1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.579 2009/10/20 13:10:15 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/27 Bugzilla 807: Improvements to LMTP delivery logging
139 TK/01 Add preliminary DKIM support. Currently requires a forked version of
140 ALT-N's libdkim that I have put here:
141 http://duncanthrax.net/exim-experimental/
143 Note to Michael Haardt: I had to rename some vars in sieve.c. They
144 were called 'true' and it seems that C99 defines that as a reserved
145 keyword to be used with 'bool' variable types. That means you could
146 not include C99-style headers which use bools without triggering
147 build errors in sieve.c.
149 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
150 as mailq or other aliases. Changed the --help handling significantly
151 to do whats expected. exim_usage() emits usage/help information.
153 SC/01 Added the -bylocaldomain option to eximstats.
155 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
157 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
159 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
165 PH/01 Another patch from the Sieve maintainer.
167 PH/02 When an IPv6 address is converted to a string for single-key lookup
168 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
169 dots are used instead of colons so that keys in lsearch files need not
170 contain colons. This was done some time before quoting was made available
171 in lsearch files. However, iplsearch files do require colons in IPv6 keys
172 (notated using the quote facility) so as to distinguish them from IPv4
173 keys. This meant that lookups for IP addresses in host lists did not work
174 for iplsearch lookups.
176 This has been fixed by arranging for IPv6 addresses to be expressed with
177 colons if the lookup type is iplsearch. This is not incompatible, because
178 previously such lookups could never work.
180 The situation is now rather anomolous, since one *can* have colons in
181 ordinary lsearch keys. However, making the change in all cases is
182 incompatible and would probably break a number of configurations.
184 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
187 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
188 conversion specification without a maximum field width, thereby enabling
189 a rogue spamd server to cause a buffer overflow. While nobody in their
190 right mind would setup Exim to query an untrusted spamd server, an
191 attacker that gains access to a server running spamd could potentially
192 exploit this vulnerability to run arbitrary code as the Exim user.
194 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
195 $primary_hostname instead of what libspf2 thinks the hosts name is.
197 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
198 a directory entry by the name of the lookup key. Previously, if a
199 symlink pointed to a non-existing file or a file in a directory that
200 Exim lacked permissions to read, a lookup for a key matching that
201 symlink would fail. Now it is enough that a matching directory entry
202 exists, symlink or not. (Bugzilla 503.)
204 PH/03 The body_linecount and body_zerocount variables are now exported in the
207 PH/04 Added the $dnslist_matched variable.
209 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
210 This means they are set thereafter only if the connection becomes
213 PH/06 Added the client_condition to authenticators so that some can be skipped
214 by clients under certain conditions.
216 PH/07 The error message for a badly-placed control=no_multiline_responses left
217 "_responses" off the end of the name.
219 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
221 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
222 (without spaces) instead of just copying the configuration text.
224 PH/10 Added the /noupdate option to the ratelimit ACL condition.
226 PH/11 Added $max_received_linelength.
228 PH/12 Added +ignore_defer and +include_defer to host lists.
230 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
231 way in which PCRE > 7.0 is built.
233 PH/14 Implemented queue_only_load_latch.
235 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
236 MAIL command. The effect was to mangle the value on 64-bit systems.
238 PH/16 Another patch from the Sieve maintainer.
240 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
242 PH/18 If a system quota error occurred while trying to create the file for
243 a maildir delivery, the message "Mailbox is full" was not appended to the
244 bounce if the delivery eventually timed out. Change 4.67/27 below applied
245 only to a quota excession during the actual writing of the file.
247 PH/19 It seems that peer DN values may contain newlines (and other non-printing
248 characters?) which causes problems in log lines. The DN values are now
249 passed through string_printing() before being added to log lines.
251 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
252 and InterBase are left for another time.)
254 PH/21 Added message_body_newlines option.
256 PH/22 Guard against possible overflow in moan_check_errorcopy().
258 PH/23 POSIX allows open() to be a macro; guard against that.
260 PH/24 If the recipient of an error message contained an @ in the local part
261 (suitably quoted, of course), incorrect values were put in $domain and
262 $local_part during the evaluation of errors_copy.
268 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
269 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
272 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
273 issue a MAIL command.
275 PH/02 In an ACL statement such as
277 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
279 if a client was not listed at all, or was listed with a value other than
280 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
281 the condition was not true (as it should be), so access was not denied.
282 The bug was that the ! inversion was incorrectly passed on to the second
283 item. This has been fixed.
285 PH/03 Added additional dnslists conditions == and =& which are different from
286 = and & when the dns lookup returns more than one IP address.
288 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
289 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
291 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
292 FSYNC, which compiles an option called disable_fsync that allows for
293 bypassing fsync(). The documentation is heavily laced with warnings.
295 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
297 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
298 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
299 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
300 including adding "make clean"; (3) Added -fPIC when compiling the test
301 dynamically loaded module, to get rid of a warning.
303 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
304 message fails, move_frozen_messages = true and ignore_bounce_errors_after
305 = 0s. The bug is otherwise harmless.
307 PH/07 There was a bug in the dovecot authenticator such that the value of
308 $auth1 could be overwritten, and so not correctly preserved, after a
309 successful authentication. This usually meant that the value preserved by
310 the server_setid option was incorrect.
312 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
314 PH/09 Installed PCRE release 7.0.
316 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
317 run for batched SMTP input. It is now run at the start of every message
318 in the batch. While fixing this I discovered that the process information
319 (output by running exiwhat) was not always getting set for -bs and -bS
320 input. This is fixed, and it now also says "batched" for BSMTP.
322 PH/11 Added control=no_pipelining.
324 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
325 patch, slightly modified), and move the expansion of helo_data till after
326 the connection is made in the smtp transport (so it can use these
329 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
331 PH/14 Added log_selector = +pid.
333 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
335 PH/16 Add ${if forany and ${if forall.
337 PH/17 Added dsn_from option to vary the From: line in DSNs.
339 PH/18 Flush SMTP output before performing a callout, unless control =
340 no_callout_flush is set.
342 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
343 was true (the default) a successful delivery failed to delete the retry
344 item, thus causing premature timeout of the address. The bug is now
347 PH/20 Added hosts_avoid_pipelining to the smtp transport.
349 PH/21 Long custom messages for fakedefer and fakereject are now split up
350 into multiline reponses in the same way that messages for "deny" and
351 other ACL rejections are.
353 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
354 with slight modification.
356 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
357 draft, changing the syntax and factoring some duplicate code".
359 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
360 for deliveries of the second and subsequent messages over the same SMTP
363 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
364 ${reduce, with only minor "tidies".
366 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
368 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
369 expansion side effects.
371 PH/27 When a message times out after an over-quota error from an Exim-imposed
372 quota, the bounce message says "mailbox is full". This message was not
373 being given when it was a system quota that was exceeded. It now should
376 MH/03 Made $recipients available in local_scan(). local_scan() already has
377 better access to the recipient list through recipients_list[], but
378 $recipients can be useful in postmaster-provided expansion strings.
380 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
381 in the case of a MAIL command with additional options following the
382 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
383 were accidentally chopped off.
385 PH/29 SMTP synchronization checks are implemented when a command is read -
386 there is a check that no more input is waiting when there shouldn't be
387 any. However, for some commands, a delay in an ACL can mean that it is
388 some time before the response is written. In this time, more input might
389 arrive, invalidly. So now there are extra checks after an ACL has run for
390 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
391 pipelining has not been advertised.
393 PH/30 MH's patch to allow iscntrl() characters to be list separators.
395 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
396 returned in the SMTP response when smtp_return_error_details was false.
399 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
400 instead of the C I/O that was originally supplied, because problems were
403 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
404 Exim which did not show up earlier: it was assuming that a call to
405 SSL_CTX_set_info_callback() might give an error value. In fact, there is
406 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
407 was a macro that became an assignment, so it seemed to work. This has
408 changed to a proper function call with a void return, hence the compile
409 error. Exim's code has been fixed.
411 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
414 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
416 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
418 PH/37 If a message is not accepted after it has had an id assigned (e.g.
419 because it turns out to be too big or there is a timeout) there is no
420 "Completed" line in the log. When some messages of this type were
421 selected by exigrep, they were listed as "not completed". Others were
422 picked up by some special patterns. I have improved the selection
423 criteria to be more general.
425 PH/38 The host_find_failed option in the manualroute router can now be set
426 to "ignore", to completely ignore a host whose IP address cannot be
427 found. If all hosts are ignored, the behaviour is controlled by the new
428 host_all_ignored option.
430 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
431 homing or because of multiple MX records with /mx) generated more than
432 one IP address, and the following item turned out to be the local host,
433 all the secondary addresses of the first item were incorrectly removed
434 from the list, along with the local host and any following hosts (which
435 is what is supposed to happen).
437 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
438 whoever called Exim into the -H file. In the case of the daemon it was
439 behaving confusingly. When first started, it used values for whoever
440 started the daemon, but after a SIGHUP it used the Exim user (because it
441 calls itself on a restart). I have changed the code so that it now always
444 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
445 message are rejected with the same error (e.g. no authentication or bad
446 sender address), and a DATA command is nevertheless sent (as can happen
447 with PIPELINING or a stupid MUA), the error message that was given to the
448 RCPT commands is included in the rejection of the DATA command. This is
449 intended to be helpful for MUAs that show only the final error to their
452 PH/42 Another patch from the Sieve maintainer.
454 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
455 Eximstats - Fixed some broken HTML links and added missing column headers
457 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
458 columns for Rejects, Temp Rejects, Ham, and Spam rows.
460 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
462 PH/43 Yet another patch from the Sieve maintainer.
464 PH/44 I found a way to check for a TCP/IP connection going away before sending
465 the response to the final '.' that terminates a message, but only in the
466 case where the client has not sent further data following the '.'
467 (unfortunately, this is allowed). However, in many cases there won't be
468 any further data because there won't be any more messages to send. A call
469 to select() can be used: if it shows that the input is "ready", there is
470 either input waiting, or the socket has been closed. An attempt to read
471 the next input character can distinguish the two cases. Previously, Exim
472 would have sent an OK response which the client would never have see.
473 This could lead to message repetition. This fix should cure that, at
474 least in a lot of common cases.
476 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
477 advertised in response to EHLO.
483 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
484 fixed by 4.65/MH/01 (is this a record?) are fixed:
486 (i) An empty string was always treated as zero by the numeric comparison
487 operators. This behaviour has been restored.
489 (ii) It is documented that the numeric comparison operators always treat
490 their arguments as decimal numbers. This was broken in that numbers
491 starting with 0 were being interpreted as octal.
493 While fixing these problems I realized that there was another issue that
494 hadn't been noticed. Values of message_size_limit (both the global option
495 and the transport option) were treated as octal if they started with 0.
496 The documentation was vague. These values are now always treated as
497 decimal, and I will make that clear in the documentation.
503 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
504 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
507 MH/01 Don't check that the operands of numeric comparison operators are
508 integers when their expansion is in "skipping" mode (fixes bug
509 introduced by 4.64-PH/07).
511 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
512 child addresses, Exim now panics and dies. Previously, because the count
513 is held in a short int, deliveries were likely to be lost. As such a
514 large number of recipients for a single message is ridiculous
515 (performance will be very, very poor), I have chosen to impose a limit
516 rather than extend the field.
522 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
523 leftover -K file (the existence of which was triggered by #402).
524 While we were at it, introduced process PID as part of the -K
525 filename. This should rule out race conditions when creating
528 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
529 processing considerably. Previous code took too long for large mails,
530 triggering a timeout which in turn triggers #401.
532 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
533 in the DK code in transports.c. sendfile() is not really portable,
534 hence the _LINUX specificness.
536 TF/01 In the add_headers option to the mail command in an Exim filter,
537 there was a bug that Exim would claim a syntax error in any
538 header after the first one which had an odd number of characters
541 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
542 callout verification, Exim cached a "reject" for the entire domain. This
543 is correct for most verifications, but it is not correct for a recipient
544 verification with use_sender or use_postmaster set, because in that case
545 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
546 case of MAIL FROM:<> rejection from other early rejections (e.g.
547 rejection of HELO). When verifying a recipient using a non-null MAIL
548 address, the cache is ignored if it shows MAIL FROM:<> rejection.
549 Whatever the result of the callout, the value of the domain cache is
550 left unchanged (for any other kind of callout, getting as far as trying
551 RCPT means that the domain itself is ok).
553 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
556 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
557 manifest itself as EPIPE rather than ECONNECT. When tidying away a
558 session, the daemon ignores ECONNECT errors and logs others; it now
559 ignores EPIPE as well.
561 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
562 (quoted-printable decoding).
564 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
565 later the small subsequent patch to fix an introduced bug.
567 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
569 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
571 PH/08 An error is now given if message_size_limit is specified negative.
573 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
574 to be given (somewhat) arbitrary names.
576 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
579 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
580 miscellaneous code fixes
582 PH/10 Added the log_reject_target ACL modifier to specify where to log
585 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
586 hostname. This is wrong, because it relates to the incoming message (and
587 probably the interface on which it is arriving) and not to the outgoing
588 callout (which could be using a different interface). This has been
589 changed to use the value of the helo_data option from the smtp transport
590 instead - this is what is used when a message is actually being sent. If
591 there is no remote transport (possible with a router that sets up host
592 addresses), $smtp_active_hostname is used.
594 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
595 tweaks were necessary in order to get it to work (see also 21 below):
596 (a) The code assumed that strncpy() returns a negative number on buffer
597 overflow, which isn't the case. Replaced with Exim's string_format()
599 (b) There were several signed/unsigned issues. I just did the minimum
600 hacking in of casts. There is scope for a larger refactoring.
601 (c) The code used strcasecmp() which is not a standard C function.
602 Replaced with Exim's strcmpic() function.
603 (d) The code set only $1; it now sets $auth1 as well.
604 (e) A simple test gave the error "authentication client didn't specify
605 service in request". It would seem that Dovecot has changed its
606 interface. Fortunately there's a specification; I followed it and
607 changed what the client sends and it appears to be working now.
609 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
612 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
613 address A is aliased to B and C, where B exists and C does not. Without
614 -v the output is "A verified" because verification stops after a
615 successful redirection if more than one address is generated. However,
616 with -v the child addresses are also verified. Exim was outputting "A
617 failed to verify" and then showing the successful verification for C,
618 with its parentage. It now outputs "B failed to verify", showing B's
619 parentage before showing the successful verification of C.
621 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
622 look up a TXT record in a specific list after matching in a combined
625 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
626 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
627 they consult the DNS. I had assumed they would set it the way they
628 wanted; and indeed my experiments on Linux seem to show that in some
629 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
630 To be on the safe side, however, I have now made the interface to
631 host_find_byname() similar to host_find_bydns(), with an argument
632 containing the DNS resolver options. The host_find_byname() function now
633 sets these options at its start, just as host_find_bydns() does. The smtp
634 transport options dns_qualify_single and dns_search_parents are passed to
635 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
636 of host_find_byname() use the default settings of RES_DEFNAMES
637 (qualify_single) but not RES_DNSRCH (search_parents).
639 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
640 spool_read_header() do less string testing, by means of a preliminary
641 switch on the second character of optional "-foo" lines. (This is
642 overdue, caused by the large number of possibilities that now exist.
643 Originally there were few.) While I was there, I also converted the
644 str(n)cmp tests so they don't re-test the leading "-" and the first
645 character, in the hope this might squeeze out yet more improvement.
647 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
648 flag allowing group syntax was set by the header_syntax check but not
649 turned off, possible causing trouble later; (2) The flag was not being
650 set at all for the header_verify test, causing "group"-style headers to
651 be rejected. I have now set it in this case, and also caused header_
652 verify to ignore an empty address taken from a group. While doing this, I
653 came across some other cases where the code for allowing group syntax
654 while scanning a header line wasn't quite right (mostly, not resetting
655 the flag correctly in the right place). These bugs could have caused
656 trouble for malformed header lines. I hope it is now all correct.
658 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
659 with the "reply" argument non-NULL. The code, however (which originally
660 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
661 but it didn't always do it. This confused somebody who was copying the
662 code for some other use. I have removed all the tests.
664 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
665 feature that was used to support insecure browsers during the U.S. crypto
666 embargo. It requires special client support, and Exim is probably the
667 only MTA that supported it -- and would never use it because real RSA is
668 always available. This code has been removed, because it had the bad
669 effect of slowing Exim down by computing (never used) parameters for the
670 RSA_EXPORT functionality.
672 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
673 authenticator to fail if there's a tab character in the incoming data
674 (there should never be unless someone is messing about, as it's supposed
675 to be base64-encoded). Also added, on Timo's advice, the "secured" option
676 if the connection is using TLS or if the remote IP is the same as the
677 local IP, and the "valid-client-cert option" if a client certificate has
680 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
681 authenticators. This can be used for authorization after authentication
682 succeeds. (In the case of plaintext, it servers for both authentication
685 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
686 if any retry times were supplied.
688 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
689 connection, where there is no client IP address to check. In this
690 situation, the verify now always succeeds.
692 PH/25 Applied John Jetmore's -Mset patch.
694 PH/26 Added -bem to be like -Mset, but loading a message from a file.
696 PH/27 In a string expansion for a processed (not raw) header when multiple
697 headers of the same name were present, leading whitespace was being
698 removed from all of them, but trailing whitespace was being removed only
699 from the last one. Now trailing whitespace is removed from each header
700 before concatenation. Completely empty headers in a concatenation (as
703 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
704 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
706 PH/29 [Removed. This was a change that I later backed out, and forgot to
707 correct the ChangeLog entry (that I had efficiently created) before
708 committing the later change.]
710 PH/30 Exim was sometimes attempting to deliver messages that had suffered
711 address errors (4xx response to RCPT) over the same connection as other
712 messages routed to the same hosts. Such deliveries are always "forced",
713 so retry times are not inspected. This resulted in far too many retries
714 for the affected addresses. The effect occurred only when there were more
715 hosts than the hosts_max_try setting in the smtp transport when it had
716 the 4xx errors. Those hosts that it had tried were not added to the list
717 of hosts for which the message was waiting, so if all were tried, there
718 was no problem. Two fixes have been applied:
720 (i) If there are any address or message errors in an SMTP delivery, none
721 of the hosts (tried or untried) are now added to the list of hosts
722 for which the message is waiting, so the message should not be a
723 candidate for sending over the same connection that was used for a
724 successful delivery of some other message. This seems entirely
725 reasonable: after all the message is NOT "waiting for some host".
726 This is so "obvious" that I'm not sure why it wasn't done
727 previously. Hope I haven't missed anything, but it can't do any
728 harm, as the worst effect is to miss an optimization.
730 (ii) If, despite (i), such a delivery is accidentally attempted, the
731 routing retry time is respected, so at least it doesn't keep
732 hammering the server.
734 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
735 in ${readsocket because some servers need this prod.
737 PH/32 Added some extra debug output when updating a wait-xxx database.
739 PH/33 The hint "could be header name not terminated by colon", which has been
740 given for certain expansion errors for a long time, was not being given
741 for the ${if def:h_colon_omitted{... case.
743 PH/34 The spec says: "With one important exception, whenever a domain list is
744 being scanned, $domain contains the subject domain." There was at least
745 one case where this was not true.
747 PH/35 The error "getsockname() failed: connection reset by peer" was being
748 written to the panic log as well as the main log, but it isn't really
749 panic-worthy as it just means the connection died rather early on. I have
750 removed the panic log writing for the ECONNRESET error when getsockname()
753 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
754 runs only) independently of the message's sender address. This meant
755 that, if the 4xx error was in fact related to the sender, a different
756 message to the same recipient with a different sender could confuse
757 things. In particualar, this can happen when sending to a greylisting
758 server, but other circumstances could also provoke similar problems.
759 I have changed the default so that the retry time for these errors is now
760 based a combination of the sender and recipient addresses. This change
761 can be overridden by setting address_retry_include_sender=false in the
764 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
765 remote server are returned as part of bounce messages. This was not
766 happening for LMTP over a pipe (the lmtp transport), but now it is the
767 same for both kinds of LMTP.
769 PH/38 Despite being documented as not happening, Exim was rewriting addresses
770 in header lines that were in fact CNAMEs. This is no longer the case.
772 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
773 and queue runs started by the daemon processed all messages. This has
774 been fixed so that -R and -S can now usefully be given with -q<time>.
776 PH/40 Import PCRE release 6.7 (fixes some bugs).
778 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
780 PH/42 Give an error if -q is specified more than once.
782 PH/43 Renamed the variables $interface_address and $interface_port as
783 $received_ip_address and $received_port, to make it clear that these
784 values apply to message reception, and not to the outgoing interface when
785 a message is delivered. (The old names remain recognized, of course.)
787 PH/44 There was no timeout on the connect() call when using a Unix domain
788 socket in the ${readsocket expansion. There now is.
790 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
791 be meaningful with "accept".
793 SC/01 Eximstats V1.43
794 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
796 SC/02 Eximstats V1.44
797 Use a glob alias rather than an array ref in the generated
798 parser. This improves both readability and performance.
800 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
801 Collect SpamAssassin and rejection statistics.
802 Don't display local sender or destination tables unless
803 there is data to show.
804 Added average volumes into the top table text output.
806 SC/04 Eximstats V1.46
807 Collect data on the number of addresses (recipients)
808 as well as the number of messages.
810 SC/05 Eximstats V1.47
811 Added 'Message too big' to the list of mail rejection
812 reasons (thanks to Marco Gaiarin).
814 SC/06 Eximstats V1.48
815 Mainlog lines which have GMT offsets and are too short to
816 have a flag are now skipped.
818 SC/07 Eximstats V1.49 (Alain Williams)
819 Added the -emptyok flag.
821 SC/08 Eximstats V1.50
822 Fixes for obtaining the IP address from reject messages.
824 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
825 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
826 whitesspace changes from 4.64-PH/27
828 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
831 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
832 are found, allow negative numbers in numeric criteria)
834 JJ/06 exipick.20061117.2, added new $message_body_missing variable
836 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
837 to match changes made in 4.64-PH/43
839 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
841 PH/47 Put in an explicit test for a DNS lookup of an address record where the
842 "domain" is actually an IP address, and force a failure. This locks out
843 those revolvers/nameservers that support "A-for-A" lookups, in
844 contravention of the specifications.
846 PH/48 When a host name was looked up from an IP address, and the subsequent
847 forward lookup of the name timed out, the host name was left in
848 $sender_host_name, contrary to the specification.
850 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
851 restricted to single-key lookups, Exim was not diagnosing an error if
852 * or *@ was used with a query-style lookup.
854 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
856 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
857 long ago, but noone interested enough thought of it. Let's just say that
858 the "1.1" means that there are some new functions that weren't there at
859 some point in the past.
861 PH/51 Error processing for expansion failure of helo_data from an smtp
862 transport during callout processing was broken.
864 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
865 tested/used via the -bh/-bhc/-bs options.
867 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
868 bug, fixed in subsequent PCRE releases).
870 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
871 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
873 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
879 SC/01 Use a glob alias rather than an array ref in eximstats generated
880 parser. This improves both readability and performance.
882 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
883 Don't display local sender or destination tables in eximstats unless
884 there is data to show.
885 Added average volumes into the eximstats top table text output.
887 SC/03 Collect data on the number of addresses (recipients) as well
888 as the number of messages in eximstats.
890 TF/01 Correct an error in the documentation for the redirect router. Exim
891 does (usually) call initgroups() when daemonizing.
893 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
894 with consistent privilege compared to when running as a daemon.
896 TF/03 Note in the spec that $authenticated_id is not set for local
897 submissions from trusted users.
899 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
900 Thanks to Dean Brooks <dean@iglou.com> for the patch.
902 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
903 by adding some example configuration directives to the default
904 configuration file. A little bit of work is required to uncomment the
905 directives and define how usernames and passwords are checked, but
906 there is now a framework to start from.
908 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
909 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
910 without this. I don't know how relevant this is to other LDAP libraries.
912 PH/02 Add the verb name to the "unknown ACL verb" error.
914 PH/03 Magnus Holmgren's patch for filter_prepend_home.
916 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
918 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
919 directory not expanded when it should be if an expanded home directory
920 was set for the address (which is overridden by the transport).
922 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
925 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
926 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
927 because it is too late at that time, and has no effect.
929 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
930 security issue with \' (bugzilla #107). I could not use the
931 PQescapeStringConn() function, because it needs a PGconn value as one of
934 PH/08 When testing addresses using -bt, indicate those final addresses that
935 are duplicates that would not cause an additional delivery. At least one
936 person was confused, thinking that -bt output corresponded to deliveries.
937 (Suppressing duplicates isn't a good idea as you lose the information
938 about possibly different redirections that led to the duplicates.)
940 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
941 systems where poll() doesn't work, in particular OS X.
943 PH/10 Added more information to debugging output for retry time not reached.
945 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
946 operations in malware.c.
948 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
951 PH/13 If write_rejectlog was set false when logging was sent to syslog with
952 syslog_duplication set false, log lines that would normally be written
953 both the the main log and to the reject log were not written to syslog at
956 PH/14 In the default configuration, change the use of "message" in ACL warn
957 statements to "add_header".
959 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
960 not followed by a command (e.g. "seen endif").
962 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
963 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
966 PH/17 Added extra conditions to the default value of delay_warning_condition
970 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
971 { match{$h_precedence:}{(?i)bulk|list|junk} } \
972 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
975 The Auto-Submitted: and various List- headers are standardised, whereas I
976 don't think Precedence: ever was.
978 PH/18 Refactored debugging code in route_finduser() to show more information,
979 in particular, the error code if getpwnam() issues one.
981 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
982 This is apparently needed in addition to the PH/07 change above to avoid
983 any possible encoding problems.
985 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
986 but not after initializing Perl.
988 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
989 output them only if debugging. By default they are written stderr,
990 apparently, which is not desirable.
992 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
995 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
998 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1000 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1001 authenticated or an ident call has been made. Suppress the default
1002 values for $authenticated_id and $authenticated_sender (but permit -oMai
1003 and -oMas) when testing with -bh.
1005 PH/24 Re-jigged the order of the tests in the default configuration so that the
1006 tests for valid domains and recipients precede the DNS black list and CSA
1007 tests, on the grounds that those ones are more expensive.
1009 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1010 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1011 This bug exists in every version of Exim that I still have, right back to
1014 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1015 However, an attempt to turn on case-sensitivity in a regex key by
1016 including (?-i) didn't work because the subject string was already
1017 lowercased, and the effects were non-intuitive. It turns out that a
1018 one-line patch can be used to allow (?-i) to work as expected.
1024 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1025 other effects) broke the use of negated acl sub-conditions.
1027 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1030 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1031 "Deny" causes Exim to reject the incoming connection with a 554 error.
1032 Unfortunately, if there is a major crisis, such as a disk failure,
1033 tcp-wrappers gives "deny", whereas what one would like would be some
1034 kind of temporary error. A kludge has been added to help with this.
1035 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1036 554 error is used if errno is still zero or contains ENOENT (which occurs
1037 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1040 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1042 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1043 errors. Otherwise a message that provokes a temporary error (when other
1044 messages do not) can cause a whole host to time out.
1046 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1047 addresses were routed directly to files or pipes from a redirect router.
1048 File deliveries just didn't batch; pipe deliveries might have suffered
1051 PH/06 A failure to get a lock for a hints database would erroneously always say
1052 "Failed to get write lock", even when it was really a read lock.
1054 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1055 of 0600. This has been changed to use the value of the lockfile_mode
1056 option (which defaults to 0600).
1058 PH/08 Applied small patch from the Sieve maintainer.
1060 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1061 folder from quota calculations, a direct delivery into this folder messed
1062 up the contents of the maildirsize file. This was because the regex was
1063 used only to exclude .Trash (or whatever) when the size of the mailbox
1064 was calculated. There was no check that a delivery was happening into an
1065 excluded directory. This bug has been fixed by ignoring all quota
1066 processing for deliveries into excluded directories.
1068 PH/10 Added the maildirfolder_create_regex option to appendfile.
1074 PH/01 The code for finding all the local interface addresses on a FreeBSD
1075 system running IPv6 was broken. This may well have applied to all BSD
1076 systems, as well as to others that have similar system calls. The broken
1077 code found IPv4 interfaces correctly, but gave incorrect values for the
1078 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1079 that it would not match correctly against @[] and not recognize the IPv6
1082 PH/02 The ipliteral router was not recognizing addresses of the form user@
1083 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1085 PH/03 Added disable_ipv6.
1087 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1088 decoded form, because it is most often used to construct To: headers
1089 lines in autoreplies, and the decoded form may well be syntactically
1090 invalid. However, $reply_address has leading white space removed, and all
1091 newlines turned into spaces so that the autoreply transport does not
1094 PH/05 If group was specified without a user on a router, and no group or user
1095 was specified on a transport, the group from the router was ignored.
1097 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1098 for visible compile-time settings that can be used to change these
1099 numbers, for those that want even more. Backwards compatibility with old
1100 spool files has been maintained. However, going back to a previous Exim
1101 release will lost any variables that are in spool files.
1103 PH/07 Two small changes when running in the test harness: increase delay when
1104 passing a TCP/IP connection to a new process, in case the original
1105 process has to generate a bounce, and remove special handling of
1106 127.0.0.2 (sic), which is no longer necessary.
1108 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1109 be the same on different OS.
1111 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1114 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1115 whether --show-vars was specified or not
1117 JJ/02 exipick: Added support for new ACL variable spool format introduced
1120 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1121 syntactically invalid From: or Reply-to: line, and a filter used this to
1122 generate an autoreply, and therefore failed to obtain an address for the
1123 autoreply, Exim could try to deliver to a non-existent relative file
1124 name, causing unrelated and misleading errors. What now happens is that
1125 it logs this as a hard delivery error, but does not attempt to create a
1128 PH/11 The exinext utility has a -C option for testing purposes, but although
1129 the given file was scanned by exinext itself; it wasn't being passed on
1130 when Exim was called.
1132 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1133 an end-of-file indication when reading a command response.
1135 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1136 compiled. In many other places in Exim, IPv6 addresses are always
1137 recognized, so I have changed this. It also means that IPv4 domain
1138 literals of the form [IPV4:n.n.n.n] are now always recognized.
1140 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1141 used if the router is not running as root, for example, when verifying at
1142 ACL time, or when using -bh. The debugging output from this situation was
1143 non-existent - all you got was a failure to exec. I have made two
1146 (a) Failures to set uid/gid, the current directory, or a process leader
1147 in a subprocess such as that created by queryprogram now generate
1148 suitable debugging ouput when -d is set.
1150 (b) The queryprogram router detects when it is not running as root,
1151 outputs suitable debugging information if -d is set, and then runs
1152 the subprocess without attempting to change uid/gid.
1154 PH/15 Minor change to Makefile for building test_host (undocumented testing
1157 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1158 additional section of a DNS packet that returns MX or SRV records.
1159 Instead, it always explicitly searches for A/AAAA records. This avoids
1160 major problems that occur when a DNS server includes only records of one
1161 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1162 fixed another bug: if SRV records were looked up and the corresponding
1163 address records were *not* found in the additional section, the port
1164 values from the SRV records were lost.
1166 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1167 using the correct key (the original address) when searching the retry
1168 rules in order to find which one to use for generating the retry hint.
1170 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1171 adding the default one. Similarly, if it contains a Reply-To: header, the
1172 errors_reply_to option, if set, is not used.
1174 PH/19 When calculating a retry time, Exim used to measure the "time since
1175 failure" by looking at the "first failed" field in the retry record. Now
1176 it does not use this if it is later than than the arrival time of the
1177 message. Instead it uses the arrival time. This makes for better
1178 behaviour in cases where some deliveries succeed, thus re-setting the
1179 "first failed" field. An example is a quota failure for a huge message
1180 when small messages continue to be delivered. Without this change, the
1181 "time since failure" will always be short, possible causing more frequent
1182 delivery attempts for the huge message than are intended.
1183 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1185 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1186 $1, $2, $3) because the numerical variables can be reset during some
1187 expansion items (e.g. "match"), thereby losing the authentication data.
1189 PH/21 Make -bV show the size of off_t variables so that the test suite can
1190 decide whether to run tests for quotas > 2G.
1192 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1193 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1194 filecount value is greater than 2G or if a quota value is greater than 2G
1195 on a system where the size of off_t is not greater than 4, a panic error
1198 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1199 never match. The debug and -bh output now contains an explicit error
1200 message indicating a malformed IPv4 address or mask.
1202 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1203 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1204 PH/23 above applies.
1206 PH/25 Do not write to syslog when running in the test harness. The only
1207 occasion when this arises is a failure to open the main or panic logs
1208 (for which there is an explicit test).
1210 PH/26 Added the /no_tell option to "control=freeze".
1212 PH/27 If a host name lookup failed very early in a connection, for example, if
1213 the IP address matched host_lookup and the reverse lookup yielded a name
1214 that did not have a forward lookup, an error message of the form "no IP
1215 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1216 could be logged. Now it outputs the IP address instead of "NULL".
1218 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1219 allows the sender and the authenticated sender to be set when
1220 submitting a message from within Exim. Since child_open_exim() is
1221 documented for local_scan(), the new function should be too.
1223 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1224 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1225 results in an empty string is now treated as unset.
1227 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1229 PH/31 Added sender_verify_fail logging option.
1231 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1232 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1233 tidied the source and removed it altogether.
1235 PH/33 When a queue run was abandoned because the load average was too high, a
1236 log line was always written; now it is written only if the queue_run log
1237 selector is set. In addition, the log line for abandonment now contains
1238 information about the queue run such as the pid. This is always present
1239 in "start" and "stop" lines but was omitted from the "abandon" line.
1241 PH/34 Omit spaces between a header name and the colon in the error message that
1242 is given when verify = headers_syntax fails (if there are lots of them,
1243 the message gets confusing).
1245 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1246 names, as there are now some PTR records that contain slashes. This check
1247 is only to protect against broken name servers that fall over on strange
1248 characters, so the fact that it applies to all lookups doesn't matter.
1250 PH/36 Now that the new test suite is complete, we can remove some of the
1251 special code in Exim that was needed for the old test suite. For example,
1252 sorting DNS records because real resolvers return them in an arbitrary
1253 order. The new test suite's fake resolver always returns records in the
1256 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1257 bounces) except when queue_only is set, to avoid logging races between
1258 the different processes.
1260 PH/38 Panic-die if .include specifies a non-absolute path.
1262 PH/39 A tweak to the "H" retry rule from its user.
1264 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1265 a label. They prevented compilation on older perls.
1267 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1268 a warning to be raised on newish perls.
1270 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1271 on queue. Changes to match documented behaviour of showing count of
1272 messages matching specified criteria.
1274 PH/40 Changed the default ident timeout from 30s to 5s.
1276 PH/41 Added support for the use of login_cap features, on those BSD systems
1277 that have them, for controlling the resources used by pipe deliveries.
1279 PH/42 The content-scanning code uses fopen() to create files in which to put
1280 message data. Previously it was not paying any attention to the mode of
1281 the files. Exim runs with umask(0) because the rest of the code creates
1282 files with open(), and sets the required mode explicitly. Thus, these
1283 files were ending up world-writeable. This was not a big issue, because,
1284 being within the spool directory, they were not world-accessible. I have
1285 created a function called modefopen, which takes an additional mode
1286 argument. It sets umask(777), creates the file, chmods it to the required
1287 mode, then resets the umask. All the relevant calls to fopen() in the
1288 content scanning code have been changed to use this function.
1290 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1291 to 24 hours. This avoids potential overflow problems when processing G
1292 and H retry rules. I suspect nobody ever tinkers with this value.
1294 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1296 PH/45 When the plaintext authenticator is running as a client, the server's
1297 challenges are checked to ensure they are valid base64 strings. By
1298 default, the authentication attempt is cancelled if an invalid string is
1299 received. Setting client_ignore_invalid_base64 true ignores these errors.
1300 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1301 they are received. Thus, the responses can be made to depend on the
1302 challenges. If an invalid string is ignored, an empty string is placed in
1305 PH/46 Messages that are created by the autoreply transport now contains a
1306 References: header, in accordance with RFCs 2822 and 3834.
1308 PH/47 Added authenticated_sender_force to the smtp transport.
1310 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1312 PH/49 Installed latest patch from the Sieve maintainer.
1314 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1315 was also set, the appendfile transport was unnecessarily scanning a
1316 directory of message files (e.g. for maildir delivery) to find the count
1317 of files (along with the size), even though it did not need this
1318 information. It now does the scan only if it needs to find either the
1319 size of the count of files.
1321 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1323 PH/52 Two bugs concerned with error handling when the smtp transport is
1326 (i) Exim was not creating retry information for temporary errors given
1327 for individual recipients after the DATA command when the smtp transport
1328 was used in LMTP mode. This meant that they could be retried too
1329 frequently, and not timed out correctly.
1331 (ii) Exim was setting the flag that allows error details to be returned
1332 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1333 recipients that were returned after the DATA command.
1335 PH/53 This is related to PH/52, but is more general: for any failing address,
1336 when detailed error information was permitted to be returned to the
1337 sender, but the error was temporary, then after the final timeout, only
1338 "retry timeout exceeded" was returned. Now it returns the full error as
1339 well as "retry timeout exceeded".
1341 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1342 do this, and (what is worse) MTAs that accept it.
1344 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1345 will now be deprecated.
1347 PH/56 New os.c-cygwin from the Cygwin maintainer.
1349 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1350 formats (previously only available in exim formats via -bpr, -bpru,
1351 and -bpra. Now also available in native and exiqgrep formats)
1353 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1354 with very large, slow to parse queues
1356 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1358 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1360 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1361 responses to authentication challenges, though it was showing the
1362 challenges; (ii) I've removed the CR characters from the debug output for
1365 PH/58 Allow for the insertion of a newline as well as a space when a string
1366 is turned into more than one encoded-word during RFC 2047 encoding. The
1367 Sieve code now uses this.
1369 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1370 data_4xx, lost_connection, tls_required.
1372 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1373 message was being sent as an SMTP response.
1375 PH/61 Add -l and -k options to exicyclog.
1377 PH/62 When verifying, if an address was redirected to one new address, so that
1378 verification continued, and the new address failed or deferred after
1379 having set something in $address_data, the value of $address_data was not
1380 passed back to the ACL. This was different to the case when no
1381 redirection occurred. The value is now passed back in both cases.
1383 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1384 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1385 use login_cap.h, so on its own it isn't the distinguishing feature. The
1386 new name refers directly to the setclassresources() function.
1388 PH/65 Added configuration files for NetBSD3.
1390 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1392 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1393 is preferred over IPv4.
1395 PH/68 The bounce_return_message and bounce_return_body options were not being
1396 honoured for bounces generated during the reception of non-SMTP messages.
1397 In particular, this applied to messages rejected by the ACL. This bug has
1398 been fixed. However, if bounce_return_message is true and bounce_return_
1399 body is false, the headers that are returned for a non-SMTP message
1400 include only those that have been read before the error was detected.
1401 (In the case of an ACL rejection, they have all been read.)
1403 PH/69 The HTML version of the specification is now built in a directory called
1404 spec_html instead of spec.html, because the latter looks like a path with
1405 a MIME-type, and this confuses some software.
1407 PH/70 Catch two compiler warnings in sieve.c.
1409 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1410 function verify_get_ident() calls ip_connect() to connect a socket, but
1411 if the "connect()" function timed out, ip_connect() used to close the
1412 socket. However, verify_get_ident() also closes the socket later, and in
1413 between Exim writes to the log, which may get opened at this point. When
1414 the socket was closed in ip_connect(), the log could get the same file
1415 descriptor number as the socket. This naturally causes chaos. The fix is
1416 not to close the socket in ip_connect(); the socket should be closed by
1417 the function that creates it. There was only one place in the code where
1418 this was missing, in the iplookup router, which I don't think anybody now
1419 uses, but I've fixed it anyway.
1421 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1422 well as to direct DNS lookups. Otherwise the handling of names in host
1423 lists is inconsistent and therefore confusing.
1429 PH/01 Two changes to the default runtime configuration:
1431 (1) Move the checks for relay_from_hosts and authenticated clients from
1432 after to before the (commented out) DNS black list checks.
1434 (2) Add control=submission to the relay_from_hosts and authenticated
1435 clients checks, on the grounds that messages accepted by these
1436 statements are most likely to be submissions.
1438 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1440 (1) Generate an error if the third argument for the ${prvs expansion is
1443 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1446 (3) Reset the variables that are obtained from the first argument of
1447 ${prvscheck and used in the second argument before leaving the code,
1448 because their memory is reclaimed, so using them afterwards may do
1451 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1452 one (it's much easier than Tom thought :-).
1454 (5) Because of (4), we can now allow for the use of $prvscheck_result
1455 inside the third argument.
1457 PH/03 For some reason, the default setting of PATH when running a command from
1458 a pipe transport was just "/usr/bin". I have changed it to
1461 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1462 anything to be listed in the output from -bV.
1464 PH/05 When a filter generated an autoreply, the entire To: header line was
1465 quoted in the delivery log line, like this:
1467 => >A.N.Other <ano@some.domain> <original@ddress> ...
1469 This has been changed so that it extracts the operative address. There
1470 may be more than one such address. If so, they are comma-separated, like
1473 => >ano@some.domain,ona@other.domain <original@ddress> ...
1475 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1476 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1477 not being looked up in the rDNS to get a host name, Exim was showing the
1478 IP address twice in Received: lines, even though the IP addresses were
1479 identical. For example:
1481 Received: from [1.2.3.4] (helo=[1.2.3.4])
1483 However, if the real host name was known, it was omitting the HELO data
1484 if it matched the actual IP address. This has been tidied up so that it
1485 doesn't show the same IP address twice.
1487 PH/07 When both +timestamp and +memory debugging was on, the value given by
1488 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1489 called by the debug printing, thereby overwriting the timestamp buffer.
1490 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1492 PH/08 When the original message was included in an autoreply transport, it
1493 always said "this is a copy of the message, including all the headers",
1494 even if body_only or headers_only was set. It now gives an appropriate
1497 PH/09 Applied a patch from the Sieve maintainer which:
1499 o fixes some comments
1500 o adds the (disabled) notify extension core
1501 o adds some debug output for the result of if/elsif tests
1502 o points to the current vacation draft in the documentation
1503 and documents the missing references header update
1507 o fixes a bug in processing the envelope test (when testing
1508 multiple envelope elements, the last element determinted the
1511 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1512 Electronic Mail") by including:
1514 Auto-submitted: auto-generated
1516 in the messages that it generates (bounce messages and others, such as
1517 warnings). In the case of bounce messages for non-SMTP mesages, there was
1518 also a typo: it was using "Auto_submitted" (underscore instead of
1519 hyphen). Since every message generated by Exim is necessarily in response
1520 to another message, thes have all been changed to:
1522 Auto-Submitted: auto-replied
1524 in accordance with these statements in the RFC:
1526 The auto-replied keyword:
1528 - SHOULD be used on messages sent in direct response to another
1529 message by an automatic process,
1531 - MUST NOT be used on manually-generated messages,
1533 - MAY be used on Delivery Status Notifications (DSNs) and Message
1534 Disposition Notifications (MDNs),
1536 - MUST NOT be used on messages generated by automatic or periodic
1537 processes, except for messages which are automatic responses to
1540 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1541 to the default Received: header definition.
1543 PH/12 Added log selector acl_warn_skipped (default on).
1545 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1546 variables because (a) they are in the wrong storage pool and (b) even if
1547 they were copied, it wouldn't work properly because of the caching.
1549 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1550 checking when decoding. Apparently there are clients that generate
1551 overlong encoded strings. Why am I not surprised?
1553 PH/15 If the first argument of "${if match_address" was not empty, but did not
1554 contain an "@" character, Exim crashed. Now it writes a panic log message
1555 and treats the condition as false.
1557 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1559 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1560 extension "envelope-auth". The code is finished and in agreement with
1561 other implementations, but there is no documentation so far and in fact,
1562 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1563 not changing the active code.
1565 Print executed "if" and "elsif" statements when debugging is used. This
1566 helps a great deal to understand what a filter does.
1568 Document more things not specified clearly in RFC3028. I had all this
1569 sorted out, when out of a sudden new issues came to my mind. Oops."
1571 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1574 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1575 canonical form (as documented). However, after a host name lookup from
1576 the IP address, check_host() was doing a simple string comparison with
1577 addresses acquired from the DNS when checking that the found name did
1578 have the original IP as one of its addresses. Since any found IPv6
1579 addresses are likely to be in abbreviated form, the comparison could
1580 fail. Luckily, there already exists a function for doing the comparison
1581 by converting both addresses to binary, so now that is used instead of
1582 the text comparison.
1584 PH/20 There was another similar case to PH/19, when a complete host name was
1585 given in a host list; looking up its IP address could give an abbreviated
1586 form, whereas the current host's name might or might not be abbreviated.
1587 The same fix has been applied.
1593 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1594 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1597 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1598 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1600 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1602 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1603 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1604 Also, for "Auto-Submitted", treat anything other than "no" as
1605 non-personal, in accordance with RFC 3834. (Previously it treated
1606 anything starting "auto-" as non-personal.)
1608 TF/01 The control=submission/name=... option had a problem with syntax
1609 errors if the name included a slash character. The /name= option
1610 now slurps the rest of the string, so it can include any characters
1611 but it must come last in the list of options (after /sender_retain
1614 PH/05 Some modifications to the interface to the fake nameserver for the new
1622 TK/01 Added the "success_on_redirect" address verification option. See
1623 NewStuff for rationale and an example.
1625 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1627 PH/02 Patch to exigrep to allow it to work on syslog lines.
1629 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1630 fread() to skip over the body file's header line, because in Cygwin the
1631 header line is locked and is inaccessible.
1633 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1634 co-exist for some time) to make it clear that it is the Exim ID that is
1635 referenced, not the Message-ID: header line.
1637 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1638 string_format() function, because snprintf() does not exist on all
1641 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1642 ${stat: expansion item.
1644 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1645 protocol synchronization error", to keep the pedants happy.
1647 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1648 well as for IRIX systems, when gcc is being used. See the host.c source
1651 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1653 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1656 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1657 transport and to the smtp transport in LMTP mode.
1659 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1661 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1662 run a filter in a subprocess. This could lead to confusion in subsequent
1663 lookups in the parent process. There should also be a search_tidyup() at
1664 the end of the subprocess.
1666 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1667 only if the host matched helo_try_verify_hosts, which caused the
1668 verification to occur when the EHLO/HELO command was issued. The ACL just
1669 tested the remembered result. Now, if a previous verification attempt has
1670 not happened, "verify = helo" does it there and then.
1672 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1674 TK/03 Fix log output including CR from clamd.
1676 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1677 exist provoked a memory error which could cause a segfault.
1679 PH/15 Installed PCRE 6.2
1681 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1683 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1684 of the problem. Specifically, suggested +O2 rather than +O1 for the
1687 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1689 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1690 record, the port was not correctly logged when the outgoing_port log
1691 selector was set (it logged the transort's default port).
1693 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1694 fallback_hosts, and "hosts" in the smtp transport.
1696 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1697 host errors such as "Connection refused".
1699 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1700 authentication with radiusclient 0.4.9:
1702 - Error returned from rc_read_config was caught wrongly
1703 - Username/password not passed on to radius server due to wrong length.
1705 The presumption is that some radiusclient API changes for 4.51/PH/17
1706 were not taken care of correctly. The code is still untested by me (my
1707 Linux distribution still has 0.3.2 of radiusclient), but it was
1708 contributed by a Radius user.
1710 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1711 expanding the "port" option of the smtp transport.
1713 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1714 while reading a MIME header. Thanks to Tom Hughes for a patch.
1716 PH/24 Include config.h inside local_scan.h so that configuration settings are
1719 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1720 that in an ACL for RCPT (for example), you can examine exactly what was
1723 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1724 commands, but it was not correctly comparing the address with the actual
1725 client host address. Thus, it would show the EHLO address in Received:
1726 header lines when this was not necessary.
1728 PH/27 Added the % operator to ${eval:}.
1730 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1731 it should be ignoring failures (because with -C, for example, it has lost
1732 privilege). It wasn't ignoring creation failures other than "already
1735 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1738 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1739 because an input line was too long, either on its own, or by virtue of
1740 too many continuations, the temporary file was not being removed, and the
1741 return code was incorrect.
1743 PH/31 Missing "BOOL" in function definition in filtertest.c.
1745 PH/32 Applied Sieve patches from the maintainer.
1747 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1749 PH/33 Added "verify = not_blind".
1751 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1752 Local/Makefile (with some defaults set). These are used in built scripts
1753 such as exicyclog, but they have never been used in the exim_install
1754 script (though there are many overriding facilities there). I have
1755 arranged that the exim_install script now takes note of these two
1758 PH/35 Installed configuration files for Dragonfly.
1760 PH/36 When a locally submitted message by a trusted user did not contain a
1761 From: header, and the sender address was obtained from -f or from an SMTP
1762 MAIL command, and the trusted user did not use -F to supply a sender
1763 name, $originator_name was incorrectly used when constructing a From:
1764 header. Furthermore, $originator_name was used for submission mode
1765 messages from external hosts without From: headers in a similar way,
1766 which is clearly wrong.
1768 PH/37 Added control=suppress_local_fixups.
1770 PH/38 When log_selector = +received_sender was set, and the addition of the
1771 sender made the log line's construction buffer exactly full, or one byte
1772 less than full, an overflow happened when the terminating "\n" was
1775 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1776 when the result of a list match is failure because a DNS lookup failed.
1778 PH/40 RM_COMMAND is now used in the building process.
1780 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1781 the "build-* directories that it finds.
1783 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1784 address in a domain literal was a prefix of an interface address.
1786 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1787 when verifying a sender address, unless rewrite_headers is false.
1789 PH/44 Wrote a long comment about why errors_to addresses are verified as
1790 recipients, not senders.
1792 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1793 the ratelimit ACL was added.
1795 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1797 PH/46 Added extra information about PostgreSQL errors to the error string.
1799 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1800 avoiding the need to install special zones in a real server. This is
1801 backwards compatible; if it can't find the fake resolver, it drops back.
1802 Thus, both old and new test suites can be run.
1804 TF/02 Added util/ratelimit.pl
1806 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1807 clock is set back in time.
1809 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1810 Candler <B.Candler@pobox.com>.
1812 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1813 OK for addresses that are the result of redirections.
1815 PH/48 A number of further additions for the benefit of the new test suite,
1816 including a fake gethostbyname() that interfaces to the fake DNS resolver
1819 TF/06 The fix for widen_domains has also been applied to qualify_single and
1820 search_parents which are the other dnslookup options that can cause
1823 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1826 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1828 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1829 service" if it encountered an MX record with an empty target hostname.
1830 The message is now "an MX or SRV record indicated no SMTP service".
1832 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1833 if the DNS of the sending site is misconfigured. This is quite a
1834 common situation. This change restores the behaviour of treating a
1835 helo verification defer as a failure.
1837 PH/51 If self=fail was set on a router, the bounce message did not include the
1838 actual error message.
1844 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1846 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1847 command itself ended in error, the underlying message about the transport
1848 filter timeout was being overwritten with the pipe command error. Now the
1849 underlying error message should be appended to the second error message.
1851 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1853 PH/02 Reduce the amount of output that "make" produces by default. Full output
1854 can still be requested.
1856 PH/03 The warning log line about a condition test deferring for a "warn" verb
1857 was being output only once per connection, rather than after each
1858 occurrence (because it was using the same function as for successful
1859 "warn" verbs). This seems wrong, so I have changed it.
1861 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1862 it should not have, which might have caused a crash in the right
1863 circumstances, but probably never did.
1865 PH/04 Installed a modified version of Tony Finch's patch to make submission
1866 mode fix the return path as well as the Sender: header line, and to
1867 add a /name= option so that you can make the user's friendly name appear
1870 TF/03 Added the control = fakedefer ACL modifier.
1872 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1873 Mark Lowes for thorough testing.
1875 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1877 TK/03 Merged latest SRS patch from Miles Wilton.
1879 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1880 with the definition in sysexits.h (which is #included earlier).
1881 Fortunately, Exim does not actually use EX_OK. The code used to try to
1882 preserve the sysexits.h value, by assumimg that macro definitions were
1883 scanned for macro replacements. I have been disabused of this notion,
1884 so now the code just undefines EX_OK before #including unistd.h.
1886 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1887 in the smtp transport. When a block could not be written in a single
1888 write() function, the timeout was being re-applied to each part-write.
1889 This seems wrong - if the receiver was accepting one byte at a time it
1890 would take for ever. The timeout is now adjusted when this happens. It
1891 doesn't have to be particularly precise.
1893 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1894 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1896 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1897 without a domain if the check to <postmaster@domain> fails.
1899 SC/01 Eximstats: added -xls and the ability to specify output files
1900 (patch written by Frank Heydlauf).
1902 SC/02 Eximstats: use FileHandles for outputing results.
1904 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1906 SC/04 Eximstats: fixed display of large numbers with -nvr option
1908 SC/05 Eximstats: fixed merging of reports with empty tables.
1910 SC/06 Eximstats: added the -include_original_destination flag
1912 SC/07 Eximstats: removed tabs and trailing whitespace.
1914 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1916 TK/06 MBOX spool code: Add real "From " MBOX separator line
1917 so the .eml file is really in mbox format (even though
1918 most programs do not really care). Patch from Alex Miller.
1920 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1921 The latter is generated from $received_to and is only set if the
1922 message has one envelope recipient. SA can use these headers,
1923 obviously out-of-the-box. Patch from Alex Miller.
1925 PH/08 The ${def test on a variable was returning false if the variable's
1926 value was "0", contrary to what the specification has always said!
1927 The result should be true unless the variable is empty.
1929 PH/09 The syntax error of a character other than { following "${if
1930 def:variable_name" (after optional whitespace) was not being diagnosed.
1931 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1932 accidental colon was present, for example, could give incorrect results.
1934 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1935 result is used (not including appendfile, where other changes are about
1938 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1939 This involved changing a lot of size variables from int to off_t. It
1940 should work with maildirs and everything.
1942 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1943 spamd dying while we are connected to it.
1945 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1948 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1949 function for BDB 4.3.
1951 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1953 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1954 the sources for PCRE have been re-arranged and more files are now
1957 PH/15 The code I had for printing potentially long long variables in PH/11
1958 above was not the best (it lost precision). The length of off_t variables
1959 is now inspected at build time, and an appropriate printing format (%ld
1960 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1961 to be "long long int" or "long int". This is needed for the internal
1962 formatting function string_vformat().
1964 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1965 the configuration file to be ":syslog", then the script "guesses" where
1966 the logs files are, rather than using the compiled in default. In our
1967 case the guess is not the same as the compiled default, so the script
1968 suddenly stopped working when I started to use syslog. The patch checks
1969 to see if log_file_path is "". If so, it attempts to read it from exim
1970 with no configuration file to get the compiled in version, before it
1971 falls back to the previous guessing code."
1973 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1974 implementing BATV in an Exim configuration. See NewStuff for the gory
1977 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1978 Makefile that are specific to HP-UX.
1980 PH/18 If the "use_postmaster" option was set for a recipient callout together
1981 with the "random" option, the postmaster address was used as the MAIL
1982 FROM address for the random test, but not for the subsequent recipient
1983 test. It is now used for both.
1985 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1986 patch removes a few documentation additions to RFC 3028, because the
1987 latest draft now contains them. It adds the new en;ascii-case comparator
1988 and a new error check for 8bit text in MIME parts. Comparator and
1989 require names are now matched exactly. I enabled the subaddress
1990 extension, but it is not well tested yet (read: it works for me)."
1992 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1993 rework some of the code of TK/09 above to avoid the hardwired use of
1994 "%lld" and "long long". Replaced the call to snprintf() with a call to
1997 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1998 records point to non-existent hosts", "retry timeout exceeded", and
1999 "retry time not reached for any host after a long failure period".
2001 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2002 experimental DomainKeys support:
2004 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2005 (2) On an error such as an illegally used "control", the wrong name for
2006 the control was given.
2008 These problems did NOT occur unless DomainKeys support was compiled.
2010 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2012 PH/24 Added ${if match_ip condition.
2014 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2015 databases so that it will be absolutely obvious if a crash occurs in the
2016 DB library. This is a regular occurrence (often caused by mis-matched
2019 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2020 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2021 picked up on a user's system that detects such things. There doesn't seem
2022 to be a gcc warning option for this - only an attribute that has to be
2023 put on the function's prototype. It seems that in Fedora Core 4 they have
2024 set this on a number of new functions. No doubt there will be more in due
2027 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2028 specify a transport. However, if an address that was verified by such a
2029 router was the subject of a callout, Exim crashed because it tried to
2030 read the rcpt_include_affixes from the non-existent transport. Now it
2031 just assumes that the setting of that option is false. This bug was
2032 introduced by 4.51/PH/31.
2034 PH/28 Changed -d+all to exclude +memory, because that information is very
2035 rarely of interest, but it makes the output a lot bigger. People tend to
2036 do -d+all out of habit.
2038 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2039 code in os-type was giving problems when libc.so lives in lib64, like on
2042 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2043 aren't the modern standard, and it seems that some systems' include files
2044 don't always have them. Exim was already checking for some of the newer
2045 ones like T_AAAA, and defining it itself. I've added checks for all the
2046 record types that Exim uses.
2048 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2049 not automatically generating a new one, as it is supposed to. This
2050 prevented TLS from working. If the file did exist, but contained invalid
2051 data, a new version was generated, as expected. It was only the case of a
2052 non-existent file that was broken.
2054 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2055 with a change in libdomainkeys > 0.64.
2057 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2058 from DNS. If the selector record carries the flag, it now has
2059 precedence over the domain-wide flag.
2061 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2063 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2064 the use of a port name that isn't defined in /etc/services) occurred, the
2065 message was deferred as in a normal delivery, and thus remained on the
2066 spool, instead of being failed because of the mua_wrapper setting. This
2067 is now fixed, and I tidied up some of the mua_wrapper messages at the
2070 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2071 the messages in a hash of arrays rather than using individual hashes.
2072 This is a bit cleaner and results in dramatic memory savings, albeit
2073 at a slight CPU cost.
2075 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2076 as requested by Marc Sherman.
2078 SC/10 Eximstats: added histograms for user specified patterns as requested
2081 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2083 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2084 fopen() in the content-scanning modules that did not already have it.
2090 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2091 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2093 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2095 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2097 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2098 to test Sieve filters that use "vacation".
2100 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2101 that changes the way the GnuTLS parameters are stored in the cache file.
2102 The new format can be generated externally. For backward compatibility,
2103 if the data in the cache doesn't make sense, Exim assumes it has read an
2104 old-format file, and it generates new data and writes a new file. This
2105 means that you can't go back to an older release without removing the
2108 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2109 work if there are any delivery delays because "one_time" forces the
2110 parent to be marked "delivered", so its unseen clone is never tried
2111 again. For this reason, Exim now forbids the simultaneous setting of
2114 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2115 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2116 entry if you want to know the details. The fix, however, neglected to
2117 consider the case when local delivery batching is involved. The test for
2118 "previously delivered" was not happening when checking to see if an
2119 address could be batched with a previous (undelivered) one; under
2120 certain circumstances this could lead to multiple deliveries to the same
2123 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2124 in its include files, and this causes problems building Exim.
2126 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2127 header_syntax) but Exim was just ignoring anything given after a slash.
2128 In particular, this caused confusion with an attempt to use "verify =
2129 reverse_host_lookup/defer_ok". An error is now given when options are
2130 supplied for verify items that do not have them. (Maybe reverse_host_
2131 lookup should have a defer_ok option, but that's a different point.)
2133 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2134 defined by RFC 821) to 2048, because there were problems with some AUTH
2135 commands, and RFC 1869 says the size should be increased for extended
2136 SMTP commands that take arguments.
2138 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2141 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2142 "unknown" error; now it says that the functionality isn't in the binary.
2144 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2145 an address' error message when a string expansion fails (syntax or
2146 whatever). Otherwise the password may appear in the log. Following change
2147 PH/42 below, there is no longer a chance of it appearing in a bounce
2150 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2152 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2153 of its addresses was ever tried. (Bugzilla bug #2.)
2155 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2156 the result incorrectly in the debug output. (It correctly added a newline
2157 to what was transported.)
2159 TF/01 Added $received_time.
2161 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2162 commented out examples of how to interface to a virus scanner and to
2163 SpamAssassin. Also added commented examples of av_scanner and
2164 spamd_address settings.
2166 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2167 and controls are allowed in which ACLs. There were a couple of minor
2168 errors. Some of the entries in the conditions table (which is a table of
2169 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2170 negation of where the condition IS allowed.
2172 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2174 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2175 header file does not have a version number, so I've had to invent a new
2176 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2177 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2178 radiusclient), but it was contributed by a Radius user.
2180 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2181 files or directories, for OpenSSL.
2183 PH/19 When an Exim process that is running as root has to create an Exim log
2184 file, it does so in a subprocess that runs as exim:exim so as to get the
2185 ownership right at creation (otherwise, other Exim processes might see
2186 the file with the wrong ownership). There was no test for failure of this
2187 fork() call, which would lead to the process getting stuck as it waited
2188 for a non-existent subprocess. Forks do occasionally fail when resources
2189 run out. I reviewed all the other calls to fork(); they all seem to check
2192 PH/20 When checking for unexpected SMTP input at connect time (before writing
2193 the banner), Exim was not dealing correctly with a non-positive return
2194 from the read() function. If the client had disconnected by this time,
2195 the result was a log entry for a synchronization error with an empty
2196 string after "input=" when read() returned zero. If read() returned -1
2197 (an event I could not check), uninitialized data bytes were printed.
2198 There were reports of junk text (parts of files, etc) appearing after
2201 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2203 PH/22 Added support for macro redefinition, and (re)definition in between
2204 driver and ACL definitions.
2206 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2207 forgetting to use the resulting value; it was using the unexpanded value.
2209 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2210 hadn't been configured. The fix is from Juergen Kreileder, who
2211 understands it better than I do:
2213 "Here's what I see happening with three configured cyrus_sasl
2214 authenticators configured (plain, login, cram-md5):
2216 On startup auth_cyrus_sasl_init() gets called for each of these.
2217 This means three calls to sasl_listmech() without a specified mech_list.
2218 => SASL tests which mechs of all available mechs actually work
2219 => three warnings about OTP not working
2220 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2222 With the patch, sasl_listmech() also gets called three times. But now
2223 SASL's mech_list option is set to the server_mech specified in the the
2224 authenticator. Or in other words, the answer from sasl_listmech()
2225 gets limited to just the mech you're testing for (which is different
2227 => the return list contains just 'plain' or 'login', 'cram-md5' or
2228 nothing depending on the value of ob->server_mech.
2230 I've just tested the patch: Authentication still works fine,
2231 unavailable mechs specified in the exim configuration are still
2232 caught, and the auth.log warnings about OTP are gone."
2234 PH/25 When debugging is enabled, the contents of the command line are added
2235 to the debugging output, even when log_selector=+arguments is not
2238 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2239 answer is "GNU", and only if the return is "GNU/something" is the answer
2242 PH/27 $acl_verify_message is now set immediately after the failure of a
2243 verification in an ACL, and so is available in subsequent modifiers. In
2244 particular, the message can be preserved by coding like this:
2246 warn !verify = sender
2247 set acl_m0 = $acl_verify_message
2249 Previously, $acl_verify_message was set only while expanding "message"
2250 and "log_message" when a very denied access.
2252 PH/28 Modified OS/os.c-Linux with
2254 -#ifndef OS_LOAD_AVERAGE
2255 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2257 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2258 nomenclature these days.)
2260 PH/29 Installed patch from the Sieve maintainer that adds the options
2261 sieve_useraddress and sieve_subaddress to the redirect router.
2263 PH/30 In these circumstances:
2264 . Two addresses routed to the same list of hosts;
2265 . First host does not offer TLS;
2266 . First host accepts first address;
2267 . First host gives temporary error to second address;
2268 . Second host offers TLS and a TLS session is established;
2269 . Second host accepts second address.
2270 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2271 and peerdn, if requested) that were in fact used only for the second
2274 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2275 attention to any local part prefix or suffix that was matched by the
2276 router that accepted the address. It now behaves in the same way as it
2277 does for delivery: the affixes are removed from the local part unless
2278 rcpt_include_affixes is set on the transport.
2280 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2281 timeout during the DATA phase of an incoming message.
2283 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2284 applied a patch sanctioned by the Sieve maintainer.
2286 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2287 the uid or gid is negative. A case of a negative gid caused this to be
2288 noticed. The fix allows for either to be negative.
2290 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2291 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2294 PH/36 The change PH/12 above was broken. Fixed it.
2296 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2297 the grounds that routing the same address twice would always produce the
2298 same answer. This might have been true once, but it is certainly no
2299 longer true now. Routing a child address may depend on the previous
2300 routing that produced that child. Some complicated redirection strategies
2301 went wrong when messages had multiple recipients, and made Exim's
2302 behaviour dependent on the order in which the addresses were given.
2304 I have moved the duplicate checking until after the routing is complete.
2305 Exim scans the addresses that are assigned to local and remote
2306 transports, and removes any duplicates. This means that more work will be
2307 done, as duplicates will always all be routed, but duplicates are
2308 presumably rare, so I don't expect this is of any significance.
2310 For deliveries to pipes, files, and autoreplies, the duplicate checking
2311 still happens during the routing process, since they are not going to be
2314 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2315 It corrects a timeout issue with spamd. This is Ian's comment: "The
2316 background is that sometimes spamd either never reads data from a
2317 connection it has accepted, or it never writes response data. The exiscan
2318 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2319 blindly assumes that writes won't block so it may never time out."
2321 PH/39 Allow G after quota size as well as K and M.
2323 PH/40 The value set for $authenticated_id in an authenticator may not contain
2324 binary zeroes or newlines because the value is written to log lines and
2325 to spool files. There was no check on this. Now the value is run through
2326 the string_printing() function so that such characters are converted to
2327 printable escape sequences.
2329 PH/41 $message_linecount is a new variable that contains the total number of
2330 lines in the message. Compare $body_linecount, which is the count for the
2333 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2334 bounce and delay warning messages, except in certain special cases, which
2337 (a) An SMTP error message from a remote host;
2338 (b) A message specified in a :fail: redirection;
2339 (c) A message specified in a "fail" command in a system filter;
2340 (d) A message specified in a FAIL return from the queryprogram router;
2341 (e) A message specified by the cannot_route_message router option.
2343 In these cases only, Exim does include the error details in bounce and
2344 warning messages. There are also a few cases where bland messages such
2345 as "unrouteable address" or "local delivery error" are given.
2347 PH/43 $value is now also set for the "else" part of a ${run expansion.
2349 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2350 being worked on, but at least Exim now implements the latest version to
2353 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2354 process to complete was treated as a delivery failure, a timeout while
2355 writing the message to the pipe was logged, but erroneously treated as a
2356 successful delivery. Such timeouts include transport filter timeouts. For
2357 consistency with the overall process timeout, these timeouts are now
2358 treated as errors, giving rise to delivery failures by default. However,
2359 there is now a new Boolean option for the pipe transport called
2360 timeout_defer, which, if set TRUE, converts the failures into defers for
2361 both kinds of timeout. A transport filter timeout is now identified in
2364 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2365 systems where "make" and "gmake" are different, calling "gmake" at top
2366 level broke things. I've arranged for the value of $(MAKE) to be passed
2367 from the Makefile to this script so that it can call the same version of
2371 A note about Exim versions 4.44 and 4.50
2372 ----------------------------------------
2374 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2375 changes of various kinds. As a consequence, a big documentation update was
2376 needed. This delayed the release for rather longer than seemed good, especially
2377 in the light of a couple of (minor) security issues. Therefore, the changes
2378 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2379 release. So 4.44 and 4.50 are in effect two different branches that both start
2382 I have left the 4.50 change log unchanged; it contains all the changes since
2383 4.43. The change log for 4.44 is below; many of its items are identical to
2384 those for 4.50. This seems to be the most sensible way to preserve the
2385 historical information.
2391 1. Minor wording change to the doc/README.SIEVE file.
2393 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2394 computation of the current number of files was incorrect.
2396 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2397 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2400 4. Give more explanation in the error message when the command for a transport
2401 filter fails to execute.
2403 5. There are several places where Exim runs a non-Exim command in a
2404 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2405 was being done only for the command run by the queryprogram router. It is
2406 now done for all such subprocesses. The other cases are: ${run, transport
2407 filters, and the commands run by the lmtp and pipe transports.
2409 6. Added CONFIGURE_GROUP build-time option.
2411 7. Some older OS have a limit of 256 on the maximum number of file
2412 descriptors. Exim was using setrlimit() to set 1000 as a large value
2413 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2414 systems. I've change it so that if it can't get 1000, it tries for 256.
2416 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2417 was an oversight, and furthermore, ever since the addition of extra
2418 controls (e.g. 4.43/32), the checks on when to allow different forms of
2419 "control" were broken. There should now be diagnostics for all cases when a
2420 control that does not make sense is encountered.
2422 9. Added the /retain_sender option to "control=submission".
2424 10. $recipients is now available in the predata ACL (oversight).
2426 11. Tidy the search cache before the fork to do a delivery from a message
2427 received from the command line. Otherwise the child will trigger a lookup
2428 failure and thereby defer the delivery if it tries to use (for example) a
2429 cached ldap connection that the parent has called unbind on.
2431 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2432 of $address_data from the recipient verification was clobbered by the
2433 sender verification.
2435 13. The value of address_data from a sender verification is now available in
2436 $sender_address_data in subsequent conditions in the ACL statement.
2438 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2440 15. Added a new option "connect=<time>" to callout options, to set a different
2443 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2444 was its contents. (It was OK if the option was not defined at all.)
2446 17. A "Completed" log line is now written for messages that are removed from
2447 the spool by the -Mrm option.
2449 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2450 information about exactly what failed.
2452 19. Added -dd to debug only the daemon process.
2454 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2455 handles timeouts, both on the server side and network timeouts. Renamed the
2456 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2458 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2459 It is now set to "smtps".
2461 22. $host_address is now set to the target address during the checking of
2462 ignore_target_hosts.
2464 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2465 being passed; this would have caused $sender_host_name to have been used if
2466 matching the list had actually called for a host name (not very likely,
2467 since this list is usually IP addresses). A host name is now passed as
2470 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2471 code that specifies a non-restarting handler (typically sigaction() in
2472 modern systems) in an attempt to fix a rare and obscure crash bug.
2474 25. Narrowed the window for a race in the daemon that could cause it to ignore
2475 SIGCHLD signals. This is not a major problem, because they are used only to
2476 wake it up if nothing else does.
2478 26. A malformed maildirsize file could cause Exim to calculate negative values
2479 for the mailbox size or file count. Odd effects could occur as a result.
2480 The maildirsize information is now recalculated if the size or filecount
2483 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2484 support for a long time. Removed HAVE_SYS_VFS_H.
2486 28. Installed the latest version of exipick from John Jetmore.
2488 29. In an address list, if the pattern was not a regular expression, an empty
2489 subject address (from a bounce message) matched only if the pattern was an
2490 empty string. Non-empty patterns were not even tested. This was the wrong
2491 because it is perfectly reasonable to use an empty address as part of a
2492 database query. An empty address is now tested by patterns that are
2493 lookups. However, all the other forms of pattern expect the subject to
2494 contain a local part and a domain, and therefore, for them, an empty
2495 address still always fails if the pattern is not itself empty.
2497 30. Exim went into a mad DNS loop when attempting to do a callout where the
2498 host was specified on an smtp transport, and looking it up yielded more
2499 than one IP address.
2501 31. Re-factored the code for checking spool and log partition space into a
2502 function that finds that data and another that does the check. The former
2503 is then used to implement four new variables: $spool_space, $log_space,
2504 $spool_inodes, and $log_inodes.
2506 32. The RFC2047 encoding function was originally intended for short strings
2507 such as real names; it was not keeping to the 75-character limit for
2508 encoded words that the RFC imposes. It now respects the limit, and
2509 generates multiple encoded words if necessary. To be on the safe side, I
2510 have increased the buffer size for the ${rfc2047: expansion operator from
2513 33. It is now permitted to omit both strings after an "if" condition; if the
2514 condition is true, the result is "true". As before, when the second string
2515 is omitted, a false condition yields an empty string. This makes it less
2516 cumbersome to write custom ACL and router conditions.
2518 34. Failure to deliver a bounce message always caused it to be frozen, even if
2519 there was an errors_to setting on the router. The errors_to setting is now
2522 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2523 canonical form (fully expanded) before being placed in
2524 $sender_host_address.
2526 36. The table in the code that translates DNS record types into text (T_A to
2527 "A" for instance) was missing entries for NS and CNAME. It is just possible
2528 that this could have caused confusion if both these types were looked up
2529 for the same domain, because the text type is used as part of Exim's
2530 per-process caching. But the chance of anyone hitting this buglet seems
2533 37. The dnsdb lookup has been extended in a number of ways.
2535 (1) There is a new type, "zns", which walks up the domain tree until it
2536 finds some nameserver records. It should be used with care.
2538 (2) There is a new type, "mxh", which is like "mx" except that it returns
2539 just the host names, not the priorities.
2541 (3) It is now possible to give a list of domains (or IP addresses) to be
2542 looked up. The behaviour when one of the lookups defers can be
2543 controlled by a keyword.
2545 (4) It is now possible to specify the separator character for use when
2546 multiple records are returned.
2548 38. The dnslists ACL condition has been extended: it is now possible to supply
2549 a list of IP addresses and/or domains to be looked up in a particular DNS
2552 39. Added log_selector=+queue_time_overall.
2554 40. When running the queue in the test harness, wait just a tad after forking a
2555 delivery process, to get repeatability of debugging output.
2557 41. Include certificate and key file names in error message when GnuTLS fails
2558 to set them up, because the GnuTLS error message doesn't include the name
2559 of the failing file when there is a problem reading it.
2561 42. Allow both -bf and -bF in the same test run.
2563 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2565 44. The "Exiscan patch" is now merged into the mainline Exim source.
2567 45. Sometimes the final signoff response after QUIT could fail to get
2568 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2569 < 0 before doing a fflush(). This bug looks as though it goes back to the
2570 introduction of TLS in release 3.20, but "sometimes" must have been rare
2571 because the tests only now provoked it.
2573 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2574 (this can affect the format of dates).
2576 47. exim_tidydb, when checking for the continued existence of a message for
2577 which it has found a message-specific retry record, was not finding
2578 messages that were in split spool directories. Consequently, it was
2579 deleting retry records that should have stayed in existence.
2581 48. Steve fixed some bugs in eximstats.
2583 49. The SPA authentication driver was not abandoning authentication and moving
2584 on to the next authenticator when an expansion was forced to fail,
2585 contradicting the general specification for all authenticators. Instead it
2586 was generating a temporary error. It now behaves as specified.
2588 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2589 (the order specifies the preference for clients). The order is now AES256,
2590 AES128, 3DES, ARCFOUR128.
2592 51. Small patch to Sieve code - explicitly set From: when generating an
2595 52. Exim crashed if a remote delivery caused a very long error message to be
2596 recorded - for instance if somebody sent an entire SpamAssassin report back
2597 as a large number of 550 error lines. This bug was coincidentally fixed by
2598 increasing the size of one of Exim's internal buffers (big_buffer) that
2599 happened as part of the Exiscan merge. However, to be on the safe side, I
2600 have made the code more robust (and fixed the comments that describe what
2603 53. Now that there can be additional text after "Completed" in log lines (if
2604 the queue_time_overall log selector is set), a one-byte patch to exigrep
2605 was needed to allow it to recognize "Completed" as not the last thing in
2608 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2609 patch that reportedly fixes this has been added. I am not expert enough to
2610 create a test for it. This is what the patch creator wrote:
2612 "I found a little strange behaviour of ldap code when working with
2613 Windows 2003 AD Domain, where users was placed in more than one
2614 Organization Units. When I tried to give exim partial DN, the exit code
2615 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2616 But simultaneously result of request was absolutely normal ldap result,
2617 so I produce this patch..."
2619 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2620 REFERENCE, so I have modified the code to exclude the patch when that macro
2623 55. Some experimental protocols are using DNS PTR records for new purposes. The
2624 keys for these records are domain names, not reversed IP addresses. The
2625 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2626 leaves it alone. Component reversal etc. now happens only for IP addresses.
2629 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2631 57. Double the size of the debug message buffer (to 2048) so that more of very
2632 long debug lines gets shown.
2634 58. The exicyclog utility now does better if the number of log files to keep
2635 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2637 59. Two changes related to the smtp_active_hostname option:
2639 (1) $smtp_active_hostname is now available as a variable.
2640 (2) The default for smtp_banner uses $smtp_active_hostname instead
2641 of $primary_hostname.
2643 60. The host_aton() function is supposed to be passed a string that is known
2644 to be a valid IP address. However, in the case of IPv6 addresses, it was
2645 not checking this. This is a hostage to fortune. Exim now panics and dies
2646 if the condition is not met. A case was found where this could be provoked
2647 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2648 components; fortuitously, this particular loophole had already been fixed
2649 by change 4.50/55 above.
2651 If there are any other similar loopholes, the new check in host_aton()
2652 itself should stop them being exploited. The report I received stated that
2653 data on the command line could provoke the exploit when Exim was running as
2654 exim, but did not say which command line option was involved. All I could
2655 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2656 running as the user.
2659 61. There was a buffer overflow vulnerability in the SPA authentication code
2660 (which came originally from the Samba project). I have added a test to the
2661 spa_base64_to_bits() function which I hope fixes it.
2664 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2665 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2667 63. The daemon start-up calls getloadavg() while still root for those OS that
2668 need the first call to be done as root, but it missed one case: when
2669 deliver_queue_load_max is set with deliver_drop_privilege. This is
2670 necessary for the benefit of the queue runner, because there is no re-exec
2671 when deliver_drop_privilege is set.
2673 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2674 This has been fixed.
2676 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2677 was in use, was not putting the data itself into the right store pool;
2678 consequently, it could be overwritten for a subsequent message in the same
2679 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2682 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2684 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2685 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2686 the function were treating the return as a boolean value, which happened to
2687 work because 0=false and not-0=true, but is not correct code.
2689 68. The host_aton() function was not handling scoped IPv6 addresses (those
2690 with, for example, "%eth0" on the end) correctly.
2692 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2693 negated items (that is, ~something) in unsigned ints. Some compilers
2694 apparently mutter when there is no cast.
2696 70. If an address verification called from an ACL failed, and did not produce a
2697 user-specific message (i.e. there was only a "system" message), nothing was
2698 put in $acl_verify_message. In this situation, it now puts the system
2701 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2702 session; change 4.31/43 added the unwanted input to the log line - except
2703 that it did not do this in the start of session case. It now does.
2705 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2706 This is wrong and can cause the other end to generate a synchronization
2707 error if it is another Exim or anything else that does the synchronization
2708 check. A QUIT command is no longer sent after a timeout.
2710 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2711 during host lookups.
2713 74. The defer_ok option of callout verification was not working if it was used
2714 when verifying addresses in header lines, that is, for this case:
2716 verify = header_sender/callout=defer_ok
2718 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2719 those file descriptors could be used for SMTP connections. If anything
2720 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2721 could be sent to the SMTP client, causing chaos. The daemon now opens
2722 stdin, stdout, and stderr to /dev/null when it puts itself into the
2725 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2726 log by default. The user can override this with suitable Perl magic.
2728 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2729 the log message when discard triggers, was not working for the DATA ACL or
2730 for the non-SMTP ACL.
2732 78. Error message wording change in sieve.c.
2734 79. If smtp_accept_max_per_host was set, the number of connections could be
2735 restricted to fewer than expected, because the daemon was trying to set up
2736 a new connection before checking whether the processes handling previous
2737 connections had finished. The check for completed processes is now done
2738 earlier. On busy systems, this bug wouldn't be noticed because something
2739 else would have woken the daemon, and it would have reaped the completed
2742 80. If a message was submitted locally by a user whose login name contained one
2743 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2744 It caused a spool format error. I have fixed the spool reading code. A
2745 related problem was that the "from" clause in the Received: line became
2746 illegal because of the space(s). It is now covered by ${quote_local_part.
2748 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2751 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2753 83. Patch from Sieve maintainer to fix unterminated string problem in
2754 "vacation" handling.
2756 84. Some minor changes to the Linux configuration files to help with other
2757 OS variants using glibc.
2759 85. One more patch for Sieve to update vacation handling to latest spec.
2762 ----------------------------------------------------
2763 See the note above about the 4.44 and 4.50 releases.
2764 ----------------------------------------------------
2770 1. Change 4.43/35 introduced a bug that caused file counts to be
2771 incorrectly computed when quota_filecount was set in an appendfile
2774 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2775 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2778 3. Give more explanation in the error message when the command for a transport
2779 filter fails to execute.
2781 4. There are several places where Exim runs a non-Exim command in a
2782 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2783 was being done only for the command run by the queryprogram router. It is
2784 now done for all such subprocesses. The other cases are: ${run, transport
2785 filters, and the commands run by the lmtp and pipe transports.
2787 5. Some older OS have a limit of 256 on the maximum number of file
2788 descriptors. Exim was using setrlimit() to set 1000 as a large value
2789 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2790 systems. I've change it so that if it can't get 1000, it tries for 256.
2792 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2793 was an oversight, and furthermore, ever since the addition of extra
2794 controls (e.g. 4.43/32), the checks on when to allow different forms of
2795 "control" were broken. There should now be diagnostics for all cases when a
2796 control that does not make sense is encountered.
2798 7. $recipients is now available in the predata ACL (oversight).
2800 8. Tidy the search cache before the fork to do a delivery from a message
2801 received from the command line. Otherwise the child will trigger a lookup
2802 failure and thereby defer the delivery if it tries to use (for example) a
2803 cached ldap connection that the parent has called unbind on.
2805 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2806 of $address_data from the recipient verification was clobbered by the
2807 sender verification.
2809 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2810 was its contents. (It was OK if the option was not defined at all.)
2812 11. A "Completed" log line is now written for messages that are removed from
2813 the spool by the -Mrm option.
2815 12. $host_address is now set to the target address during the checking of
2816 ignore_target_hosts.
2818 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2819 being passed; this would have caused $sender_host_name to have been used if
2820 matching the list had actually called for a host name (not very likely,
2821 since this list is usually IP addresses). A host name is now passed as
2824 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2825 code that specifies a non-restarting handler (typically sigaction() in
2826 modern systems) in an attempt to fix a rare and obscure crash bug.
2828 15. Narrowed the window for a race in the daemon that could cause it to ignore
2829 SIGCHLD signals. This is not a major problem, because they are used only to
2830 wake it up if nothing else does.
2832 16. A malformed maildirsize file could cause Exim to calculate negative values
2833 for the mailbox size or file count. Odd effects could occur as a result.
2834 The maildirsize information is now recalculated if the size or filecount
2837 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2838 support for a long time. Removed HAVE_SYS_VFS_H.
2840 18. Updated exipick to current release from John Jetmore.
2842 19. Allow an empty sender to be matched against a lookup in an address list.
2843 Previously the only cases considered were a regular expression, or an
2846 20. Exim went into a mad DNS lookup loop when doing a callout where the
2847 host was specified on the transport, if the DNS lookup yielded more than
2850 21. The RFC2047 encoding function was originally intended for short strings
2851 such as real names; it was not keeping to the 75-character limit for
2852 encoded words that the RFC imposes. It now respects the limit, and
2853 generates multiple encoded words if necessary. To be on the safe side, I
2854 have increased the buffer size for the ${rfc2047: expansion operator from
2857 22. Failure to deliver a bounce message always caused it to be frozen, even if
2858 there was an errors_to setting on the router. The errors_to setting is now
2861 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2862 canonical form (fully expanded) before being placed in
2863 $sender_host_address.
2865 24. Updated eximstats to version 1.33
2867 25. Include certificate and key file names in error message when GnuTLS fails
2868 to set them up, because the GnuTLS error message doesn't include the name
2869 of the failing file when there is a problem reading it.
2871 26. Expand error message when OpenSSL has problems setting up cert/key files.
2874 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2875 (this can affect the format of dates).
2877 28. exim_tidydb, when checking for the continued existence of a message for
2878 which it has found a message-specific retry record, was not finding
2879 messages that were in split spool directories. Consequently, it was
2880 deleting retry records that should have stayed in existence.
2882 29. eximstats updated to version 1.35
2883 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2884 1.35 - bugfix such that pie charts by volume are generated correctly
2886 30. The SPA authentication driver was not abandoning authentication and moving
2887 on to the next authenticator when an expansion was forced to fail,
2888 contradicting the general specification for all authenticators. Instead it
2889 was generating a temporary error. It now behaves as specified.
2891 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2892 (the order specifies the preference for clients). The order is now AES256,
2893 AES128, 3DES, ARCFOUR128.
2895 31. Small patch to Sieve code - explicitly set From: when generating an
2898 32. Exim crashed if a remote delivery caused a very long error message to be
2899 recorded - for instance if somebody sent an entire SpamAssassin report back
2900 as a large number of 550 error lines. This bug was coincidentally fixed by
2901 increasing the size of one of Exim's internal buffers (big_buffer) that
2902 happened as part of the Exiscan merge. However, to be on the safe side, I
2903 have made the code more robust (and fixed the comments that describe what
2906 33. Some experimental protocols are using DNS PTR records for new purposes. The
2907 keys for these records are domain names, not reversed IP addresses. The
2908 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2909 leaves it alone. Component reversal etc. now happens only for IP addresses.
2912 34. The host_aton() function is supposed to be passed a string that is known
2913 to be a valid IP address. However, in the case of IPv6 addresses, it was
2914 not checking this. This is a hostage to fortune. Exim now panics and dies
2915 if the condition is not met. A case was found where this could be provoked
2916 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2917 components; fortuitously, this particular loophole had already been fixed
2918 by change 4.50/55 or 4.44/33 above.
2920 If there are any other similar loopholes, the new check in host_aton()
2921 itself should stop them being exploited. The report I received stated that
2922 data on the command line could provoke the exploit when Exim was running as
2923 exim, but did not say which command line option was involved. All I could
2924 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2925 running as the user.
2928 35. There was a buffer overflow vulnerability in the SPA authentication code
2929 (which came originally from the Samba project). I have added a test to the
2930 spa_base64_to_bits() function which I hope fixes it.
2933 36. The daemon start-up calls getloadavg() while still root for those OS that
2934 need the first call to be done as root, but it missed one case: when
2935 deliver_queue_load_max is set with deliver_drop_privilege. This is
2936 necessary for the benefit of the queue runner, because there is no re-exec
2937 when deliver_drop_privilege is set.
2939 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2940 was in use, was not putting the data itself into the right store pool;
2941 consequently, it could be overwritten for a subsequent message in the same
2942 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2945 38. Sometimes the final signoff response after QUIT could fail to get
2946 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2947 < 0 before doing a fflush(). This bug looks as though it goes back to the
2948 introduction of TLS in release 3.20, but "sometimes" must have been rare
2949 because the tests only now provoked it.
2955 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2956 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2957 Now it returns an integer. A number of calls were still expecting a T/F
2958 return. Fortuitously, in all cases, the tests worked in OK situations,
2959 which is the norm. However, things would have gone wrong on any write
2960 failures on the smtp file descriptor. This function is used when sending
2961 messages over SMTP and also when doing verify callouts.
2963 2. When Exim is called to do synchronous delivery of a locally submitted
2964 message (the -odf or -odi options), it no longer closes stderr before doing
2967 3. Implemented the mua_wrapper option.
2969 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2971 5. Implemented the functions header_remove(), header_testname(),
2972 header_add_at_position(), and receive_remove_recipient(), and exported them
2975 6. If an ACL "warn" statement specified the addition of headers, Exim already
2976 inserted X-ACL-Warn: at the start if there was no header name. However, it
2977 was not making this test for the second and subsequent header lines if
2978 there were newlines in the string. This meant that an invalid header could
2979 be inserted if Exim was badly configured.
2981 7. Allow an ACL "warn" statement to add header lines at the start or after all
2982 the Received: headers, as well as at the end.
2984 8. Added the rcpt_4xx retry error code.
2986 9. Added postmaster_mailfrom=xxx to callout verification option.
2988 10. Added mailfrom=xxxx to the callout verification option, for verify=
2991 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2992 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2994 12. Inserted some casts to stop certain compilers warning when using pointer
2995 differences as field lengths or precisions in printf-type calls (mostly
2996 affecting debugging statements).
2998 13. Added optional readline() support for -be (dynamically loaded).
3000 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3001 same clock tick as a message's arrival, so that its received time was the
3002 same as the "first fail" time on the retry record, and that message
3003 remained on the queue past the ultimate address timeout, every queue runner
3004 would try a delivery (because it was past the ultimate address timeout) but
3005 after another failure, the ultimate address timeout, which should have then
3006 bounced the address, did not kick in. This was a "< instead of <=" error;
3007 in most cases the first failure would have been in the next clock tick
3008 after the received time, and all would be well.
3010 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3011 being recognized when the domain list was tested by the match_domain
3012 condition in an expansion string.
3014 16. Added the ${str2b64: operator.
3016 17. Exim was always calling setrlimit() to set a large limit for the number of
3017 processes, without checking whether the existing limit was already
3018 adequate. (It did check for the limit on file descriptors.) Furthermore,
3019 errors from getrlimit() and setrlimit() were being ignored. Now they are
3020 logged to the main and panic logs, but Exim does carry on, to try to do its
3021 job under whatever limits there are.
3023 18. Imported PCRE 5.0.
3025 19. Trivial typo in log message " temporarily refused connection" (the leading
3028 20. If the log selector return_path_on_delivery was set and an address was
3029 redirected to /dev/null, the delivery process crashed because it assumed
3030 that a return path would always be set for a "successful" delivery. In this
3031 case, the whole delivery is bypassed as an optimization, and therefore no
3034 21. Internal re-arrangement: the function for sending a challenge and reading
3035 a response while authentication was assuming a zero-terminated challenge
3036 string. It's now changed to take a pointer and a length, to allow for
3037 binary data in such strings.
3039 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3041 23. Exim was not respecting finduser_retries when seeking the login of the
3042 uid under which it was called; it was always trying 10 times. (The default
3043 setting of finduser_retries is zero.) Also, it was sleeping after the final
3044 failure, which is pointless.
3046 24. Implemented tls_on_connect_ports.
3048 25. Implemented acl_smtp_predata.
3050 26. If the domain in control=submission is set empty, Exim assumes that the
3051 authenticated id is a complete email address when it generates From: or
3052 Sender: header lines.
3054 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3055 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3056 chown and chgrp in /bin and hostname in /usr/bin.
3058 28. Exim was keeping the "process log" file open after each use, just as it
3059 does for the main log. This opens the possibility of it remaining open for
3060 long periods when the USR1 signal hits a daemon. Occasional processlog
3061 errors were reported, that could have been caused by this. Anyway, it seems
3062 much more sensible not to leave this file open at all, so that is what now
3065 29. The long-running daemon process does not normally write to the log once it
3066 has entered its main loop, and it closes the log before doing so. This is
3067 so that log files can straightforwardly be renamed and moved. However,
3068 there are a couple of unusual error situations where the daemon does write
3069 log entries, and I had neglected to close the log afterwards.
3071 30. The text of an SMTP error response that was received during a remote
3072 delivery was being truncated at 512 bytes. This is too short for some of
3073 the long messages that one sometimes sees. I've increased the limit to
3076 31. It is now possible to make retry rules that apply only when a message has a
3077 specific sender, in particular, an empty sender.
3079 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3080 it possible to be selective about when SMTP synchronization is enforced.
3082 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3084 32. Implemented hosts_connection_nolog.
3086 33. Added an ACL for QUIT.
3088 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3091 35. Added mailbox_size and mailbox_filecount to appendfile.
3093 36. Added control = no_multiline_responses to ACLs.
3095 37. There was a bug in the logic of the code that waits for the clock to tick
3096 in the case where the clock went backwards by a substantial amount such
3097 that the microsecond fraction of "now" was more than the microsecond
3098 fraction of "then" (but the whole seconds number was less).
3100 38. Added support for the libradius Radius client library this is found on
3101 FreeBSD (previously only the radiusclient library was supported).
3107 1. When certain lookups returned multiple values in the form name=value, the
3108 quoting of the values was not always being done properly. Specifically:
3109 (a) If the value started with a double quote, but contained no whitespace,
3111 (b) If the value contained whitespace other than a space character (i.e.
3112 tabs or newlines or carriage returns) it was not quoted.
3113 This fix has been applied to the mysql and pgsql lookups by writing a
3114 separate quoting function and calling it from the lookup code. The fix
3115 should probably also be applied to nisplus, ibase and oracle lookups, but
3116 since I cannot test any of those, I have not disturbed their existing code.
3118 2. A hit in the callout cache for a specific address caused a log line with no
3119 reason for rejecting RCPT. Now it says "Previous (cached) callout
3120 verification failure".
3122 3. There was an off-by-one bug in the queryprogram router. An over-long
3123 return line was truncated at 256 instead of 255 characters, thereby
3124 overflowing its buffer with the terminating zero. As well as fixing this, I
3125 have increased the buffer size to 1024 (and made a note to document this).
3127 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3128 when Exim is waiting for an SMTP response from a remote server, Exim
3129 restarts its select() call on the socket, thereby resetting its timeout.
3130 This is not a problem when such interrupts are rare. Somebody set up a cron
3131 job to run exiwhat every 2 minutes, which is less than the normal select()
3132 timeout (5 or 10 minutes). This meant that the select() timeout never
3133 kicked in because it was always reset. I have fixed this by comparing the
3134 time when an interrupt arrives with the time at the start of the first call
3135 to select(). If more time than the timeout has elapsed, the interrupt is
3136 treated as a timeout.
3138 5. Some internal re-factoring in preparation for the addition of Sieve
3139 extensions (by MH). In particular, the "personal" test is moved to a
3140 separate function, and given an option for scanning Cc: and Bcc: (which is
3141 not set for Exim filters).
3143 6. When Exim created an email address using the login of the caller as the
3144 local part (e.g. when creating a From: or Sender: header line), it was not
3145 quoting the local part when it contained special characters such as @.
3147 7. Installed new OpenBSD configuration files.
3149 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3150 try to make them clearer.
3152 9. Callout options, other than the timeout value, were being ignored when
3153 verifying sender addresses in header lines. For example, when using
3155 verify = header_sender/callout=no_cache
3157 the cache was (incorrectly) being used.
3159 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3160 only the Cygwin environment.
3162 11. When return_path_on_delivery was set as a log selector, if different remote
3163 addresses in the same message used different return paths and parallel
3164 remote delivery occurred, the wrong values would sometimes be logged.
3165 (Whenever a remote delivery process finished, the return path value from
3166 the most recently started remote delivery process was logged.)
3168 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3169 lines when AUTH and/or TLS are in use. This is the "received protocol"
3170 field. Exim used to use "asmtp" for authenticated SMTP, without any
3171 indication (in the protocol name) for TLS use. Now it follows the RFC and
3172 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3173 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3174 names appear in log lines as well as in Received: header lines.
3176 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3177 extensions, and comparison tests, and to fix some bugs.
3179 14. Changes to the "personal" filter test:
3181 (1) The test was buggy in that it was just doing the equivalent of
3182 "contains" tests on header lines. For example, if a user's address was
3183 anne@some.where, the "personal" test would incorrectly be true for
3185 To: susanne@some.where
3187 This test is now done by extracting each address from the header in turn,
3188 and checking the entire address. Other tests that are part of "personal"
3189 are now done using regular expressions (for example, to check local parts
3190 of addresses in From: header lines).
3192 (2) The list of non-personal local parts in From: addresses has been
3193 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3194 taken from the Sieve specification recommendations.
3196 (3) If the message contains any header line starting with "List-" it is
3197 treated as non-personal.
3199 (4) The test for "circular" in the Subject: header line has been removed
3200 because it now seems ill-conceived.
3202 15. Minor typos in src/EDITME comments corrected.
3204 16. Installed latest exipick from John Jetmore.
3206 17. If headers_add on a router specified a text string that was too long for
3207 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3208 of string_sprintf() is now avoided.
3210 18. $message_body_size was not set (it was always zero) when running the DATA
3211 ACL and the local_scan() function.
3213 19. For the "mail" command in an Exim filter, no default was being set for
3214 the once_repeat time, causing a random time value to be used if "once" was
3215 specified. (If the value happened to be <= 0, no repeat happened.) The
3216 default is now 0s, meaning "never repeat". The "vacation" command was OK
3217 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3218 (I found it when inspecting the code).
3220 20. There is now an overall timeout for performing a callout verification. It
3221 defaults to 4 times the callout timeout, which applies to individual SMTP
3222 commands during the callout. The overall timeout applies when there is more
3223 than one host that can be tried. The timeout is checked before trying the
3224 next host. This prevents very long delays if there are a large number of
3225 hosts and all are timing out (e.g. when the network connections are timing
3226 out). The value of the overall timeout can be changed by specifying an
3227 additional sub-option for "callout", called "maxwait". For example:
3229 verify = sender/callout=5s,maxwait=20s
3231 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3232 to the end before writing, but this should make it even safer).
3234 22. Exim was forgetting that it had advertised PIPELINING for the second and
3235 subsequent messages on an SMTP connection. It was also not resetting its
3236 memory on STARTTLS and an internal HELO.
3238 23. When Exim logs an SMTP synchronization error within a session, it now
3239 records whether PIPELINING has been advertised or not.
3241 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3242 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3243 rather than long int.
3245 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3247 26. Added the never_mail option to autoreply.
3253 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3254 crash if the getsockname() call failed; this can happen if a connection is
3255 closed very soon after it is established. The problem was simply in the
3256 order in which certain operations were done, causing Exim to try to write
3257 to the SMTP stream before it had set up the file descriptor. The bug has
3258 been fixed by making things happen in the correct order.
3264 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3265 before the connection was closed, thus losing the rejection response.
3267 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3268 some early Solaris releases, but causes trouble in current releases where
3269 socklen_t is defined.
3271 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3274 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3277 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3278 that releasing the top of it at the end releases what was used for sub-
3279 expansions (unless the block got too big). However, discard this block if
3280 the first thing is a variable or header, so that we can use its block when
3281 it is dynamic (useful for very large $message_headers, for example).
3283 6. Lookups now cache *every* query, not just the most recent. A new, separate
3284 store pool is used for this. It can be recovered when all lookup caches are
3285 flushed. Lookups now release memory at the end of their result strings.
3286 This has involved some general refactoring of the lookup sources.
3288 7. Some code has been added to the store_xxx() functions to reduce the amount
3289 of flapping under certain conditions.
3291 8. log_incoming_interface used to affect only the <= reception log lines. Now
3292 it causes the local interface and port to be added to several more SMTP log
3293 lines, for example "SMTP connection from", and rejection lines.
3295 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3297 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3299 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3300 could be overwritten at the end of the current message (or the start of a
3301 new message if it was set in a HELO ACL). The value is now preserved for
3302 the duration of the SMTP connection.
3304 12. If a transport had a headers_rewrite setting, and a matching header line
3305 contained an unqualified address, that address was qualified, even if it
3306 did not match any rewriting rules. The underlying bug was that the values
3307 of the flags that permit the existence of unqualified sender and recipient
3308 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3309 non-local messages, and by -bnq for local messages) were not being
3310 preserved with the message after it was received.
3312 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3313 "next input=" as part of the text comprising the host identity instead of
3314 the correct text. The code was using the same buffer for two different
3315 strings. However, depending on which order the printing function evaluated
3316 its arguments, the bug did not always show up. Under Linux, for example, my
3317 test suite worked just fine.
3319 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3320 doesn't work with some older versions of Perl. It has been changed to "my",
3321 which in any case is probably the better facility to use.
3323 15. A really picky compiler found some instances of statements for creating
3324 error messages that either had too many or two few arguments for the format
3327 16. The size of the buffer for calls to the DNS resolver has been increased
3328 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3329 for addresses that have a lot of PTR records. This alleviates a problem; it
3330 does not fully solve it.
3332 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3333 buffer now truncates the list and logs the incident, which is the same
3334 action as happens when Exim is looking up a host name and its aliases.
3335 Previously in this situation something unpredictable would happen;
3336 sometimes it was "internal error: store_reset failed".
3338 18. If a server dropped the connection unexpectedly when an Exim client was
3339 using GnuTLS and trying to read a response, the client delivery process
3340 crashed while trying to generate an error log message.
3342 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3343 string, for example:
3345 warn message = H1: something\nH2: something
3347 the text was added as a single header line from Exim's point of view
3348 though it ended up OK in the delivered message. However, searching for the
3349 second and subsequent header lines using $h_h2: did not work. This has been
3350 fixed. Similarly, if a system filter added multiple headers in this way,
3351 the routers could not see them.
3353 20. Expanded the error message when iplsearch is called with an invalid key to
3354 suggest using net-iplsearch in a host list.
3356 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3357 ACLs are no longer actually imposed (and a message to that effect is
3360 22. If a "gecos" field in a passwd entry contained escaped characters, in
3361 particular, if it contained a \" sequence, Exim got it wrong when building
3362 a From: or a Sender: header from that name. A second bug also caused
3363 incorrect handling when an unquoted " was present following a character
3364 that needed quoting.
3366 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3367 was not being matched caselessly.
3369 24. Arranged for all hyphens in the exim.8 source to be escaped with
3372 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3373 properly. Recipient callout cache records were still being keyed to include
3374 the sender, even when use_sender was set false. This led to far more
3375 callouts that were necessary. The sender is no longer included in the key
3376 when use_sender is false.
3378 26. Added "control = submission" modifier to ACLs.
3380 27. Added the ${base62d: operator to decode base 62 numbers.
3382 28. dnsdb lookups can now access SRV records.
3384 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3385 the configuration file.
3387 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3388 (-v) mode. This makes the output for a verbose queue run more intelligible.
3390 31. Added a use_postmaster feature to recipient callouts.
3392 32. Added the $body_zerocount variable, containing the number of binary zero
3393 bytes in the message body.
3395 33. The time of last modification of the "new" subdirectory is now used as the
3396 "mailbox time last read" when there is a quota error for a maildir
3399 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3401 35. Added +ignore_unknown as a special item in host lists.
3403 36. Code for decoding IPv6 addresses in host lists is now included, even if
3404 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3405 address was recognized as an IP address, but was then not correctly decoded
3406 into binary, causing unexpected and incorrect effects when compared with
3413 1. Very minor rewording of debugging text in manualroute to say "list of
3414 hosts" instead of "hostlist".
3416 2. If verify=header_syntax was set, and a header line with an unqualified
3417 address (no domain) and a large number of spaces between the end of the
3418 name and the colon was received, the reception process suffered a buffer
3419 overflow, and (when I tested it) crashed. This was caused by some obsolete
3420 code that should have been removed. The fix is to remove it!
3422 3. When running in the test harness, delay a bit after writing a bounce
3423 message to get a bit more predictability in the log output.
3425 4. Added a call to search_tidyup() just before forking a reception process. In
3426 theory, someone could use a lookup in the expansion of smtp_accept_max_
3427 per_host which, without the tidyup, could leave open a database connection.
3429 5. Added the variables $recipient_data and $sender_data which get set from a
3430 lookup success in an ACL "recipients" or "senders" condition, or a router
3431 "senders" option, similar to $domain_data and $local_part_data.
3433 6. Moved the writing of debug_print from before to after the "senders" test
3436 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3437 problems for message scanning, either using a data ACL, or using
3438 local_scan() because the Received: header was not generated till after they
3439 were called (in order to set the time as the time of reception completion).
3440 I have revised the way this works. The header is now generated after the
3441 body is received, but before the ACL or local_scan() are called. After they
3442 are run, the timestamp in the header is updated.
3448 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3449 before starting a queue runner without re-exec. This happened only when
3450 deliver_drop_privilege was set or when the Exim user was set to root. The
3451 effect of the bug was that timeouts during subsequent deliveries caused
3452 crashes instead of being properly handled. The handler is now left at its
3453 default (and expected) setting.
3455 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3456 message, again when deliver_drop_privilege is set or Exim is run as root.
3457 The bug described in (1) was not present in this case, but the tidying up
3458 of the other signals was missing. I have made the two cases consistent.
3460 3. The ignore_target_hosts setting on a manualroute router was being ignored
3461 for hosts that were looked up using the /MX notation.
3463 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3466 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3467 operated on the sender address. After changing the $sender_address to <>
3468 for the sender address verify, Exim was re-instated it as the original
3469 (before rewriting) address, but remembering that it had rewritten it, so it
3470 wasn't rewriting it again. This bug also had the effect of breaking the
3471 sender address verification caching when the sender address was rewritten.
3473 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3474 This has been changed so that if the ip literal address matches
3475 ignore_target_hosts, the router declines.
3477 7. Added expansion conditions match_domain, match_address, and match_local_
3478 part (NOT match_host).
3480 8. The placeholder for the Received: header didn't have a length field set.
3482 9. Added code to Exim itself and to exim_lock to test for a specific race
3483 condition that could lead to file corruption when using MBX delivery. The
3484 issue is with the lockfile that is created in /tmp. If this file is removed
3485 after a process has opened it but before that process has acquired a lock,
3486 there is the potential for a second process to recreate the file and also
3487 acquire a lock. This could lead to two Exim processes writing to the file
3488 at the same time. The added code performs the same test as UW imapd; it
3489 checks after acquiring the lock that its file descriptor still refers to
3490 the same named file.
3492 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3493 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3496 11. Added the smtp_active_hostname option. If used, this will typically be made
3497 to depend on the incoming interface address. Because $interface_address is
3498 not set up until the daemon has forked a reception process, error responses
3499 that can happen earlier (such as "too many connections") no longer contain
3502 12. If an expansion in a condition on a "warn" statement fails because a lookup
3503 defers, the "warn" statement is abandoned, and the next ACL statement is
3504 processed. Previously this caused the whole ACL to be aborted.
3506 13. Added the iplsearch lookup type.
3508 14. Added ident_timeout as a log selector.
3510 15. Added tls_certificate_verified as a log selector.
3512 16. Added a global option tls_require_ciphers (compare the smtp transport
3513 option of the same name). This controls incoming TLS connections.
3515 17. I finally figured out how to make tls_require_ciphers do a similar thing
3516 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3517 before starting the TLS session.
3519 18. Tabs are now shown as \t in -bP output.
3521 19. If the log selector return_path_on_delivery was set, Exim crashed when
3522 bouncing a message because it had too many Received: header lines.
3524 20. If two routers both had headers_remove settings, and the first one included
3525 a superfluous trailing colon, the final name in the first list and the
3526 first name in the second list were incorrectly joined into one item (with a
3527 colon in the middle).
3533 1. Added -C and -D options to the exinext utility, mainly to make it easier
3534 to include in the automated testing, but these could be helpful when
3535 multiple configurations are in use.
3537 2. The exinext utility was not formatting the output nicely when there was
3538 an alternate port involved in the retry record key, nor when there was a
3539 message id as well (for retries that were specific to a specific message
3540 and a specific host). It was also confused by IPv6 addresses, because of
3541 the additional colons they contain. I have fixed the IPv4 problem, and
3542 patched it up to do a reasonable job for IPv6.
3544 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3545 delivery, the log line now contains "pipelined" if PIPELINING was used.
3547 4. An SMTP transport process used to panic and die if the bind() call to set
3548 an explicit outgoing interface failed. This has been changed; it is now
3549 treated in the same way as a connect() failure.
3551 5. A reference to $sender_host_name in the part of a conditional expansion
3552 that was being skipped was still causing a DNS lookup. This no longer
3555 6. The def: expansion condition was not recognizing references to header lines
3556 that used bh_ and bheader_.
3558 7. Added the _cache feature to named lists.
3560 8. The code for checking quota_filecount in the appendfile transport was
3561 allowing one more file than it should have been.
3563 9. For compatibility with Sendmail, the command line option
3571 and sets the incoming protocol and host name (for trusted callers). The
3572 host name and its colon can be omitted when only the protocol is to be set.
3573 Note the Exim already has two private options, -pd and -ps, that refer to
3574 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3575 "s", but I don't think that's a major issue.
3577 10. A number of refactoring changes to the code, none of which should affect
3580 (a) The number of logging options was getting close to filling up the
3581 32-bit word that was used as a bit map. I have split them into two classes:
3582 those that are passed in the argument to log_write(), and those that are
3583 only ever tested independently outside of that function. These are now in
3584 separate 32-bit words, so there is plenty of room for expansion again.
3585 There is no change in the user interface or the logging behaviour.
3587 (b) When building, for example, log lines, the code previously used a
3588 macro that called string_cat() twice, in order to add two strings. This is
3589 not really sufficiently general. Furthermore, there was one instance where
3590 it was actually wrong because one of the argument was used twice, and in
3591 one call a function was used. (As it happened, calling the function twice
3592 did not affect the overall behaviour.) The macro has been replaced by a
3593 function that can join an arbitrary number of extra strings onto a growing
3596 (c) The code for expansion conditions now uses a table and a binary chop
3597 instead of a serial search (which was left over from when there were very
3598 few conditions). Also, it now recognizes conditions like "pam" even when
3599 the relevant support is not compiled in: a suitably worded error message is
3600 given if an attempt is made to use such a condition.
3602 11. Added ${time_interval:xxxxx}.
3604 12. A bug was causing one of the ddress fields not to be passed back correctly
3605 from remote delivery subprocesses. The field in question was not being
3606 subsequently used, so this caused to problems in practice.
3608 13. Added new log selectors queue_time and deliver_time.
3610 14. Might have fixed a bug in maildirsizefile handling that threw up
3611 "unexpected character" debug warnings, and recalculated the data
3612 unnecessarily. In any case, I expanded the warning message to give more
3615 15. Added the message "Restricted characters in address" to the statements in
3616 the default ACL that block characters like @ and % in local parts.
3618 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3619 Three changes have been made:
3621 (a) There was a serious bug; a negative response to MAIL caused the whole
3622 recipient domain to be cached as invalid, thereby blocking all messages
3623 to all local parts at the same domain, from all senders. This bug has
3624 been fixed. The domain is no longer cached after a negative response to
3625 MAIL if the sender used is not empty.
3627 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3630 (c) A new callout option, "use_sender" has been added for people who want
3631 the modified behaviour.
3637 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3640 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3641 indeed breaks things for older releases.
3643 3. Added additional logging to the case where there is a problem reading data
3644 from a filter that is running in a subprocess using a pipe, in order to
3645 try to track down a specific problem.
3647 4. Testing facility fudge: when running in the test harness and attempting
3648 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3649 getting "No route to host". Convert this to a timeout.
3651 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3654 6. Some OS don't have socklen_t but use size_t instead. This affects the
3655 fifth argument of getsockopt() amongst other things. This is now
3656 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3657 can be set for individual OS. I have set it for SunOS5, OSF1, and
3658 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3659 some earlier ones do not.
3661 7. Change 4.30/15 was not doing the test caselessly.
3663 8. The standard form for an IPv6 address literal was being rejected by address
3664 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3665 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3666 this, as well as the form without the "IPv6" on the front (but only when
3667 address literals are enabled, of course).
3669 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3671 10. Exim crashed if a message with an empty sender address specified by -f
3672 encountered a router with an errors_to setting. This could be provoked only
3673 by a command such as
3677 where an empty string was supplied; "<>" did not hit this bug.
3679 11. Installed PCRE release 4.5.
3681 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3682 remained set. It is now erased.
3684 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3685 times from message ids (which are base 36 rather than the normal 62).
3687 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3688 were being counted as actual protocol errors, and logged if the log
3689 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3690 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3691 and DATA following a set of rejected RCPTs do not count as protocol errors.
3692 In other words, Exim assumes they were pipelined, though this may not
3693 actually be the case. Of course, in all cases the client gets an
3694 appropriate error code.
3696 15. If a lookup fails in an ACL condition, a message about the failure may
3697 be available; it is used if testing the ACL cannot continue, because most
3698 such messages specify what the cause of the deferral is. However, some
3699 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3700 that caused an old message to be retained and used if a later statement
3701 caused a defer, replacing the real cause of the deferral.
3703 16. If an IP address had so many PTR records that the DNS lookup buffer
3704 was not large enough to hold them, Exim could crash while trying to process
3705 the truncated data. It now detects and logs this case.
3707 17. Further to 4.21/58, another change has been made: if (and only if) the
3708 first line of a message (the first header line) ends with CRLF, a bare LF
3709 in a subsequent header line has a space inserted after it, so as not to
3710 terminate the header.
3712 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3713 unnecessarily, used atoi() instead of strtol(), and didn't check the
3714 termination when getting file sizes from file names by regex.
3716 19. Completely re-implemented the support for maildirsize files, in the light
3717 of a number of problems with the previous contributed implementation
3718 (4.30/29). In particular:
3720 . If the quota is zero, the maildirsize file is maintained, but no quota is
3723 . If the maildir directory does not exist, it is created before any attempt
3724 to write a maildirsize file.
3726 . The quota value in the file is just a cache; if the quota is changed in
3727 the transport, the new value overrides.
3729 . A regular expression is available for excluding directories from the
3732 20. The autoreply transport checks the characters in options that define the
3733 message's headers; it allows continued headers, but it was checking with
3734 isspace() after an embedded newline instead of explicitly looking for a
3737 21. If all the "regular" hosts to which an address was routed had passed their
3738 expiry times, and had not reached their retry times, the address was
3739 bounced, even if fallback hosts were defined. Now Exim should go on to try
3742 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3743 equivalent code in the SMTP transport. Some hosts send humungous responses
3744 to HELO/EHLO, more than 1024 it seems.
3746 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3747 gives compiler warnings in some environments. I've now done it "properly",
3750 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3751 (because of problems with the built-in one) was declared to return uschar *
3752 instead of char *, causing compiler failure.
3754 25. Fixed a file descriptor leak when processing alias/forward files.
3756 26. Fixed a minor format string issue in dbfn.c.
3758 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3760 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3761 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3764 29. When a sender address is verified, it is cached, to save repeating the test
3765 when there is more than one recipient in a message. However, when the
3766 verification involves a callout, it is possible for different callout
3767 options to be set for different recipients. It is too complicated to keep
3768 track of this in the cache, so now Exim always runs a verification when a
3769 callout is required, relying on the callout cache for the optimization.
3770 The overhead is duplication of the address routing, but this should not be
3773 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3774 to be verified with callout=postmaster, and the main callout worked but the
3775 postmaster check failed, the verification correctly failed. However, if a
3776 subsequent RCPT command asked for sender verification *without* the
3777 postmaster check, incorrect caching caused this verification also to fail,
3780 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3781 it was not caching the DNS options (qualify_single, search_parents) that
3782 were used when the lookup failed. A subsequent lookup with different
3783 options therefore always gave the same answer, though there were cases
3784 where it should not have. (Example: a "domains = !$mx_any" option on a
3785 dnslookup router: the "domains" option is always processed without any
3786 widening, but the router might have qualify_single set.) Now Exim uses the
3787 cached value only when the same options are set.
3789 32. Added John Jetmore's "exipick" utility to the distribution.
3791 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3792 than a timeout (e.g. a certificate is required, and is not provided), an
3793 Exim server now closes the connection immediately. Previously it waited for
3794 the client to close - but if the client is SSL, it seems that they each
3795 wait for each other, leading to a delay before one of them times out.
3797 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3798 maintained 0.8.x compatibility because I don't think many are using it, and
3799 it is clearly obsolete.
3801 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3804 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3805 client certificate was expired. A simple patch fixes this, though I don't
3806 understand the full logic of why the verify callback is called multiple
3809 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3810 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3811 which causes problems with some clients (such as the Certicom SSL Plus
3812 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3813 disables the coutermeasure allowing Eudora to connect."
3815 38. Exim was not checking that a write() to a log file succeeded. This could
3816 lead to Bad Things if a log got too big, in particular if it hit a file
3817 size limit. Exim now panics and dies if it cannot write to a log file, just
3818 as it does if it cannot open a log file.
3820 39. Modified OS/Makefile-Linux so that it now contains
3822 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3824 The two -D definitions ensure that Exim is compiled with large file
3825 support, which makes it possible to handle log files that are bigger than
3828 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3829 instance) a domain was checked against a named list that involved a lookup,
3830 causing $domain_data to be set, then another domain was checked against the
3831 same list, then the first domain was re-checked, the value of $domain_data
3832 after the final check could be wrong. In particular, if the second check
3833 failed, it could be set empty. This bug probably also applied to
3836 41. The strip_trailing_dot option was not being applied to the address given
3837 with the -f command-line option.
3839 42. The code for reading a message's header from the spool was incrementing
3840 $received_count, but never initializing it. This meant that the value was
3841 incorrect (doubled) while delivering a message in the same process in which
3842 it was received. In the most common configuration of Exim, this never
3843 happens - a fresh exec is done - but it can happen when
3844 deliver_drop_privilege is set.
3846 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3847 it now includes up to 150 characters of the unexpected data in the log
3850 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3851 and building data strings. The size of both of these buffers was 10 000
3852 bytes - far larger than anybody would *ever* want, thought I. Needless to
3853 say, somebody hit the limit. I have increased the maximum line length to
3854 20 000 and the maximum data length of concatenated lines to 100 000. I have
3855 also fixed two bugs, because there was no checking on these buffers. Tsk,
3856 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3857 buffer is too small.
3859 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3860 lsearch lookups. Now it does.
3862 46. When parsing a route_list item in a manualroute router, a fixed-length
3863 buffer was used for the list of hosts. I made this 1024 bytes long,
3864 thinking that nobody would ever have a list of hosts that long. Wrong.
3865 Somebody had a whole pile of complicated expansion conditions, and the
3866 string was silently truncated, leading to an expansion error. It turns out
3867 that it is easier to change to an unlimited length (owing to other changes
3868 that have happened since this code was originally written) than to build
3869 structure for giving a limitation error. The length of the item that
3870 expands into the list of hosts is now unlimited.
3872 47. The lsearch lookup could not handle data where the length of text line was
3873 more than 4095 characters. Such lines were truncated, leading to shortened
3874 data being returned. It should now handle lines of any length.
3876 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3877 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3880 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3881 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3882 no longer happens in comment lines. A list of replacements is now placed
3883 at the head of all of the source files, except those whose only change is
3884 to replace PERL_COMMAND in the very first #! line.
3886 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3887 messages on the queue, with a bottom-up merge sort, using code contributed
3888 by Michael Haardt. This should make operations like -bp somewhat faster on
3889 large queues. It won't affect queue runners, except when queue_run_in_order
3892 51. Installed eximstats 1.31 in the distribution.
3894 52. Added support for SRV lookups to the dnslookup router.
3896 53. If an ACL referred to $message_body or $message_body_end, the value was not
3897 reset for any messages that followed in the same SMTP session.
3899 54. The store-handling optimization for building very long strings was not
3900 differentiating between the different store pools. I don't think this
3901 actually made any difference in practice, but I've tidied it.
3903 55. While running the routers to verify a sender address, $sender_address
3904 was still set to the sender address. This is wrong, because when routing to
3905 send a bounce to the sender, it would be empty. Therefore, I have changed
3906 it so that, while verifying a sender address, $sender_address is set to <>.
3907 (There is no change to what happens when verifying a recipient address.)
3909 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3910 target A or AAAA records (if not already returned) without resetting the
3911 qualify_single or search_parents options of the DNS resolver. These are
3912 inappropriate in this case because the targets of MX and SRV records must
3913 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3914 target that, when qualified, matched something in the local domain. These
3915 two options are now turned off when doing these lookups.
3917 57. It seems that at least some releases of Reiserfs (which does not have the
3918 concept of a fixed number of inodes) returns zero and not -1 for the
3919 number of available inodes. This interacted badly with check_spool_inodes,
3920 which assumed that -1 was the "no such thing" setting. What I have done is
3921 to check that the total number of inodes is greater than zero before doing
3922 the test of how many are available.
3924 58. When a "warn" ACL statement has a log_message modifier, the message is
3925 remembered, and not repeated. This is to avoid a lot of repetition when a
3926 message has many recipients that cause the same warning to be written.
3927 Howewer, Exim was preserving the list of already written lines for an
3928 entire SMTP session, which doesn't seem right. The memory is now reset if a
3929 new message is started.
3931 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3932 debugging output unless log_rewrite was also set.
3934 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3935 of (int)(handle) when we know that handle contains (void *)(-1).
3937 61. The Exim daemon panic-logs an error return when it closes the incoming
3938 connection. However "connection reset by peer" seems to be common, and
3939 isn't really an error worthy of noting specially, so that particular error
3942 62. When Exim is trying to find all the local interfaces, it used to panic and
3943 die if the ioctl to get the interface flags failed. However, it seems that
3944 on at least one OS (Solaris 9) it is possible to have an interface that is
3945 included in the list of interfaces, but for which you get a failure error
3946 for this call. This happens when the interface is not "plumbed" into a
3947 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3948 failure of the "get flags" call assumes that the interface is down.
3950 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3951 makes life easier for people who are doing arithmetic on fields extracted
3952 from dates, where you often get leading zeros that should not be
3953 interpreted as octal.
3955 64. Added qualify_domain to the redirect router, to override the global
3958 65. If a pathologically long header line contained very many addresses (the
3959 report of this problem mentioned 10 000) and each of them was rewritten,
3960 Exim could use up a very large amount of memory. (It kept on making new
3961 copies of the header line as it rewrote, and never released the old ones.)
3962 At the expense of a bit more processing, the header rewriting function has
3963 been changed so that it no longer eats memory in this way.
3965 66. The generation of the Received: header has been moved from the time that a
3966 message starts to be received, to the time that it finishes. The timestamp
3967 in the Received: header should now be very close to that of the <= log
3968 line. There are two side-effects of this change:
3970 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3971 logged header lines no longer include the local Received: line, because
3972 it has not yet been created. The same applies to a copy of the message
3973 that is returned to a non-SMTP sender when a message is rejected.
3975 (b) When a filter file is tested using -bf, no additional Received: header
3976 is added to the test message. After some thought, I decided that this
3979 This change does not affect the value of $received_for. It is still set
3980 after address rewriting, but before local_scan() is called.
3982 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3984 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3985 gave an unhelpful panic error message, and a defer error. I have managed to
3986 change this behaviour so that it now rejects any supplied certificate,
3987 which seems right, as the list of acceptable certificates is empty.
3989 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3990 gave an unhelpful defer error. I have not managed to make this reject any
3991 supplied certificates, but the error message it gives is "no certificate
3992 supplied", which is not helpful.
3994 70. exigrep's output now also includes lines that are not associated with any
3995 message, but which match the given pattern. Implemented by a patch from
3996 Martin Sluka, which also tidied up the Perl a bit.
3998 71. Recipient callout verification, like sender verification, was using <> in
3999 the MAIL FROM command. This isn't really the right thing, since the actual
4000 sender may affect whether the remote host accepts the recipient or not. I
4001 have changed it to use the actual sender in the callout; this means that
4002 the cache record is now keyed on a recipient/sender pair, not just the
4003 recipient address. There doesn't seem to be a real danger of callout loops,
4004 since a callout by the remote host to check the sender would use <>.
4005 [SEE ABOVE: changed after hitting problems.]
4007 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4008 temporary errors. However, in the case of such a code being given after
4009 the end of a data transmission (i.e. after ".") Exim was failing to write
4010 a retry record for the message. (Yes, there was some broken host that was
4011 actually sending 8xx at this point.)
4013 73. An unknown lookup type in a host list could cause Exim to panic-die when
4014 the list was checked. (An example that provoked this was putting <; in the
4015 middle of a list instead of at the start.) If this happened during a DATA
4016 ACL check, a -D file could be left lying around. This kind of configuration
4017 error no longer causes Exim to die; instead it causes a defer errror. The
4018 incident is still logged to the main and panic logs.
4020 74. Buglet left over from Exim 3 conversion. The message "too many messages
4021 in one connection" was written to the rejectlog but not the mainlog, except
4022 when address rewriting (yes!) was being logged.
4024 75. Added write_rejectlog option.
4026 76. When a system filter was run not as root (that is, when system_filter_user
4027 was set), the values of the $n variables were not being returned to the
4028 main process; thus, they were not subsequently available in the $sn
4031 77. Added +return_path_on_delivery log selector.
4033 78. A connection timeout was being treated differently from recipients deferred
4034 when testing hosts_max_try with a message that was older than the host's
4035 retry timeout. (The host should not be counted, thus allowing all hosts to
4036 be tried at least once before bouncing.) This may have been the cause of an
4037 occasionally reported bug whereby a message would remain on the queue
4038 longer than the retry timeout, but would be bounced if a delivery was
4039 forced. I say "may" because I never totally pinned down the problem;
4040 setting up timeout/retry tests is difficult. See also the next item.
4042 79. The ultimate address timeout was not being applied to errors that involved
4043 a combination of host plus message (for example, a timeout on a MAIL
4044 command). When an address resolved to a number of possible hosts, and they
4045 were not all tried for each delivery (e.g. because of hosts_max_try), a
4046 message could remain on the queue longer than the retry timeout.
4048 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4051 81. Fixed an obscure SMTP outgoing bug which required at least the following
4052 conditions: (a) there was another message waiting for the same server;
4053 (b) the server returned 5xx to all RCPT commands in the first message so
4054 that the message was not completed; (c) the server dropped the connection
4055 or gave a negative response to the RSET that Exim sends to abort the
4056 transaction. The observed case was a dropped connection after DATA that had
4057 been sent in pipelining mode. That is, the server had advertised PIPELINING
4058 but was not implementing it correctly. The effect of the bug was incorrect
4059 behaviour, such as trying another host, and this could lead to a crash.
4065 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4066 and daemon.c were passed as pointers to ints; they should have been
4067 pointers to socklen_t variables (which are typically unsigned ints).
4069 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4072 3. Fixed a really odd bug that affected only the testing scheme; patching a
4073 certain fixed string in the binary changed the value of another string that
4074 happened to be identical to the end of the original first string.
4076 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4077 name", it returns that address as the IP address. On some operating
4078 systems (e.g. Solaris), it also passes back the IP address string as the
4079 "host name". However, on others (e.g. Linux), it passes back an empty
4080 string. Exim wasn't checking for this, and was changing the host name to an
4081 empty string, assuming it had been canonicized.
4083 5. Although rare, it is permitted to have more than one PTR record for a given
4084 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4085 all the names associated with an address, because they do in Solaris.
4086 However, it seems that they do not in Linux for data that comes from the
4087 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4088 I found this out when I moved to a new Linux workstation and tried to run
4089 the Exim test suite.
4091 To get round this problem I have changed the code so that it now does its
4092 own call to the DNS to look up PTR records when searching for a host name.
4093 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4094 addresses that are only in /etc/hosts are still found.
4096 This behaviour is, however, controlled by an option called host_lookup_
4097 order, which defaults to "bydns:byaddr". If people want to use the other
4098 order, or indeed, just use one or the other means of lookup, they can
4099 specify it in this variable.
4101 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4102 some operating systems, this comes back from gethostbyaddr() as an empty
4103 string, and this is what Exim used to test for. However, it seems that in
4104 other systems, "." is yielded. Exim now tests for this case too.
4106 7. The values of check_spool_space and check_log_space are now held internally
4107 as a number of kilobytes instead of an absolute number of bytes. If a
4108 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4109 kilobyte. This means that much larger values can be stored.
4111 8. Exim monitor: an attempt to get the action menu when not actually pointing
4112 at a message produces an empty menu entitled "No message selected". This
4113 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4114 no entries in it ("Shell widget menu has zero width and/or height"). So I
4115 have added a single, blank menu entry in this case.
4117 9. Added ${quote_local_part.
4119 10. MIME decoding is now applied to the contents of Subject: header lines when
4122 11. Now that a reference to $sender_host_address automatically causes a reverse
4123 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4124 host lookup before query-style lookups in lists that might use this
4125 variable. This has therefore been abolished, and the "net-" prefix is no
4126 longer necessary for query-style lookups.
4128 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4129 have been a typo for LFLAGS, so it has been changed.
4131 13. The install script calls Exim with "-C /dev/null" in order to find the
4132 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4133 to be output. Howeve, since Exim outputs its version number before the
4134 error, it didn't break the script. It just looked ugly. I fixed this by
4135 always allowing "-C /dev/null" if the caller is root.
4137 14. Ignore overlarge ACL variable number when reading spool file - insurance
4138 against a later release with more variables having written the file.
4140 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4141 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4142 this, as well as the form without the "IPv6" on the front.
4144 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4145 OS/Makefile-Darwin file.
4147 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4148 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4150 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4151 variable. However, some OS use ints for inodes. I've added cast to long int
4152 to get rid of the compiler warning.
4154 19. I had forgotten to lock out "/../" in configuration file names when
4155 ALT_CONFIG_PREFIX was set.
4157 20. Routers used for verification do not need to specify transports. However,
4158 if such a router generated a host list, and callout was configured, Exim
4159 crashed, because it could not find a port number from the (non-existent)
4160 transport. It now assumes port 25 in this circumstance.
4162 21. Added the -t option to exigrep.
4164 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4165 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4166 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4167 them is set, LOOKUP_LSEARCH is forced.
4169 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4171 24. Added sender and host information to the "rejected by local_scan()" log
4172 line; previously there was no indication of these.
4174 25. Added .include_if_exists.
4176 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4177 Linux. It turns out that not all file systems support this. Apparently some
4178 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4179 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4180 not supported on the file descriptor, is now ignored when Exim is trying to
4181 sync a directory. This applies only to Linux.
4183 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4185 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4186 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4189 29. Added support for maildirsize files from supplied patch (modified a bit).
4191 30. The use of :fail: followed by an empty string could lead Exim to respond to
4192 sender verification failures with (e.g.):
4194 550 Verification failed for <xxx>
4195 550 Sender verify failed
4197 where the first response line was missing the '-' that indicates it is not
4198 the final line of the response.
4200 31. The loop for finding the name of the user that called Exim had a hardwired
4201 limit of 10; it now uses the value of finduser_retries, which is used for
4202 all other user lookups.
4204 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4207 33. Exim was neglecting to zero errno before one call of strtol() when
4208 expanding a string and expecting an integer value. On some systems this
4209 resulted in spurious "integer overflow" errors. Also, it was casting the
4210 result into an int without checking.
4212 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4213 did not work. The code looks as if it has *never* worked, though it appears
4214 to have been documented since at least releast 1.62. I have made it work.
4216 35. The "timeout_DNS" error in retry rules, also documented since at least
4217 1.62, also never worked. As it isn't clear exactly what this means, and
4218 clearly it isn't a major issue, I have abolished the feature by treating it
4219 as "timeout", and writing a warning to the main and panic logs.
4221 36. The display of retry rules for -brt wasn't always showing the error code
4224 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4225 timeout_connect_A, timeout_connect_MX.
4227 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4228 to the empty sender.
4230 39. The daemon was not analysing the content of -oX till after it had closed
4231 stderr and disconnected from the controlling terminal. This meant that any
4232 syntax errors were only noted on the panic log, and the return code from
4233 the command was 0. By re-arranging the code a little, I've made the
4234 decoding happen first, so such errors now appear on stderr, and the return
4235 code is 1. However, the actual setting up of the sockets still happens in
4236 the disconnected process, so errors there are still only recorded on the
4239 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4240 connections (as happens on some IP stacks) was logged at start up time as
4241 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4242 it from "IPv6 and IPv4", which means that two separate sockets are being
4245 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4246 says whether AF_INET or AF_INET6 was passed as an argument.
4248 42. Exiwhat output was messed up when time zones were included in log
4251 43. Exiwhat now gives more information about the daemon's listening ports,
4252 and whether -tls-on-connect was used.
4254 44. The "port" option of the smtp transport is now expanded.
4256 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4257 silently ignored. Now an error message is written to the main and panic
4260 46. There's a new ACL modifier called "logwrite" which writes to a log file
4261 as soon as it is encountered.
4263 47. Added $local_user_uid and $local_user_gid at routing time.
4265 48. Exim crashed when trying to verify a sender address that was being
4268 49. Exim was recognizing only a space character after ".include". It now also
4269 recognizes a tab character.
4271 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4272 extracting the relevant information from the specification. The man page no
4273 longer contains scrambled data for the -d option, and I've added a section
4274 at the front about calling Exim under different names.
4276 51. Added "extra_headers" argument to the "mail" command in filter files.
4278 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4281 53. Installed eximstats 1.29.
4283 54. Added transport_filter_timeout as a generic transport option.
4285 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4286 Cc: header lines. This was required by RFC 822, but it not required by RFC
4289 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4290 incoming messages that did not have them. Now it does so only if the
4291 message originates locally, that is, if there is no associated remote host
4292 address. When Resent- header lines are present, this applies to the Resent-
4293 lines rather than the non-Resent- lines.
4295 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4296 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4298 58. Messages for configuration errors now include the name of the main
4299 configuration file - useful now that there may be more than one file in a
4300 list (.included file names were always shown).
4302 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4303 for those rare installations that do not start the daemon as root or run it
4304 setuid root. I've cut out the call to initgroups() if the daemon is not
4307 60. The Exim user and group can now be bound into the binary as text strings
4308 that are looked up at the start of Exim's processing.
4310 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4312 62. Added $mailstore_basename variable.
4314 63. Installed patch to sieve.c from Michael Haardt.
4316 64. When Exim failed to open the panic log after failing to open the main log,
4317 the original message it was trying to log was written to stderr and debug
4318 output, but if they were not available (the usual case in production), it
4319 was lost. Now it is written to syslog before the two lines that record the
4320 failures to open the logs.
4322 65. Users' Exim filters run in subprocesses under the user's uid. It is
4323 possible for a "deliver" command or an alias in a "personal" command to
4324 provoke an address rewrite. If logging of address rewriting is configured,
4325 this fails because the process is not running as root or exim. There may be
4326 a better way of dealing with this, but for the moment (because 4.30 needs
4327 to be released), I have disabled address rewrite logging when running a
4328 filter in a non-root, non-exim process.
4334 1. The buildconfig auxiliary program wasn't quoting the value set for
4335 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4336 not defined. This bug was masked in 4.22 by the effect that was fixed in
4339 2. Some messages that were rejected after a message id was allocated were
4340 shown as "incomplete" by exigrep. It no longer does this for messages that
4341 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4343 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4344 have allow_domain_literals set, the ID did not get logged in the <= line.
4345 Domain literals are now always recognized in Message-ID: header lines.
4347 4. The first argument for a ${extract expansion item is the key name or field
4348 number. Leading and trailing spaces in this item were not being ignored,
4349 causing some misleading effects.
4351 5. When deliver_drop_privilege was set, single queue runner processes started
4352 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4353 same command in the process it spins off) were not dropping privilege.
4355 6. When the daemon running as "exim" started a queue runner, it always
4356 re-executed Exim in the spun-off process. This is a waste of effort when
4357 deliver_drop_privilege is set. The new process now just calls the
4358 queue-runner function directly.
4364 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4367 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4368 ignored. Though the use of -oP was forcing the writing of a pid file, it
4369 was always written to the default place.
4371 3. If the message "no IP address found for host xxxx" is generated during
4372 incoming verification, it is now followed by identification of the incoming
4373 connection (so you can more easily find what provoked it).
4375 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4377 5. Added some features to "harden" Exim a bit more against certain attacks:
4379 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4380 be put in Local/Makefile. This is like the never_users runtime option,
4381 but it cannot be overridden. The default setting is "root".
4383 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4384 prefix string with which any file named in a -C command line option
4387 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4388 is retained for -C and -D only if the caller of Exim is root. Without
4389 it, the exim user may also use -C and -D and retain privilege.
4391 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4392 command line option is disabled.
4394 6. Macro names set by the -D option must start with an upper case letter, just
4395 like macro names defined in the configuration file.
4397 7. Added "dereference=" facility to LDAP.
4399 8. Two instances of the typo "uknown" in the source files are fixed.
4401 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4402 the Configure-Makefile script screwed up while processing it.
4404 10. Incorporated PCRE 4.4.
4406 11. The SMTP synchronization check was not operating right at the start of an
4407 SMTP session. For example, it could not catch a HELO sent before the client
4408 waited for the greeting. There is now a check for outstanding input at the
4409 point when the greeting is written. Because of the duplex, asynchronous
4410 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4411 way, but not yet received, when the check is performed.
4413 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4414 on TCP/IP sockets, because this apparently causes some broken clients to
4417 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4418 unchanged) from the Cygwin maintainer.
4420 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4421 was supported instead of testing for mbx. Effectively a typo.
4423 15. The spa authenticator server code was not checking that the input it
4424 received was valid base64.
4426 16. The debug output line for the "set" modifier in ACLs was not showing the
4427 name of the variable that was being set.
4429 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4431 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4432 lookup on its own. Something else was needed to trigger the lookup. For
4433 example, a match in host_lookup or the need for a host name in a host list.
4434 Now, if $sender_host_name is referenced and the host name has not yet been
4435 looked up, a lookup is performed. If the lookup fails, the variable remains
4436 empty, and $host_lookup_failed is set to "1".
4438 19. Added "eqi" as a case-independent comparison operator.
4440 20. The saslauthd authentication condition could segfault if neither service
4441 nor realm was specified.
4443 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4444 error message that was logged was misleading, and incoming SMTP
4445 connections were dropped. The message is now more accurate, and temporary
4446 errors are given to SMTP connections.
4448 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4449 (except RSET) with a 421 error, until QUIT is received. However, it was
4450 failing to send a response to QUIT.
4452 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4453 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4454 ACL is now run after the helo_try_verify_hosts code.
4456 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4457 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4458 case-independent, so other case variants are also recognized). Apparently
4459 some systems use these upper case variants.
4461 25. If more than two messages were waiting for the same host, and a transport
4462 filter was specified for the transport, Exim sent two messages over the
4463 same TCP/IP connection, and then failed with "socket operation on non-
4464 socket" when it tried to send the third.
4466 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4468 27. The extern definition of crypt16() in expand.c was not being excluded when
4469 the OS had its own crypt16() function.
4471 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4472 as a preferred synonym for return_size_limit, both as an option and as an
4475 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4477 30. Changed the default configuration ACL to relax the local part checking rule
4478 for addresses that are not in any local domains. For these addresses,
4479 slashes and pipe symbols are allowed within local parts, but the sequence
4480 /../ is explicitly forbidden.
4482 31. SPA server authentication was not clearing the challenge buffer before
4485 32. log_message in a "warn" ACL statement was writing to the reject log as
4486 well as to the main log, which contradicts the documentation and doesn't
4487 seem right (because no rejection is happening). So I have stopped it.
4489 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4490 However, I am unable to do any testing of this.
4492 34. Fixed an infelicity in the appendfile transport. When checking directories
4493 for a mailbox, to see if any needed to be created, it was accidentally
4494 using path names with one or more superfluous leading slashes; tracing
4495 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4497 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4498 contents are added to the log line that is written for every discarded
4499 recipient. (Previously a log_message setting was ignored.)
4501 36. The ${quote: operator now quotes the string if it is empty.
4503 37. The install script runs exim in order to find its version number. If for
4504 some reason other than non-existence or emptiness, which it checks, it
4505 could not run './exim', it was installing it with an empty version number,
4506 i.e. as "exim-". This error state is now caught, and the installation is
4509 38. An argument was missing from the function that creates an error message
4510 when Exim fails to connect to the socket for saslauthd authentication.
4511 This could cause Exim to crash, or give a corrupted message.
4513 39. Added isip, isip4, and isip6 to ${if conditions.
4515 40. The ACL variables $acl_xx are now saved with the message, and can be
4516 accessed later in routers, transports, and filters.
4518 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4519 strings in the file are not string-expanded.
4521 42. If a MAIL command specified a SIZE value that was too large to fit into an
4522 int variable, the check against message_size_limit failed. Such values are
4523 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4524 day this will have to be increased, but I don't think I want to be around
4525 when emails are that large.
4532 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4533 iconv() is not standard in FreeBSD.
4535 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4536 IPv6 enabled. The observed symptom was a segmentation fault on return from
4537 the function os_common_find_running_interfaces() in src/os.c.
4539 3. In the check_special_case() function in daemon.c I had used "errno" as an
4540 argument name, which causes warnings on some systems. This was basically a
4541 typo, since it was named "eno" in the comments!
4543 4. The code that waits for the clock to tick (at a resolution of some fraction
4544 of a second) so as to ensure message-id uniqueness was always waiting for
4545 at least one whole tick, when it could have waited for less. [This is
4546 almost certainly not relevant at current processor speeds, where it is
4547 unlikely to ever wait at all. But we try to future-proof.]
4549 5. The function that sleeps for a time interval that includes fractions of a
4550 second contained a race. It did not block SIGALRM between setting the
4551 timer, and suspending (a couple of lines later). If the interval was short
4552 and the sigsuspend() was delayed until after it had expired, the suspension
4553 never ended. On busy systems this could lead to processes getting stuck for
4556 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4557 process, before it forks any delivery processes. The open lookup caching
4558 mechanism meant that the open file or database connection was passed into
4559 the delivery process. The problem was that delivery processes always tidy
4560 up cached lookup data. This could cause a problem for the next delivery
4561 process started by the queue runner, because the external queue runner
4562 process does not know about the closure. So the next delivery process
4563 still has data in the lookup cache. In the case of a file lookup, there was
4564 no problem because closing a file descriptor in a subprocess doesn't affect
4565 the parent. However, if the lookup was caching a connection to a database,
4566 the connection was closed, and the second delivery process was likely to
4567 see errors such as "PGSQL: query failed: server closed the connection
4568 unexpectedly". The problem has been fixed by closing all cached lookups
4569 in a queue runner before running a delivery process.
4571 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4572 seem to have the "const" qualifier which it has on other OS. I've
4575 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4576 *of the same type* (client or server) with the same public name that an
4577 error should be diagnosed.
4579 9. When Exim looked up a host name for an IP address, but failed to find the
4580 original IP address when looking up the host name (a safety check), it
4581 output the message "<ip address> does not match any IP for NULL", which was
4582 confusing, to say the least. The bug was that the host name should have
4583 appeared instead of "NULL".
4585 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4586 user that is built into the binary, and the -C or -D options is used, root
4587 privilege is dropped before the configuration file is read. In addition,
4588 logging is switched to stderr instead of the normal log files. If the
4589 configuration then re-defines the Exim user, the unprivileged environment
4590 is probably not what is expected, so Exim logs a panic warning message (but
4593 However, if deliver_drop_privilege is set, the unprivileged state may well
4594 be exactly what is intended, so the warning has been cut out in that case,
4595 and Exim is allowed to try to write to its normal log files.
4601 1. smtp_return_error_details was not giving details for temporary sender
4602 or receiver verification errors.
4604 2. Diagnose a configuration error if two authenticators have the same public
4607 3. Exim used not to create the message log file for a message until the first
4608 delivery attempt. This could be confusing when incoming messages were held
4609 for policy or load reasons. The message log file is now created at the time
4610 the message is received, and an initial "Received" line is written to it.
4612 4. The automatically generated man page for command line options had a minor
4613 bug that caused no ill effects; however, a more serious problem was that
4614 the procedure for building the man page automatically didn't always
4615 operate. Consequently, release 4.20 contains an out-of-date version. This
4616 shouldn't happen again.
4618 5. When building Exim with embedded Perl support, the script that builds the
4619 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4620 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4622 6. The freeze_tell option was not being used for messages that were frozen on
4623 arrival, either by an ACL or by local_scan().
4625 7. Added the smtp_incomplete_transaction log selector.
4627 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4628 was accepting AUTH without a new EHLO.
4630 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4631 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4634 10. Logging of TCP/IP connections (when configured) now happens in the main
4635 daemon process instead of the child process, so that the TCP/IP connection
4636 count is more accurate (but it can never be perfect).
4638 11. The use of "drop" in a nested ACL was not being handled correctly in the
4639 outer ACL. Now, if condition failure induced by the nested "drop" causes
4640 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4641 or "require"), the connection is dropped.
4643 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4644 that yield "discard" can now be used with an "accept" or a "discard" verb,
4645 but an error is generated for any others (because I can't see a useful way
4646 to define what should happen).
4648 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4649 are now processed in the same way as lines in the Exim configuration file.
4650 In particular, continuation lines are supported.
4652 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4654 15. Added -ti meaning -t -i.
4656 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4657 domains, and warn by logging if others are found.
4659 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4660 returned from GIFCONF when Exim is trying to find the list of interfaces on
4661 a host. The code in os.c has been modified to copy each ifreq to an aligned
4662 structure in all cases.
4664 Also, in some cases, the returned ifreq's were being copied to a 'struct
4665 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4666 means the last couple of bytes of an IPv6 address could be chopped if the
4667 ifreq contained only a normal sockaddr (14 bytes storage).
4669 18. Named domain lists were not supported in the hosts_treat_as_local option.
4670 An entry such as +xxxx was not recognized, and was treated as a literal
4673 19. Ensure that header lines added by a DATA ACL are included in the reject log
4674 if the ACL subsequently rejects the message.
4676 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4677 MD5 (which is deprecated).
4679 21. When testing a filter file using -bf, Exim was writing a message when it
4680 took the sender from a "From " line in the message, but it was not doing so
4681 when it took $return_path from a Return-Path: header line. It now does.
4683 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4684 with a valid header line field name (a series of printing characters
4685 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4687 23. Changed "disc" in the source to "disk" to conform to the documentation and
4688 the book and for uniformity.
4690 24. Ignore Sendmail's -Ooption=value command line item.
4692 25. When execve() failed while trying to run a command in a pipe transport,
4693 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4694 could be confused with a return value of 69 from the command itself. This
4695 has been changed to 127, the value the shell returns if it is asked to run
4696 a non-existent command. The wording for the related log line suggests a
4697 non-existent command as the problem.
4699 26. If received_header_text expands to an empty string, do not add a Received:
4700 header line to the message. (Well, it adds a token one on the spool, but
4701 marks it "old" so that it doesn't get used or transmitted.)
4703 27. Installed eximstats 1.28 (addition of -nt option).
4705 28. There was no check for failure on the call to getsockname() in the daemon
4706 code. This can fail if there is a shortage of resources on the system, with
4707 ENOMEM, for example. A temporary error is now given on failure.
4709 29. Contrary to the C standard, it seems that in some environments, the
4710 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4711 program. Exim now does this explicitly; it affects the formatting of
4712 timestamps using strftime().
4714 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4715 complaints. I've now initialized all the variables, to avoid this.
4717 32. Header lines added by a system filter were not being "seen" during
4718 transport-time rewrites.
4720 33. The info_callback() function passed to OpenSSL is set up with type void
4721 (*)(SSL *, int, int), as described somewhere. However, when calling the
4722 function (actually a macro) that sets it up, the type void(*)() is
4723 expected. I've put in a cast to prevent warnings from picky compilers.
4725 34. If a DNS black list lookup found a CNAME record, but there were no A
4726 records associated with the domain it pointed at, Exim crashed.
4728 35. If a DNS black list lookup returned more than one A record, Exim ignored
4729 all but the first. It now scans all returned addresses if a particular IP
4730 value is being sought. In this situation, the contents of the
4731 $dnslist_value variable are a list of all the addresses, separated by a
4734 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4735 to accept a host name and all its aliases if the forward lookup for any of
4736 them yielded the IP address of the incoming connection. Now it accepts only
4737 those names whose forward lookup yields the correct IP address. Any other
4738 names are discarded. This closes a loophole whereby a rogue DNS
4739 administrator could create reverse DNS records to break through a
4740 wildcarded host restriction in an ACL.
4742 37. If a user filter or a system filter that ran in a subprocess used any of
4743 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4744 the wrong values were passed to the pipe command ($thisaddress had the
4745 value of $0, $0 had the value of $1, etc). This bug was introduced by
4746 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4748 38. Improved the line breaking for long SMTP error messages from ACLs.
4749 Previously, if there was no break point between 40 and 75 characters, Exim
4750 left the rest of the message alone. Two changes have been made: (a) I've
4751 reduced the minimum length to 35 characters; (b) if it can't find a break
4752 point between 35 and 75 characters, it looks ahead and uses the first one
4753 that it finds. This may give the occasional overlong line, but at least the
4754 remaining text gets split now.
4756 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4757 file descriptors might be low, and that setting 1000 would always raise it.
4758 It turns out that in some environments, the limit is already over 1000 and
4759 that lowering it causes trouble. So now Exim takes care not to decrease it.
4761 40. When delivering a message, the value of $return_path is set to $sender_
4762 address at the start of routing (routers may change the value). By an
4763 oversight, this default was not being set up when an address was tested by
4764 -bt or -bv, which affected the outcome if any router or filter referred to
4767 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4768 log only when "message" or "log_message" are set. However, if one of the
4769 conditions was an address verification, or a call to a nested ACL, the
4770 messages generated by the underlying test were being passed through. This
4771 no longer happens. The underlying message is available in $acl_verify_
4772 message for both "message" and "log_message" expansions, so it can be
4773 passed through if needed.
4775 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4776 new expansion $bh_ to give the encoded byte string without charset
4777 translation. Translation happens only if iconv() is available; HAVE_ICONV
4778 indicates this at build time. HEADERS_CHARSET gives the charset to
4779 translate to; headers_charset can change it in the configuration, and
4780 "headers charset" can change it in an individual filter file.
4782 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4783 that creates RFC 2047 encoded "words" labels them as that charset instead
4784 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4785 expansion operator; (ii) when Exim creates a From: line for a local
4786 message; (iii) when a header line is rewritten to include a "phrase" part.
4788 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4789 buggy, causing it to skip the first lines of messages whose message ID
4790 ended in 'D'. This would not have bitten before Exim release 4.14, because
4791 message IDs were unlikely to end in 'D' before then. The effect was to have
4792 incorrect size information for certain domains.
4794 45. #include "config.h" was missing at the start of the crypt16.c module. This
4795 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4798 46. If there was a timeout during a "random" callout check, Exim treated it as
4799 a failure of the random address, and carried on sending RSET and the real
4800 address. If the delay was just some slowness somewhere, the response to the
4801 original RCPT would be taken as a response to RSET and so on, causing
4802 mayhem of various kinds.
4804 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4805 when I implemented it. It didn't allow for the fact that some option values
4806 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4807 the right test for positive values.
4809 48. Domain names in DNS records are case-independent. Exim always looks them up
4810 in lower case. Some resolvers return domain names in exactly the case they
4811 appear in the zone file, that is, they may contain uppercase letters. Not
4812 all resolvers do this - some return always lower case. Exim was treating a
4813 change of case by a resolver as a change of domain, similar to a widening
4814 of a domain abbreviation. This triggered its re-routing code and so it was
4815 trying to route what was effectively the same domain again. This normally
4816 caused routing to fail (because the router wouldn't handle the domain
4817 twice). Now Exim checks for this case specially, and just changes the
4818 casing of the domain that it ultimately uses when it transmits the message
4821 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4824 50. If a filter generated a file delivery with a non-absolute name (possible if
4825 no home directory exists for the router), the forbid_file option was not
4828 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4829 the existing equality matching.
4831 52. Exim was using ints instead of ino_t variables in some places where it was
4832 dealing with inode numbers.
4834 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4835 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4836 called TMPDIR, and if it finds it is different, it changes its value.
4838 54. The smtp_printf() function is now made available to local_scan() so
4839 additional output lines can be written before returning. There is also an
4840 smtp_fflush() function to enable the detection of a dropped connection.
4841 The variables smtp_input and smtp_batched_input are exported to
4844 55. Changed the default runtime configuration: the message "Unknown user"
4845 has been removed from the ACL, and instead placed on the localuser router,
4846 using the cannot_route_message feature. This means that any verification
4847 failures that generate their own messages won't get overridden. Similarly,
4848 the "Unrouteable address" message that was in the ACL for unverifiable
4849 relay addresses has also been removed.
4851 56. Added hosts_avoid_esmtp to the smtp transport.
4853 57. The exicyclog script was not checking for the esoteric option
4854 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4855 will work only if exicyclog is run under the appropriate euid.
4857 58. Following a discussion on the list, the rules by which Exim recognises line
4858 endings on incoming messages have been changed. The -dropcr and drop_cr
4859 options are now no-ops, retained only for backwards compatibility. The
4860 following line terminators are recognized: LF CRLF CR. However, special
4861 processing applies to CR:
4863 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4864 nor a local message in the state where . is a terminator.
4866 (ii) If a bare CR is encountered in a header line, an extra space is added
4867 after the line terminator so as not to end the header. The reasoning
4868 behind this is that bare CRs in header lines are most likely either
4869 to be mistakes, or people trying to play silly games.
4871 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4872 was being incorrectly given as 18 bytes larger than it should have been.
4873 This is a VOB (very old bug).
4875 60. This may never have affected anything current, but just in case it has:
4876 When the local host is found other than at the start of a list of hosts,
4877 the local host, those with the same MX, and any that follow, are discarded.
4878 When the list in question was part of a longer list of hosts, the following
4879 hosts (not currently being processed) were also being discarded. This no
4880 longer happens. I'm not sure if this situation could ever has previously
4883 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4886 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4887 header. This is something that is recommended in a new Internet Draft, and
4888 is something that is documented as being done by Sendmail. There are two
4889 possible values. For messages generated by the autoreply transport, Exim
4892 Auto-Submitted: auto-replied
4894 whereas for all other generated messages (e.g. bounces) it adds
4896 Auto-Submitted: auto-generated
4898 63. The "personal" condition in filters now includes a test for the
4899 Auto-Submitted: header. If it contains the string "auto-" the message it
4900 not considered personal.
4902 64. Added rcpt_include_affixes as a generic transport option.
4904 65. Added queue_only_override (default true).
4906 66. Added the syslog_duplication option.
4908 67. If what should have been the first header line of a message consisted of
4909 a space followed by a colon, Exim was mis-interpreting it as a header line.
4910 It isn't of course - it is syntactically invalid and should therefore be
4911 treated as the start of the message body. The misbehaviour could have
4912 caused a number of strange effects, including loss of data in subsequent
4913 header lines, and spool format errors.
4915 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4916 client host had authenticated. This control can now be exercised by an ACL
4917 for more flexibility.
4919 69. By default, callouts do not happen when testing with -bh. There is now a
4920 variant, -bhc, which does actually run the callout code, including
4921 consulting and updating the callout cache.
4923 70. Added support for saslauthd authentication, courtesy of Alexander
4926 71. If statvfs() failed on the spool or log directories while checking their
4927 size for availability, Exim confusingly gave the error "space shortage".
4928 Furthermore, in debugging mode it crashed with a floating point exception.
4929 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4930 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4931 serious problem, Exim now writes to the main and panic logs when this
4932 happens, with details of the failure. It then refuses to accept the
4933 incoming message, giving the message "spool directory problem" or "log
4934 directory problem" with a 421 code for SMTP messages.
4936 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4937 0, 1, and 2 exist, because some OS complain for execs without them (see
4938 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4939 descriptors. However, the code omitted to check that the open succeeded,
4940 causing mysterious errors if for some reason the permissions on /dev/null
4941 got screwed. Now Exim writes a message to the main and panic logs, and
4942 bombs out if it can't open /dev/null.
4944 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4945 interact so that it is all more flexible. It is supposed to remain
4946 backwards compatible. Also added extra_local_interfaces.
4948 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4949 to bomb out with an assertion failure - to the client this appears as a
4950 connection drop. This problem occurs in the part of the code that was taken
4951 from the Samba project. Fortunately, the assertion is in a very simple
4952 function, so I have fixed this by reproducing the function inline in the
4953 one place where it is called, and arranging for authentication to fail
4954 instead of killing the process with assert().
4956 75. The SPA client code was not working when the server requested OEM rather
4957 than Unicode encoding.
4959 76. Added code to make require_files with a specific uid setting more usable in
4960 the case where statting the file as root fails - usually a non-root-mounted
4961 NFS file system. When this happens and the failure is EACCES, Exim now
4962 forks a subprocess and does the per-uid checking as the relevant uid.
4964 77. Added process_log_path.
4966 78. If log_file_path was not explicitly set, a setting of check_log_space or
4967 check_log_inodes was ignored.
4969 79. If a space check for the spool or log partitions fails, the incident is now
4970 logged. Of course, in the latter case the data may get lost...
4972 80. Added the %p formatting code to string_format() so that it can be used to
4973 print addresses in debug_print(). Adjusted all the address printing in the
4974 debugging in store.c to use %p rather than %d.
4976 81. There was a concern that a line of code in smtp_in.c could overflow a
4977 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4978 initially expressed, the concern was not well-founded, because trailing
4979 spaces are removed early. However, if the trailing spaces were followed by
4980 a NULL, they did not get removed, so the overflow was possible. Two fixes
4983 (a) I re-wrote the offending code in a cleaner fashion.
4984 (b) If an incoming SMTP command contains a NULL character, it is rejected
4987 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4988 runs initgroups(), so that if the Exim user is in any additional groups,
4989 they will be used during message reception.
4995 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob