only, Exim does include the error details in bounce and
2346 warning messages. There are also a few cases where bland messages such
2347 as "unrouteable address" or "local delivery error" are given.
2349 PH/43 $value is now also set for the "else" part of a ${run expansion.
2351 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2352 being worked on, but at least Exim now implements the latest version to
2355 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2356 process to complete was treated as a delivery failure, a timeout while
2357 writing the message to the pipe was logged, but erroneously treated as a
2358 successful delivery. Such timeouts include transport filter timeouts. For
2359 consistency with the overall process timeout, these timeouts are now
2360 treated as errors, giving rise to delivery failures by default. However,
2361 there is now a new Boolean option for the pipe transport called
2362 timeout_defer, which, if set TRUE, converts the failures into defers for
2363 both kinds of timeout. A transport filter timeout is now identified in
2366 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2367 systems where "make" and "gmake" are different, calling "gmake" at top
2368 level broke things. I've arranged for the value of $(MAKE) to be passed
2369 from the Makefile to this script so that it can call the same version of
2373 A note about Exim versions 4.44 and 4.50
2374 ----------------------------------------
2376 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2377 changes of various kinds. As a consequence, a big documentation update was
2378 needed. This delayed the release for rather longer than seemed good, especially
2379 in the light of a couple of (minor) security issues. Therefore, the changes
2380 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2381 release. So 4.44 and 4.50 are in effect two different branches that both start
2384 I have left the 4.50 change log unchanged; it contains all the changes since
2385 4.43. The change log for 4.44 is below; many of its items are identical to
2386 those for 4.50. This seems to be the most sensible way to preserve the
2387 historical information.
2393 1. Minor wording change to the doc/README.SIEVE file.
2395 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2396 computation of the current number of files was incorrect.
2398 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2399 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2402 4. Give more explanation in the error message when the command for a transport
2403 filter fails to execute.
2405 5. There are several places where Exim runs a non-Exim command in a
2406 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2407 was being done only for the command run by the queryprogram router. It is
2408 now done for all such subprocesses. The other cases are: ${run, transport
2409 filters, and the commands run by the lmtp and pipe transports.
2411 6. Added CONFIGURE_GROUP build-time option.
2413 7. Some older OS have a limit of 256 on the maximum number of file
2414 descriptors. Exim was using setrlimit() to set 1000 as a large value
2415 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2416 systems. I've change it so that if it can't get 1000, it tries for 256.
2418 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2419 was an oversight, and furthermore, ever since the addition of extra
2420 controls (e.g. 4.43/32), the checks on when to allow different forms of
2421 "control" were broken. There should now be diagnostics for all cases when a
2422 control that does not make sense is encountered.
2424 9. Added the /retain_sender option to "control=submission".
2426 10. $recipients is now available in the predata ACL (oversight).
2428 11. Tidy the search cache before the fork to do a delivery from a message
2429 received from the command line. Otherwise the child will trigger a lookup
2430 failure and thereby defer the delivery if it tries to use (for example) a
2431 cached ldap connection that the parent has called unbind on.
2433 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2434 of $address_data from the recipient verification was clobbered by the
2435 sender verification.
2437 13. The value of address_data from a sender verification is now available in
2438 $sender_address_data in subsequent conditions in the ACL statement.
2440 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2442 15. Added a new option "connect=<time>" to callout options, to set a different
2445 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2446 was its contents. (It was OK if the option was not defined at all.)
2448 17. A "Completed" log line is now written for messages that are removed from
2449 the spool by the -Mrm option.
2451 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2452 information about exactly what failed.
2454 19. Added -dd to debug only the daemon process.
2456 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2457 handles timeouts, both on the server side and network timeouts. Renamed the
2458 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2460 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2461 It is now set to "smtps".
2463 22. $host_address is now set to the target address during the checking of
2464 ignore_target_hosts.
2466 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2467 being passed; this would have caused $sender_host_name to have been used if
2468 matching the list had actually called for a host name (not very likely,
2469 since this list is usually IP addresses). A host name is now passed as
2472 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2473 code that specifies a non-restarting handler (typically sigaction() in
2474 modern systems) in an attempt to fix a rare and obscure crash bug.
2476 25. Narrowed the window for a race in the daemon that could cause it to ignore
2477 SIGCHLD signals. This is not a major problem, because they are used only to
2478 wake it up if nothing else does.
2480 26. A malformed maildirsize file could cause Exim to calculate negative values
2481 for the mailbox size or file count. Odd effects could occur as a result.
2482 The maildirsize information is now recalculated if the size or filecount
2485 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2486 support for a long time. Removed HAVE_SYS_VFS_H.
2488 28. Installed the latest version of exipick from John Jetmore.
2490 29. In an address list, if the pattern was not a regular expression, an empty
2491 subject address (from a bounce message) matched only if the pattern was an
2492 empty string. Non-empty patterns were not even tested. This was the wrong
2493 because it is perfectly reasonable to use an empty address as part of a
2494 database query. An empty address is now tested by patterns that are
2495 lookups. However, all the other forms of pattern expect the subject to
2496 contain a local part and a domain, and therefore, for them, an empty
2497 address still always fails if the pattern is not itself empty.
2499 30. Exim went into a mad DNS loop when attempting to do a callout where the
2500 host was specified on an smtp transport, and looking it up yielded more
2501 than one IP address.
2503 31. Re-factored the code for checking spool and log partition space into a
2504 function that finds that data and another that does the check. The former
2505 is then used to implement four new variables: $spool_space, $log_space,
2506 $spool_inodes, and $log_inodes.
2508 32. The RFC2047 encoding function was originally intended for short strings
2509 such as real names; it was not keeping to the 75-character limit for
2510 encoded words that the RFC imposes. It now respects the limit, and
2511 generates multiple encoded words if necessary. To be on the safe side, I
2512 have increased the buffer size for the ${rfc2047: expansion operator from
2515 33. It is now permitted to omit both strings after an "if" condition; if the
2516 condition is true, the result is "true". As before, when the second string
2517 is omitted, a false condition yields an empty string. This makes it less
2518 cumbersome to write custom ACL and router conditions.
2520 34. Failure to deliver a bounce message always caused it to be frozen, even if
2521 there was an errors_to setting on the router. The errors_to setting is now
2524 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2525 canonical form (fully expanded) before being placed in
2526 $sender_host_address.
2528 36. The table in the code that translates DNS record types into text (T_A to
2529 "A" for instance) was missing entries for NS and CNAME. It is just possible
2530 that this could have caused confusion if both these types were looked up
2531 for the same domain, because the text type is used as part of Exim's
2532 per-process caching. But the chance of anyone hitting this buglet seems
2535 37. The dnsdb lookup has been extended in a number of ways.
2537 (1) There is a new type, "zns", which walks up the domain tree until it
2538 finds some nameserver records. It should be used with care.
2540 (2) There is a new type, "mxh", which is like "mx" except that it returns
2541 just the host names, not the priorities.
2543 (3) It is now possible to give a list of domains (or IP addresses) to be
2544 looked up. The behaviour when one of the lookups defers can be
2545 controlled by a keyword.
2547 (4) It is now possible to specify the separator character for use when
2548 multiple records are returned.
2550 38. The dnslists ACL condition has been extended: it is now possible to supply
2551 a list of IP addresses and/or domains to be looked up in a particular DNS
2554 39. Added log_selector=+queue_time_overall.
2556 40. When running the queue in the test harness, wait just a tad after forking a
2557 delivery process, to get repeatability of debugging output.
2559 41. Include certificate and key file names in error message when GnuTLS fails
2560 to set them up, because the GnuTLS error message doesn't include the name
2561 of the failing file when there is a problem reading it.
2563 42. Allow both -bf and -bF in the same test run.
2565 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2567 44. The "Exiscan patch" is now merged into the mainline Exim source.
2569 45. Sometimes the final signoff response after QUIT could fail to get
2570 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2571 < 0 before doing a fflush(). This bug looks as though it goes back to the
2572 introduction of TLS in release 3.20, but "sometimes" must have been rare
2573 because the tests only now provoked it.
2575 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2576 (this can affect the format of dates).
2578 47. exim_tidydb, when checking for the continued existence of a message for
2579 which it has found a message-specific retry record, was not finding
2580 messages that were in split spool directories. Consequently, it was
2581 deleting retry records that should have stayed in existence.
2583 48. Steve fixed some bugs in eximstats.
2585 49. The SPA authentication driver was not abandoning authentication and moving
2586 on to the next authenticator when an expansion was forced to fail,
2587 contradicting the general specification for all authenticators. Instead it
2588 was generating a temporary error. It now behaves as specified.
2590 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2591 (the order specifies the preference for clients). The order is now AES256,
2592 AES128, 3DES, ARCFOUR128.
2594 51. Small patch to Sieve code - explicitly set From: when generating an
2597 52. Exim crashed if a remote delivery caused a very long error message to be
2598 recorded - for instance if somebody sent an entire SpamAssassin report back
2599 as a large number of 550 error lines. This bug was coincidentally fixed by
2600 increasing the size of one of Exim's internal buffers (big_buffer) that
2601 happened as part of the Exiscan merge. However, to be on the safe side, I
2602 have made the code more robust (and fixed the comments that describe what
2605 53. Now that there can be additional text after "Completed" in log lines (if
2606 the queue_time_overall log selector is set), a one-byte patch to exigrep
2607 was needed to allow it to recognize "Completed" as not the last thing in
2610 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2611 patch that reportedly fixes this has been added. I am not expert enough to
2612 create a test for it. This is what the patch creator wrote:
2614 "I found a little strange behaviour of ldap code when working with
2615 Windows 2003 AD Domain, where users was placed in more than one
2616 Organization Units. When I tried to give exim partial DN, the exit code
2617 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2618 But simultaneously result of request was absolutely normal ldap result,
2619 so I produce this patch..."
2621 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2622 REFERENCE, so I have modified the code to exclude the patch when that macro
2625 55. Some experimental protocols are using DNS PTR records for new purposes. The
2626 keys for these records are domain names, not reversed IP addresses. The
2627 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2628 leaves it alone. Component reversal etc. now happens only for IP addresses.
2631 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2633 57. Double the size of the debug message buffer (to 2048) so that more of very
2634 long debug lines gets shown.
2636 58. The exicyclog utility now does better if the number of log files to keep
2637 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2639 59. Two changes related to the smtp_active_hostname option:
2641 (1) $smtp_active_hostname is now available as a variable.
2642 (2) The default for smtp_banner uses $smtp_active_hostname instead
2643 of $primary_hostname.
2645 60. The host_aton() function is supposed to be passed a string that is known
2646 to be a valid IP address. However, in the case of IPv6 addresses, it was
2647 not checking this. This is a hostage to fortune. Exim now panics and dies
2648 if the condition is not met. A case was found where this could be provoked
2649 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2650 components; fortuitously, this particular loophole had already been fixed
2651 by change 4.50/55 above.
2653 If there are any other similar loopholes, the new check in host_aton()
2654 itself should stop them being exploited. The report I received stated that
2655 data on the command line could provoke the exploit when Exim was running as
2656 exim, but did not say which command line option was involved. All I could
2657 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2658 running as the user.
2661 61. There was a buffer overflow vulnerability in the SPA authentication code
2662 (which came originally from the Samba project). I have added a test to the
2663 spa_base64_to_bits() function which I hope fixes it.
2666 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2667 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2669 63. The daemon start-up calls getloadavg() while still root for those OS that
2670 need the first call to be done as root, but it missed one case: when
2671 deliver_queue_load_max is set with deliver_drop_privilege. This is
2672 necessary for the benefit of the queue runner, because there is no re-exec
2673 when deliver_drop_privilege is set.
2675 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2676 This has been fixed.
2678 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2679 was in use, was not putting the data itself into the right store pool;
2680 consequently, it could be overwritten for a subsequent message in the same
2681 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2684 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2686 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2687 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2688 the function were treating the return as a boolean value, which happened to
2689 work because 0=false and not-0=true, but is not correct code.
2691 68. The host_aton() function was not handling scoped IPv6 addresses (those
2692 with, for example, "%eth0" on the end) correctly.
2694 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2695 negated items (that is, ~something) in unsigned ints. Some compilers
2696 apparently mutter when there is no cast.
2698 70. If an address verification called from an ACL failed, and did not produce a
2699 user-specific message (i.e. there was only a "system" message), nothing was
2700 put in $acl_verify_message. In this situation, it now puts the system
2703 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2704 session; change 4.31/43 added the unwanted input to the log line - except
2705 that it did not do this in the start of session case. It now does.
2707 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2708 This is wrong and can cause the other end to generate a synchronization
2709 error if it is another Exim or anything else that does the synchronization
2710 check. A QUIT command is no longer sent after a timeout.
2712 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2713 during host lookups.
2715 74. The defer_ok option of callout verification was not working if it was used
2716 when verifying addresses in header lines, that is, for this case:
2718 verify = header_sender/callout=defer_ok
2720 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2721 those file descriptors could be used for SMTP connections. If anything
2722 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2723 could be sent to the SMTP client, causing chaos. The daemon now opens
2724 stdin, stdout, and stderr to /dev/null when it puts itself into the
2727 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2728 log by default. The user can override this with suitable Perl magic.
2730 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2731 the log message when discard triggers, was not working for the DATA ACL or
2732 for the non-SMTP ACL.
2734 78. Error message wording change in sieve.c.
2736 79. If smtp_accept_max_per_host was set, the number of connections could be
2737 restricted to fewer than expected, because the daemon was trying to set up
2738 a new connection before checking whether the processes handling previous
2739 connections had finished. The check for completed processes is now done
2740 earlier. On busy systems, this bug wouldn't be noticed because something
2741 else would have woken the daemon, and it would have reaped the completed
2744 80. If a message was submitted locally by a user whose login name contained one
2745 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2746 It caused a spool format error. I have fixed the spool reading code. A
2747 related problem was that the "from" clause in the Received: line became
2748 illegal because of the space(s). It is now covered by ${quote_local_part.
2750 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2753 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2755 83. Patch from Sieve maintainer to fix unterminated string problem in
2756 "vacation" handling.
2758 84. Some minor changes to the Linux configuration files to help with other
2759 OS variants using glibc.
2761 85. One more patch for Sieve to update vacation handling to latest spec.
2764 ----------------------------------------------------
2765 See the note above about the 4.44 and 4.50 releases.
2766 ----------------------------------------------------
2772 1. Change 4.43/35 introduced a bug that caused file counts to be
2773 incorrectly computed when quota_filecount was set in an appendfile
2776 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2777 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2780 3. Give more explanation in the error message when the command for a transport
2781 filter fails to execute.
2783 4. There are several places where Exim runs a non-Exim command in a
2784 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2785 was being done only for the command run by the queryprogram router. It is
2786 now done for all such subprocesses. The other cases are: ${run, transport
2787 filters, and the commands run by the lmtp and pipe transports.
2789 5. Some older OS have a limit of 256 on the maximum number of file
2790 descriptors. Exim was using setrlimit() to set 1000 as a large value
2791 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2792 systems. I've change it so that if it can't get 1000, it tries for 256.
2794 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2795 was an oversight, and furthermore, ever since the addition of extra
2796 controls (e.g. 4.43/32), the checks on when to allow different forms of
2797 "control" were broken. There should now be diagnostics for all cases when a
2798 control that does not make sense is encountered.
2800 7. $recipients is now available in the predata ACL (oversight).
2802 8. Tidy the search cache before the fork to do a delivery from a message
2803 received from the command line. Otherwise the child will trigger a lookup
2804 failure and thereby defer the delivery if it tries to use (for example) a
2805 cached ldap connection that the parent has called unbind on.
2807 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2808 of $address_data from the recipient verification was clobbered by the
2809 sender verification.
2811 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2812 was its contents. (It was OK if the option was not defined at all.)
2814 11. A "Completed" log line is now written for messages that are removed from
2815 the spool by the -Mrm option.
2817 12. $host_address is now set to the target address during the checking of
2818 ignore_target_hosts.
2820 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2821 being passed; this would have caused $sender_host_name to have been used if
2822 matching the list had actually called for a host name (not very likely,
2823 since this list is usually IP addresses). A host name is now passed as
2826 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2827 code that specifies a non-restarting handler (typically sigaction() in
2828 modern systems) in an attempt to fix a rare and obscure crash bug.
2830 15. Narrowed the window for a race in the daemon that could cause it to ignore
2831 SIGCHLD signals. This is not a major problem, because they are used only to
2832 wake it up if nothing else does.
2834 16. A malformed maildirsize file could cause Exim to calculate negative values
2835 for the mailbox size or file count. Odd effects could occur as a result.
2836 The maildirsize information is now recalculated if the size or filecount
2839 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2840 support for a long time. Removed HAVE_SYS_VFS_H.
2842 18. Updated exipick to current release from John Jetmore.
2844 19. Allow an empty sender to be matched against a lookup in an address list.
2845 Previously the only cases considered were a regular expression, or an
2848 20. Exim went into a mad DNS lookup loop when doing a callout where the
2849 host was specified on the transport, if the DNS lookup yielded more than
2852 21. The RFC2047 encoding function was originally intended for short strings
2853 such as real names; it was not keeping to the 75-character limit for
2854 encoded words that the RFC imposes. It now respects the limit, and
2855 generates multiple encoded words if necessary. To be on the safe side, I
2856 have increased the buffer size for the ${rfc2047: expansion operator from
2859 22. Failure to deliver a bounce message always caused it to be frozen, even if
2860 there was an errors_to setting on the router. The errors_to setting is now
2863 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2864 canonical form (fully expanded) before being placed in
2865 $sender_host_address.
2867 24. Updated eximstats to version 1.33
2869 25. Include certificate and key file names in error message when GnuTLS fails
2870 to set them up, because the GnuTLS error message doesn't include the name
2871 of the failing file when there is a problem reading it.
2873 26. Expand error message when OpenSSL has problems setting up cert/key files.
2876 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2877 (this can affect the format of dates).
2879 28. exim_tidydb, when checking for the continued existence of a message for
2880 which it has found a message-specific retry record, was not finding
2881 messages that were in split spool directories. Consequently, it was
2882 deleting retry records that should have stayed in existence.
2884 29. eximstats updated to version 1.35
2885 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2886 1.35 - bugfix such that pie charts by volume are generated correctly
2888 30. The SPA authentication driver was not abandoning authentication and moving
2889 on to the next authenticator when an expansion was forced to fail,
2890 contradicting the general specification for all authenticators. Instead it
2891 was generating a temporary error. It now behaves as specified.
2893 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2894 (the order specifies the preference for clients). The order is now AES256,
2895 AES128, 3DES, ARCFOUR128.
2897 31. Small patch to Sieve code - explicitly set From: when generating an
2900 32. Exim crashed if a remote delivery caused a very long error message to be
2901 recorded - for instance if somebody sent an entire SpamAssassin report back
2902 as a large number of 550 error lines. This bug was coincidentally fixed by
2903 increasing the size of one of Exim's internal buffers (big_buffer) that
2904 happened as part of the Exiscan merge. However, to be on the safe side, I
2905 have made the code more robust (and fixed the comments that describe what
2908 33. Some experimental protocols are using DNS PTR records for new purposes. The
2909 keys for these records are domain names, not reversed IP addresses. The
2910 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2911 leaves it alone. Component reversal etc. now happens only for IP addresses.
2914 34. The host_aton() function is supposed to be passed a string that is known
2915 to be a valid IP address. However, in the case of IPv6 addresses, it was
2916 not checking this. This is a hostage to fortune. Exim now panics and dies
2917 if the condition is not met. A case was found where this could be provoked
2918 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2919 components; fortuitously, this particular loophole had already been fixed
2920 by change 4.50/55 or 4.44/33 above.
2922 If there are any other similar loopholes, the new check in host_aton()
2923 itself should stop them being exploited. The report I received stated that
2924 data on the command line could provoke the exploit when Exim was running as
2925 exim, but did not say which command line option was involved. All I could
2926 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2927 running as the user.
2930 35. There was a buffer overflow vulnerability in the SPA authentication code
2931 (which came originally from the Samba project). I have added a test to the
2932 spa_base64_to_bits() function which I hope fixes it.
2935 36. The daemon start-up calls getloadavg() while still root for those OS that
2936 need the first call to be done as root, but it missed one case: when
2937 deliver_queue_load_max is set with deliver_drop_privilege. This is
2938 necessary for the benefit of the queue runner, because there is no re-exec
2939 when deliver_drop_privilege is set.
2941 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2942 was in use, was not putting the data itself into the right store pool;
2943 consequently, it could be overwritten for a subsequent message in the same
2944 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2947 38. Sometimes the final signoff response after QUIT could fail to get
2948 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2949 < 0 before doing a fflush(). This bug looks as though it goes back to the
2950 introduction of TLS in release 3.20, but "sometimes" must have been rare
2951 because the tests only now provoked it.
2957 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2958 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2959 Now it returns an integer. A number of calls were still expecting a T/F
2960 return. Fortuitously, in all cases, the tests worked in OK situations,
2961 which is the norm. However, things would have gone wrong on any write
2962 failures on the smtp file descriptor. This function is used when sending
2963 messages over SMTP and also when doing verify callouts.
2965 2. When Exim is called to do synchronous delivery of a locally submitted
2966 message (the -odf or -odi options), it no longer closes stderr before doing
2969 3. Implemented the mua_wrapper option.
2971 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2973 5. Implemented the functions header_remove(), header_testname(),
2974 header_add_at_position(), and receive_remove_recipient(), and exported them
2977 6. If an ACL "warn" statement specified the addition of headers, Exim already
2978 inserted X-ACL-Warn: at the start if there was no header name. However, it
2979 was not making this test for the second and subsequent header lines if
2980 there were newlines in the string. This meant that an invalid header could
2981 be inserted if Exim was badly configured.
2983 7. Allow an ACL "warn" statement to add header lines at the start or after all
2984 the Received: headers, as well as at the end.
2986 8. Added the rcpt_4xx retry error code.
2988 9. Added postmaster_mailfrom=xxx to callout verification option.
2990 10. Added mailfrom=xxxx to the callout verification option, for verify=
2993 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2994 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2996 12. Inserted some casts to stop certain compilers warning when using pointer
2997 differences as field lengths or precisions in printf-type calls (mostly
2998 affecting debugging statements).
3000 13. Added optional readline() support for -be (dynamically loaded).
3002 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3003 same clock tick as a message's arrival, so that its received time was the
3004 same as the "first fail" time on the retry record, and that message
3005 remained on the queue past the ultimate address timeout, every queue runner
3006 would try a delivery (because it was past the ultimate address timeout) but
3007 after another failure, the ultimate address timeout, which should have then
3008 bounced the address, did not kick in. This was a "< instead of <=" error;
3009 in most cases the first failure would have been in the next clock tick
3010 after the received time, and all would be well.
3012 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3013 being recognized when the domain list was tested by the match_domain
3014 condition in an expansion string.
3016 16. Added the ${str2b64: operator.
3018 17. Exim was always calling setrlimit() to set a large limit for the number of
3019 processes, without checking whether the existing limit was already
3020 adequate. (It did check for the limit on file descriptors.) Furthermore,
3021 errors from getrlimit() and setrlimit() were being ignored. Now they are
3022 logged to the main and panic logs, but Exim does carry on, to try to do its
3023 job under whatever limits there are.
3025 18. Imported PCRE 5.0.
3027 19. Trivial typo in log message " temporarily refused connection" (the leading
3030 20. If the log selector return_path_on_delivery was set and an address was
3031 redirected to /dev/null, the delivery process crashed because it assumed
3032 that a return path would always be set for a "successful" delivery. In this
3033 case, the whole delivery is bypassed as an optimization, and therefore no
3036 21. Internal re-arrangement: the function for sending a challenge and reading
3037 a response while authentication was assuming a zero-terminated challenge
3038 string. It's now changed to take a pointer and a length, to allow for
3039 binary data in such strings.
3041 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3043 23. Exim was not respecting finduser_retries when seeking the login of the
3044 uid under which it was called; it was always trying 10 times. (The default
3045 setting of finduser_retries is zero.) Also, it was sleeping after the final
3046 failure, which is pointless.
3048 24. Implemented tls_on_connect_ports.
3050 25. Implemented acl_smtp_predata.
3052 26. If the domain in control=submission is set empty, Exim assumes that the
3053 authenticated id is a complete email address when it generates From: or
3054 Sender: header lines.
3056 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3057 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3058 chown and chgrp in /bin and hostname in /usr/bin.
3060 28. Exim was keeping the "process log" file open after each use, just as it
3061 does for the main log. This opens the possibility of it remaining open for
3062 long periods when the USR1 signal hits a daemon. Occasional processlog
3063 errors were reported, that could have been caused by this. Anyway, it seems
3064 much more sensible not to leave this file open at all, so that is what now
3067 29. The long-running daemon process does not normally write to the log once it
3068 has entered its main loop, and it closes the log before doing so. This is
3069 so that log files can straightforwardly be renamed and moved. However,
3070 there are a couple of unusual error situations where the daemon does write
3071 log entries, and I had neglected to close the log afterwards.
3073 30. The text of an SMTP error response that was received during a remote
3074 delivery was being truncated at 512 bytes. This is too short for some of
3075 the long messages that one sometimes sees. I've increased the limit to
3078 31. It is now possible to make retry rules that apply only when a message has a
3079 specific sender, in particular, an empty sender.
3081 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3082 it possible to be selective about when SMTP synchronization is enforced.
3084 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3086 32. Implemented hosts_connection_nolog.
3088 33. Added an ACL for QUIT.
3090 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3093 35. Added mailbox_size and mailbox_filecount to appendfile.
3095 36. Added control = no_multiline_responses to ACLs.
3097 37. There was a bug in the logic of the code that waits for the clock to tick
3098 in the case where the clock went backwards by a substantial amount such
3099 that the microsecond fraction of "now" was more than the microsecond
3100 fraction of "then" (but the whole seconds number was less).
3102 38. Added support for the libradius Radius client library this is found on
3103 FreeBSD (previously only the radiusclient library was supported).
3109 1. When certain lookups returned multiple values in the form name=value, the
3110 quoting of the values was not always being done properly. Specifically:
3111 (a) If the value started with a double quote, but contained no whitespace,
3113 (b) If the value contained whitespace other than a space character (i.e.
3114 tabs or newlines or carriage returns) it was not quoted.
3115 This fix has been applied to the mysql and pgsql lookups by writing a
3116 separate quoting function and calling it from the lookup code. The fix
3117 should probably also be applied to nisplus, ibase and oracle lookups, but
3118 since I cannot test any of those, I have not disturbed their existing code.
3120 2. A hit in the callout cache for a specific address caused a log line with no
3121 reason for rejecting RCPT. Now it says "Previous (cached) callout
3122 verification failure".
3124 3. There was an off-by-one bug in the queryprogram router. An over-long
3125 return line was truncated at 256 instead of 255 characters, thereby
3126 overflowing its buffer with the terminating zero. As well as fixing this, I
3127 have increased the buffer size to 1024 (and made a note to document this).
3129 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3130 when Exim is waiting for an SMTP response from a remote server, Exim
3131 restarts its select() call on the socket, thereby resetting its timeout.
3132 This is not a problem when such interrupts are rare. Somebody set up a cron
3133 job to run exiwhat every 2 minutes, which is less than the normal select()
3134 timeout (5 or 10 minutes). This meant that the select() timeout never
3135 kicked in because it was always reset. I have fixed this by comparing the
3136 time when an interrupt arrives with the time at the start of the first call
3137 to select(). If more time than the timeout has elapsed, the interrupt is
3138 treated as a timeout.
3140 5. Some internal re-factoring in preparation for the addition of Sieve
3141 extensions (by MH). In particular, the "personal" test is moved to a
3142 separate function, and given an option for scanning Cc: and Bcc: (which is
3143 not set for Exim filters).
3145 6. When Exim created an email address using the login of the caller as the
3146 local part (e.g. when creating a From: or Sender: header line), it was not
3147 quoting the local part when it contained special characters such as @.
3149 7. Installed new OpenBSD configuration files.
3151 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3152 try to make them clearer.
3154 9. Callout options, other than the timeout value, were being ignored when
3155 verifying sender addresses in header lines. For example, when using
3157 verify = header_sender/callout=no_cache
3159 the cache was (incorrectly) being used.
3161 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3162 only the Cygwin environment.
3164 11. When return_path_on_delivery was set as a log selector, if different remote
3165 addresses in the same message used different return paths and parallel
3166 remote delivery occurred, the wrong values would sometimes be logged.
3167 (Whenever a remote delivery process finished, the return path value from
3168 the most recently started remote delivery process was logged.)
3170 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3171 lines when AUTH and/or TLS are in use. This is the "received protocol"
3172 field. Exim used to use "asmtp" for authenticated SMTP, without any
3173 indication (in the protocol name) for TLS use. Now it follows the RFC and
3174 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3175 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3176 names appear in log lines as well as in Received: header lines.
3178 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3179 extensions, and comparison tests, and to fix some bugs.
3181 14. Changes to the "personal" filter test:
3183 (1) The test was buggy in that it was just doing the equivalent of
3184 "contains" tests on header lines. For example, if a user's address was
3185 anne@some.where, the "personal" test would incorrectly be true for
3187 To: susanne@some.where
3189 This test is now done by extracting each address from the header in turn,
3190 and checking the entire address. Other tests that are part of "personal"
3191 are now done using regular expressions (for example, to check local parts
3192 of addresses in From: header lines).
3194 (2) The list of non-personal local parts in From: addresses has been
3195 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3196 taken from the Sieve specification recommendations.
3198 (3) If the message contains any header line starting with "List-" it is
3199 treated as non-personal.
3201 (4) The test for "circular" in the Subject: header line has been removed
3202 because it now seems ill-conceived.
3204 15. Minor typos in src/EDITME comments corrected.
3206 16. Installed latest exipick from John Jetmore.
3208 17. If headers_add on a router specified a text string that was too long for
3209 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3210 of string_sprintf() is now avoided.
3212 18. $message_body_size was not set (it was always zero) when running the DATA
3213 ACL and the local_scan() function.
3215 19. For the "mail" command in an Exim filter, no default was being set for
3216 the once_repeat time, causing a random time value to be used if "once" was
3217 specified. (If the value happened to be <= 0, no repeat happened.) The
3218 default is now 0s, meaning "never repeat". The "vacation" command was OK
3219 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3220 (I found it when inspecting the code).
3222 20. There is now an overall timeout for performing a callout verification. It
3223 defaults to 4 times the callout timeout, which applies to individual SMTP
3224 commands during the callout. The overall timeout applies when there is more
3225 than one host that can be tried. The timeout is checked before trying the
3226 next host. This prevents very long delays if there are a large number of
3227 hosts and all are timing out (e.g. when the network connections are timing
3228 out). The value of the overall timeout can be changed by specifying an
3229 additional sub-option for "callout", called "maxwait". For example:
3231 verify = sender/callout=5s,maxwait=20s
3233 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3234 to the end before writing, but this should make it even safer).
3236 22. Exim was forgetting that it had advertised PIPELINING for the second and
3237 subsequent messages on an SMTP connection. It was also not resetting its
3238 memory on STARTTLS and an internal HELO.
3240 23. When Exim logs an SMTP synchronization error within a session, it now
3241 records whether PIPELINING has been advertised or not.
3243 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3244 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3245 rather than long int.
3247 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3249 26. Added the never_mail option to autoreply.
3255 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3256 crash if the getsockname() call failed; this can happen if a connection is
3257 closed very soon after it is established. The problem was simply in the
3258 order in which certain operations were done, causing Exim to try to write
3259 to the SMTP stream before it had set up the file descriptor. The bug has
3260 been fixed by making things happen in the correct order.
3266 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3267 before the connection was closed, thus losing the rejection response.
3269 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3270 some early Solaris releases, but causes trouble in current releases where
3271 socklen_t is defined.
3273 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3276 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3279 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3280 that releasing the top of it at the end releases what was used for sub-
3281 expansions (unless the block got too big). However, discard this block if
3282 the first thing is a variable or header, so that we can use its block when
3283 it is dynamic (useful for very large $message_headers, for example).
3285 6. Lookups now cache *every* query, not just the most recent. A new, separate
3286 store pool is used for this. It can be recovered when all lookup caches are
3287 flushed. Lookups now release memory at the end of their result strings.
3288 This has involved some general refactoring of the lookup sources.
3290 7. Some code has been added to the store_xxx() functions to reduce the amount
3291 of flapping under certain conditions.
3293 8. log_incoming_interface used to affect only the <= reception log lines. Now
3294 it causes the local interface and port to be added to several more SMTP log
3295 lines, for example "SMTP connection from", and rejection lines.
3297 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3299 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3301 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3302 could be overwritten at the end of the current message (or the start of a
3303 new message if it was set in a HELO ACL). The value is now preserved for
3304 the duration of the SMTP connection.
3306 12. If a transport had a headers_rewrite setting, and a matching header line
3307 contained an unqualified address, that address was qualified, even if it
3308 did not match any rewriting rules. The underlying bug was that the values
3309 of the flags that permit the existence of unqualified sender and recipient
3310 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3311 non-local messages, and by -bnq for local messages) were not being
3312 preserved with the message after it was received.
3314 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3315 "next input=" as part of the text comprising the host identity instead of
3316 the correct text. The code was using the same buffer for two different
3317 strings. However, depending on which order the printing function evaluated
3318 its arguments, the bug did not always show up. Under Linux, for example, my
3319 test suite worked just fine.
3321 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3322 doesn't work with some older versions of Perl. It has been changed to "my",
3323 which in any case is probably the better facility to use.
3325 15. A really picky compiler found some instances of statements for creating
3326 error messages that either had too many or two few arguments for the format
3329 16. The size of the buffer for calls to the DNS resolver has been increased
3330 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3331 for addresses that have a lot of PTR records. This alleviates a problem; it
3332 does not fully solve it.
3334 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3335 buffer now truncates the list and logs the incident, which is the same
3336 action as happens when Exim is looking up a host name and its aliases.
3337 Previously in this situation something unpredictable would happen;
3338 sometimes it was "internal error: store_reset failed".
3340 18. If a server dropped the connection unexpectedly when an Exim client was
3341 using GnuTLS and trying to read a response, the client delivery process
3342 crashed while trying to generate an error log message.
3344 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3345 string, for example:
3347 warn message = H1: something\nH2: something
3349 the text was added as a single header line from Exim's point of view
3350 though it ended up OK in the delivered message. However, searching for the
3351 second and subsequent header lines using $h_h2: did not work. This has been
3352 fixed. Similarly, if a system filter added multiple headers in this way,
3353 the routers could not see them.
3355 20. Expanded the error message when iplsearch is called with an invalid key to
3356 suggest using net-iplsearch in a host list.
3358 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3359 ACLs are no longer actually imposed (and a message to that effect is
3362 22. If a "gecos" field in a passwd entry contained escaped characters, in
3363 particular, if it contained a \" sequence, Exim got it wrong when building
3364 a From: or a Sender: header from that name. A second bug also caused
3365 incorrect handling when an unquoted " was present following a character
3366 that needed quoting.
3368 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3369 was not being matched caselessly.
3371 24. Arranged for all hyphens in the exim.8 source to be escaped with
3374 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3375 properly. Recipient callout cache records were still being keyed to include
3376 the sender, even when use_sender was set false. This led to far more
3377 callouts that were necessary. The sender is no longer included in the key
3378 when use_sender is false.
3380 26. Added "control = submission" modifier to ACLs.
3382 27. Added the ${base62d: operator to decode base 62 numbers.
3384 28. dnsdb lookups can now access SRV records.
3386 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3387 the configuration file.
3389 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3390 (-v) mode. This makes the output for a verbose queue run more intelligible.
3392 31. Added a use_postmaster feature to recipient callouts.
3394 32. Added the $body_zerocount variable, containing the number of binary zero
3395 bytes in the message body.
3397 33. The time of last modification of the "new" subdirectory is now used as the
3398 "mailbox time last read" when there is a quota error for a maildir
3401 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3403 35. Added +ignore_unknown as a special item in host lists.
3405 36. Code for decoding IPv6 addresses in host lists is now included, even if
3406 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3407 address was recognized as an IP address, but was then not correctly decoded
3408 into binary, causing unexpected and incorrect effects when compared with
3415 1. Very minor rewording of debugging text in manualroute to say "list of
3416 hosts" instead of "hostlist".
3418 2. If verify=header_syntax was set, and a header line with an unqualified
3419 address (no domain) and a large number of spaces between the end of the
3420 name and the colon was received, the reception process suffered a buffer
3421 overflow, and (when I tested it) crashed. This was caused by some obsolete
3422 code that should have been removed. The fix is to remove it!
3424 3. When running in the test harness, delay a bit after writing a bounce
3425 message to get a bit more predictability in the log output.
3427 4. Added a call to search_tidyup() just before forking a reception process. In
3428 theory, someone could use a lookup in the expansion of smtp_accept_max_
3429 per_host which, without the tidyup, could leave open a database connection.
3431 5. Added the variables $recipient_data and $sender_data which get set from a
3432 lookup success in an ACL "recipients" or "senders" condition, or a router
3433 "senders" option, similar to $domain_data and $local_part_data.
3435 6. Moved the writing of debug_print from before to after the "senders" test
3438 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3439 problems for message scanning, either using a data ACL, or using
3440 local_scan() because the Received: header was not generated till after they
3441 were called (in order to set the time as the time of reception completion).
3442 I have revised the way this works. The header is now generated after the
3443 body is received, but before the ACL or local_scan() are called. After they
3444 are run, the timestamp in the header is updated.
3450 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3451 before starting a queue runner without re-exec. This happened only when
3452 deliver_drop_privilege was set or when the Exim user was set to root. The
3453 effect of the bug was that timeouts during subsequent deliveries caused
3454 crashes instead of being properly handled. The handler is now left at its
3455 default (and expected) setting.
3457 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3458 message, again when deliver_drop_privilege is set or Exim is run as root.
3459 The bug described in (1) was not present in this case, but the tidying up
3460 of the other signals was missing. I have made the two cases consistent.
3462 3. The ignore_target_hosts setting on a manualroute router was being ignored
3463 for hosts that were looked up using the /MX notation.
3465 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3468 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3469 operated on the sender address. After changing the $sender_address to <>
3470 for the sender address verify, Exim was re-instated it as the original
3471 (before rewriting) address, but remembering that it had rewritten it, so it
3472 wasn't rewriting it again. This bug also had the effect of breaking the
3473 sender address verification caching when the sender address was rewritten.
3475 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3476 This has been changed so that if the ip literal address matches
3477 ignore_target_hosts, the router declines.
3479 7. Added expansion conditions match_domain, match_address, and match_local_
3480 part (NOT match_host).
3482 8. The placeholder for the Received: header didn't have a length field set.
3484 9. Added code to Exim itself and to exim_lock to test for a specific race
3485 condition that could lead to file corruption when using MBX delivery. The
3486 issue is with the lockfile that is created in /tmp. If this file is removed
3487 after a process has opened it but before that process has acquired a lock,
3488 there is the potential for a second process to recreate the file and also
3489 acquire a lock. This could lead to two Exim processes writing to the file
3490 at the same time. The added code performs the same test as UW imapd; it
3491 checks after acquiring the lock that its file descriptor still refers to
3492 the same named file.
3494 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3495 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3498 11. Added the smtp_active_hostname option. If used, this will typically be made
3499 to depend on the incoming interface address. Because $interface_address is
3500 not set up until the daemon has forked a reception process, error responses
3501 that can happen earlier (such as "too many connections") no longer contain
3504 12. If an expansion in a condition on a "warn" statement fails because a lookup
3505 defers, the "warn" statement is abandoned, and the next ACL statement is
3506 processed. Previously this caused the whole ACL to be aborted.
3508 13. Added the iplsearch lookup type.
3510 14. Added ident_timeout as a log selector.
3512 15. Added tls_certificate_verified as a log selector.
3514 16. Added a global option tls_require_ciphers (compare the smtp transport
3515 option of the same name). This controls incoming TLS connections.
3517 17. I finally figured out how to make tls_require_ciphers do a similar thing
3518 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3519 before starting the TLS session.
3521 18. Tabs are now shown as \t in -bP output.
3523 19. If the log selector return_path_on_delivery was set, Exim crashed when
3524 bouncing a message because it had too many Received: header lines.
3526 20. If two routers both had headers_remove settings, and the first one included
3527 a superfluous trailing colon, the final name in the first list and the
3528 first name in the second list were incorrectly joined into one item (with a
3529 colon in the middle).
3535 1. Added -C and -D options to the exinext utility, mainly to make it easier
3536 to include in the automated testing, but these could be helpful when
3537 multiple configurations are in use.
3539 2. The exinext utility was not formatting the output nicely when there was
3540 an alternate port involved in the retry record key, nor when there was a
3541 message id as well (for retries that were specific to a specific message
3542 and a specific host). It was also confused by IPv6 addresses, because of
3543 the additional colons they contain. I have fixed the IPv4 problem, and
3544 patched it up to do a reasonable job for IPv6.
3546 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3547 delivery, the log line now contains "pipelined" if PIPELINING was used.
3549 4. An SMTP transport process used to panic and die if the bind() call to set
3550 an explicit outgoing interface failed. This has been changed; it is now
3551 treated in the same way as a connect() failure.
3553 5. A reference to $sender_host_name in the part of a conditional expansion
3554 that was being skipped was still causing a DNS lookup. This no longer
3557 6. The def: expansion condition was not recognizing references to header lines
3558 that used bh_ and bheader_.
3560 7. Added the _cache feature to named lists.
3562 8. The code for checking quota_filecount in the appendfile transport was
3563 allowing one more file than it should have been.
3565 9. For compatibility with Sendmail, the command line option
3573 and sets the incoming protocol and host name (for trusted callers). The
3574 host name and its colon can be omitted when only the protocol is to be set.
3575 Note the Exim already has two private options, -pd and -ps, that refer to
3576 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3577 "s", but I don't think that's a major issue.
3579 10. A number of refactoring changes to the code, none of which should affect
3582 (a) The number of logging options was getting close to filling up the
3583 32-bit word that was used as a bit map. I have split them into two classes:
3584 those that are passed in the argument to log_write(), and those that are
3585 only ever tested independently outside of that function. These are now in
3586 separate 32-bit words, so there is plenty of room for expansion again.
3587 There is no change in the user interface or the logging behaviour.
3589 (b) When building, for example, log lines, the code previously used a
3590 macro that called string_cat() twice, in order to add two strings. This is
3591 not really sufficiently general. Furthermore, there was one instance where
3592 it was actually wrong because one of the argument was used twice, and in
3593 one call a function was used. (As it happened, calling the function twice
3594 did not affect the overall behaviour.) The macro has been replaced by a
3595 function that can join an arbitrary number of extra strings onto a growing
3598 (c) The code for expansion conditions now uses a table and a binary chop
3599 instead of a serial search (which was left over from when there were very
3600 few conditions). Also, it now recognizes conditions like "pam" even when
3601 the relevant support is not compiled in: a suitably worded error message is
3602 given if an attempt is made to use such a condition.
3604 11. Added ${time_interval:xxxxx}.
3606 12. A bug was causing one of the ddress fields not to be passed back correctly
3607 from remote delivery subprocesses. The field in question was not being
3608 subsequently used, so this caused to problems in practice.
3610 13. Added new log selectors queue_time and deliver_time.
3612 14. Might have fixed a bug in maildirsizefile handling that threw up
3613 "unexpected character" debug warnings, and recalculated the data
3614 unnecessarily. In any case, I expanded the warning message to give more
3617 15. Added the message "Restricted characters in address" to the statements in
3618 the default ACL that block characters like @ and % in local parts.
3620 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3621 Three changes have been made:
3623 (a) There was a serious bug; a negative response to MAIL caused the whole
3624 recipient domain to be cached as invalid, thereby blocking all messages
3625 to all local parts at the same domain, from all senders. This bug has
3626 been fixed. The domain is no longer cached after a negative response to
3627 MAIL if the sender used is not empty.
3629 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3632 (c) A new callout option, "use_sender" has been added for people who want
3633 the modified behaviour.
3639 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3642 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3643 indeed breaks things for older releases.
3645 3. Added additional logging to the case where there is a problem reading data
3646 from a filter that is running in a subprocess using a pipe, in order to
3647 try to track down a specific problem.
3649 4. Testing facility fudge: when running in the test harness and attempting
3650 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3651 getting "No route to host". Convert this to a timeout.
3653 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3656 6. Some OS don't have socklen_t but use size_t instead. This affects the
3657 fifth argument of getsockopt() amongst other things. This is now
3658 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3659 can be set for individual OS. I have set it for SunOS5, OSF1, and
3660 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3661 some earlier ones do not.
3663 7. Change 4.30/15 was not doing the test caselessly.
3665 8. The standard form for an IPv6 address literal was being rejected by address
3666 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3667 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3668 this, as well as the form without the "IPv6" on the front (but only when
3669 address literals are enabled, of course).
3671 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3673 10. Exim crashed if a message with an empty sender address specified by -f
3674 encountered a router with an errors_to setting. This could be provoked only
3675 by a command such as
3679 where an empty string was supplied; "<>" did not hit this bug.
3681 11. Installed PCRE release 4.5.
3683 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3684 remained set. It is now erased.
3686 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3687 times from message ids (which are base 36 rather than the normal 62).
3689 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3690 were being counted as actual protocol errors, and logged if the log
3691 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3692 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3693 and DATA following a set of rejected RCPTs do not count as protocol errors.
3694 In other words, Exim assumes they were pipelined, though this may not
3695 actually be the case. Of course, in all cases the client gets an
3696 appropriate error code.
3698 15. If a lookup fails in an ACL condition, a message about the failure may
3699 be available; it is used if testing the ACL cannot continue, because most
3700 such messages specify what the cause of the deferral is. However, some
3701 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3702 that caused an old message to be retained and used if a later statement
3703 caused a defer, replacing the real cause of the deferral.
3705 16. If an IP address had so many PTR records that the DNS lookup buffer
3706 was not large enough to hold them, Exim could crash while trying to process
3707 the truncated data. It now detects and logs this case.
3709 17. Further to 4.21/58, another change has been made: if (and only if) the
3710 first line of a message (the first header line) ends with CRLF, a bare LF
3711 in a subsequent header line has a space inserted after it, so as not to
3712 terminate the header.
3714 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3715 unnecessarily, used atoi() instead of strtol(), and didn't check the
3716 termination when getting file sizes from file names by regex.
3718 19. Completely re-implemented the support for maildirsize files, in the light
3719 of a number of problems with the previous contributed implementation
3720 (4.30/29). In particular:
3722 . If the quota is zero, the maildirsize file is maintained, but no quota is
3725 . If the maildir directory does not exist, it is created before any attempt
3726 to write a maildirsize file.
3728 . The quota value in the file is just a cache; if the quota is changed in
3729 the transport, the new value overrides.
3731 . A regular expression is available for excluding directories from the
3734 20. The autoreply transport checks the characters in options that define the
3735 message's headers; it allows continued headers, but it was checking with
3736 isspace() after an embedded newline instead of explicitly looking for a
3739 21. If all the "regular" hosts to which an address was routed had passed their
3740 expiry times, and had not reached their retry times, the address was
3741 bounced, even if fallback hosts were defined. Now Exim should go on to try
3744 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3745 equivalent code in the SMTP transport. Some hosts send humungous responses
3746 to HELO/EHLO, more than 1024 it seems.
3748 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3749 gives compiler warnings in some environments. I've now done it "properly",
3752 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3753 (because of problems with the built-in one) was declared to return uschar *
3754 instead of char *, causing compiler failure.
3756 25. Fixed a file descriptor leak when processing alias/forward files.
3758 26. Fixed a minor format string issue in dbfn.c.
3760 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3762 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3763 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3766 29. When a sender address is verified, it is cached, to save repeating the test
3767 when there is more than one recipient in a message. However, when the
3768 verification involves a callout, it is possible for different callout
3769 options to be set for different recipients. It is too complicated to keep
3770 track of this in the cache, so now Exim always runs a verification when a
3771 callout is required, relying on the callout cache for the optimization.
3772 The overhead is duplication of the address routing, but this should not be
3775 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3776 to be verified with callout=postmaster, and the main callout worked but the
3777 postmaster check failed, the verification correctly failed. However, if a
3778 subsequent RCPT command asked for sender verification *without* the
3779 postmaster check, incorrect caching caused this verification also to fail,
3782 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3783 it was not caching the DNS options (qualify_single, search_parents) that
3784 were used when the lookup failed. A subsequent lookup with different
3785 options therefore always gave the same answer, though there were cases
3786 where it should not have. (Example: a "domains = !$mx_any" option on a
3787 dnslookup router: the "domains" option is always processed without any
3788 widening, but the router might have qualify_single set.) Now Exim uses the
3789 cached value only when the same options are set.
3791 32. Added John Jetmore's "exipick" utility to the distribution.
3793 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3794 than a timeout (e.g. a certificate is required, and is not provided), an
3795 Exim server now closes the connection immediately. Previously it waited for
3796 the client to close - but if the client is SSL, it seems that they each
3797 wait for each other, leading to a delay before one of them times out.
3799 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3800 maintained 0.8.x compatibility because I don't think many are using it, and
3801 it is clearly obsolete.
3803 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3806 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3807 client certificate was expired. A simple patch fixes this, though I don't
3808 understand the full logic of why the verify callback is called multiple
3811 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3812 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3813 which causes problems with some clients (such as the Certicom SSL Plus
3814 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3815 disables the coutermeasure allowing Eudora to connect."
3817 38. Exim was not checking that a write() to a log file succeeded. This could
3818 lead to Bad Things if a log got too big, in particular if it hit a file
3819 size limit. Exim now panics and dies if it cannot write to a log file, just
3820 as it does if it cannot open a log file.
3822 39. Modified OS/Makefile-Linux so that it now contains
3824 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3826 The two -D definitions ensure that Exim is compiled with large file
3827 support, which makes it possible to handle log files that are bigger than
3830 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3831 instance) a domain was checked against a named list that involved a lookup,
3832 causing $domain_data to be set, then another domain was checked against the
3833 same list, then the first domain was re-checked, the value of $domain_data
3834 after the final check could be wrong. In particular, if the second check
3835 failed, it could be set empty. This bug probably also applied to
3838 41. The strip_trailing_dot option was not being applied to the address given
3839 with the -f command-line option.
3841 42. The code for reading a message's header from the spool was incrementing
3842 $received_count, but never initializing it. This meant that the value was
3843 incorrect (doubled) while delivering a message in the same process in which
3844 it was received. In the most common configuration of Exim, this never
3845 happens - a fresh exec is done - but it can happen when
3846 deliver_drop_privilege is set.
3848 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3849 it now includes up to 150 characters of the unexpected data in the log
3852 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3853 and building data strings. The size of both of these buffers was 10 000
3854 bytes - far larger than anybody would *ever* want, thought I. Needless to
3855 say, somebody hit the limit. I have increased the maximum line length to
3856 20 000 and the maximum data length of concatenated lines to 100 000. I have
3857 also fixed two bugs, because there was no checking on these buffers. Tsk,
3858 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3859 buffer is too small.
3861 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3862 lsearch lookups. Now it does.
3864 46. When parsing a route_list item in a manualroute router, a fixed-length
3865 buffer was used for the list of hosts. I made this 1024 bytes long,
3866 thinking that nobody would ever have a list of hosts that long. Wrong.
3867 Somebody had a whole pile of complicated expansion conditions, and the
3868 string was silently truncated, leading to an expansion error. It turns out
3869 that it is easier to change to an unlimited length (owing to other changes
3870 that have happened since this code was originally written) than to build
3871 structure for giving a limitation error. The length of the item that
3872 expands into the list of hosts is now unlimited.
3874 47. The lsearch lookup could not handle data where the length of text line was
3875 more than 4095 characters. Such lines were truncated, leading to shortened
3876 data being returned. It should now handle lines of any length.
3878 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3879 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3882 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3883 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3884 no longer happens in comment lines. A list of replacements is now placed
3885 at the head of all of the source files, except those whose only change is
3886 to replace PERL_COMMAND in the very first #! line.
3888 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3889 messages on the queue, with a bottom-up merge sort, using code contributed
3890 by Michael Haardt. This should make operations like -bp somewhat faster on
3891 large queues. It won't affect queue runners, except when queue_run_in_order
3894 51. Installed eximstats 1.31 in the distribution.
3896 52. Added support for SRV lookups to the dnslookup router.
3898 53. If an ACL referred to $message_body or $message_body_end, the value was not
3899 reset for any messages that followed in the same SMTP session.
3901 54. The store-handling optimization for building very long strings was not
3902 differentiating between the different store pools. I don't think this
3903 actually made any difference in practice, but I've tidied it.
3905 55. While running the routers to verify a sender address, $sender_address
3906 was still set to the sender address. This is wrong, because when routing to
3907 send a bounce to the sender, it would be empty. Therefore, I have changed
3908 it so that, while verifying a sender address, $sender_address is set to <>.
3909 (There is no change to what happens when verifying a recipient address.)
3911 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3912 target A or AAAA records (if not already returned) without resetting the
3913 qualify_single or search_parents options of the DNS resolver. These are
3914 inappropriate in this case because the targets of MX and SRV records must
3915 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3916 target that, when qualified, matched something in the local domain. These
3917 two options are now turned off when doing these lookups.
3919 57. It seems that at least some releases of Reiserfs (which does not have the
3920 concept of a fixed number of inodes) returns zero and not -1 for the
3921 number of available inodes. This interacted badly with check_spool_inodes,
3922 which assumed that -1 was the "no such thing" setting. What I have done is
3923 to check that the total number of inodes is greater than zero before doing
3924 the test of how many are available.
3926 58. When a "warn" ACL statement has a log_message modifier, the message is
3927 remembered, and not repeated. This is to avoid a lot of repetition when a
3928 message has many recipients that cause the same warning to be written.
3929 Howewer, Exim was preserving the list of already written lines for an
3930 entire SMTP session, which doesn't seem right. The memory is now reset if a
3931 new message is started.
3933 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3934 debugging output unless log_rewrite was also set.
3936 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3937 of (int)(handle) when we know that handle contains (void *)(-1).
3939 61. The Exim daemon panic-logs an error return when it closes the incoming
3940 connection. However "connection reset by peer" seems to be common, and
3941 isn't really an error worthy of noting specially, so that particular error
3944 62. When Exim is trying to find all the local interfaces, it used to panic and
3945 die if the ioctl to get the interface flags failed. However, it seems that
3946 on at least one OS (Solaris 9) it is possible to have an interface that is
3947 included in the list of interfaces, but for which you get a failure error
3948 for this call. This happens when the interface is not "plumbed" into a
3949 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3950 failure of the "get flags" call assumes that the interface is down.
3952 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3953 makes life easier for people who are doing arithmetic on fields extracted
3954 from dates, where you often get leading zeros that should not be
3955 interpreted as octal.
3957 64. Added qualify_domain to the redirect router, to override the global
3960 65. If a pathologically long header line contained very many addresses (the
3961 report of this problem mentioned 10 000) and each of them was rewritten,
3962 Exim could use up a very large amount of memory. (It kept on making new
3963 copies of the header line as it rewrote, and never released the old ones.)
3964 At the expense of a bit more processing, the header rewriting function has
3965 been changed so that it no longer eats memory in this way.
3967 66. The generation of the Received: header has been moved from the time that a
3968 message starts to be received, to the time that it finishes. The timestamp
3969 in the Received: header should now be very close to that of the <= log
3970 line. There are two side-effects of this change:
3972 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3973 logged header lines no longer include the local Received: line, because
3974 it has not yet been created. The same applies to a copy of the message
3975 that is returned to a non-SMTP sender when a message is rejected.
3977 (b) When a filter file is tested using -bf, no additional Received: header
3978 is added to the test message. After some thought, I decided that this
3981 This change does not affect the value of $received_for. It is still set
3982 after address rewriting, but before local_scan() is called.
3984 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3986 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3987 gave an unhelpful panic error message, and a defer error. I have managed to
3988 change this behaviour so that it now rejects any supplied certificate,
3989 which seems right, as the list of acceptable certificates is empty.
3991 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3992 gave an unhelpful defer error. I have not managed to make this reject any
3993 supplied certificates, but the error message it gives is "no certificate
3994 supplied", which is not helpful.
3996 70. exigrep's output now also includes lines that are not associated with any
3997 message, but which match the given pattern. Implemented by a patch from
3998 Martin Sluka, which also tidied up the Perl a bit.
4000 71. Recipient callout verification, like sender verification, was using <> in
4001 the MAIL FROM command. This isn't really the right thing, since the actual
4002 sender may affect whether the remote host accepts the recipient or not. I
4003 have changed it to use the actual sender in the callout; this means that
4004 the cache record is now keyed on a recipient/sender pair, not just the
4005 recipient address. There doesn't seem to be a real danger of callout loops,
4006 since a callout by the remote host to check the sender would use <>.
4007 [SEE ABOVE: changed after hitting problems.]
4009 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4010 temporary errors. However, in the case of such a code being given after
4011 the end of a data transmission (i.e. after ".") Exim was failing to write
4012 a retry record for the message. (Yes, there was some broken host that was
4013 actually sending 8xx at this point.)
4015 73. An unknown lookup type in a host list could cause Exim to panic-die when
4016 the list was checked. (An example that provoked this was putting <; in the
4017 middle of a list instead of at the start.) If this happened during a DATA
4018 ACL check, a -D file could be left lying around. This kind of configuration
4019 error no longer causes Exim to die; instead it causes a defer errror. The
4020 incident is still logged to the main and panic logs.
4022 74. Buglet left over from Exim 3 conversion. The message "too many messages
4023 in one connection" was written to the rejectlog but not the mainlog, except
4024 when address rewriting (yes!) was being logged.
4026 75. Added write_rejectlog option.
4028 76. When a system filter was run not as root (that is, when system_filter_user
4029 was set), the values of the $n variables were not being returned to the
4030 main process; thus, they were not subsequently available in the $sn
4033 77. Added +return_path_on_delivery log selector.
4035 78. A connection timeout was being treated differently from recipients deferred
4036 when testing hosts_max_try with a message that was older than the host's
4037 retry timeout. (The host should not be counted, thus allowing all hosts to
4038 be tried at least once before bouncing.) This may have been the cause of an
4039 occasionally reported bug whereby a message would remain on the queue
4040 longer than the retry timeout, but would be bounced if a delivery was
4041 forced. I say "may" because I never totally pinned down the problem;
4042 setting up timeout/retry tests is difficult. See also the next item.
4044 79. The ultimate address timeout was not being applied to errors that involved
4045 a combination of host plus message (for example, a timeout on a MAIL
4046 command). When an address resolved to a number of possible hosts, and they
4047 were not all tried for each delivery (e.g. because of hosts_max_try), a
4048 message could remain on the queue longer than the retry timeout.
4050 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4053 81. Fixed an obscure SMTP outgoing bug which required at least the following
4054 conditions: (a) there was another message waiting for the same server;
4055 (b) the server returned 5xx to all RCPT commands in the first message so
4056 that the message was not completed; (c) the server dropped the connection
4057 or gave a negative response to the RSET that Exim sends to abort the
4058 transaction. The observed case was a dropped connection after DATA that had
4059 been sent in pipelining mode. That is, the server had advertised PIPELINING
4060 but was not implementing it correctly. The effect of the bug was incorrect
4061 behaviour, such as trying another host, and this could lead to a crash.
4067 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4068 and daemon.c were passed as pointers to ints; they should have been
4069 pointers to socklen_t variables (which are typically unsigned ints).
4071 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4074 3. Fixed a really odd bug that affected only the testing scheme; patching a
4075 certain fixed string in the binary changed the value of another string that
4076 happened to be identical to the end of the original first string.
4078 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4079 name", it returns that address as the IP address. On some operating
4080 systems (e.g. Solaris), it also passes back the IP address string as the
4081 "host name". However, on others (e.g. Linux), it passes back an empty
4082 string. Exim wasn't checking for this, and was changing the host name to an
4083 empty string, assuming it had been canonicized.
4085 5. Although rare, it is permitted to have more than one PTR record for a given
4086 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4087 all the names associated with an address, because they do in Solaris.
4088 However, it seems that they do not in Linux for data that comes from the
4089 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4090 I found this out when I moved to a new Linux workstation and tried to run
4091 the Exim test suite.
4093 To get round this problem I have changed the code so that it now does its
4094 own call to the DNS to look up PTR records when searching for a host name.
4095 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4096 addresses that are only in /etc/hosts are still found.
4098 This behaviour is, however, controlled by an option called host_lookup_
4099 order, which defaults to "bydns:byaddr". If people want to use the other
4100 order, or indeed, just use one or the other means of lookup, they can
4101 specify it in this variable.
4103 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4104 some operating systems, this comes back from gethostbyaddr() as an empty
4105 string, and this is what Exim used to test for. However, it seems that in
4106 other systems, "." is yielded. Exim now tests for this case too.
4108 7. The values of check_spool_space and check_log_space are now held internally
4109 as a number of kilobytes instead of an absolute number of bytes. If a
4110 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4111 kilobyte. This means that much larger values can be stored.
4113 8. Exim monitor: an attempt to get the action menu when not actually pointing
4114 at a message produces an empty menu entitled "No message selected". This
4115 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4116 no entries in it ("Shell widget menu has zero width and/or height"). So I
4117 have added a single, blank menu entry in this case.
4119 9. Added ${quote_local_part.
4121 10. MIME decoding is now applied to the contents of Subject: header lines when
4124 11. Now that a reference to $sender_host_address automatically causes a reverse
4125 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4126 host lookup before query-style lookups in lists that might use this
4127 variable. This has therefore been abolished, and the "net-" prefix is no
4128 longer necessary for query-style lookups.
4130 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4131 have been a typo for LFLAGS, so it has been changed.
4133 13. The install script calls Exim with "-C /dev/null" in order to find the
4134 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4135 to be output. Howeve, since Exim outputs its version number before the
4136 error, it didn't break the script. It just looked ugly. I fixed this by
4137 always allowing "-C /dev/null" if the caller is root.
4139 14. Ignore overlarge ACL variable number when reading spool file - insurance
4140 against a later release with more variables having written the file.
4142 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4143 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4144 this, as well as the form without the "IPv6" on the front.
4146 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4147 OS/Makefile-Darwin file.
4149 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4150 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4152 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4153 variable. However, some OS use ints for inodes. I've added cast to long int
4154 to get rid of the compiler warning.
4156 19. I had forgotten to lock out "/../" in configuration file names when
4157 ALT_CONFIG_PREFIX was set.
4159 20. Routers used for verification do not need to specify transports. However,
4160 if such a router generated a host list, and callout was configured, Exim
4161 crashed, because it could not find a port number from the (non-existent)
4162 transport. It now assumes port 25 in this circumstance.
4164 21. Added the -t option to exigrep.
4166 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4167 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4168 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4169 them is set, LOOKUP_LSEARCH is forced.
4171 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4173 24. Added sender and host information to the "rejected by local_scan()" log
4174 line; previously there was no indication of these.
4176 25. Added .include_if_exists.
4178 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4179 Linux. It turns out that not all file systems support this. Apparently some
4180 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4181 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4182 not supported on the file descriptor, is now ignored when Exim is trying to
4183 sync a directory. This applies only to Linux.
4185 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4187 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4188 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4191 29. Added support for maildirsize files from supplied patch (modified a bit).
4193 30. The use of :fail: followed by an empty string could lead Exim to respond to
4194 sender verification failures with (e.g.):
4196 550 Verification failed for <xxx>
4197 550 Sender verify failed
4199 where the first response line was missing the '-' that indicates it is not
4200 the final line of the response.
4202 31. The loop for finding the name of the user that called Exim had a hardwired
4203 limit of 10; it now uses the value of finduser_retries, which is used for
4204 all other user lookups.
4206 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4209 33. Exim was neglecting to zero errno before one call of strtol() when
4210 expanding a string and expecting an integer value. On some systems this
4211 resulted in spurious "integer overflow" errors. Also, it was casting the
4212 result into an int without checking.
4214 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4215 did not work. The code looks as if it has *never* worked, though it appears
4216 to have been documented since at least releast 1.62. I have made it work.
4218 35. The "timeout_DNS" error in retry rules, also documented since at least
4219 1.62, also never worked. As it isn't clear exactly what this means, and
4220 clearly it isn't a major issue, I have abolished the feature by treating it
4221 as "timeout", and writing a warning to the main and panic logs.
4223 36. The display of retry rules for -brt wasn't always showing the error code
4226 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4227 timeout_connect_A, timeout_connect_MX.
4229 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4230 to the empty sender.
4232 39. The daemon was not analysing the content of -oX till after it had closed
4233 stderr and disconnected from the controlling terminal. This meant that any
4234 syntax errors were only noted on the panic log, and the return code from
4235 the command was 0. By re-arranging the code a little, I've made the
4236 decoding happen first, so such errors now appear on stderr, and the return
4237 code is 1. However, the actual setting up of the sockets still happens in
4238 the disconnected process, so errors there are still only recorded on the
4241 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4242 connections (as happens on some IP stacks) was logged at start up time as
4243 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4244 it from "IPv6 and IPv4", which means that two separate sockets are being
4247 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4248 says whether AF_INET or AF_INET6 was passed as an argument.
4250 42. Exiwhat output was messed up when time zones were included in log
4253 43. Exiwhat now gives more information about the daemon's listening ports,
4254 and whether -tls-on-connect was used.
4256 44. The "port" option of the smtp transport is now expanded.
4258 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4259 silently ignored. Now an error message is written to the main and panic
4262 46. There's a new ACL modifier called "logwrite" which writes to a log file
4263 as soon as it is encountered.
4265 47. Added $local_user_uid and $local_user_gid at routing time.
4267 48. Exim crashed when trying to verify a sender address that was being
4270 49. Exim was recognizing only a space character after ".include". It now also
4271 recognizes a tab character.
4273 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4274 extracting the relevant information from the specification. The man page no
4275 longer contains scrambled data for the -d option, and I've added a section
4276 at the front about calling Exim under different names.
4278 51. Added "extra_headers" argument to the "mail" command in filter files.
4280 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4283 53. Installed eximstats 1.29.
4285 54. Added transport_filter_timeout as a generic transport option.
4287 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4288 Cc: header lines. This was required by RFC 822, but it not required by RFC
4291 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4292 incoming messages that did not have them. Now it does so only if the
4293 message originates locally, that is, if there is no associated remote host
4294 address. When Resent- header lines are present, this applies to the Resent-
4295 lines rather than the non-Resent- lines.
4297 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4298 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4300 58. Messages for configuration errors now include the name of the main
4301 configuration file - useful now that there may be more than one file in a
4302 list (.included file names were always shown).
4304 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4305 for those rare installations that do not start the daemon as root or run it
4306 setuid root. I've cut out the call to initgroups() if the daemon is not
4309 60. The Exim user and group can now be bound into the binary as text strings
4310 that are looked up at the start of Exim's processing.
4312 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4314 62. Added $mailstore_basename variable.
4316 63. Installed patch to sieve.c from Michael Haardt.
4318 64. When Exim failed to open the panic log after failing to open the main log,
4319 the original message it was trying to log was written to stderr and debug
4320 output, but if they were not available (the usual case in production), it
4321 was lost. Now it is written to syslog before the two lines that record the
4322 failures to open the logs.
4324 65. Users' Exim filters run in subprocesses under the user's uid. It is
4325 possible for a "deliver" command or an alias in a "personal" command to
4326 provoke an address rewrite. If logging of address rewriting is configured,
4327 this fails because the process is not running as root or exim. There may be
4328 a better way of dealing with this, but for the moment (because 4.30 needs
4329 to be released), I have disabled address rewrite logging when running a
4330 filter in a non-root, non-exim process.
4336 1. The buildconfig auxiliary program wasn't quoting the value set for
4337 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4338 not defined. This bug was masked in 4.22 by the effect that was fixed in
4341 2. Some messages that were rejected after a message id was allocated were
4342 shown as "incomplete" by exigrep. It no longer does this for messages that
4343 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4345 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4346 have allow_domain_literals set, the ID did not get logged in the <= line.
4347 Domain literals are now always recognized in Message-ID: header lines.
4349 4. The first argument for a ${extract expansion item is the key name or field
4350 number. Leading and trailing spaces in this item were not being ignored,
4351 causing some misleading effects.
4353 5. When deliver_drop_privilege was set, single queue runner processes started
4354 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4355 same command in the process it spins off) were not dropping privilege.
4357 6. When the daemon running as "exim" started a queue runner, it always
4358 re-executed Exim in the spun-off process. This is a waste of effort when
4359 deliver_drop_privilege is set. The new process now just calls the
4360 queue-runner function directly.
4366 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4369 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4370 ignored. Though the use of -oP was forcing the writing of a pid file, it
4371 was always written to the default place.
4373 3. If the message "no IP address found for host xxxx" is generated during
4374 incoming verification, it is now followed by identification of the incoming
4375 connection (so you can more easily find what provoked it).
4377 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4379 5. Added some features to "harden" Exim a bit more against certain attacks:
4381 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4382 be put in Local/Makefile. This is like the never_users runtime option,
4383 but it cannot be overridden. The default setting is "root".
4385 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4386 prefix string with which any file named in a -C command line option
4389 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4390 is retained for -C and -D only if the caller of Exim is root. Without
4391 it, the exim user may also use -C and -D and retain privilege.
4393 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4394 command line option is disabled.
4396 6. Macro names set by the -D option must start with an upper case letter, just
4397 like macro names defined in the configuration file.
4399 7. Added "dereference=" facility to LDAP.
4401 8. Two instances of the typo "uknown" in the source files are fixed.
4403 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4404 the Configure-Makefile script screwed up while processing it.
4406 10. Incorporated PCRE 4.4.
4408 11. The SMTP synchronization check was not operating right at the start of an
4409 SMTP session. For example, it could not catch a HELO sent before the client
4410 waited for the greeting. There is now a check for outstanding input at the
4411 point when the greeting is written. Because of the duplex, asynchronous
4412 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4413 way, but not yet received, when the check is performed.
4415 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4416 on TCP/IP sockets, because this apparently causes some broken clients to
4419 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4420 unchanged) from the Cygwin maintainer.
4422 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4423 was supported instead of testing for mbx. Effectively a typo.
4425 15. The spa authenticator server code was not checking that the input it
4426 received was valid base64.
4428 16. The debug output line for the "set" modifier in ACLs was not showing the
4429 name of the variable that was being set.
4431 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4433 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4434 lookup on its own. Something else was needed to trigger the lookup. For
4435 example, a match in host_lookup or the need for a host name in a host list.
4436 Now, if $sender_host_name is referenced and the host name has not yet been
4437 looked up, a lookup is performed. If the lookup fails, the variable remains
4438 empty, and $host_lookup_failed is set to "1".
4440 19. Added "eqi" as a case-independent comparison operator.
4442 20. The saslauthd authentication condition could segfault if neither service
4443 nor realm was specified.
4445 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4446 error message that was logged was misleading, and incoming SMTP
4447 connections were dropped. The message is now more accurate, and temporary
4448 errors are given to SMTP connections.
4450 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4451 (except RSET) with a 421 error, until QUIT is received. However, it was
4452 failing to send a response to QUIT.
4454 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4455 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4456 ACL is now run after the helo_try_verify_hosts code.
4458 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4459 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4460 case-independent, so other case variants are also recognized). Apparently
4461 some systems use these upper case variants.
4463 25. If more than two messages were waiting for the same host, and a transport
4464 filter was specified for the transport, Exim sent two messages over the
4465 same TCP/IP connection, and then failed with "socket operation on non-
4466 socket" when it tried to send the third.
4468 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4470 27. The extern definition of crypt16() in expand.c was not being excluded when
4471 the OS had its own crypt16() function.
4473 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4474 as a preferred synonym for return_size_limit, both as an option and as an
4477 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4479 30. Changed the default configuration ACL to relax the local part checking rule
4480 for addresses that are not in any local domains. For these addresses,
4481 slashes and pipe symbols are allowed within local parts, but the sequence
4482 /../ is explicitly forbidden.
4484 31. SPA server authentication was not clearing the challenge buffer before
4487 32. log_message in a "warn" ACL statement was writing to the reject log as
4488 well as to the main log, which contradicts the documentation and doesn't
4489 seem right (because no rejection is happening). So I have stopped it.
4491 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4492 However, I am unable to do any testing of this.
4494 34. Fixed an infelicity in the appendfile transport. When checking directories
4495 for a mailbox, to see if any needed to be created, it was accidentally
4496 using path names with one or more superfluous leading slashes; tracing
4497 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4499 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4500 contents are added to the log line that is written for every discarded
4501 recipient. (Previously a log_message setting was ignored.)
4503 36. The ${quote: operator now quotes the string if it is empty.
4505 37. The install script runs exim in order to find its version number. If for
4506 some reason other than non-existence or emptiness, which it checks, it
4507 could not run './exim', it was installing it with an empty version number,
4508 i.e. as "exim-". This error state is now caught, and the installation is
4511 38. An argument was missing from the function that creates an error message
4512 when Exim fails to connect to the socket for saslauthd authentication.
4513 This could cause Exim to crash, or give a corrupted message.
4515 39. Added isip, isip4, and isip6 to ${if conditions.
4517 40. The ACL variables $acl_xx are now saved with the message, and can be
4518 accessed later in routers, transports, and filters.
4520 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4521 strings in the file are not string-expanded.
4523 42. If a MAIL command specified a SIZE value that was too large to fit into an
4524 int variable, the check against message_size_limit failed. Such values are
4525 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4526 day this will have to be increased, but I don't think I want to be around
4527 when emails are that large.
4534 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4535 iconv() is not standard in FreeBSD.
4537 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4538 IPv6 enabled. The observed symptom was a segmentation fault on return from
4539 the function os_common_find_running_interfaces() in src/os.c.
4541 3. In the check_special_case() function in daemon.c I had used "errno" as an
4542 argument name, which causes warnings on some systems. This was basically a
4543 typo, since it was named "eno" in the comments!
4545 4. The code that waits for the clock to tick (at a resolution of some fraction
4546 of a second) so as to ensure message-id uniqueness was always waiting for
4547 at least one whole tick, when it could have waited for less. [This is
4548 almost certainly not relevant at current processor speeds, where it is
4549 unlikely to ever wait at all. But we try to future-proof.]
4551 5. The function that sleeps for a time interval that includes fractions of a
4552 second contained a race. It did not block SIGALRM between setting the
4553 timer, and suspending (a couple of lines later). If the interval was short
4554 and the sigsuspend() was delayed until after it had expired, the suspension
4555 never ended. On busy systems this could lead to processes getting stuck for
4558 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4559 process, before it forks any delivery processes. The open lookup caching
4560 mechanism meant that the open file or database connection was passed into
4561 the delivery process. The problem was that delivery processes always tidy
4562 up cached lookup data. This could cause a problem for the next delivery
4563 process started by the queue runner, because the external queue runner
4564 process does not know about the closure. So the next delivery process
4565 still has data in the lookup cache. In the case of a file lookup, there was
4566 no problem because closing a file descriptor in a subprocess doesn't affect
4567 the parent. However, if the lookup was caching a connection to a database,
4568 the connection was closed, and the second delivery process was likely to
4569 see errors such as "PGSQL: query failed: server closed the connection
4570 unexpectedly". The problem has been fixed by closing all cached lookups
4571 in a queue runner before running a delivery process.
4573 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4574 seem to have the "const" qualifier which it has on other OS. I've
4577 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4578 *of the same type* (client or server) with the same public name that an
4579 error should be diagnosed.
4581 9. When Exim looked up a host name for an IP address, but failed to find the
4582 original IP address when looking up the host name (a safety check), it
4583 output the message "<ip address> does not match any IP for NULL", which was
4584 confusing, to say the least. The bug was that the host name should have
4585 appeared instead of "NULL".
4587 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4588 user that is built into the binary, and the -C or -D options is used, root
4589 privilege is dropped before the configuration file is read. In addition,
4590 logging is switched to stderr instead of the normal log files. If the
4591 configuration then re-defines the Exim user, the unprivileged environment
4592 is probably not what is expected, so Exim logs a panic warning message (but
4595 However, if deliver_drop_privilege is set, the unprivileged state may well
4596 be exactly what is intended, so the warning has been cut out in that case,
4597 and Exim is allowed to try to write to its normal log files.
4603 1. smtp_return_error_details was not giving details for temporary sender
4604 or receiver verification errors.
4606 2. Diagnose a configuration error if two authenticators have the same public
4609 3. Exim used not to create the message log file for a message until the first
4610 delivery attempt. This could be confusing when incoming messages were held
4611 for policy or load reasons. The message log file is now created at the time
4612 the message is received, and an initial "Received" line is written to it.
4614 4. The automatically generated man page for command line options had a minor
4615 bug that caused no ill effects; however, a more serious problem was that
4616 the procedure for building the man page automatically didn't always
4617 operate. Consequently, release 4.20 contains an out-of-date version. This
4618 shouldn't happen again.
4620 5. When building Exim with embedded Perl support, the script that builds the
4621 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4622 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4624 6. The freeze_tell option was not being used for messages that were frozen on
4625 arrival, either by an ACL or by local_scan().
4627 7. Added the smtp_incomplete_transaction log selector.
4629 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4630 was accepting AUTH without a new EHLO.
4632 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4633 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4636 10. Logging of TCP/IP connections (when configured) now happens in the main
4637 daemon process instead of the child process, so that the TCP/IP connection
4638 count is more accurate (but it can never be perfect).
4640 11. The use of "drop" in a nested ACL was not being handled correctly in the
4641 outer ACL. Now, if condition failure induced by the nested "drop" causes
4642 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4643 or "require"), the connection is dropped.
4645 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4646 that yield "discard" can now be used with an "accept" or a "discard" verb,
4647 but an error is generated for any others (because I can't see a useful way
4648 to define what should happen).
4650 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4651 are now processed in the same way as lines in the Exim configuration file.
4652 In particular, continuation lines are supported.
4654 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4656 15. Added -ti meaning -t -i.
4658 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4659 domains, and warn by logging if others are found.
4661 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4662 returned from GIFCONF when Exim is trying to find the list of interfaces on
4663 a host. The code in os.c has been modified to copy each ifreq to an aligned
4664 structure in all cases.
4666 Also, in some cases, the returned ifreq's were being copied to a 'struct
4667 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4668 means the last couple of bytes of an IPv6 address could be chopped if the
4669 ifreq contained only a normal sockaddr (14 bytes storage).
4671 18. Named domain lists were not supported in the hosts_treat_as_local option.
4672 An entry such as +xxxx was not recognized, and was treated as a literal
4675 19. Ensure that header lines added by a DATA ACL are included in the reject log
4676 if the ACL subsequently rejects the message.
4678 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4679 MD5 (which is deprecated).
4681 21. When testing a filter file using -bf, Exim was writing a message when it
4682 took the sender from a "From " line in the message, but it was not doing so
4683 when it took $return_path from a Return-Path: header line. It now does.
4685 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4686 with a valid header line field name (a series of printing characters
4687 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4689 23. Changed "disc" in the source to "disk" to conform to the documentation and
4690 the book and for uniformity.
4692 24. Ignore Sendmail's -Ooption=value command line item.
4694 25. When execve() failed while trying to run a command in a pipe transport,
4695 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4696 could be confused with a return value of 69 from the command itself. This
4697 has been changed to 127, the value the shell returns if it is asked to run
4698 a non-existent command. The wording for the related log line suggests a
4699 non-existent command as the problem.
4701 26. If received_header_text expands to an empty string, do not add a Received:
4702 header line to the message. (Well, it adds a token one on the spool, but
4703 marks it "old" so that it doesn't get used or transmitted.)
4705 27. Installed eximstats 1.28 (addition of -nt option).
4707 28. There was no check for failure on the call to getsockname() in the daemon
4708 code. This can fail if there is a shortage of resources on the system, with
4709 ENOMEM, for example. A temporary error is now given on failure.
4711 29. Contrary to the C standard, it seems that in some environments, the
4712 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4713 program. Exim now does this explicitly; it affects the formatting of
4714 timestamps using strftime().
4716 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4717 complaints. I've now initialized all the variables, to avoid this.
4719 32. Header lines added by a system filter were not being "seen" during
4720 transport-time rewrites.
4722 33. The info_callback() function passed to OpenSSL is set up with type void
4723 (*)(SSL *, int, int), as described somewhere. However, when calling the
4724 function (actually a macro) that sets it up, the type void(*)() is
4725 expected. I've put in a cast to prevent warnings from picky compilers.
4727 34. If a DNS black list lookup found a CNAME record, but there were no A
4728 records associated with the domain it pointed at, Exim crashed.
4730 35. If a DNS black list lookup returned more than one A record, Exim ignored
4731 all but the first. It now scans all returned addresses if a particular IP
4732 value is being sought. In this situation, the contents of the
4733 $dnslist_value variable are a list of all the addresses, separated by a
4736 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4737 to accept a host name and all its aliases if the forward lookup for any of
4738 them yielded the IP address of the incoming connection. Now it accepts only
4739 those names whose forward lookup yields the correct IP address. Any other
4740 names are discarded. This closes a loophole whereby a rogue DNS
4741 administrator could create reverse DNS records to break through a
4742 wildcarded host restriction in an ACL.
4744 37. If a user filter or a system filter that ran in a subprocess used any of
4745 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4746 the wrong values were passed to the pipe command ($thisaddress had the
4747 value of $0, $0 had the value of $1, etc). This bug was introduced by
4748 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4750 38. Improved the line breaking for long SMTP error messages from ACLs.
4751 Previously, if there was no break point between 40 and 75 characters, Exim
4752 left the rest of the message alone. Two changes have been made: (a) I've
4753 reduced the minimum length to 35 characters; (b) if it can't find a break
4754 point between 35 and 75 characters, it looks ahead and uses the first one
4755 that it finds. This may give the occasional overlong line, but at least the
4756 remaining text gets split now.
4758 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4759 file descriptors might be low, and that setting 1000 would always raise it.
4760 It turns out that in some environments, the limit is already over 1000 and
4761 that lowering it causes trouble. So now Exim takes care not to decrease it.
4763 40. When delivering a message, the value of $return_path is set to $sender_
4764 address at the start of routing (routers may change the value). By an
4765 oversight, this default was not being set up when an address was tested by
4766 -bt or -bv, which affected the outcome if any router or filter referred to
4769 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4770 log only when "message" or "log_message" are set. However, if one of the
4771 conditions was an address verification, or a call to a nested ACL, the
4772 messages generated by the underlying test were being passed through. This
4773 no longer happens. The underlying message is available in $acl_verify_
4774 message for both "message" and "log_message" expansions, so it can be
4775 passed through if needed.
4777 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4778 new expansion $bh_ to give the encoded byte string without charset
4779 translation. Translation happens only if iconv() is available; HAVE_ICONV
4780 indicates this at build time. HEADERS_CHARSET gives the charset to
4781 translate to; headers_charset can change it in the configuration, and
4782 "headers charset" can change it in an individual filter file.
4784 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4785 that creates RFC 2047 encoded "words" labels them as that charset instead
4786 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4787 expansion operator; (ii) when Exim creates a From: line for a local
4788 message; (iii) when a header line is rewritten to include a "phrase" part.
4790 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4791 buggy, causing it to skip the first lines of messages whose message ID
4792 ended in 'D'. This would not have bitten before Exim release 4.14, because
4793 message IDs were unlikely to end in 'D' before then. The effect was to have
4794 incorrect size information for certain domains.
4796 45. #include "config.h" was missing at the start of the crypt16.c module. This
4797 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4800 46. If there was a timeout during a "random" callout check, Exim treated it as
4801 a failure of the random address, and carried on sending RSET and the real
4802 address. If the delay was just some slowness somewhere, the response to the
4803 original RCPT would be taken as a response to RSET and so on, causing
4804 mayhem of various kinds.
4806 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4807 when I implemented it. It didn't allow for the fact that some option values
4808 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4809 the right test for positive values.
4811 48. Domain names in DNS records are case-independent. Exim always looks them up
4812 in lower case. Some resolvers return domain names in exactly the case they
4813 appear in the zone file, that is, they may contain uppercase letters. Not
4814 all resolvers do this - some return always lower case. Exim was treating a
4815 change of case by a resolver as a change of domain, similar to a widening
4816 of a domain abbreviation. This triggered its re-routing code and so it was
4817 trying to route what was effectively the same domain again. This normally
4818 caused routing to fail (because the router wouldn't handle the domain
4819 twice). Now Exim checks for this case specially, and just changes the
4820 casing of the domain that it ultimately uses when it transmits the message
4823 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4826 50. If a filter generated a file delivery with a non-absolute name (possible if
4827 no home directory exists for the router), the forbid_file option was not
4830 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4831 the existing equality matching.
4833 52. Exim was using ints instead of ino_t variables in some places where it was
4834 dealing with inode numbers.
4836 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4837 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4838 called TMPDIR, and if it finds it is different, it changes its value.
4840 54. The smtp_printf() function is now made available to local_scan() so
4841 additional output lines can be written before returning. There is also an
4842 smtp_fflush() function to enable the detection of a dropped connection.
4843 The variables smtp_input and smtp_batched_input are exported to
4846 55. Changed the default runtime configuration: the message "Unknown user"
4847 has been removed from the ACL, and instead placed on the localuser router,
4848 using the cannot_route_message feature. This means that any verification
4849 failures that generate their own messages won't get overridden. Similarly,
4850 the "Unrouteable address" message that was in the ACL for unverifiable
4851 relay addresses has also been removed.
4853 56. Added hosts_avoid_esmtp to the smtp transport.
4855 57. The exicyclog script was not checking for the esoteric option
4856 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4857 will work only if exicyclog is run under the appropriate euid.
4859 58. Following a discussion on the list, the rules by which Exim recognises line
4860 endings on incoming messages have been changed. The -dropcr and drop_cr
4861 options are now no-ops, retained only for backwards compatibility. The
4862 following line terminators are recognized: LF CRLF CR. However, special
4863 processing applies to CR:
4865 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4866 nor a local message in the state where . is a terminator.
4868 (ii) If a bare CR is encountered in a header line, an extra space is added
4869 after the line terminator so as not to end the header. The reasoning
4870 behind this is that bare CRs in header lines are most likely either
4871 to be mistakes, or people trying to play silly games.
4873 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4874 was being incorrectly given as 18 bytes larger than it should have been.
4875 This is a VOB (very old bug).
4877 60. This may never have affected anything current, but just in case it has:
4878 When the local host is found other than at the start of a list of hosts,
4879 the local host, those with the same MX, and any that follow, are discarded.
4880 When the list in question was part of a longer list of hosts, the following
4881 hosts (not currently being processed) were also being discarded. This no
4882 longer happens. I'm not sure if this situation could ever has previously
4885 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4888 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4889 header. This is something that is recommended in a new Internet Draft, and
4890 is something that is documented as being done by Sendmail. There are two
4891 possible values. For messages generated by the autoreply transport, Exim
4894 Auto-Submitted: auto-replied
4896 whereas for all other generated messages (e.g. bounces) it adds
4898 Auto-Submitted: auto-generated
4900 63. The "personal" condition in filters now includes a test for the
4901 Auto-Submitted: header. If it contains the string "auto-" the message it
4902 not considered personal.
4904 64. Added rcpt_include_affixes as a generic transport option.
4906 65. Added queue_only_override (default true).
4908 66. Added the syslog_duplication option.
4910 67. If what should have been the first header line of a message consisted of
4911 a space followed by a colon, Exim was mis-interpreting it as a header line.
4912 It isn't of course - it is syntactically invalid and should therefore be
4913 treated as the start of the message body. The misbehaviour could have
4914 caused a number of strange effects, including loss of data in subsequent
4915 header lines, and spool format errors.
4917 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4918 client host had authenticated. This control can now be exercised by an ACL
4919 for more flexibility.
4921 69. By default, callouts do not happen when testing with -bh. There is now a
4922 variant, -bhc, which does actually run the callout code, including
4923 consulting and updating the callout cache.
4925 70. Added support for saslauthd authentication, courtesy of Alexander
4928 71. If statvfs() failed on the spool or log directories while checking their
4929 size for availability, Exim confusingly gave the error "space shortage".
4930 Furthermore, in debugging mode it crashed with a floating point exception.
4931 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4932 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4933 serious problem, Exim now writes to the main and panic logs when this
4934 happens, with details of the failure. It then refuses to accept the
4935 incoming message, giving the message "spool directory problem" or "log
4936 directory problem" with a 421 code for SMTP messages.
4938 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4939 0, 1, and 2 exist, because some OS complain for execs without them (see
4940 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4941 descriptors. However, the code omitted to check that the open succeeded,
4942 causing mysterious errors if for some reason the permissions on /dev/null
4943 got screwed. Now Exim writes a message to the main and panic logs, and
4944 bombs out if it can't open /dev/null.
4946 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4947 interact so that it is all more flexible. It is supposed to remain
4948 backwards compatible. Also added extra_local_interfaces.
4950 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4951 to bomb out with an assertion failure - to the client this appears as a
4952 connection drop. This problem occurs in the part of the code that was taken
4953 from the Samba project. Fortunately, the assertion is in a very simple
4954 function, so I have fixed this by reproducing the function inline in the
4955 one place where it is called, and arranging for authentication to fail
4956 instead of killing the process with assert().
4958 75. The SPA client code was not working when the server requested OEM rather
4959 than Unicode encoding.
4961 76. Added code to make require_files with a specific uid setting more usable in
4962 the case where statting the file as root fails - usually a non-root-mounted
4963 NFS file system. When this happens and the failure is EACCES, Exim now
4964 forks a subprocess and does the per-uid checking as the relevant uid.
4966 77. Added process_log_path.
4968 78. If log_file_path was not explicitly set, a setting of check_log_space or
4969 check_log_inodes was ignored.
4971 79. If a space check for the spool or log partitions fails, the incident is now
4972 logged. Of course, in the latter case the data may get lost...
4974 80. Added the %p formatting code to string_format() so that it can be used to
4975 print addresses in debug_print(). Adjusted all the address printing in the
4976 debugging in store.c to use %p rather than %d.
4978 81. There was a concern that a line of code in smtp_in.c could overflow a
4979 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4980 initially expressed, the concern was not well-founded, because trailing
4981 spaces are removed early. However, if the trailing spaces were followed by
4982 a NULL, they did not get removed, so the overflow was possible. Two fixes
4985 (a) I re-wrote the offending code in a cleaner fashion.
4986 (b) If an incoming SMTP command contains a NULL character, it is rejected
4989 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4990 runs initgroups(), so that if the Exim user is in any additional groups,
4991 they will be used during message reception.
4997 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob