1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.587 2009/11/05 19:40:51 nm4 Exp $
3 Change log file for Exim from version 4.21
4 -------------------------------------------
9 TK/01 Added patch by Johannes Berg that expands the main option
10 "spamd_servers" if it starts with a dollar sign.
12 TK/02 Write list of recipients to X-Envelope-Sender header when building
13 the mbox-format spool file for content scanning (suggested by Jakob
16 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
17 (http://www.dcc-servers.net/) support via dccifd. Activated by
18 setting EXPERIMENTAL_DCC=yes in Local/Makefile. Check out
19 experimental_spec.txt for more documentation.
21 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
22 by Mark Daniel Reidel <mr@df.eu>.
24 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
25 When building exim an external PCRE library is now needed -
26 PCRE is a system library on the majority of modern systems.
27 See entry on PCRE_LIBS in EDITME file.
29 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
30 conversation. Added nologin parameter to request.
31 Patch contributed by Kirill Miazine
33 TF/01 Do not log submission mode rewrites if they do not change the address.
35 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
37 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
38 log files in place. Contributed by Roberto Lima
40 NM/04 Bugzilla 667: close socket used by dovecot authenticator
42 TF/03 Bugzilla 615: When checking the local_parts router precondition
43 after a local_part_suffix or local_part_prefix option, Exim now
44 does not use the address's named list lookup cache, since this
45 contains cached lookups for the whole local part.
47 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
48 Robert Millan. Documentation is in experimental-spec.txt
50 TF/04 Bugzilla 668: Fix parallel build (make -j).
52 NM/05 Bugzilla 437: Prevent Maildix aux files being created with mode 000
54 NM/05 Bugzilla 598: Improvement to Dovecot authenticator handling.
55 Patch provided by Jan Srzednicki
57 TF/05 Leading white space used to be stripped from $spam_report which
58 wrecked the formatting. Now it is preserved.
60 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
61 that they are available at delivery time.
63 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
65 TF/08 TLS error reporting now respects the incoming_interface and
66 incoming_port log selectors.
68 TF/09 Produce a more useful error message if an SMTP transport's hosts
69 setting expands to an empty string.
71 NM/06 Bugzilla 744: EXPN did not work under TLS.
72 Patch provided by Phil Pennock
74 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
75 Patch provided by Richard Godbee
77 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
78 acl_smtp_notquit, added index entry.
80 NM/09 Bugzilla 787: Potential buffer overflow in string_format
81 Patch provided by Eugene Bujak
83 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to accept()
84 Patch provided by Maxim Dounin
86 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
87 Patch provided by Phil Pennock
89 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
91 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
92 Patch provided by Brad "anomie" Jorsch
94 NM/14 Bugzilla 622: Added timeout setting to transport filter.
95 Patch provided by Dean Brooks
97 TK/05 Add native DKIM support (does not depend on external libraries).
99 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
100 Patch provided by Graeme Fowler
102 NM/16 Bugzilla 851: Documentation example syntax fix.
104 NM/17 Changed NOTICE file to remove references to embedded PCRE.
106 NM/18 Bugzilla 894: Fix issue with very long lines including comments in lsearch
108 NM/19 Bugzilla 745: TLS version reporting
109 Patch provided by Phil Pennock
111 NM/20 Bugzilla 167: bool: condition support
112 Patch provided by Phil Pennock
114 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken clients
115 Patch provided by Phil Pennock
117 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date
118 Patch provided by Brad "anomie" Jorsch
120 NM/23 Bugzilla 687: Fix misparses in eximstats
121 Patch provided by Heiko Schlittermann
123 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid
124 Patch provided by Heiko Schlittermann
126 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file
127 plus update to original patch.
129 NM/26 Bugzilla 799: Documentation correction for ratelimit
131 NM/27 Bugzilla 802: Improvements to local interface IP addr detection
132 Patch provided by David Brownlee
134 NM/28 Bugzilla 807: Improvements to LMTP delivery logging
136 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes
138 NM/30 Bugzilla 888: TLS documentation bugfixes
140 NM/31 Bugzilla 896: Dovecot buffer overrun fix
142 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
143 Unlike the original bugzilla I have changed all shell scripts in src tree
145 NM/33 Bugzilla 898: Transport filter timeout fix
146 Patch by Todd Rinaldo
148 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mistmatches
149 Patch by Serge Demonchaux
155 TK/01 Add preliminary DKIM support. Currently requires a forked version of
156 ALT-N's libdkim that I have put here:
157 http://duncanthrax.net/exim-experimental/
159 Note to Michael Haardt: I had to rename some vars in sieve.c. They
160 were called 'true' and it seems that C99 defines that as a reserved
161 keyword to be used with 'bool' variable types. That means you could
162 not include C99-style headers which use bools without triggering
163 build errors in sieve.c.
165 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
166 as mailq or other aliases. Changed the --help handling significantly
167 to do whats expected. exim_usage() emits usage/help information.
169 SC/01 Added the -bylocaldomain option to eximstats.
171 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
173 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
175 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
181 PH/01 Another patch from the Sieve maintainer.
183 PH/02 When an IPv6 address is converted to a string for single-key lookup
184 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
185 dots are used instead of colons so that keys in lsearch files need not
186 contain colons. This was done some time before quoting was made available
187 in lsearch files. However, iplsearch files do require colons in IPv6 keys
188 (notated using the quote facility) so as to distinguish them from IPv4
189 keys. This meant that lookups for IP addresses in host lists did not work
190 for iplsearch lookups.
192 This has been fixed by arranging for IPv6 addresses to be expressed with
193 colons if the lookup type is iplsearch. This is not incompatible, because
194 previously such lookups could never work.
196 The situation is now rather anomolous, since one *can* have colons in
197 ordinary lsearch keys. However, making the change in all cases is
198 incompatible and would probably break a number of configurations.
200 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
203 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
204 conversion specification without a maximum field width, thereby enabling
205 a rogue spamd server to cause a buffer overflow. While nobody in their
206 right mind would setup Exim to query an untrusted spamd server, an
207 attacker that gains access to a server running spamd could potentially
208 exploit this vulnerability to run arbitrary code as the Exim user.
210 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
211 $primary_hostname instead of what libspf2 thinks the hosts name is.
213 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
214 a directory entry by the name of the lookup key. Previously, if a
215 symlink pointed to a non-existing file or a file in a directory that
216 Exim lacked permissions to read, a lookup for a key matching that
217 symlink would fail. Now it is enough that a matching directory entry
218 exists, symlink or not. (Bugzilla 503.)
220 PH/03 The body_linecount and body_zerocount variables are now exported in the
223 PH/04 Added the $dnslist_matched variable.
225 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
226 This means they are set thereafter only if the connection becomes
229 PH/06 Added the client_condition to authenticators so that some can be skipped
230 by clients under certain conditions.
232 PH/07 The error message for a badly-placed control=no_multiline_responses left
233 "_responses" off the end of the name.
235 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
237 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
238 (without spaces) instead of just copying the configuration text.
240 PH/10 Added the /noupdate option to the ratelimit ACL condition.
242 PH/11 Added $max_received_linelength.
244 PH/12 Added +ignore_defer and +include_defer to host lists.
246 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
247 way in which PCRE > 7.0 is built.
249 PH/14 Implemented queue_only_load_latch.
251 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
252 MAIL command. The effect was to mangle the value on 64-bit systems.
254 PH/16 Another patch from the Sieve maintainer.
256 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
258 PH/18 If a system quota error occurred while trying to create the file for
259 a maildir delivery, the message "Mailbox is full" was not appended to the
260 bounce if the delivery eventually timed out. Change 4.67/27 below applied
261 only to a quota excession during the actual writing of the file.
263 PH/19 It seems that peer DN values may contain newlines (and other non-printing
264 characters?) which causes problems in log lines. The DN values are now
265 passed through string_printing() before being added to log lines.
267 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
268 and InterBase are left for another time.)
270 PH/21 Added message_body_newlines option.
272 PH/22 Guard against possible overflow in moan_check_errorcopy().
274 PH/23 POSIX allows open() to be a macro; guard against that.
276 PH/24 If the recipient of an error message contained an @ in the local part
277 (suitably quoted, of course), incorrect values were put in $domain and
278 $local_part during the evaluation of errors_copy.
284 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
285 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
288 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
289 issue a MAIL command.
291 PH/02 In an ACL statement such as
293 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
295 if a client was not listed at all, or was listed with a value other than
296 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
297 the condition was not true (as it should be), so access was not denied.
298 The bug was that the ! inversion was incorrectly passed on to the second
299 item. This has been fixed.
301 PH/03 Added additional dnslists conditions == and =& which are different from
302 = and & when the dns lookup returns more than one IP address.
304 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
305 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
307 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
308 FSYNC, which compiles an option called disable_fsync that allows for
309 bypassing fsync(). The documentation is heavily laced with warnings.
311 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
313 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
314 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
315 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
316 including adding "make clean"; (3) Added -fPIC when compiling the test
317 dynamically loaded module, to get rid of a warning.
319 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
320 message fails, move_frozen_messages = true and ignore_bounce_errors_after
321 = 0s. The bug is otherwise harmless.
323 PH/07 There was a bug in the dovecot authenticator such that the value of
324 $auth1 could be overwritten, and so not correctly preserved, after a
325 successful authentication. This usually meant that the value preserved by
326 the server_setid option was incorrect.
328 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
330 PH/09 Installed PCRE release 7.0.
332 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
333 run for batched SMTP input. It is now run at the start of every message
334 in the batch. While fixing this I discovered that the process information
335 (output by running exiwhat) was not always getting set for -bs and -bS
336 input. This is fixed, and it now also says "batched" for BSMTP.
338 PH/11 Added control=no_pipelining.
340 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
341 patch, slightly modified), and move the expansion of helo_data till after
342 the connection is made in the smtp transport (so it can use these
345 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
347 PH/14 Added log_selector = +pid.
349 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
351 PH/16 Add ${if forany and ${if forall.
353 PH/17 Added dsn_from option to vary the From: line in DSNs.
355 PH/18 Flush SMTP output before performing a callout, unless control =
356 no_callout_flush is set.
358 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
359 was true (the default) a successful delivery failed to delete the retry
360 item, thus causing premature timeout of the address. The bug is now
363 PH/20 Added hosts_avoid_pipelining to the smtp transport.
365 PH/21 Long custom messages for fakedefer and fakereject are now split up
366 into multiline reponses in the same way that messages for "deny" and
367 other ACL rejections are.
369 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
370 with slight modification.
372 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
373 draft, changing the syntax and factoring some duplicate code".
375 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
376 for deliveries of the second and subsequent messages over the same SMTP
379 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
380 ${reduce, with only minor "tidies".
382 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
384 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
385 expansion side effects.
387 PH/27 When a message times out after an over-quota error from an Exim-imposed
388 quota, the bounce message says "mailbox is full". This message was not
389 being given when it was a system quota that was exceeded. It now should
392 MH/03 Made $recipients available in local_scan(). local_scan() already has
393 better access to the recipient list through recipients_list[], but
394 $recipients can be useful in postmaster-provided expansion strings.
396 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
397 in the case of a MAIL command with additional options following the
398 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
399 were accidentally chopped off.
401 PH/29 SMTP synchronization checks are implemented when a command is read -
402 there is a check that no more input is waiting when there shouldn't be
403 any. However, for some commands, a delay in an ACL can mean that it is
404 some time before the response is written. In this time, more input might
405 arrive, invalidly. So now there are extra checks after an ACL has run for
406 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
407 pipelining has not been advertised.
409 PH/30 MH's patch to allow iscntrl() characters to be list separators.
411 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
412 returned in the SMTP response when smtp_return_error_details was false.
415 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
416 instead of the C I/O that was originally supplied, because problems were
419 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
420 Exim which did not show up earlier: it was assuming that a call to
421 SSL_CTX_set_info_callback() might give an error value. In fact, there is
422 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
423 was a macro that became an assignment, so it seemed to work. This has
424 changed to a proper function call with a void return, hence the compile
425 error. Exim's code has been fixed.
427 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
430 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
432 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
434 PH/37 If a message is not accepted after it has had an id assigned (e.g.
435 because it turns out to be too big or there is a timeout) there is no
436 "Completed" line in the log. When some messages of this type were
437 selected by exigrep, they were listed as "not completed". Others were
438 picked up by some special patterns. I have improved the selection
439 criteria to be more general.
441 PH/38 The host_find_failed option in the manualroute router can now be set
442 to "ignore", to completely ignore a host whose IP address cannot be
443 found. If all hosts are ignored, the behaviour is controlled by the new
444 host_all_ignored option.
446 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
447 homing or because of multiple MX records with /mx) generated more than
448 one IP address, and the following item turned out to be the local host,
449 all the secondary addresses of the first item were incorrectly removed
450 from the list, along with the local host and any following hosts (which
451 is what is supposed to happen).
453 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
454 whoever called Exim into the -H file. In the case of the daemon it was
455 behaving confusingly. When first started, it used values for whoever
456 started the daemon, but after a SIGHUP it used the Exim user (because it
457 calls itself on a restart). I have changed the code so that it now always
460 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
461 message are rejected with the same error (e.g. no authentication or bad
462 sender address), and a DATA command is nevertheless sent (as can happen
463 with PIPELINING or a stupid MUA), the error message that was given to the
464 RCPT commands is included in the rejection of the DATA command. This is
465 intended to be helpful for MUAs that show only the final error to their
468 PH/42 Another patch from the Sieve maintainer.
470 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
471 Eximstats - Fixed some broken HTML links and added missing column headers
473 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
474 columns for Rejects, Temp Rejects, Ham, and Spam rows.
476 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
478 PH/43 Yet another patch from the Sieve maintainer.
480 PH/44 I found a way to check for a TCP/IP connection going away before sending
481 the response to the final '.' that terminates a message, but only in the
482 case where the client has not sent further data following the '.'
483 (unfortunately, this is allowed). However, in many cases there won't be
484 any further data because there won't be any more messages to send. A call
485 to select() can be used: if it shows that the input is "ready", there is
486 either input waiting, or the socket has been closed. An attempt to read
487 the next input character can distinguish the two cases. Previously, Exim
488 would have sent an OK response which the client would never have see.
489 This could lead to message repetition. This fix should cure that, at
490 least in a lot of common cases.
492 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
493 advertised in response to EHLO.
499 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
500 fixed by 4.65/MH/01 (is this a record?) are fixed:
502 (i) An empty string was always treated as zero by the numeric comparison
503 operators. This behaviour has been restored.
505 (ii) It is documented that the numeric comparison operators always treat
506 their arguments as decimal numbers. This was broken in that numbers
507 starting with 0 were being interpreted as octal.
509 While fixing these problems I realized that there was another issue that
510 hadn't been noticed. Values of message_size_limit (both the global option
511 and the transport option) were treated as octal if they started with 0.
512 The documentation was vague. These values are now always treated as
513 decimal, and I will make that clear in the documentation.
519 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
520 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
523 MH/01 Don't check that the operands of numeric comparison operators are
524 integers when their expansion is in "skipping" mode (fixes bug
525 introduced by 4.64-PH/07).
527 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
528 child addresses, Exim now panics and dies. Previously, because the count
529 is held in a short int, deliveries were likely to be lost. As such a
530 large number of recipients for a single message is ridiculous
531 (performance will be very, very poor), I have chosen to impose a limit
532 rather than extend the field.
538 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
539 leftover -K file (the existence of which was triggered by #402).
540 While we were at it, introduced process PID as part of the -K
541 filename. This should rule out race conditions when creating
544 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
545 processing considerably. Previous code took too long for large mails,
546 triggering a timeout which in turn triggers #401.
548 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
549 in the DK code in transports.c. sendfile() is not really portable,
550 hence the _LINUX specificness.
552 TF/01 In the add_headers option to the mail command in an Exim filter,
553 there was a bug that Exim would claim a syntax error in any
554 header after the first one which had an odd number of characters
557 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
558 callout verification, Exim cached a "reject" for the entire domain. This
559 is correct for most verifications, but it is not correct for a recipient
560 verification with use_sender or use_postmaster set, because in that case
561 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
562 case of MAIL FROM:<> rejection from other early rejections (e.g.
563 rejection of HELO). When verifying a recipient using a non-null MAIL
564 address, the cache is ignored if it shows MAIL FROM:<> rejection.
565 Whatever the result of the callout, the value of the domain cache is
566 left unchanged (for any other kind of callout, getting as far as trying
567 RCPT means that the domain itself is ok).
569 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
572 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
573 manifest itself as EPIPE rather than ECONNECT. When tidying away a
574 session, the daemon ignores ECONNECT errors and logs others; it now
575 ignores EPIPE as well.
577 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
578 (quoted-printable decoding).
580 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
581 later the small subsequent patch to fix an introduced bug.
583 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
585 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
587 PH/08 An error is now given if message_size_limit is specified negative.
589 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
590 to be given (somewhat) arbitrary names.
592 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
595 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
596 miscellaneous code fixes
598 PH/10 Added the log_reject_target ACL modifier to specify where to log
601 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
602 hostname. This is wrong, because it relates to the incoming message (and
603 probably the interface on which it is arriving) and not to the outgoing
604 callout (which could be using a different interface). This has been
605 changed to use the value of the helo_data option from the smtp transport
606 instead - this is what is used when a message is actually being sent. If
607 there is no remote transport (possible with a router that sets up host
608 addresses), $smtp_active_hostname is used.
610 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
611 tweaks were necessary in order to get it to work (see also 21 below):
612 (a) The code assumed that strncpy() returns a negative number on buffer
613 overflow, which isn't the case. Replaced with Exim's string_format()
615 (b) There were several signed/unsigned issues. I just did the minimum
616 hacking in of casts. There is scope for a larger refactoring.
617 (c) The code used strcasecmp() which is not a standard C function.
618 Replaced with Exim's strcmpic() function.
619 (d) The code set only $1; it now sets $auth1 as well.
620 (e) A simple test gave the error "authentication client didn't specify
621 service in request". It would seem that Dovecot has changed its
622 interface. Fortunately there's a specification; I followed it and
623 changed what the client sends and it appears to be working now.
625 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
628 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
629 address A is aliased to B and C, where B exists and C does not. Without
630 -v the output is "A verified" because verification stops after a
631 successful redirection if more than one address is generated. However,
632 with -v the child addresses are also verified. Exim was outputting "A
633 failed to verify" and then showing the successful verification for C,
634 with its parentage. It now outputs "B failed to verify", showing B's
635 parentage before showing the successful verification of C.
637 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
638 look up a TXT record in a specific list after matching in a combined
641 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
642 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
643 they consult the DNS. I had assumed they would set it the way they
644 wanted; and indeed my experiments on Linux seem to show that in some
645 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
646 To be on the safe side, however, I have now made the interface to
647 host_find_byname() similar to host_find_bydns(), with an argument
648 containing the DNS resolver options. The host_find_byname() function now
649 sets these options at its start, just as host_find_bydns() does. The smtp
650 transport options dns_qualify_single and dns_search_parents are passed to
651 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
652 of host_find_byname() use the default settings of RES_DEFNAMES
653 (qualify_single) but not RES_DNSRCH (search_parents).
655 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
656 spool_read_header() do less string testing, by means of a preliminary
657 switch on the second character of optional "-foo" lines. (This is
658 overdue, caused by the large number of possibilities that now exist.
659 Originally there were few.) While I was there, I also converted the
660 str(n)cmp tests so they don't re-test the leading "-" and the first
661 character, in the hope this might squeeze out yet more improvement.
663 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
664 flag allowing group syntax was set by the header_syntax check but not
665 turned off, possible causing trouble later; (2) The flag was not being
666 set at all for the header_verify test, causing "group"-style headers to
667 be rejected. I have now set it in this case, and also caused header_
668 verify to ignore an empty address taken from a group. While doing this, I
669 came across some other cases where the code for allowing group syntax
670 while scanning a header line wasn't quite right (mostly, not resetting
671 the flag correctly in the right place). These bugs could have caused
672 trouble for malformed header lines. I hope it is now all correct.
674 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
675 with the "reply" argument non-NULL. The code, however (which originally
676 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
677 but it didn't always do it. This confused somebody who was copying the
678 code for some other use. I have removed all the tests.
680 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
681 feature that was used to support insecure browsers during the U.S. crypto
682 embargo. It requires special client support, and Exim is probably the
683 only MTA that supported it -- and would never use it because real RSA is
684 always available. This code has been removed, because it had the bad
685 effect of slowing Exim down by computing (never used) parameters for the
686 RSA_EXPORT functionality.
688 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
689 authenticator to fail if there's a tab character in the incoming data
690 (there should never be unless someone is messing about, as it's supposed
691 to be base64-encoded). Also added, on Timo's advice, the "secured" option
692 if the connection is using TLS or if the remote IP is the same as the
693 local IP, and the "valid-client-cert option" if a client certificate has
696 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
697 authenticators. This can be used for authorization after authentication
698 succeeds. (In the case of plaintext, it servers for both authentication
701 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
702 if any retry times were supplied.
704 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
705 connection, where there is no client IP address to check. In this
706 situation, the verify now always succeeds.
708 PH/25 Applied John Jetmore's -Mset patch.
710 PH/26 Added -bem to be like -Mset, but loading a message from a file.
712 PH/27 In a string expansion for a processed (not raw) header when multiple
713 headers of the same name were present, leading whitespace was being
714 removed from all of them, but trailing whitespace was being removed only
715 from the last one. Now trailing whitespace is removed from each header
716 before concatenation. Completely empty headers in a concatenation (as
719 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
720 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
722 PH/29 [Removed. This was a change that I later backed out, and forgot to
723 correct the ChangeLog entry (that I had efficiently created) before
724 committing the later change.]
726 PH/30 Exim was sometimes attempting to deliver messages that had suffered
727 address errors (4xx response to RCPT) over the same connection as other
728 messages routed to the same hosts. Such deliveries are always "forced",
729 so retry times are not inspected. This resulted in far too many retries
730 for the affected addresses. The effect occurred only when there were more
731 hosts than the hosts_max_try setting in the smtp transport when it had
732 the 4xx errors. Those hosts that it had tried were not added to the list
733 of hosts for which the message was waiting, so if all were tried, there
734 was no problem. Two fixes have been applied:
736 (i) If there are any address or message errors in an SMTP delivery, none
737 of the hosts (tried or untried) are now added to the list of hosts
738 for which the message is waiting, so the message should not be a
739 candidate for sending over the same connection that was used for a
740 successful delivery of some other message. This seems entirely
741 reasonable: after all the message is NOT "waiting for some host".
742 This is so "obvious" that I'm not sure why it wasn't done
743 previously. Hope I haven't missed anything, but it can't do any
744 harm, as the worst effect is to miss an optimization.
746 (ii) If, despite (i), such a delivery is accidentally attempted, the
747 routing retry time is respected, so at least it doesn't keep
748 hammering the server.
750 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
751 in ${readsocket because some servers need this prod.
753 PH/32 Added some extra debug output when updating a wait-xxx database.
755 PH/33 The hint "could be header name not terminated by colon", which has been
756 given for certain expansion errors for a long time, was not being given
757 for the ${if def:h_colon_omitted{... case.
759 PH/34 The spec says: "With one important exception, whenever a domain list is
760 being scanned, $domain contains the subject domain." There was at least
761 one case where this was not true.
763 PH/35 The error "getsockname() failed: connection reset by peer" was being
764 written to the panic log as well as the main log, but it isn't really
765 panic-worthy as it just means the connection died rather early on. I have
766 removed the panic log writing for the ECONNRESET error when getsockname()
769 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
770 runs only) independently of the message's sender address. This meant
771 that, if the 4xx error was in fact related to the sender, a different
772 message to the same recipient with a different sender could confuse
773 things. In particualar, this can happen when sending to a greylisting
774 server, but other circumstances could also provoke similar problems.
775 I have changed the default so that the retry time for these errors is now
776 based a combination of the sender and recipient addresses. This change
777 can be overridden by setting address_retry_include_sender=false in the
780 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
781 remote server are returned as part of bounce messages. This was not
782 happening for LMTP over a pipe (the lmtp transport), but now it is the
783 same for both kinds of LMTP.
785 PH/38 Despite being documented as not happening, Exim was rewriting addresses
786 in header lines that were in fact CNAMEs. This is no longer the case.
788 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
789 and queue runs started by the daemon processed all messages. This has
790 been fixed so that -R and -S can now usefully be given with -q<time>.
792 PH/40 Import PCRE release 6.7 (fixes some bugs).
794 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
796 PH/42 Give an error if -q is specified more than once.
798 PH/43 Renamed the variables $interface_address and $interface_port as
799 $received_ip_address and $received_port, to make it clear that these
800 values apply to message reception, and not to the outgoing interface when
801 a message is delivered. (The old names remain recognized, of course.)
803 PH/44 There was no timeout on the connect() call when using a Unix domain
804 socket in the ${readsocket expansion. There now is.
806 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
807 be meaningful with "accept".
809 SC/01 Eximstats V1.43
810 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
812 SC/02 Eximstats V1.44
813 Use a glob alias rather than an array ref in the generated
814 parser. This improves both readability and performance.
816 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
817 Collect SpamAssassin and rejection statistics.
818 Don't display local sender or destination tables unless
819 there is data to show.
820 Added average volumes into the top table text output.
822 SC/04 Eximstats V1.46
823 Collect data on the number of addresses (recipients)
824 as well as the number of messages.
826 SC/05 Eximstats V1.47
827 Added 'Message too big' to the list of mail rejection
828 reasons (thanks to Marco Gaiarin).
830 SC/06 Eximstats V1.48
831 Mainlog lines which have GMT offsets and are too short to
832 have a flag are now skipped.
834 SC/07 Eximstats V1.49 (Alain Williams)
835 Added the -emptyok flag.
837 SC/08 Eximstats V1.50
838 Fixes for obtaining the IP address from reject messages.
840 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
841 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
842 whitesspace changes from 4.64-PH/27
844 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
847 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
848 are found, allow negative numbers in numeric criteria)
850 JJ/06 exipick.20061117.2, added new $message_body_missing variable
852 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
853 to match changes made in 4.64-PH/43
855 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
857 PH/47 Put in an explicit test for a DNS lookup of an address record where the
858 "domain" is actually an IP address, and force a failure. This locks out
859 those revolvers/nameservers that support "A-for-A" lookups, in
860 contravention of the specifications.
862 PH/48 When a host name was looked up from an IP address, and the subsequent
863 forward lookup of the name timed out, the host name was left in
864 $sender_host_name, contrary to the specification.
866 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
867 restricted to single-key lookups, Exim was not diagnosing an error if
868 * or *@ was used with a query-style lookup.
870 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
872 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
873 long ago, but noone interested enough thought of it. Let's just say that
874 the "1.1" means that there are some new functions that weren't there at
875 some point in the past.
877 PH/51 Error processing for expansion failure of helo_data from an smtp
878 transport during callout processing was broken.
880 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
881 tested/used via the -bh/-bhc/-bs options.
883 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
884 bug, fixed in subsequent PCRE releases).
886 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
887 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
889 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
895 SC/01 Use a glob alias rather than an array ref in eximstats generated
896 parser. This improves both readability and performance.
898 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
899 Don't display local sender or destination tables in eximstats unless
900 there is data to show.
901 Added average volumes into the eximstats top table text output.
903 SC/03 Collect data on the number of addresses (recipients) as well
904 as the number of messages in eximstats.
906 TF/01 Correct an error in the documentation for the redirect router. Exim
907 does (usually) call initgroups() when daemonizing.
909 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
910 with consistent privilege compared to when running as a daemon.
912 TF/03 Note in the spec that $authenticated_id is not set for local
913 submissions from trusted users.
915 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
916 Thanks to Dean Brooks <dean@iglou.com> for the patch.
918 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
919 by adding some example configuration directives to the default
920 configuration file. A little bit of work is required to uncomment the
921 directives and define how usernames and passwords are checked, but
922 there is now a framework to start from.
924 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
925 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
926 without this. I don't know how relevant this is to other LDAP libraries.
928 PH/02 Add the verb name to the "unknown ACL verb" error.
930 PH/03 Magnus Holmgren's patch for filter_prepend_home.
932 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
934 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
935 directory not expanded when it should be if an expanded home directory
936 was set for the address (which is overridden by the transport).
938 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
941 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
942 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
943 because it is too late at that time, and has no effect.
945 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
946 security issue with \' (bugzilla #107). I could not use the
947 PQescapeStringConn() function, because it needs a PGconn value as one of
950 PH/08 When testing addresses using -bt, indicate those final addresses that
951 are duplicates that would not cause an additional delivery. At least one
952 person was confused, thinking that -bt output corresponded to deliveries.
953 (Suppressing duplicates isn't a good idea as you lose the information
954 about possibly different redirections that led to the duplicates.)
956 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
957 systems where poll() doesn't work, in particular OS X.
959 PH/10 Added more information to debugging output for retry time not reached.
961 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
962 operations in malware.c.
964 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
967 PH/13 If write_rejectlog was set false when logging was sent to syslog with
968 syslog_duplication set false, log lines that would normally be written
969 both the the main log and to the reject log were not written to syslog at
972 PH/14 In the default configuration, change the use of "message" in ACL warn
973 statements to "add_header".
975 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
976 not followed by a command (e.g. "seen endif").
978 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
979 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
982 PH/17 Added extra conditions to the default value of delay_warning_condition
986 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
987 { match{$h_precedence:}{(?i)bulk|list|junk} } \
988 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
991 The Auto-Submitted: and various List- headers are standardised, whereas I
992 don't think Precedence: ever was.
994 PH/18 Refactored debugging code in route_finduser() to show more information,
995 in particular, the error code if getpwnam() issues one.
997 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
998 This is apparently needed in addition to the PH/07 change above to avoid
999 any possible encoding problems.
1001 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
1002 but not after initializing Perl.
1004 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
1005 output them only if debugging. By default they are written stderr,
1006 apparently, which is not desirable.
1008 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
1011 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
1014 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
1016 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
1017 authenticated or an ident call has been made. Suppress the default
1018 values for $authenticated_id and $authenticated_sender (but permit -oMai
1019 and -oMas) when testing with -bh.
1021 PH/24 Re-jigged the order of the tests in the default configuration so that the
1022 tests for valid domains and recipients precede the DNS black list and CSA
1023 tests, on the grounds that those ones are more expensive.
1025 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
1026 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
1027 This bug exists in every version of Exim that I still have, right back to
1030 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
1031 However, an attempt to turn on case-sensitivity in a regex key by
1032 including (?-i) didn't work because the subject string was already
1033 lowercased, and the effects were non-intuitive. It turns out that a
1034 one-line patch can be used to allow (?-i) to work as expected.
1040 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1041 other effects) broke the use of negated acl sub-conditions.
1043 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1046 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1047 "Deny" causes Exim to reject the incoming connection with a 554 error.
1048 Unfortunately, if there is a major crisis, such as a disk failure,
1049 tcp-wrappers gives "deny", whereas what one would like would be some
1050 kind of temporary error. A kludge has been added to help with this.
1051 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1052 554 error is used if errno is still zero or contains ENOENT (which occurs
1053 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1056 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1058 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1059 errors. Otherwise a message that provokes a temporary error (when other
1060 messages do not) can cause a whole host to time out.
1062 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1063 addresses were routed directly to files or pipes from a redirect router.
1064 File deliveries just didn't batch; pipe deliveries might have suffered
1067 PH/06 A failure to get a lock for a hints database would erroneously always say
1068 "Failed to get write lock", even when it was really a read lock.
1070 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1071 of 0600. This has been changed to use the value of the lockfile_mode
1072 option (which defaults to 0600).
1074 PH/08 Applied small patch from the Sieve maintainer.
1076 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1077 folder from quota calculations, a direct delivery into this folder messed
1078 up the contents of the maildirsize file. This was because the regex was
1079 used only to exclude .Trash (or whatever) when the size of the mailbox
1080 was calculated. There was no check that a delivery was happening into an
1081 excluded directory. This bug has been fixed by ignoring all quota
1082 processing for deliveries into excluded directories.
1084 PH/10 Added the maildirfolder_create_regex option to appendfile.
1090 PH/01 The code for finding all the local interface addresses on a FreeBSD
1091 system running IPv6 was broken. This may well have applied to all BSD
1092 systems, as well as to others that have similar system calls. The broken
1093 code found IPv4 interfaces correctly, but gave incorrect values for the
1094 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1095 that it would not match correctly against @[] and not recognize the IPv6
1098 PH/02 The ipliteral router was not recognizing addresses of the form user@
1099 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1101 PH/03 Added disable_ipv6.
1103 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1104 decoded form, because it is most often used to construct To: headers
1105 lines in autoreplies, and the decoded form may well be syntactically
1106 invalid. However, $reply_address has leading white space removed, and all
1107 newlines turned into spaces so that the autoreply transport does not
1110 PH/05 If group was specified without a user on a router, and no group or user
1111 was specified on a transport, the group from the router was ignored.
1113 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1114 for visible compile-time settings that can be used to change these
1115 numbers, for those that want even more. Backwards compatibility with old
1116 spool files has been maintained. However, going back to a previous Exim
1117 release will lost any variables that are in spool files.
1119 PH/07 Two small changes when running in the test harness: increase delay when
1120 passing a TCP/IP connection to a new process, in case the original
1121 process has to generate a bounce, and remove special handling of
1122 127.0.0.2 (sic), which is no longer necessary.
1124 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1125 be the same on different OS.
1127 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1130 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1131 whether --show-vars was specified or not
1133 JJ/02 exipick: Added support for new ACL variable spool format introduced
1136 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1137 syntactically invalid From: or Reply-to: line, and a filter used this to
1138 generate an autoreply, and therefore failed to obtain an address for the
1139 autoreply, Exim could try to deliver to a non-existent relative file
1140 name, causing unrelated and misleading errors. What now happens is that
1141 it logs this as a hard delivery error, but does not attempt to create a
1144 PH/11 The exinext utility has a -C option for testing purposes, but although
1145 the given file was scanned by exinext itself; it wasn't being passed on
1146 when Exim was called.
1148 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1149 an end-of-file indication when reading a command response.
1151 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1152 compiled. In many other places in Exim, IPv6 addresses are always
1153 recognized, so I have changed this. It also means that IPv4 domain
1154 literals of the form [IPV4:n.n.n.n] are now always recognized.
1156 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1157 used if the router is not running as root, for example, when verifying at
1158 ACL time, or when using -bh. The debugging output from this situation was
1159 non-existent - all you got was a failure to exec. I have made two
1162 (a) Failures to set uid/gid, the current directory, or a process leader
1163 in a subprocess such as that created by queryprogram now generate
1164 suitable debugging ouput when -d is set.
1166 (b) The queryprogram router detects when it is not running as root,
1167 outputs suitable debugging information if -d is set, and then runs
1168 the subprocess without attempting to change uid/gid.
1170 PH/15 Minor change to Makefile for building test_host (undocumented testing
1173 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1174 additional section of a DNS packet that returns MX or SRV records.
1175 Instead, it always explicitly searches for A/AAAA records. This avoids
1176 major problems that occur when a DNS server includes only records of one
1177 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1178 fixed another bug: if SRV records were looked up and the corresponding
1179 address records were *not* found in the additional section, the port
1180 values from the SRV records were lost.
1182 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1183 using the correct key (the original address) when searching the retry
1184 rules in order to find which one to use for generating the retry hint.
1186 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1187 adding the default one. Similarly, if it contains a Reply-To: header, the
1188 errors_reply_to option, if set, is not used.
1190 PH/19 When calculating a retry time, Exim used to measure the "time since
1191 failure" by looking at the "first failed" field in the retry record. Now
1192 it does not use this if it is later than than the arrival time of the
1193 message. Instead it uses the arrival time. This makes for better
1194 behaviour in cases where some deliveries succeed, thus re-setting the
1195 "first failed" field. An example is a quota failure for a huge message
1196 when small messages continue to be delivered. Without this change, the
1197 "time since failure" will always be short, possible causing more frequent
1198 delivery attempts for the huge message than are intended.
1199 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1201 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1202 $1, $2, $3) because the numerical variables can be reset during some
1203 expansion items (e.g. "match"), thereby losing the authentication data.
1205 PH/21 Make -bV show the size of off_t variables so that the test suite can
1206 decide whether to run tests for quotas > 2G.
1208 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1209 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1210 filecount value is greater than 2G or if a quota value is greater than 2G
1211 on a system where the size of off_t is not greater than 4, a panic error
1214 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1215 never match. The debug and -bh output now contains an explicit error
1216 message indicating a malformed IPv4 address or mask.
1218 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1219 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1220 PH/23 above applies.
1222 PH/25 Do not write to syslog when running in the test harness. The only
1223 occasion when this arises is a failure to open the main or panic logs
1224 (for which there is an explicit test).
1226 PH/26 Added the /no_tell option to "control=freeze".
1228 PH/27 If a host name lookup failed very early in a connection, for example, if
1229 the IP address matched host_lookup and the reverse lookup yielded a name
1230 that did not have a forward lookup, an error message of the form "no IP
1231 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1232 could be logged. Now it outputs the IP address instead of "NULL".
1234 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1235 allows the sender and the authenticated sender to be set when
1236 submitting a message from within Exim. Since child_open_exim() is
1237 documented for local_scan(), the new function should be too.
1239 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1240 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1241 results in an empty string is now treated as unset.
1243 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1245 PH/31 Added sender_verify_fail logging option.
1247 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1248 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1249 tidied the source and removed it altogether.
1251 PH/33 When a queue run was abandoned because the load average was too high, a
1252 log line was always written; now it is written only if the queue_run log
1253 selector is set. In addition, the log line for abandonment now contains
1254 information about the queue run such as the pid. This is always present
1255 in "start" and "stop" lines but was omitted from the "abandon" line.
1257 PH/34 Omit spaces between a header name and the colon in the error message that
1258 is given when verify = headers_syntax fails (if there are lots of them,
1259 the message gets confusing).
1261 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1262 names, as there are now some PTR records that contain slashes. This check
1263 is only to protect against broken name servers that fall over on strange
1264 characters, so the fact that it applies to all lookups doesn't matter.
1266 PH/36 Now that the new test suite is complete, we can remove some of the
1267 special code in Exim that was needed for the old test suite. For example,
1268 sorting DNS records because real resolvers return them in an arbitrary
1269 order. The new test suite's fake resolver always returns records in the
1272 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1273 bounces) except when queue_only is set, to avoid logging races between
1274 the different processes.
1276 PH/38 Panic-die if .include specifies a non-absolute path.
1278 PH/39 A tweak to the "H" retry rule from its user.
1280 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1281 a label. They prevented compilation on older perls.
1283 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1284 a warning to be raised on newish perls.
1286 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1287 on queue. Changes to match documented behaviour of showing count of
1288 messages matching specified criteria.
1290 PH/40 Changed the default ident timeout from 30s to 5s.
1292 PH/41 Added support for the use of login_cap features, on those BSD systems
1293 that have them, for controlling the resources used by pipe deliveries.
1295 PH/42 The content-scanning code uses fopen() to create files in which to put
1296 message data. Previously it was not paying any attention to the mode of
1297 the files. Exim runs with umask(0) because the rest of the code creates
1298 files with open(), and sets the required mode explicitly. Thus, these
1299 files were ending up world-writeable. This was not a big issue, because,
1300 being within the spool directory, they were not world-accessible. I have
1301 created a function called modefopen, which takes an additional mode
1302 argument. It sets umask(777), creates the file, chmods it to the required
1303 mode, then resets the umask. All the relevant calls to fopen() in the
1304 content scanning code have been changed to use this function.
1306 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1307 to 24 hours. This avoids potential overflow problems when processing G
1308 and H retry rules. I suspect nobody ever tinkers with this value.
1310 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1312 PH/45 When the plaintext authenticator is running as a client, the server's
1313 challenges are checked to ensure they are valid base64 strings. By
1314 default, the authentication attempt is cancelled if an invalid string is
1315 received. Setting client_ignore_invalid_base64 true ignores these errors.
1316 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1317 they are received. Thus, the responses can be made to depend on the
1318 challenges. If an invalid string is ignored, an empty string is placed in
1321 PH/46 Messages that are created by the autoreply transport now contains a
1322 References: header, in accordance with RFCs 2822 and 3834.
1324 PH/47 Added authenticated_sender_force to the smtp transport.
1326 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1328 PH/49 Installed latest patch from the Sieve maintainer.
1330 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1331 was also set, the appendfile transport was unnecessarily scanning a
1332 directory of message files (e.g. for maildir delivery) to find the count
1333 of files (along with the size), even though it did not need this
1334 information. It now does the scan only if it needs to find either the
1335 size of the count of files.
1337 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1339 PH/52 Two bugs concerned with error handling when the smtp transport is
1342 (i) Exim was not creating retry information for temporary errors given
1343 for individual recipients after the DATA command when the smtp transport
1344 was used in LMTP mode. This meant that they could be retried too
1345 frequently, and not timed out correctly.
1347 (ii) Exim was setting the flag that allows error details to be returned
1348 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1349 recipients that were returned after the DATA command.
1351 PH/53 This is related to PH/52, but is more general: for any failing address,
1352 when detailed error information was permitted to be returned to the
1353 sender, but the error was temporary, then after the final timeout, only
1354 "retry timeout exceeded" was returned. Now it returns the full error as
1355 well as "retry timeout exceeded".
1357 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1358 do this, and (what is worse) MTAs that accept it.
1360 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1361 will now be deprecated.
1363 PH/56 New os.c-cygwin from the Cygwin maintainer.
1365 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1366 formats (previously only available in exim formats via -bpr, -bpru,
1367 and -bpra. Now also available in native and exiqgrep formats)
1369 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1370 with very large, slow to parse queues
1372 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1374 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1376 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1377 responses to authentication challenges, though it was showing the
1378 challenges; (ii) I've removed the CR characters from the debug output for
1381 PH/58 Allow for the insertion of a newline as well as a space when a string
1382 is turned into more than one encoded-word during RFC 2047 encoding. The
1383 Sieve code now uses this.
1385 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1386 data_4xx, lost_connection, tls_required.
1388 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1389 message was being sent as an SMTP response.
1391 PH/61 Add -l and -k options to exicyclog.
1393 PH/62 When verifying, if an address was redirected to one new address, so that
1394 verification continued, and the new address failed or deferred after
1395 having set something in $address_data, the value of $address_data was not
1396 passed back to the ACL. This was different to the case when no
1397 redirection occurred. The value is now passed back in both cases.
1399 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1400 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1401 use login_cap.h, so on its own it isn't the distinguishing feature. The
1402 new name refers directly to the setclassresources() function.
1404 PH/65 Added configuration files for NetBSD3.
1406 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1408 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1409 is preferred over IPv4.
1411 PH/68 The bounce_return_message and bounce_return_body options were not being
1412 honoured for bounces generated during the reception of non-SMTP messages.
1413 In particular, this applied to messages rejected by the ACL. This bug has
1414 been fixed. However, if bounce_return_message is true and bounce_return_
1415 body is false, the headers that are returned for a non-SMTP message
1416 include only those that have been read before the error was detected.
1417 (In the case of an ACL rejection, they have all been read.)
1419 PH/69 The HTML version of the specification is now built in a directory called
1420 spec_html instead of spec.html, because the latter looks like a path with
1421 a MIME-type, and this confuses some software.
1423 PH/70 Catch two compiler warnings in sieve.c.
1425 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1426 function verify_get_ident() calls ip_connect() to connect a socket, but
1427 if the "connect()" function timed out, ip_connect() used to close the
1428 socket. However, verify_get_ident() also closes the socket later, and in
1429 between Exim writes to the log, which may get opened at this point. When
1430 the socket was closed in ip_connect(), the log could get the same file
1431 descriptor number as the socket. This naturally causes chaos. The fix is
1432 not to close the socket in ip_connect(); the socket should be closed by
1433 the function that creates it. There was only one place in the code where
1434 this was missing, in the iplookup router, which I don't think anybody now
1435 uses, but I've fixed it anyway.
1437 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1438 well as to direct DNS lookups. Otherwise the handling of names in host
1439 lists is inconsistent and therefore confusing.
1445 PH/01 Two changes to the default runtime configuration:
1447 (1) Move the checks for relay_from_hosts and authenticated clients from
1448 after to before the (commented out) DNS black list checks.
1450 (2) Add control=submission to the relay_from_hosts and authenticated
1451 clients checks, on the grounds that messages accepted by these
1452 statements are most likely to be submissions.
1454 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1456 (1) Generate an error if the third argument for the ${prvs expansion is
1459 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1462 (3) Reset the variables that are obtained from the first argument of
1463 ${prvscheck and used in the second argument before leaving the code,
1464 because their memory is reclaimed, so using them afterwards may do
1467 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1468 one (it's much easier than Tom thought :-).
1470 (5) Because of (4), we can now allow for the use of $prvscheck_result
1471 inside the third argument.
1473 PH/03 For some reason, the default setting of PATH when running a command from
1474 a pipe transport was just "/usr/bin". I have changed it to
1477 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1478 anything to be listed in the output from -bV.
1480 PH/05 When a filter generated an autoreply, the entire To: header line was
1481 quoted in the delivery log line, like this:
1483 => >A.N.Other <ano@some.domain> <original@ddress> ...
1485 This has been changed so that it extracts the operative address. There
1486 may be more than one such address. If so, they are comma-separated, like
1489 => >ano@some.domain,ona@other.domain <original@ddress> ...
1491 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1492 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1493 not being looked up in the rDNS to get a host name, Exim was showing the
1494 IP address twice in Received: lines, even though the IP addresses were
1495 identical. For example:
1497 Received: from [1.2.3.4] (helo=[1.2.3.4])
1499 However, if the real host name was known, it was omitting the HELO data
1500 if it matched the actual IP address. This has been tidied up so that it
1501 doesn't show the same IP address twice.
1503 PH/07 When both +timestamp and +memory debugging was on, the value given by
1504 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1505 called by the debug printing, thereby overwriting the timestamp buffer.
1506 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1508 PH/08 When the original message was included in an autoreply transport, it
1509 always said "this is a copy of the message, including all the headers",
1510 even if body_only or headers_only was set. It now gives an appropriate
1513 PH/09 Applied a patch from the Sieve maintainer which:
1515 o fixes some comments
1516 o adds the (disabled) notify extension core
1517 o adds some debug output for the result of if/elsif tests
1518 o points to the current vacation draft in the documentation
1519 and documents the missing references header update
1523 o fixes a bug in processing the envelope test (when testing
1524 multiple envelope elements, the last element determinted the
1527 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1528 Electronic Mail") by including:
1530 Auto-submitted: auto-generated
1532 in the messages that it generates (bounce messages and others, such as
1533 warnings). In the case of bounce messages for non-SMTP mesages, there was
1534 also a typo: it was using "Auto_submitted" (underscore instead of
1535 hyphen). Since every message generated by Exim is necessarily in response
1536 to another message, thes have all been changed to:
1538 Auto-Submitted: auto-replied
1540 in accordance with these statements in the RFC:
1542 The auto-replied keyword:
1544 - SHOULD be used on messages sent in direct response to another
1545 message by an automatic process,
1547 - MUST NOT be used on manually-generated messages,
1549 - MAY be used on Delivery Status Notifications (DSNs) and Message
1550 Disposition Notifications (MDNs),
1552 - MUST NOT be used on messages generated by automatic or periodic
1553 processes, except for messages which are automatic responses to
1556 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1557 to the default Received: header definition.
1559 PH/12 Added log selector acl_warn_skipped (default on).
1561 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1562 variables because (a) they are in the wrong storage pool and (b) even if
1563 they were copied, it wouldn't work properly because of the caching.
1565 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1566 checking when decoding. Apparently there are clients that generate
1567 overlong encoded strings. Why am I not surprised?
1569 PH/15 If the first argument of "${if match_address" was not empty, but did not
1570 contain an "@" character, Exim crashed. Now it writes a panic log message
1571 and treats the condition as false.
1573 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1575 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1576 extension "envelope-auth". The code is finished and in agreement with
1577 other implementations, but there is no documentation so far and in fact,
1578 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1579 not changing the active code.
1581 Print executed "if" and "elsif" statements when debugging is used. This
1582 helps a great deal to understand what a filter does.
1584 Document more things not specified clearly in RFC3028. I had all this
1585 sorted out, when out of a sudden new issues came to my mind. Oops."
1587 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1590 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1591 canonical form (as documented). However, after a host name lookup from
1592 the IP address, check_host() was doing a simple string comparison with
1593 addresses acquired from the DNS when checking that the found name did
1594 have the original IP as one of its addresses. Since any found IPv6
1595 addresses are likely to be in abbreviated form, the comparison could
1596 fail. Luckily, there already exists a function for doing the comparison
1597 by converting both addresses to binary, so now that is used instead of
1598 the text comparison.
1600 PH/20 There was another similar case to PH/19, when a complete host name was
1601 given in a host list; looking up its IP address could give an abbreviated
1602 form, whereas the current host's name might or might not be abbreviated.
1603 The same fix has been applied.
1609 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1610 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1613 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1614 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1616 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1618 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1619 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1620 Also, for "Auto-Submitted", treat anything other than "no" as
1621 non-personal, in accordance with RFC 3834. (Previously it treated
1622 anything starting "auto-" as non-personal.)
1624 TF/01 The control=submission/name=... option had a problem with syntax
1625 errors if the name included a slash character. The /name= option
1626 now slurps the rest of the string, so it can include any characters
1627 but it must come last in the list of options (after /sender_retain
1630 PH/05 Some modifications to the interface to the fake nameserver for the new
1638 TK/01 Added the "success_on_redirect" address verification option. See
1639 NewStuff for rationale and an example.
1641 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1643 PH/02 Patch to exigrep to allow it to work on syslog lines.
1645 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1646 fread() to skip over the body file's header line, because in Cygwin the
1647 header line is locked and is inaccessible.
1649 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1650 co-exist for some time) to make it clear that it is the Exim ID that is
1651 referenced, not the Message-ID: header line.
1653 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1654 string_format() function, because snprintf() does not exist on all
1657 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1658 ${stat: expansion item.
1660 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1661 protocol synchronization error", to keep the pedants happy.
1663 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1664 well as for IRIX systems, when gcc is being used. See the host.c source
1667 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1669 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1672 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1673 transport and to the smtp transport in LMTP mode.
1675 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1677 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1678 run a filter in a subprocess. This could lead to confusion in subsequent
1679 lookups in the parent process. There should also be a search_tidyup() at
1680 the end of the subprocess.
1682 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1683 only if the host matched helo_try_verify_hosts, which caused the
1684 verification to occur when the EHLO/HELO command was issued. The ACL just
1685 tested the remembered result. Now, if a previous verification attempt has
1686 not happened, "verify = helo" does it there and then.
1688 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1690 TK/03 Fix log output including CR from clamd.
1692 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1693 exist provoked a memory error which could cause a segfault.
1695 PH/15 Installed PCRE 6.2
1697 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1699 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1700 of the problem. Specifically, suggested +O2 rather than +O1 for the
1703 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1705 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1706 record, the port was not correctly logged when the outgoing_port log
1707 selector was set (it logged the transort's default port).
1709 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1710 fallback_hosts, and "hosts" in the smtp transport.
1712 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1713 host errors such as "Connection refused".
1715 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1716 authentication with radiusclient 0.4.9:
1718 - Error returned from rc_read_config was caught wrongly
1719 - Username/password not passed on to radius server due to wrong length.
1721 The presumption is that some radiusclient API changes for 4.51/PH/17
1722 were not taken care of correctly. The code is still untested by me (my
1723 Linux distribution still has 0.3.2 of radiusclient), but it was
1724 contributed by a Radius user.
1726 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1727 expanding the "port" option of the smtp transport.
1729 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1730 while reading a MIME header. Thanks to Tom Hughes for a patch.
1732 PH/24 Include config.h inside local_scan.h so that configuration settings are
1735 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1736 that in an ACL for RCPT (for example), you can examine exactly what was
1739 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1740 commands, but it was not correctly comparing the address with the actual
1741 client host address. Thus, it would show the EHLO address in Received:
1742 header lines when this was not necessary.
1744 PH/27 Added the % operator to ${eval:}.
1746 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1747 it should be ignoring failures (because with -C, for example, it has lost
1748 privilege). It wasn't ignoring creation failures other than "already
1751 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1754 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1755 because an input line was too long, either on its own, or by virtue of
1756 too many continuations, the temporary file was not being removed, and the
1757 return code was incorrect.
1759 PH/31 Missing "BOOL" in function definition in filtertest.c.
1761 PH/32 Applied Sieve patches from the maintainer.
1763 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1765 PH/33 Added "verify = not_blind".
1767 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1768 Local/Makefile (with some defaults set). These are used in built scripts
1769 such as exicyclog, but they have never been used in the exim_install
1770 script (though there are many overriding facilities there). I have
1771 arranged that the exim_install script now takes note of these two
1774 PH/35 Installed configuration files for Dragonfly.
1776 PH/36 When a locally submitted message by a trusted user did not contain a
1777 From: header, and the sender address was obtained from -f or from an SMTP
1778 MAIL command, and the trusted user did not use -F to supply a sender
1779 name, $originator_name was incorrectly used when constructing a From:
1780 header. Furthermore, $originator_name was used for submission mode
1781 messages from external hosts without From: headers in a similar way,
1782 which is clearly wrong.
1784 PH/37 Added control=suppress_local_fixups.
1786 PH/38 When log_selector = +received_sender was set, and the addition of the
1787 sender made the log line's construction buffer exactly full, or one byte
1788 less than full, an overflow happened when the terminating "\n" was
1791 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1792 when the result of a list match is failure because a DNS lookup failed.
1794 PH/40 RM_COMMAND is now used in the building process.
1796 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1797 the "build-* directories that it finds.
1799 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1800 address in a domain literal was a prefix of an interface address.
1802 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1803 when verifying a sender address, unless rewrite_headers is false.
1805 PH/44 Wrote a long comment about why errors_to addresses are verified as
1806 recipients, not senders.
1808 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1809 the ratelimit ACL was added.
1811 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1813 PH/46 Added extra information about PostgreSQL errors to the error string.
1815 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1816 avoiding the need to install special zones in a real server. This is
1817 backwards compatible; if it can't find the fake resolver, it drops back.
1818 Thus, both old and new test suites can be run.
1820 TF/02 Added util/ratelimit.pl
1822 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1823 clock is set back in time.
1825 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1826 Candler <B.Candler@pobox.com>.
1828 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1829 OK for addresses that are the result of redirections.
1831 PH/48 A number of further additions for the benefit of the new test suite,
1832 including a fake gethostbyname() that interfaces to the fake DNS resolver
1835 TF/06 The fix for widen_domains has also been applied to qualify_single and
1836 search_parents which are the other dnslookup options that can cause
1839 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1842 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1844 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1845 service" if it encountered an MX record with an empty target hostname.
1846 The message is now "an MX or SRV record indicated no SMTP service".
1848 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1849 if the DNS of the sending site is misconfigured. This is quite a
1850 common situation. This change restores the behaviour of treating a
1851 helo verification defer as a failure.
1853 PH/51 If self=fail was set on a router, the bounce message did not include the
1854 actual error message.
1860 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1862 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1863 command itself ended in error, the underlying message about the transport
1864 filter timeout was being overwritten with the pipe command error. Now the
1865 underlying error message should be appended to the second error message.
1867 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1869 PH/02 Reduce the amount of output that "make" produces by default. Full output
1870 can still be requested.
1872 PH/03 The warning log line about a condition test deferring for a "warn" verb
1873 was being output only once per connection, rather than after each
1874 occurrence (because it was using the same function as for successful
1875 "warn" verbs). This seems wrong, so I have changed it.
1877 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1878 it should not have, which might have caused a crash in the right
1879 circumstances, but probably never did.
1881 PH/04 Installed a modified version of Tony Finch's patch to make submission
1882 mode fix the return path as well as the Sender: header line, and to
1883 add a /name= option so that you can make the user's friendly name appear
1886 TF/03 Added the control = fakedefer ACL modifier.
1888 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1889 Mark Lowes for thorough testing.
1891 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1893 TK/03 Merged latest SRS patch from Miles Wilton.
1895 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1896 with the definition in sysexits.h (which is #included earlier).
1897 Fortunately, Exim does not actually use EX_OK. The code used to try to
1898 preserve the sysexits.h value, by assumimg that macro definitions were
1899 scanned for macro replacements. I have been disabused of this notion,
1900 so now the code just undefines EX_OK before #including unistd.h.
1902 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1903 in the smtp transport. When a block could not be written in a single
1904 write() function, the timeout was being re-applied to each part-write.
1905 This seems wrong - if the receiver was accepting one byte at a time it
1906 would take for ever. The timeout is now adjusted when this happens. It
1907 doesn't have to be particularly precise.
1909 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1910 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1912 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1913 without a domain if the check to <postmaster@domain> fails.
1915 SC/01 Eximstats: added -xls and the ability to specify output files
1916 (patch written by Frank Heydlauf).
1918 SC/02 Eximstats: use FileHandles for outputing results.
1920 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1922 SC/04 Eximstats: fixed display of large numbers with -nvr option
1924 SC/05 Eximstats: fixed merging of reports with empty tables.
1926 SC/06 Eximstats: added the -include_original_destination flag
1928 SC/07 Eximstats: removed tabs and trailing whitespace.
1930 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1932 TK/06 MBOX spool code: Add real "From " MBOX separator line
1933 so the .eml file is really in mbox format (even though
1934 most programs do not really care). Patch from Alex Miller.
1936 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1937 The latter is generated from $received_to and is only set if the
1938 message has one envelope recipient. SA can use these headers,
1939 obviously out-of-the-box. Patch from Alex Miller.
1941 PH/08 The ${def test on a variable was returning false if the variable's
1942 value was "0", contrary to what the specification has always said!
1943 The result should be true unless the variable is empty.
1945 PH/09 The syntax error of a character other than { following "${if
1946 def:variable_name" (after optional whitespace) was not being diagnosed.
1947 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1948 accidental colon was present, for example, could give incorrect results.
1950 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1951 result is used (not including appendfile, where other changes are about
1954 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1955 This involved changing a lot of size variables from int to off_t. It
1956 should work with maildirs and everything.
1958 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1959 spamd dying while we are connected to it.
1961 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1964 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1965 function for BDB 4.3.
1967 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1969 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1970 the sources for PCRE have been re-arranged and more files are now
1973 PH/15 The code I had for printing potentially long long variables in PH/11
1974 above was not the best (it lost precision). The length of off_t variables
1975 is now inspected at build time, and an appropriate printing format (%ld
1976 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1977 to be "long long int" or "long int". This is needed for the internal
1978 formatting function string_vformat().
1980 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1981 the configuration file to be ":syslog", then the script "guesses" where
1982 the logs files are, rather than using the compiled in default. In our
1983 case the guess is not the same as the compiled default, so the script
1984 suddenly stopped working when I started to use syslog. The patch checks
1985 to see if log_file_path is "". If so, it attempts to read it from exim
1986 with no configuration file to get the compiled in version, before it
1987 falls back to the previous guessing code."
1989 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1990 implementing BATV in an Exim configuration. See NewStuff for the gory
1993 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1994 Makefile that are specific to HP-UX.
1996 PH/18 If the "use_postmaster" option was set for a recipient callout together
1997 with the "random" option, the postmaster address was used as the MAIL
1998 FROM address for the random test, but not for the subsequent recipient
1999 test. It is now used for both.
2001 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
2002 patch removes a few documentation additions to RFC 3028, because the
2003 latest draft now contains them. It adds the new en;ascii-case comparator
2004 and a new error check for 8bit text in MIME parts. Comparator and
2005 require names are now matched exactly. I enabled the subaddress
2006 extension, but it is not well tested yet (read: it works for me)."
2008 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
2009 rework some of the code of TK/09 above to avoid the hardwired use of
2010 "%lld" and "long long". Replaced the call to snprintf() with a call to
2013 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
2014 records point to non-existent hosts", "retry timeout exceeded", and
2015 "retry time not reached for any host after a long failure period".
2017 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
2018 experimental DomainKeys support:
2020 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
2021 (2) On an error such as an illegally used "control", the wrong name for
2022 the control was given.
2024 These problems did NOT occur unless DomainKeys support was compiled.
2026 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
2028 PH/24 Added ${if match_ip condition.
2030 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
2031 databases so that it will be absolutely obvious if a crash occurs in the
2032 DB library. This is a regular occurrence (often caused by mis-matched
2035 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
2036 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
2037 picked up on a user's system that detects such things. There doesn't seem
2038 to be a gcc warning option for this - only an attribute that has to be
2039 put on the function's prototype. It seems that in Fedora Core 4 they have
2040 set this on a number of new functions. No doubt there will be more in due
2043 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2044 specify a transport. However, if an address that was verified by such a
2045 router was the subject of a callout, Exim crashed because it tried to
2046 read the rcpt_include_affixes from the non-existent transport. Now it
2047 just assumes that the setting of that option is false. This bug was
2048 introduced by 4.51/PH/31.
2050 PH/28 Changed -d+all to exclude +memory, because that information is very
2051 rarely of interest, but it makes the output a lot bigger. People tend to
2052 do -d+all out of habit.
2054 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2055 code in os-type was giving problems when libc.so lives in lib64, like on
2058 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2059 aren't the modern standard, and it seems that some systems' include files
2060 don't always have them. Exim was already checking for some of the newer
2061 ones like T_AAAA, and defining it itself. I've added checks for all the
2062 record types that Exim uses.
2064 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2065 not automatically generating a new one, as it is supposed to. This
2066 prevented TLS from working. If the file did exist, but contained invalid
2067 data, a new version was generated, as expected. It was only the case of a
2068 non-existent file that was broken.
2070 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2071 with a change in libdomainkeys > 0.64.
2073 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2074 from DNS. If the selector record carries the flag, it now has
2075 precedence over the domain-wide flag.
2077 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2079 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2080 the use of a port name that isn't defined in /etc/services) occurred, the
2081 message was deferred as in a normal delivery, and thus remained on the
2082 spool, instead of being failed because of the mua_wrapper setting. This
2083 is now fixed, and I tidied up some of the mua_wrapper messages at the
2086 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2087 the messages in a hash of arrays rather than using individual hashes.
2088 This is a bit cleaner and results in dramatic memory savings, albeit
2089 at a slight CPU cost.
2091 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2092 as requested by Marc Sherman.
2094 SC/10 Eximstats: added histograms for user specified patterns as requested
2097 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2099 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2100 fopen() in the content-scanning modules that did not already have it.
2106 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2107 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2109 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2111 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2113 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2114 to test Sieve filters that use "vacation".
2116 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2117 that changes the way the GnuTLS parameters are stored in the cache file.
2118 The new format can be generated externally. For backward compatibility,
2119 if the data in the cache doesn't make sense, Exim assumes it has read an
2120 old-format file, and it generates new data and writes a new file. This
2121 means that you can't go back to an older release without removing the
2124 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2125 work if there are any delivery delays because "one_time" forces the
2126 parent to be marked "delivered", so its unseen clone is never tried
2127 again. For this reason, Exim now forbids the simultaneous setting of
2130 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2131 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2132 entry if you want to know the details. The fix, however, neglected to
2133 consider the case when local delivery batching is involved. The test for
2134 "previously delivered" was not happening when checking to see if an
2135 address could be batched with a previous (undelivered) one; under
2136 certain circumstances this could lead to multiple deliveries to the same
2139 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2140 in its include files, and this causes problems building Exim.
2142 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2143 header_syntax) but Exim was just ignoring anything given after a slash.
2144 In particular, this caused confusion with an attempt to use "verify =
2145 reverse_host_lookup/defer_ok". An error is now given when options are
2146 supplied for verify items that do not have them. (Maybe reverse_host_
2147 lookup should have a defer_ok option, but that's a different point.)
2149 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2150 defined by RFC 821) to 2048, because there were problems with some AUTH
2151 commands, and RFC 1869 says the size should be increased for extended
2152 SMTP commands that take arguments.
2154 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2157 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2158 "unknown" error; now it says that the functionality isn't in the binary.
2160 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2161 an address' error message when a string expansion fails (syntax or
2162 whatever). Otherwise the password may appear in the log. Following change
2163 PH/42 below, there is no longer a chance of it appearing in a bounce
2166 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2168 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2169 of its addresses was ever tried. (Bugzilla bug #2.)
2171 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2172 the result incorrectly in the debug output. (It correctly added a newline
2173 to what was transported.)
2175 TF/01 Added $received_time.
2177 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2178 commented out examples of how to interface to a virus scanner and to
2179 SpamAssassin. Also added commented examples of av_scanner and
2180 spamd_address settings.
2182 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2183 and controls are allowed in which ACLs. There were a couple of minor
2184 errors. Some of the entries in the conditions table (which is a table of
2185 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2186 negation of where the condition IS allowed.
2188 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2190 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2191 header file does not have a version number, so I've had to invent a new
2192 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2193 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2194 radiusclient), but it was contributed by a Radius user.
2196 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2197 files or directories, for OpenSSL.
2199 PH/19 When an Exim process that is running as root has to create an Exim log
2200 file, it does so in a subprocess that runs as exim:exim so as to get the
2201 ownership right at creation (otherwise, other Exim processes might see
2202 the file with the wrong ownership). There was no test for failure of this
2203 fork() call, which would lead to the process getting stuck as it waited
2204 for a non-existent subprocess. Forks do occasionally fail when resources
2205 run out. I reviewed all the other calls to fork(); they all seem to check
2208 PH/20 When checking for unexpected SMTP input at connect time (before writing
2209 the banner), Exim was not dealing correctly with a non-positive return
2210 from the read() function. If the client had disconnected by this time,
2211 the result was a log entry for a synchronization error with an empty
2212 string after "input=" when read() returned zero. If read() returned -1
2213 (an event I could not check), uninitialized data bytes were printed.
2214 There were reports of junk text (parts of files, etc) appearing after
2217 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2219 PH/22 Added support for macro redefinition, and (re)definition in between
2220 driver and ACL definitions.
2222 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2223 forgetting to use the resulting value; it was using the unexpanded value.
2225 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2226 hadn't been configured. The fix is from Juergen Kreileder, who
2227 understands it better than I do:
2229 "Here's what I see happening with three configured cyrus_sasl
2230 authenticators configured (plain, login, cram-md5):
2232 On startup auth_cyrus_sasl_init() gets called for each of these.
2233 This means three calls to sasl_listmech() without a specified mech_list.
2234 => SASL tests which mechs of all available mechs actually work
2235 => three warnings about OTP not working
2236 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2238 With the patch, sasl_listmech() also gets called three times. But now
2239 SASL's mech_list option is set to the server_mech specified in the the
2240 authenticator. Or in other words, the answer from sasl_listmech()
2241 gets limited to just the mech you're testing for (which is different
2243 => the return list contains just 'plain' or 'login', 'cram-md5' or
2244 nothing depending on the value of ob->server_mech.
2246 I've just tested the patch: Authentication still works fine,
2247 unavailable mechs specified in the exim configuration are still
2248 caught, and the auth.log warnings about OTP are gone."
2250 PH/25 When debugging is enabled, the contents of the command line are added
2251 to the debugging output, even when log_selector=+arguments is not
2254 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2255 answer is "GNU", and only if the return is "GNU/something" is the answer
2258 PH/27 $acl_verify_message is now set immediately after the failure of a
2259 verification in an ACL, and so is available in subsequent modifiers. In
2260 particular, the message can be preserved by coding like this:
2262 warn !verify = sender
2263 set acl_m0 = $acl_verify_message
2265 Previously, $acl_verify_message was set only while expanding "message"
2266 and "log_message" when a very denied access.
2268 PH/28 Modified OS/os.c-Linux with
2270 -#ifndef OS_LOAD_AVERAGE
2271 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2273 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2274 nomenclature these days.)
2276 PH/29 Installed patch from the Sieve maintainer that adds the options
2277 sieve_useraddress and sieve_subaddress to the redirect router.
2279 PH/30 In these circumstances:
2280 . Two addresses routed to the same list of hosts;
2281 . First host does not offer TLS;
2282 . First host accepts first address;
2283 . First host gives temporary error to second address;
2284 . Second host offers TLS and a TLS session is established;
2285 . Second host accepts second address.
2286 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2287 and peerdn, if requested) that were in fact used only for the second
2290 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2291 attention to any local part prefix or suffix that was matched by the
2292 router that accepted the address. It now behaves in the same way as it
2293 does for delivery: the affixes are removed from the local part unless
2294 rcpt_include_affixes is set on the transport.
2296 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2297 timeout during the DATA phase of an incoming message.
2299 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2300 applied a patch sanctioned by the Sieve maintainer.
2302 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2303 the uid or gid is negative. A case of a negative gid caused this to be
2304 noticed. The fix allows for either to be negative.
2306 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2307 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2310 PH/36 The change PH/12 above was broken. Fixed it.
2312 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2313 the grounds that routing the same address twice would always produce the
2314 same answer. This might have been true once, but it is certainly no
2315 longer true now. Routing a child address may depend on the previous
2316 routing that produced that child. Some complicated redirection strategies
2317 went wrong when messages had multiple recipients, and made Exim's
2318 behaviour dependent on the order in which the addresses were given.
2320 I have moved the duplicate checking until after the routing is complete.
2321 Exim scans the addresses that are assigned to local and remote
2322 transports, and removes any duplicates. This means that more work will be
2323 done, as duplicates will always all be routed, but duplicates are
2324 presumably rare, so I don't expect this is of any significance.
2326 For deliveries to pipes, files, and autoreplies, the duplicate checking
2327 still happens during the routing process, since they are not going to be
2330 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2331 It corrects a timeout issue with spamd. This is Ian's comment: "The
2332 background is that sometimes spamd either never reads data from a
2333 connection it has accepted, or it never writes response data. The exiscan
2334 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2335 blindly assumes that writes won't block so it may never time out."
2337 PH/39 Allow G after quota size as well as K and M.
2339 PH/40 The value set for $authenticated_id in an authenticator may not contain
2340 binary zeroes or newlines because the value is written to log lines and
2341 to spool files. There was no check on this. Now the value is run through
2342 the string_printing() function so that such characters are converted to
2343 printable escape sequences.
2345 PH/41 $message_linecount is a new variable that contains the total number of
2346 lines in the message. Compare $body_linecount, which is the count for the
2349 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2350 bounce and delay warning messages, except in certain special cases, which
2353 (a) An SMTP error message from a remote host;
2354 (b) A message specified in a :fail: redirection;
2355 (c) A message specified in a "fail" command in a system filter;
2356 (d) A message specified in a FAIL return from the queryprogram router;
2357 (e) A message specified by the cannot_route_message router option.
2359 In these cases only, Exim does include the error details in bounce and
2360 warning messages. There are also a few cases where bland messages such
2361 as "unrouteable address" or "local delivery error" are given.
2363 PH/43 $value is now also set for the "else" part of a ${run expansion.
2365 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2366 being worked on, but at least Exim now implements the latest version to
2369 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2370 process to complete was treated as a delivery failure, a timeout while
2371 writing the message to the pipe was logged, but erroneously treated as a
2372 successful delivery. Such timeouts include transport filter timeouts. For
2373 consistency with the overall process timeout, these timeouts are now
2374 treated as errors, giving rise to delivery failures by default. However,
2375 there is now a new Boolean option for the pipe transport called
2376 timeout_defer, which, if set TRUE, converts the failures into defers for
2377 both kinds of timeout. A transport filter timeout is now identified in
2380 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2381 systems where "make" and "gmake" are different, calling "gmake" at top
2382 level broke things. I've arranged for the value of $(MAKE) to be passed
2383 from the Makefile to this script so that it can call the same version of
2387 A note about Exim versions 4.44 and 4.50
2388 ----------------------------------------
2390 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2391 changes of various kinds. As a consequence, a big documentation update was
2392 needed. This delayed the release for rather longer than seemed good, especially
2393 in the light of a couple of (minor) security issues. Therefore, the changes
2394 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2395 release. So 4.44 and 4.50 are in effect two different branches that both start
2398 I have left the 4.50 change log unchanged; it contains all the changes since
2399 4.43. The change log for 4.44 is below; many of its items are identical to
2400 those for 4.50. This seems to be the most sensible way to preserve the
2401 historical information.
2407 1. Minor wording change to the doc/README.SIEVE file.
2409 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2410 computation of the current number of files was incorrect.
2412 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2413 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2416 4. Give more explanation in the error message when the command for a transport
2417 filter fails to execute.
2419 5. There are several places where Exim runs a non-Exim command in a
2420 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2421 was being done only for the command run by the queryprogram router. It is
2422 now done for all such subprocesses. The other cases are: ${run, transport
2423 filters, and the commands run by the lmtp and pipe transports.
2425 6. Added CONFIGURE_GROUP build-time option.
2427 7. Some older OS have a limit of 256 on the maximum number of file
2428 descriptors. Exim was using setrlimit() to set 1000 as a large value
2429 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2430 systems. I've change it so that if it can't get 1000, it tries for 256.
2432 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2433 was an oversight, and furthermore, ever since the addition of extra
2434 controls (e.g. 4.43/32), the checks on when to allow different forms of
2435 "control" were broken. There should now be diagnostics for all cases when a
2436 control that does not make sense is encountered.
2438 9. Added the /retain_sender option to "control=submission".
2440 10. $recipients is now available in the predata ACL (oversight).
2442 11. Tidy the search cache before the fork to do a delivery from a message
2443 received from the command line. Otherwise the child will trigger a lookup
2444 failure and thereby defer the delivery if it tries to use (for example) a
2445 cached ldap connection that the parent has called unbind on.
2447 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2448 of $address_data from the recipient verification was clobbered by the
2449 sender verification.
2451 13. The value of address_data from a sender verification is now available in
2452 $sender_address_data in subsequent conditions in the ACL statement.
2454 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2456 15. Added a new option "connect=<time>" to callout options, to set a different
2459 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2460 was its contents. (It was OK if the option was not defined at all.)
2462 17. A "Completed" log line is now written for messages that are removed from
2463 the spool by the -Mrm option.
2465 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2466 information about exactly what failed.
2468 19. Added -dd to debug only the daemon process.
2470 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2471 handles timeouts, both on the server side and network timeouts. Renamed the
2472 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2474 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2475 It is now set to "smtps".
2477 22. $host_address is now set to the target address during the checking of
2478 ignore_target_hosts.
2480 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2481 being passed; this would have caused $sender_host_name to have been used if
2482 matching the list had actually called for a host name (not very likely,
2483 since this list is usually IP addresses). A host name is now passed as
2486 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2487 code that specifies a non-restarting handler (typically sigaction() in
2488 modern systems) in an attempt to fix a rare and obscure crash bug.
2490 25. Narrowed the window for a race in the daemon that could cause it to ignore
2491 SIGCHLD signals. This is not a major problem, because they are used only to
2492 wake it up if nothing else does.
2494 26. A malformed maildirsize file could cause Exim to calculate negative values
2495 for the mailbox size or file count. Odd effects could occur as a result.
2496 The maildirsize information is now recalculated if the size or filecount
2499 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2500 support for a long time. Removed HAVE_SYS_VFS_H.
2502 28. Installed the latest version of exipick from John Jetmore.
2504 29. In an address list, if the pattern was not a regular expression, an empty
2505 subject address (from a bounce message) matched only if the pattern was an
2506 empty string. Non-empty patterns were not even tested. This was the wrong
2507 because it is perfectly reasonable to use an empty address as part of a
2508 database query. An empty address is now tested by patterns that are
2509 lookups. However, all the other forms of pattern expect the subject to
2510 contain a local part and a domain, and therefore, for them, an empty
2511 address still always fails if the pattern is not itself empty.
2513 30. Exim went into a mad DNS loop when attempting to do a callout where the
2514 host was specified on an smtp transport, and looking it up yielded more
2515 than one IP address.
2517 31. Re-factored the code for checking spool and log partition space into a
2518 function that finds that data and another that does the check. The former
2519 is then used to implement four new variables: $spool_space, $log_space,
2520 $spool_inodes, and $log_inodes.
2522 32. The RFC2047 encoding function was originally intended for short strings
2523 such as real names; it was not keeping to the 75-character limit for
2524 encoded words that the RFC imposes. It now respects the limit, and
2525 generates multiple encoded words if necessary. To be on the safe side, I
2526 have increased the buffer size for the ${rfc2047: expansion operator from
2529 33. It is now permitted to omit both strings after an "if" condition; if the
2530 condition is true, the result is "true". As before, when the second string
2531 is omitted, a false condition yields an empty string. This makes it less
2532 cumbersome to write custom ACL and router conditions.
2534 34. Failure to deliver a bounce message always caused it to be frozen, even if
2535 there was an errors_to setting on the router. The errors_to setting is now
2538 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2539 canonical form (fully expanded) before being placed in
2540 $sender_host_address.
2542 36. The table in the code that translates DNS record types into text (T_A to
2543 "A" for instance) was missing entries for NS and CNAME. It is just possible
2544 that this could have caused confusion if both these types were looked up
2545 for the same domain, because the text type is used as part of Exim's
2546 per-process caching. But the chance of anyone hitting this buglet seems
2549 37. The dnsdb lookup has been extended in a number of ways.
2551 (1) There is a new type, "zns", which walks up the domain tree until it
2552 finds some nameserver records. It should be used with care.
2554 (2) There is a new type, "mxh", which is like "mx" except that it returns
2555 just the host names, not the priorities.
2557 (3) It is now possible to give a list of domains (or IP addresses) to be
2558 looked up. The behaviour when one of the lookups defers can be
2559 controlled by a keyword.
2561 (4) It is now possible to specify the separator character for use when
2562 multiple records are returned.
2564 38. The dnslists ACL condition has been extended: it is now possible to supply
2565 a list of IP addresses and/or domains to be looked up in a particular DNS
2568 39. Added log_selector=+queue_time_overall.
2570 40. When running the queue in the test harness, wait just a tad after forking a
2571 delivery process, to get repeatability of debugging output.
2573 41. Include certificate and key file names in error message when GnuTLS fails
2574 to set them up, because the GnuTLS error message doesn't include the name
2575 of the failing file when there is a problem reading it.
2577 42. Allow both -bf and -bF in the same test run.
2579 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2581 44. The "Exiscan patch" is now merged into the mainline Exim source.
2583 45. Sometimes the final signoff response after QUIT could fail to get
2584 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2585 < 0 before doing a fflush(). This bug looks as though it goes back to the
2586 introduction of TLS in release 3.20, but "sometimes" must have been rare
2587 because the tests only now provoked it.
2589 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2590 (this can affect the format of dates).
2592 47. exim_tidydb, when checking for the continued existence of a message for
2593 which it has found a message-specific retry record, was not finding
2594 messages that were in split spool directories. Consequently, it was
2595 deleting retry records that should have stayed in existence.
2597 48. Steve fixed some bugs in eximstats.
2599 49. The SPA authentication driver was not abandoning authentication and moving
2600 on to the next authenticator when an expansion was forced to fail,
2601 contradicting the general specification for all authenticators. Instead it
2602 was generating a temporary error. It now behaves as specified.
2604 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2605 (the order specifies the preference for clients). The order is now AES256,
2606 AES128, 3DES, ARCFOUR128.
2608 51. Small patch to Sieve code - explicitly set From: when generating an
2611 52. Exim crashed if a remote delivery caused a very long error message to be
2612 recorded - for instance if somebody sent an entire SpamAssassin report back
2613 as a large number of 550 error lines. This bug was coincidentally fixed by
2614 increasing the size of one of Exim's internal buffers (big_buffer) that
2615 happened as part of the Exiscan merge. However, to be on the safe side, I
2616 have made the code more robust (and fixed the comments that describe what
2619 53. Now that there can be additional text after "Completed" in log lines (if
2620 the queue_time_overall log selector is set), a one-byte patch to exigrep
2621 was needed to allow it to recognize "Completed" as not the last thing in
2624 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2625 patch that reportedly fixes this has been added. I am not expert enough to
2626 create a test for it. This is what the patch creator wrote:
2628 "I found a little strange behaviour of ldap code when working with
2629 Windows 2003 AD Domain, where users was placed in more than one
2630 Organization Units. When I tried to give exim partial DN, the exit code
2631 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2632 But simultaneously result of request was absolutely normal ldap result,
2633 so I produce this patch..."
2635 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2636 REFERENCE, so I have modified the code to exclude the patch when that macro
2639 55. Some experimental protocols are using DNS PTR records for new purposes. The
2640 keys for these records are domain names, not reversed IP addresses. The
2641 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2642 leaves it alone. Component reversal etc. now happens only for IP addresses.
2645 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2647 57. Double the size of the debug message buffer (to 2048) so that more of very
2648 long debug lines gets shown.
2650 58. The exicyclog utility now does better if the number of log files to keep
2651 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2653 59. Two changes related to the smtp_active_hostname option:
2655 (1) $smtp_active_hostname is now available as a variable.
2656 (2) The default for smtp_banner uses $smtp_active_hostname instead
2657 of $primary_hostname.
2659 60. The host_aton() function is supposed to be passed a string that is known
2660 to be a valid IP address. However, in the case of IPv6 addresses, it was
2661 not checking this. This is a hostage to fortune. Exim now panics and dies
2662 if the condition is not met. A case was found where this could be provoked
2663 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2664 components; fortuitously, this particular loophole had already been fixed
2665 by change 4.50/55 above.
2667 If there are any other similar loopholes, the new check in host_aton()
2668 itself should stop them being exploited. The report I received stated that
2669 data on the command line could provoke the exploit when Exim was running as
2670 exim, but did not say which command line option was involved. All I could
2671 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2672 running as the user.
2675 61. There was a buffer overflow vulnerability in the SPA authentication code
2676 (which came originally from the Samba project). I have added a test to the
2677 spa_base64_to_bits() function which I hope fixes it.
2680 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2681 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2683 63. The daemon start-up calls getloadavg() while still root for those OS that
2684 need the first call to be done as root, but it missed one case: when
2685 deliver_queue_load_max is set with deliver_drop_privilege. This is
2686 necessary for the benefit of the queue runner, because there is no re-exec
2687 when deliver_drop_privilege is set.
2689 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2690 This has been fixed.
2692 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2693 was in use, was not putting the data itself into the right store pool;
2694 consequently, it could be overwritten for a subsequent message in the same
2695 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2698 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2700 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2701 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2702 the function were treating the return as a boolean value, which happened to
2703 work because 0=false and not-0=true, but is not correct code.
2705 68. The host_aton() function was not handling scoped IPv6 addresses (those
2706 with, for example, "%eth0" on the end) correctly.
2708 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2709 negated items (that is, ~something) in unsigned ints. Some compilers
2710 apparently mutter when there is no cast.
2712 70. If an address verification called from an ACL failed, and did not produce a
2713 user-specific message (i.e. there was only a "system" message), nothing was
2714 put in $acl_verify_message. In this situation, it now puts the system
2717 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2718 session; change 4.31/43 added the unwanted input to the log line - except
2719 that it did not do this in the start of session case. It now does.
2721 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2722 This is wrong and can cause the other end to generate a synchronization
2723 error if it is another Exim or anything else that does the synchronization
2724 check. A QUIT command is no longer sent after a timeout.
2726 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2727 during host lookups.
2729 74. The defer_ok option of callout verification was not working if it was used
2730 when verifying addresses in header lines, that is, for this case:
2732 verify = header_sender/callout=defer_ok
2734 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2735 those file descriptors could be used for SMTP connections. If anything
2736 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2737 could be sent to the SMTP client, causing chaos. The daemon now opens
2738 stdin, stdout, and stderr to /dev/null when it puts itself into the
2741 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2742 log by default. The user can override this with suitable Perl magic.
2744 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2745 the log message when discard triggers, was not working for the DATA ACL or
2746 for the non-SMTP ACL.
2748 78. Error message wording change in sieve.c.
2750 79. If smtp_accept_max_per_host was set, the number of connections could be
2751 restricted to fewer than expected, because the daemon was trying to set up
2752 a new connection before checking whether the processes handling previous
2753 connections had finished. The check for completed processes is now done
2754 earlier. On busy systems, this bug wouldn't be noticed because something
2755 else would have woken the daemon, and it would have reaped the completed
2758 80. If a message was submitted locally by a user whose login name contained one
2759 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2760 It caused a spool format error. I have fixed the spool reading code. A
2761 related problem was that the "from" clause in the Received: line became
2762 illegal because of the space(s). It is now covered by ${quote_local_part.
2764 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2767 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2769 83. Patch from Sieve maintainer to fix unterminated string problem in
2770 "vacation" handling.
2772 84. Some minor changes to the Linux configuration files to help with other
2773 OS variants using glibc.
2775 85. One more patch for Sieve to update vacation handling to latest spec.
2778 ----------------------------------------------------
2779 See the note above about the 4.44 and 4.50 releases.
2780 ----------------------------------------------------
2786 1. Change 4.43/35 introduced a bug that caused file counts to be
2787 incorrectly computed when quota_filecount was set in an appendfile
2790 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2791 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2794 3. Give more explanation in the error message when the command for a transport
2795 filter fails to execute.
2797 4. There are several places where Exim runs a non-Exim command in a
2798 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2799 was being done only for the command run by the queryprogram router. It is
2800 now done for all such subprocesses. The other cases are: ${run, transport
2801 filters, and the commands run by the lmtp and pipe transports.
2803 5. Some older OS have a limit of 256 on the maximum number of file
2804 descriptors. Exim was using setrlimit() to set 1000 as a large value
2805 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2806 systems. I've change it so that if it can't get 1000, it tries for 256.
2808 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2809 was an oversight, and furthermore, ever since the addition of extra
2810 controls (e.g. 4.43/32), the checks on when to allow different forms of
2811 "control" were broken. There should now be diagnostics for all cases when a
2812 control that does not make sense is encountered.
2814 7. $recipients is now available in the predata ACL (oversight).
2816 8. Tidy the search cache before the fork to do a delivery from a message
2817 received from the command line. Otherwise the child will trigger a lookup
2818 failure and thereby defer the delivery if it tries to use (for example) a
2819 cached ldap connection that the parent has called unbind on.
2821 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2822 of $address_data from the recipient verification was clobbered by the
2823 sender verification.
2825 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2826 was its contents. (It was OK if the option was not defined at all.)
2828 11. A "Completed" log line is now written for messages that are removed from
2829 the spool by the -Mrm option.
2831 12. $host_address is now set to the target address during the checking of
2832 ignore_target_hosts.
2834 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2835 being passed; this would have caused $sender_host_name to have been used if
2836 matching the list had actually called for a host name (not very likely,
2837 since this list is usually IP addresses). A host name is now passed as
2840 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2841 code that specifies a non-restarting handler (typically sigaction() in
2842 modern systems) in an attempt to fix a rare and obscure crash bug.
2844 15. Narrowed the window for a race in the daemon that could cause it to ignore
2845 SIGCHLD signals. This is not a major problem, because they are used only to
2846 wake it up if nothing else does.
2848 16. A malformed maildirsize file could cause Exim to calculate negative values
2849 for the mailbox size or file count. Odd effects could occur as a result.
2850 The maildirsize information is now recalculated if the size or filecount
2853 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2854 support for a long time. Removed HAVE_SYS_VFS_H.
2856 18. Updated exipick to current release from John Jetmore.
2858 19. Allow an empty sender to be matched against a lookup in an address list.
2859 Previously the only cases considered were a regular expression, or an
2862 20. Exim went into a mad DNS lookup loop when doing a callout where the
2863 host was specified on the transport, if the DNS lookup yielded more than
2866 21. The RFC2047 encoding function was originally intended for short strings
2867 such as real names; it was not keeping to the 75-character limit for
2868 encoded words that the RFC imposes. It now respects the limit, and
2869 generates multiple encoded words if necessary. To be on the safe side, I
2870 have increased the buffer size for the ${rfc2047: expansion operator from
2873 22. Failure to deliver a bounce message always caused it to be frozen, even if
2874 there was an errors_to setting on the router. The errors_to setting is now
2877 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2878 canonical form (fully expanded) before being placed in
2879 $sender_host_address.
2881 24. Updated eximstats to version 1.33
2883 25. Include certificate and key file names in error message when GnuTLS fails
2884 to set them up, because the GnuTLS error message doesn't include the name
2885 of the failing file when there is a problem reading it.
2887 26. Expand error message when OpenSSL has problems setting up cert/key files.
2890 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2891 (this can affect the format of dates).
2893 28. exim_tidydb, when checking for the continued existence of a message for
2894 which it has found a message-specific retry record, was not finding
2895 messages that were in split spool directories. Consequently, it was
2896 deleting retry records that should have stayed in existence.
2898 29. eximstats updated to version 1.35
2899 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2900 1.35 - bugfix such that pie charts by volume are generated correctly
2902 30. The SPA authentication driver was not abandoning authentication and moving
2903 on to the next authenticator when an expansion was forced to fail,
2904 contradicting the general specification for all authenticators. Instead it
2905 was generating a temporary error. It now behaves as specified.
2907 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2908 (the order specifies the preference for clients). The order is now AES256,
2909 AES128, 3DES, ARCFOUR128.
2911 31. Small patch to Sieve code - explicitly set From: when generating an
2914 32. Exim crashed if a remote delivery caused a very long error message to be
2915 recorded - for instance if somebody sent an entire SpamAssassin report back
2916 as a large number of 550 error lines. This bug was coincidentally fixed by
2917 increasing the size of one of Exim's internal buffers (big_buffer) that
2918 happened as part of the Exiscan merge. However, to be on the safe side, I
2919 have made the code more robust (and fixed the comments that describe what
2922 33. Some experimental protocols are using DNS PTR records for new purposes. The
2923 keys for these records are domain names, not reversed IP addresses. The
2924 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2925 leaves it alone. Component reversal etc. now happens only for IP addresses.
2928 34. The host_aton() function is supposed to be passed a string that is known
2929 to be a valid IP address. However, in the case of IPv6 addresses, it was
2930 not checking this. This is a hostage to fortune. Exim now panics and dies
2931 if the condition is not met. A case was found where this could be provoked
2932 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2933 components; fortuitously, this particular loophole had already been fixed
2934 by change 4.50/55 or 4.44/33 above.
2936 If there are any other similar loopholes, the new check in host_aton()
2937 itself should stop them being exploited. The report I received stated that
2938 data on the command line could provoke the exploit when Exim was running as
2939 exim, but did not say which command line option was involved. All I could
2940 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2941 running as the user.
2944 35. There was a buffer overflow vulnerability in the SPA authentication code
2945 (which came originally from the Samba project). I have added a test to the
2946 spa_base64_to_bits() function which I hope fixes it.
2949 36. The daemon start-up calls getloadavg() while still root for those OS that
2950 need the first call to be done as root, but it missed one case: when
2951 deliver_queue_load_max is set with deliver_drop_privilege. This is
2952 necessary for the benefit of the queue runner, because there is no re-exec
2953 when deliver_drop_privilege is set.
2955 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2956 was in use, was not putting the data itself into the right store pool;
2957 consequently, it could be overwritten for a subsequent message in the same
2958 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2961 38. Sometimes the final signoff response after QUIT could fail to get
2962 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2963 < 0 before doing a fflush(). This bug looks as though it goes back to the
2964 introduction of TLS in release 3.20, but "sometimes" must have been rare
2965 because the tests only now provoked it.
2971 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2972 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2973 Now it returns an integer. A number of calls were still expecting a T/F
2974 return. Fortuitously, in all cases, the tests worked in OK situations,
2975 which is the norm. However, things would have gone wrong on any write
2976 failures on the smtp file descriptor. This function is used when sending
2977 messages over SMTP and also when doing verify callouts.
2979 2. When Exim is called to do synchronous delivery of a locally submitted
2980 message (the -odf or -odi options), it no longer closes stderr before doing
2983 3. Implemented the mua_wrapper option.
2985 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2987 5. Implemented the functions header_remove(), header_testname(),
2988 header_add_at_position(), and receive_remove_recipient(), and exported them
2991 6. If an ACL "warn" statement specified the addition of headers, Exim already
2992 inserted X-ACL-Warn: at the start if there was no header name. However, it
2993 was not making this test for the second and subsequent header lines if
2994 there were newlines in the string. This meant that an invalid header could
2995 be inserted if Exim was badly configured.
2997 7. Allow an ACL "warn" statement to add header lines at the start or after all
2998 the Received: headers, as well as at the end.
3000 8. Added the rcpt_4xx retry error code.
3002 9. Added postmaster_mailfrom=xxx to callout verification option.
3004 10. Added mailfrom=xxxx to the callout verification option, for verify=
3007 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
3008 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
3010 12. Inserted some casts to stop certain compilers warning when using pointer
3011 differences as field lengths or precisions in printf-type calls (mostly
3012 affecting debugging statements).
3014 13. Added optional readline() support for -be (dynamically loaded).
3016 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
3017 same clock tick as a message's arrival, so that its received time was the
3018 same as the "first fail" time on the retry record, and that message
3019 remained on the queue past the ultimate address timeout, every queue runner
3020 would try a delivery (because it was past the ultimate address timeout) but
3021 after another failure, the ultimate address timeout, which should have then
3022 bounced the address, did not kick in. This was a "< instead of <=" error;
3023 in most cases the first failure would have been in the next clock tick
3024 after the received time, and all would be well.
3026 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
3027 being recognized when the domain list was tested by the match_domain
3028 condition in an expansion string.
3030 16. Added the ${str2b64: operator.
3032 17. Exim was always calling setrlimit() to set a large limit for the number of
3033 processes, without checking whether the existing limit was already
3034 adequate. (It did check for the limit on file descriptors.) Furthermore,
3035 errors from getrlimit() and setrlimit() were being ignored. Now they are
3036 logged to the main and panic logs, but Exim does carry on, to try to do its
3037 job under whatever limits there are.
3039 18. Imported PCRE 5.0.
3041 19. Trivial typo in log message " temporarily refused connection" (the leading
3044 20. If the log selector return_path_on_delivery was set and an address was
3045 redirected to /dev/null, the delivery process crashed because it assumed
3046 that a return path would always be set for a "successful" delivery. In this
3047 case, the whole delivery is bypassed as an optimization, and therefore no
3050 21. Internal re-arrangement: the function for sending a challenge and reading
3051 a response while authentication was assuming a zero-terminated challenge
3052 string. It's now changed to take a pointer and a length, to allow for
3053 binary data in such strings.
3055 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3057 23. Exim was not respecting finduser_retries when seeking the login of the
3058 uid under which it was called; it was always trying 10 times. (The default
3059 setting of finduser_retries is zero.) Also, it was sleeping after the final
3060 failure, which is pointless.
3062 24. Implemented tls_on_connect_ports.
3064 25. Implemented acl_smtp_predata.
3066 26. If the domain in control=submission is set empty, Exim assumes that the
3067 authenticated id is a complete email address when it generates From: or
3068 Sender: header lines.
3070 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3071 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3072 chown and chgrp in /bin and hostname in /usr/bin.
3074 28. Exim was keeping the "process log" file open after each use, just as it
3075 does for the main log. This opens the possibility of it remaining open for
3076 long periods when the USR1 signal hits a daemon. Occasional processlog
3077 errors were reported, that could have been caused by this. Anyway, it seems
3078 much more sensible not to leave this file open at all, so that is what now
3081 29. The long-running daemon process does not normally write to the log once it
3082 has entered its main loop, and it closes the log before doing so. This is
3083 so that log files can straightforwardly be renamed and moved. However,
3084 there are a couple of unusual error situations where the daemon does write
3085 log entries, and I had neglected to close the log afterwards.
3087 30. The text of an SMTP error response that was received during a remote
3088 delivery was being truncated at 512 bytes. This is too short for some of
3089 the long messages that one sometimes sees. I've increased the limit to
3092 31. It is now possible to make retry rules that apply only when a message has a
3093 specific sender, in particular, an empty sender.
3095 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3096 it possible to be selective about when SMTP synchronization is enforced.
3098 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3100 32. Implemented hosts_connection_nolog.
3102 33. Added an ACL for QUIT.
3104 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3107 35. Added mailbox_size and mailbox_filecount to appendfile.
3109 36. Added control = no_multiline_responses to ACLs.
3111 37. There was a bug in the logic of the code that waits for the clock to tick
3112 in the case where the clock went backwards by a substantial amount such
3113 that the microsecond fraction of "now" was more than the microsecond
3114 fraction of "then" (but the whole seconds number was less).
3116 38. Added support for the libradius Radius client library this is found on
3117 FreeBSD (previously only the radiusclient library was supported).
3123 1. When certain lookups returned multiple values in the form name=value, the
3124 quoting of the values was not always being done properly. Specifically:
3125 (a) If the value started with a double quote, but contained no whitespace,
3127 (b) If the value contained whitespace other than a space character (i.e.
3128 tabs or newlines or carriage returns) it was not quoted.
3129 This fix has been applied to the mysql and pgsql lookups by writing a
3130 separate quoting function and calling it from the lookup code. The fix
3131 should probably also be applied to nisplus, ibase and oracle lookups, but
3132 since I cannot test any of those, I have not disturbed their existing code.
3134 2. A hit in the callout cache for a specific address caused a log line with no
3135 reason for rejecting RCPT. Now it says "Previous (cached) callout
3136 verification failure".
3138 3. There was an off-by-one bug in the queryprogram router. An over-long
3139 return line was truncated at 256 instead of 255 characters, thereby
3140 overflowing its buffer with the terminating zero. As well as fixing this, I
3141 have increased the buffer size to 1024 (and made a note to document this).
3143 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3144 when Exim is waiting for an SMTP response from a remote server, Exim
3145 restarts its select() call on the socket, thereby resetting its timeout.
3146 This is not a problem when such interrupts are rare. Somebody set up a cron
3147 job to run exiwhat every 2 minutes, which is less than the normal select()
3148 timeout (5 or 10 minutes). This meant that the select() timeout never
3149 kicked in because it was always reset. I have fixed this by comparing the
3150 time when an interrupt arrives with the time at the start of the first call
3151 to select(). If more time than the timeout has elapsed, the interrupt is
3152 treated as a timeout.
3154 5. Some internal re-factoring in preparation for the addition of Sieve
3155 extensions (by MH). In particular, the "personal" test is moved to a
3156 separate function, and given an option for scanning Cc: and Bcc: (which is
3157 not set for Exim filters).
3159 6. When Exim created an email address using the login of the caller as the
3160 local part (e.g. when creating a From: or Sender: header line), it was not
3161 quoting the local part when it contained special characters such as @.
3163 7. Installed new OpenBSD configuration files.
3165 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3166 try to make them clearer.
3168 9. Callout options, other than the timeout value, were being ignored when
3169 verifying sender addresses in header lines. For example, when using
3171 verify = header_sender/callout=no_cache
3173 the cache was (incorrectly) being used.
3175 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3176 only the Cygwin environment.
3178 11. When return_path_on_delivery was set as a log selector, if different remote
3179 addresses in the same message used different return paths and parallel
3180 remote delivery occurred, the wrong values would sometimes be logged.
3181 (Whenever a remote delivery process finished, the return path value from
3182 the most recently started remote delivery process was logged.)
3184 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3185 lines when AUTH and/or TLS are in use. This is the "received protocol"
3186 field. Exim used to use "asmtp" for authenticated SMTP, without any
3187 indication (in the protocol name) for TLS use. Now it follows the RFC and
3188 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3189 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3190 names appear in log lines as well as in Received: header lines.
3192 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3193 extensions, and comparison tests, and to fix some bugs.
3195 14. Changes to the "personal" filter test:
3197 (1) The test was buggy in that it was just doing the equivalent of
3198 "contains" tests on header lines. For example, if a user's address was
3199 anne@some.where, the "personal" test would incorrectly be true for
3201 To: susanne@some.where
3203 This test is now done by extracting each address from the header in turn,
3204 and checking the entire address. Other tests that are part of "personal"
3205 are now done using regular expressions (for example, to check local parts
3206 of addresses in From: header lines).
3208 (2) The list of non-personal local parts in From: addresses has been
3209 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3210 taken from the Sieve specification recommendations.
3212 (3) If the message contains any header line starting with "List-" it is
3213 treated as non-personal.
3215 (4) The test for "circular" in the Subject: header line has been removed
3216 because it now seems ill-conceived.
3218 15. Minor typos in src/EDITME comments corrected.
3220 16. Installed latest exipick from John Jetmore.
3222 17. If headers_add on a router specified a text string that was too long for
3223 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3224 of string_sprintf() is now avoided.
3226 18. $message_body_size was not set (it was always zero) when running the DATA
3227 ACL and the local_scan() function.
3229 19. For the "mail" command in an Exim filter, no default was being set for
3230 the once_repeat time, causing a random time value to be used if "once" was
3231 specified. (If the value happened to be <= 0, no repeat happened.) The
3232 default is now 0s, meaning "never repeat". The "vacation" command was OK
3233 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3234 (I found it when inspecting the code).
3236 20. There is now an overall timeout for performing a callout verification. It
3237 defaults to 4 times the callout timeout, which applies to individual SMTP
3238 commands during the callout. The overall timeout applies when there is more
3239 than one host that can be tried. The timeout is checked before trying the
3240 next host. This prevents very long delays if there are a large number of
3241 hosts and all are timing out (e.g. when the network connections are timing
3242 out). The value of the overall timeout can be changed by specifying an
3243 additional sub-option for "callout", called "maxwait". For example:
3245 verify = sender/callout=5s,maxwait=20s
3247 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3248 to the end before writing, but this should make it even safer).
3250 22. Exim was forgetting that it had advertised PIPELINING for the second and
3251 subsequent messages on an SMTP connection. It was also not resetting its
3252 memory on STARTTLS and an internal HELO.
3254 23. When Exim logs an SMTP synchronization error within a session, it now
3255 records whether PIPELINING has been advertised or not.
3257 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3258 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3259 rather than long int.
3261 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3263 26. Added the never_mail option to autoreply.
3269 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3270 crash if the getsockname() call failed; this can happen if a connection is
3271 closed very soon after it is established. The problem was simply in the
3272 order in which certain operations were done, causing Exim to try to write
3273 to the SMTP stream before it had set up the file descriptor. The bug has
3274 been fixed by making things happen in the correct order.
3280 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3281 before the connection was closed, thus losing the rejection response.
3283 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3284 some early Solaris releases, but causes trouble in current releases where
3285 socklen_t is defined.
3287 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3290 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3293 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3294 that releasing the top of it at the end releases what was used for sub-
3295 expansions (unless the block got too big). However, discard this block if
3296 the first thing is a variable or header, so that we can use its block when
3297 it is dynamic (useful for very large $message_headers, for example).
3299 6. Lookups now cache *every* query, not just the most recent. A new, separate
3300 store pool is used for this. It can be recovered when all lookup caches are
3301 flushed. Lookups now release memory at the end of their result strings.
3302 This has involved some general refactoring of the lookup sources.
3304 7. Some code has been added to the store_xxx() functions to reduce the amount
3305 of flapping under certain conditions.
3307 8. log_incoming_interface used to affect only the <= reception log lines. Now
3308 it causes the local interface and port to be added to several more SMTP log
3309 lines, for example "SMTP connection from", and rejection lines.
3311 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3313 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3315 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3316 could be overwritten at the end of the current message (or the start of a
3317 new message if it was set in a HELO ACL). The value is now preserved for
3318 the duration of the SMTP connection.
3320 12. If a transport had a headers_rewrite setting, and a matching header line
3321 contained an unqualified address, that address was qualified, even if it
3322 did not match any rewriting rules. The underlying bug was that the values
3323 of the flags that permit the existence of unqualified sender and recipient
3324 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3325 non-local messages, and by -bnq for local messages) were not being
3326 preserved with the message after it was received.
3328 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3329 "next input=" as part of the text comprising the host identity instead of
3330 the correct text. The code was using the same buffer for two different
3331 strings. However, depending on which order the printing function evaluated
3332 its arguments, the bug did not always show up. Under Linux, for example, my
3333 test suite worked just fine.
3335 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3336 doesn't work with some older versions of Perl. It has been changed to "my",
3337 which in any case is probably the better facility to use.
3339 15. A really picky compiler found some instances of statements for creating
3340 error messages that either had too many or two few arguments for the format
3343 16. The size of the buffer for calls to the DNS resolver has been increased
3344 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3345 for addresses that have a lot of PTR records. This alleviates a problem; it
3346 does not fully solve it.
3348 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3349 buffer now truncates the list and logs the incident, which is the same
3350 action as happens when Exim is looking up a host name and its aliases.
3351 Previously in this situation something unpredictable would happen;
3352 sometimes it was "internal error: store_reset failed".
3354 18. If a server dropped the connection unexpectedly when an Exim client was
3355 using GnuTLS and trying to read a response, the client delivery process
3356 crashed while trying to generate an error log message.
3358 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3359 string, for example:
3361 warn message = H1: something\nH2: something
3363 the text was added as a single header line from Exim's point of view
3364 though it ended up OK in the delivered message. However, searching for the
3365 second and subsequent header lines using $h_h2: did not work. This has been
3366 fixed. Similarly, if a system filter added multiple headers in this way,
3367 the routers could not see them.
3369 20. Expanded the error message when iplsearch is called with an invalid key to
3370 suggest using net-iplsearch in a host list.
3372 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3373 ACLs are no longer actually imposed (and a message to that effect is
3376 22. If a "gecos" field in a passwd entry contained escaped characters, in
3377 particular, if it contained a \" sequence, Exim got it wrong when building
3378 a From: or a Sender: header from that name. A second bug also caused
3379 incorrect handling when an unquoted " was present following a character
3380 that needed quoting.
3382 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3383 was not being matched caselessly.
3385 24. Arranged for all hyphens in the exim.8 source to be escaped with
3388 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3389 properly. Recipient callout cache records were still being keyed to include
3390 the sender, even when use_sender was set false. This led to far more
3391 callouts that were necessary. The sender is no longer included in the key
3392 when use_sender is false.
3394 26. Added "control = submission" modifier to ACLs.
3396 27. Added the ${base62d: operator to decode base 62 numbers.
3398 28. dnsdb lookups can now access SRV records.
3400 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3401 the configuration file.
3403 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3404 (-v) mode. This makes the output for a verbose queue run more intelligible.
3406 31. Added a use_postmaster feature to recipient callouts.
3408 32. Added the $body_zerocount variable, containing the number of binary zero
3409 bytes in the message body.
3411 33. The time of last modification of the "new" subdirectory is now used as the
3412 "mailbox time last read" when there is a quota error for a maildir
3415 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3417 35. Added +ignore_unknown as a special item in host lists.
3419 36. Code for decoding IPv6 addresses in host lists is now included, even if
3420 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3421 address was recognized as an IP address, but was then not correctly decoded
3422 into binary, causing unexpected and incorrect effects when compared with
3429 1. Very minor rewording of debugging text in manualroute to say "list of
3430 hosts" instead of "hostlist".
3432 2. If verify=header_syntax was set, and a header line with an unqualified
3433 address (no domain) and a large number of spaces between the end of the
3434 name and the colon was received, the reception process suffered a buffer
3435 overflow, and (when I tested it) crashed. This was caused by some obsolete
3436 code that should have been removed. The fix is to remove it!
3438 3. When running in the test harness, delay a bit after writing a bounce
3439 message to get a bit more predictability in the log output.
3441 4. Added a call to search_tidyup() just before forking a reception process. In
3442 theory, someone could use a lookup in the expansion of smtp_accept_max_
3443 per_host which, without the tidyup, could leave open a database connection.
3445 5. Added the variables $recipient_data and $sender_data which get set from a
3446 lookup success in an ACL "recipients" or "senders" condition, or a router
3447 "senders" option, similar to $domain_data and $local_part_data.
3449 6. Moved the writing of debug_print from before to after the "senders" test
3452 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3453 problems for message scanning, either using a data ACL, or using
3454 local_scan() because the Received: header was not generated till after they
3455 were called (in order to set the time as the time of reception completion).
3456 I have revised the way this works. The header is now generated after the
3457 body is received, but before the ACL or local_scan() are called. After they
3458 are run, the timestamp in the header is updated.
3464 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3465 before starting a queue runner without re-exec. This happened only when
3466 deliver_drop_privilege was set or when the Exim user was set to root. The
3467 effect of the bug was that timeouts during subsequent deliveries caused
3468 crashes instead of being properly handled. The handler is now left at its
3469 default (and expected) setting.
3471 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3472 message, again when deliver_drop_privilege is set or Exim is run as root.
3473 The bug described in (1) was not present in this case, but the tidying up
3474 of the other signals was missing. I have made the two cases consistent.
3476 3. The ignore_target_hosts setting on a manualroute router was being ignored
3477 for hosts that were looked up using the /MX notation.
3479 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3482 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3483 operated on the sender address. After changing the $sender_address to <>
3484 for the sender address verify, Exim was re-instated it as the original
3485 (before rewriting) address, but remembering that it had rewritten it, so it
3486 wasn't rewriting it again. This bug also had the effect of breaking the
3487 sender address verification caching when the sender address was rewritten.
3489 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3490 This has been changed so that if the ip literal address matches
3491 ignore_target_hosts, the router declines.
3493 7. Added expansion conditions match_domain, match_address, and match_local_
3494 part (NOT match_host).
3496 8. The placeholder for the Received: header didn't have a length field set.
3498 9. Added code to Exim itself and to exim_lock to test for a specific race
3499 condition that could lead to file corruption when using MBX delivery. The
3500 issue is with the lockfile that is created in /tmp. If this file is removed
3501 after a process has opened it but before that process has acquired a lock,
3502 there is the potential for a second process to recreate the file and also
3503 acquire a lock. This could lead to two Exim processes writing to the file
3504 at the same time. The added code performs the same test as UW imapd; it
3505 checks after acquiring the lock that its file descriptor still refers to
3506 the same named file.
3508 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3509 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3512 11. Added the smtp_active_hostname option. If used, this will typically be made
3513 to depend on the incoming interface address. Because $interface_address is
3514 not set up until the daemon has forked a reception process, error responses
3515 that can happen earlier (such as "too many connections") no longer contain
3518 12. If an expansion in a condition on a "warn" statement fails because a lookup
3519 defers, the "warn" statement is abandoned, and the next ACL statement is
3520 processed. Previously this caused the whole ACL to be aborted.
3522 13. Added the iplsearch lookup type.
3524 14. Added ident_timeout as a log selector.
3526 15. Added tls_certificate_verified as a log selector.
3528 16. Added a global option tls_require_ciphers (compare the smtp transport
3529 option of the same name). This controls incoming TLS connections.
3531 17. I finally figured out how to make tls_require_ciphers do a similar thing
3532 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3533 before starting the TLS session.
3535 18. Tabs are now shown as \t in -bP output.
3537 19. If the log selector return_path_on_delivery was set, Exim crashed when
3538 bouncing a message because it had too many Received: header lines.
3540 20. If two routers both had headers_remove settings, and the first one included
3541 a superfluous trailing colon, the final name in the first list and the
3542 first name in the second list were incorrectly joined into one item (with a
3543 colon in the middle).
3549 1. Added -C and -D options to the exinext utility, mainly to make it easier
3550 to include in the automated testing, but these could be helpful when
3551 multiple configurations are in use.
3553 2. The exinext utility was not formatting the output nicely when there was
3554 an alternate port involved in the retry record key, nor when there was a
3555 message id as well (for retries that were specific to a specific message
3556 and a specific host). It was also confused by IPv6 addresses, because of
3557 the additional colons they contain. I have fixed the IPv4 problem, and
3558 patched it up to do a reasonable job for IPv6.
3560 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3561 delivery, the log line now contains "pipelined" if PIPELINING was used.
3563 4. An SMTP transport process used to panic and die if the bind() call to set
3564 an explicit outgoing interface failed. This has been changed; it is now
3565 treated in the same way as a connect() failure.
3567 5. A reference to $sender_host_name in the part of a conditional expansion
3568 that was being skipped was still causing a DNS lookup. This no longer
3571 6. The def: expansion condition was not recognizing references to header lines
3572 that used bh_ and bheader_.
3574 7. Added the _cache feature to named lists.
3576 8. The code for checking quota_filecount in the appendfile transport was
3577 allowing one more file than it should have been.
3579 9. For compatibility with Sendmail, the command line option
3587 and sets the incoming protocol and host name (for trusted callers). The
3588 host name and its colon can be omitted when only the protocol is to be set.
3589 Note the Exim already has two private options, -pd and -ps, that refer to
3590 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3591 "s", but I don't think that's a major issue.
3593 10. A number of refactoring changes to the code, none of which should affect
3596 (a) The number of logging options was getting close to filling up the
3597 32-bit word that was used as a bit map. I have split them into two classes:
3598 those that are passed in the argument to log_write(), and those that are
3599 only ever tested independently outside of that function. These are now in
3600 separate 32-bit words, so there is plenty of room for expansion again.
3601 There is no change in the user interface or the logging behaviour.
3603 (b) When building, for example, log lines, the code previously used a
3604 macro that called string_cat() twice, in order to add two strings. This is
3605 not really sufficiently general. Furthermore, there was one instance where
3606 it was actually wrong because one of the argument was used twice, and in
3607 one call a function was used. (As it happened, calling the function twice
3608 did not affect the overall behaviour.) The macro has been replaced by a
3609 function that can join an arbitrary number of extra strings onto a growing
3612 (c) The code for expansion conditions now uses a table and a binary chop
3613 instead of a serial search (which was left over from when there were very
3614 few conditions). Also, it now recognizes conditions like "pam" even when
3615 the relevant support is not compiled in: a suitably worded error message is
3616 given if an attempt is made to use such a condition.
3618 11. Added ${time_interval:xxxxx}.
3620 12. A bug was causing one of the ddress fields not to be passed back correctly
3621 from remote delivery subprocesses. The field in question was not being
3622 subsequently used, so this caused to problems in practice.
3624 13. Added new log selectors queue_time and deliver_time.
3626 14. Might have fixed a bug in maildirsizefile handling that threw up
3627 "unexpected character" debug warnings, and recalculated the data
3628 unnecessarily. In any case, I expanded the warning message to give more
3631 15. Added the message "Restricted characters in address" to the statements in
3632 the default ACL that block characters like @ and % in local parts.
3634 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3635 Three changes have been made:
3637 (a) There was a serious bug; a negative response to MAIL caused the whole
3638 recipient domain to be cached as invalid, thereby blocking all messages
3639 to all local parts at the same domain, from all senders. This bug has
3640 been fixed. The domain is no longer cached after a negative response to
3641 MAIL if the sender used is not empty.
3643 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3646 (c) A new callout option, "use_sender" has been added for people who want
3647 the modified behaviour.
3653 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3656 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3657 indeed breaks things for older releases.
3659 3. Added additional logging to the case where there is a problem reading data
3660 from a filter that is running in a subprocess using a pipe, in order to
3661 try to track down a specific problem.
3663 4. Testing facility fudge: when running in the test harness and attempting
3664 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3665 getting "No route to host". Convert this to a timeout.
3667 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3670 6. Some OS don't have socklen_t but use size_t instead. This affects the
3671 fifth argument of getsockopt() amongst other things. This is now
3672 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3673 can be set for individual OS. I have set it for SunOS5, OSF1, and
3674 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3675 some earlier ones do not.
3677 7. Change 4.30/15 was not doing the test caselessly.
3679 8. The standard form for an IPv6 address literal was being rejected by address
3680 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3681 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3682 this, as well as the form without the "IPv6" on the front (but only when
3683 address literals are enabled, of course).
3685 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3687 10. Exim crashed if a message with an empty sender address specified by -f
3688 encountered a router with an errors_to setting. This could be provoked only
3689 by a command such as
3693 where an empty string was supplied; "<>" did not hit this bug.
3695 11. Installed PCRE release 4.5.
3697 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3698 remained set. It is now erased.
3700 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3701 times from message ids (which are base 36 rather than the normal 62).
3703 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3704 were being counted as actual protocol errors, and logged if the log
3705 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3706 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3707 and DATA following a set of rejected RCPTs do not count as protocol errors.
3708 In other words, Exim assumes they were pipelined, though this may not
3709 actually be the case. Of course, in all cases the client gets an
3710 appropriate error code.
3712 15. If a lookup fails in an ACL condition, a message about the failure may
3713 be available; it is used if testing the ACL cannot continue, because most
3714 such messages specify what the cause of the deferral is. However, some
3715 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3716 that caused an old message to be retained and used if a later statement
3717 caused a defer, replacing the real cause of the deferral.
3719 16. If an IP address had so many PTR records that the DNS lookup buffer
3720 was not large enough to hold them, Exim could crash while trying to process
3721 the truncated data. It now detects and logs this case.
3723 17. Further to 4.21/58, another change has been made: if (and only if) the
3724 first line of a message (the first header line) ends with CRLF, a bare LF
3725 in a subsequent header line has a space inserted after it, so as not to
3726 terminate the header.
3728 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3729 unnecessarily, used atoi() instead of strtol(), and didn't check the
3730 termination when getting file sizes from file names by regex.
3732 19. Completely re-implemented the support for maildirsize files, in the light
3733 of a number of problems with the previous contributed implementation
3734 (4.30/29). In particular:
3736 . If the quota is zero, the maildirsize file is maintained, but no quota is
3739 . If the maildir directory does not exist, it is created before any attempt
3740 to write a maildirsize file.
3742 . The quota value in the file is just a cache; if the quota is changed in
3743 the transport, the new value overrides.
3745 . A regular expression is available for excluding directories from the
3748 20. The autoreply transport checks the characters in options that define the
3749 message's headers; it allows continued headers, but it was checking with
3750 isspace() after an embedded newline instead of explicitly looking for a
3753 21. If all the "regular" hosts to which an address was routed had passed their
3754 expiry times, and had not reached their retry times, the address was
3755 bounced, even if fallback hosts were defined. Now Exim should go on to try
3758 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3759 equivalent code in the SMTP transport. Some hosts send humungous responses
3760 to HELO/EHLO, more than 1024 it seems.
3762 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3763 gives compiler warnings in some environments. I've now done it "properly",
3766 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3767 (because of problems with the built-in one) was declared to return uschar *
3768 instead of char *, causing compiler failure.
3770 25. Fixed a file descriptor leak when processing alias/forward files.
3772 26. Fixed a minor format string issue in dbfn.c.
3774 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3776 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3777 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3780 29. When a sender address is verified, it is cached, to save repeating the test
3781 when there is more than one recipient in a message. However, when the
3782 verification involves a callout, it is possible for different callout
3783 options to be set for different recipients. It is too complicated to keep
3784 track of this in the cache, so now Exim always runs a verification when a
3785 callout is required, relying on the callout cache for the optimization.
3786 The overhead is duplication of the address routing, but this should not be
3789 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3790 to be verified with callout=postmaster, and the main callout worked but the
3791 postmaster check failed, the verification correctly failed. However, if a
3792 subsequent RCPT command asked for sender verification *without* the
3793 postmaster check, incorrect caching caused this verification also to fail,
3796 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3797 it was not caching the DNS options (qualify_single, search_parents) that
3798 were used when the lookup failed. A subsequent lookup with different
3799 options therefore always gave the same answer, though there were cases
3800 where it should not have. (Example: a "domains = !$mx_any" option on a
3801 dnslookup router: the "domains" option is always processed without any
3802 widening, but the router might have qualify_single set.) Now Exim uses the
3803 cached value only when the same options are set.
3805 32. Added John Jetmore's "exipick" utility to the distribution.
3807 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3808 than a timeout (e.g. a certificate is required, and is not provided), an
3809 Exim server now closes the connection immediately. Previously it waited for
3810 the client to close - but if the client is SSL, it seems that they each
3811 wait for each other, leading to a delay before one of them times out.
3813 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3814 maintained 0.8.x compatibility because I don't think many are using it, and
3815 it is clearly obsolete.
3817 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3820 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3821 client certificate was expired. A simple patch fixes this, though I don't
3822 understand the full logic of why the verify callback is called multiple
3825 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3826 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3827 which causes problems with some clients (such as the Certicom SSL Plus
3828 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3829 disables the coutermeasure allowing Eudora to connect."
3831 38. Exim was not checking that a write() to a log file succeeded. This could
3832 lead to Bad Things if a log got too big, in particular if it hit a file
3833 size limit. Exim now panics and dies if it cannot write to a log file, just
3834 as it does if it cannot open a log file.
3836 39. Modified OS/Makefile-Linux so that it now contains
3838 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3840 The two -D definitions ensure that Exim is compiled with large file
3841 support, which makes it possible to handle log files that are bigger than
3844 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3845 instance) a domain was checked against a named list that involved a lookup,
3846 causing $domain_data to be set, then another domain was checked against the
3847 same list, then the first domain was re-checked, the value of $domain_data
3848 after the final check could be wrong. In particular, if the second check
3849 failed, it could be set empty. This bug probably also applied to
3852 41. The strip_trailing_dot option was not being applied to the address given
3853 with the -f command-line option.
3855 42. The code for reading a message's header from the spool was incrementing
3856 $received_count, but never initializing it. This meant that the value was
3857 incorrect (doubled) while delivering a message in the same process in which
3858 it was received. In the most common configuration of Exim, this never
3859 happens - a fresh exec is done - but it can happen when
3860 deliver_drop_privilege is set.
3862 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3863 it now includes up to 150 characters of the unexpected data in the log
3866 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3867 and building data strings. The size of both of these buffers was 10 000
3868 bytes - far larger than anybody would *ever* want, thought I. Needless to
3869 say, somebody hit the limit. I have increased the maximum line length to
3870 20 000 and the maximum data length of concatenated lines to 100 000. I have
3871 also fixed two bugs, because there was no checking on these buffers. Tsk,
3872 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3873 buffer is too small.
3875 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3876 lsearch lookups. Now it does.
3878 46. When parsing a route_list item in a manualroute router, a fixed-length
3879 buffer was used for the list of hosts. I made this 1024 bytes long,
3880 thinking that nobody would ever have a list of hosts that long. Wrong.
3881 Somebody had a whole pile of complicated expansion conditions, and the
3882 string was silently truncated, leading to an expansion error. It turns out
3883 that it is easier to change to an unlimited length (owing to other changes
3884 that have happened since this code was originally written) than to build
3885 structure for giving a limitation error. The length of the item that
3886 expands into the list of hosts is now unlimited.
3888 47. The lsearch lookup could not handle data where the length of text line was
3889 more than 4095 characters. Such lines were truncated, leading to shortened
3890 data being returned. It should now handle lines of any length.
3892 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3893 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3896 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3897 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3898 no longer happens in comment lines. A list of replacements is now placed
3899 at the head of all of the source files, except those whose only change is
3900 to replace PERL_COMMAND in the very first #! line.
3902 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3903 messages on the queue, with a bottom-up merge sort, using code contributed
3904 by Michael Haardt. This should make operations like -bp somewhat faster on
3905 large queues. It won't affect queue runners, except when queue_run_in_order
3908 51. Installed eximstats 1.31 in the distribution.
3910 52. Added support for SRV lookups to the dnslookup router.
3912 53. If an ACL referred to $message_body or $message_body_end, the value was not
3913 reset for any messages that followed in the same SMTP session.
3915 54. The store-handling optimization for building very long strings was not
3916 differentiating between the different store pools. I don't think this
3917 actually made any difference in practice, but I've tidied it.
3919 55. While running the routers to verify a sender address, $sender_address
3920 was still set to the sender address. This is wrong, because when routing to
3921 send a bounce to the sender, it would be empty. Therefore, I have changed
3922 it so that, while verifying a sender address, $sender_address is set to <>.
3923 (There is no change to what happens when verifying a recipient address.)
3925 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3926 target A or AAAA records (if not already returned) without resetting the
3927 qualify_single or search_parents options of the DNS resolver. These are
3928 inappropriate in this case because the targets of MX and SRV records must
3929 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3930 target that, when qualified, matched something in the local domain. These
3931 two options are now turned off when doing these lookups.
3933 57. It seems that at least some releases of Reiserfs (which does not have the
3934 concept of a fixed number of inodes) returns zero and not -1 for the
3935 number of available inodes. This interacted badly with check_spool_inodes,
3936 which assumed that -1 was the "no such thing" setting. What I have done is
3937 to check that the total number of inodes is greater than zero before doing
3938 the test of how many are available.
3940 58. When a "warn" ACL statement has a log_message modifier, the message is
3941 remembered, and not repeated. This is to avoid a lot of repetition when a
3942 message has many recipients that cause the same warning to be written.
3943 Howewer, Exim was preserving the list of already written lines for an
3944 entire SMTP session, which doesn't seem right. The memory is now reset if a
3945 new message is started.
3947 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3948 debugging output unless log_rewrite was also set.
3950 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3951 of (int)(handle) when we know that handle contains (void *)(-1).
3953 61. The Exim daemon panic-logs an error return when it closes the incoming
3954 connection. However "connection reset by peer" seems to be common, and
3955 isn't really an error worthy of noting specially, so that particular error
3958 62. When Exim is trying to find all the local interfaces, it used to panic and
3959 die if the ioctl to get the interface flags failed. However, it seems that
3960 on at least one OS (Solaris 9) it is possible to have an interface that is
3961 included in the list of interfaces, but for which you get a failure error
3962 for this call. This happens when the interface is not "plumbed" into a
3963 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3964 failure of the "get flags" call assumes that the interface is down.
3966 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3967 makes life easier for people who are doing arithmetic on fields extracted
3968 from dates, where you often get leading zeros that should not be
3969 interpreted as octal.
3971 64. Added qualify_domain to the redirect router, to override the global
3974 65. If a pathologically long header line contained very many addresses (the
3975 report of this problem mentioned 10 000) and each of them was rewritten,
3976 Exim could use up a very large amount of memory. (It kept on making new
3977 copies of the header line as it rewrote, and never released the old ones.)
3978 At the expense of a bit more processing, the header rewriting function has
3979 been changed so that it no longer eats memory in this way.
3981 66. The generation of the Received: header has been moved from the time that a
3982 message starts to be received, to the time that it finishes. The timestamp
3983 in the Received: header should now be very close to that of the <= log
3984 line. There are two side-effects of this change:
3986 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3987 logged header lines no longer include the local Received: line, because
3988 it has not yet been created. The same applies to a copy of the message
3989 that is returned to a non-SMTP sender when a message is rejected.
3991 (b) When a filter file is tested using -bf, no additional Received: header
3992 is added to the test message. After some thought, I decided that this
3995 This change does not affect the value of $received_for. It is still set
3996 after address rewriting, but before local_scan() is called.
3998 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
4000 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
4001 gave an unhelpful panic error message, and a defer error. I have managed to
4002 change this behaviour so that it now rejects any supplied certificate,
4003 which seems right, as the list of acceptable certificates is empty.
4005 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
4006 gave an unhelpful defer error. I have not managed to make this reject any
4007 supplied certificates, but the error message it gives is "no certificate
4008 supplied", which is not helpful.
4010 70. exigrep's output now also includes lines that are not associated with any
4011 message, but which match the given pattern. Implemented by a patch from
4012 Martin Sluka, which also tidied up the Perl a bit.
4014 71. Recipient callout verification, like sender verification, was using <> in
4015 the MAIL FROM command. This isn't really the right thing, since the actual
4016 sender may affect whether the remote host accepts the recipient or not. I
4017 have changed it to use the actual sender in the callout; this means that
4018 the cache record is now keyed on a recipient/sender pair, not just the
4019 recipient address. There doesn't seem to be a real danger of callout loops,
4020 since a callout by the remote host to check the sender would use <>.
4021 [SEE ABOVE: changed after hitting problems.]
4023 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
4024 temporary errors. However, in the case of such a code being given after
4025 the end of a data transmission (i.e. after ".") Exim was failing to write
4026 a retry record for the message. (Yes, there was some broken host that was
4027 actually sending 8xx at this point.)
4029 73. An unknown lookup type in a host list could cause Exim to panic-die when
4030 the list was checked. (An example that provoked this was putting <; in the
4031 middle of a list instead of at the start.) If this happened during a DATA
4032 ACL check, a -D file could be left lying around. This kind of configuration
4033 error no longer causes Exim to die; instead it causes a defer errror. The
4034 incident is still logged to the main and panic logs.
4036 74. Buglet left over from Exim 3 conversion. The message "too many messages
4037 in one connection" was written to the rejectlog but not the mainlog, except
4038 when address rewriting (yes!) was being logged.
4040 75. Added write_rejectlog option.
4042 76. When a system filter was run not as root (that is, when system_filter_user
4043 was set), the values of the $n variables were not being returned to the
4044 main process; thus, they were not subsequently available in the $sn
4047 77. Added +return_path_on_delivery log selector.
4049 78. A connection timeout was being treated differently from recipients deferred
4050 when testing hosts_max_try with a message that was older than the host's
4051 retry timeout. (The host should not be counted, thus allowing all hosts to
4052 be tried at least once before bouncing.) This may have been the cause of an
4053 occasionally reported bug whereby a message would remain on the queue
4054 longer than the retry timeout, but would be bounced if a delivery was
4055 forced. I say "may" because I never totally pinned down the problem;
4056 setting up timeout/retry tests is difficult. See also the next item.
4058 79. The ultimate address timeout was not being applied to errors that involved
4059 a combination of host plus message (for example, a timeout on a MAIL
4060 command). When an address resolved to a number of possible hosts, and they
4061 were not all tried for each delivery (e.g. because of hosts_max_try), a
4062 message could remain on the queue longer than the retry timeout.
4064 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4067 81. Fixed an obscure SMTP outgoing bug which required at least the following
4068 conditions: (a) there was another message waiting for the same server;
4069 (b) the server returned 5xx to all RCPT commands in the first message so
4070 that the message was not completed; (c) the server dropped the connection
4071 or gave a negative response to the RSET that Exim sends to abort the
4072 transaction. The observed case was a dropped connection after DATA that had
4073 been sent in pipelining mode. That is, the server had advertised PIPELINING
4074 but was not implementing it correctly. The effect of the bug was incorrect
4075 behaviour, such as trying another host, and this could lead to a crash.
4081 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4082 and daemon.c were passed as pointers to ints; they should have been
4083 pointers to socklen_t variables (which are typically unsigned ints).
4085 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4088 3. Fixed a really odd bug that affected only the testing scheme; patching a
4089 certain fixed string in the binary changed the value of another string that
4090 happened to be identical to the end of the original first string.
4092 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4093 name", it returns that address as the IP address. On some operating
4094 systems (e.g. Solaris), it also passes back the IP address string as the
4095 "host name". However, on others (e.g. Linux), it passes back an empty
4096 string. Exim wasn't checking for this, and was changing the host name to an
4097 empty string, assuming it had been canonicized.
4099 5. Although rare, it is permitted to have more than one PTR record for a given
4100 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4101 all the names associated with an address, because they do in Solaris.
4102 However, it seems that they do not in Linux for data that comes from the
4103 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4104 I found this out when I moved to a new Linux workstation and tried to run
4105 the Exim test suite.
4107 To get round this problem I have changed the code so that it now does its
4108 own call to the DNS to look up PTR records when searching for a host name.
4109 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4110 addresses that are only in /etc/hosts are still found.
4112 This behaviour is, however, controlled by an option called host_lookup_
4113 order, which defaults to "bydns:byaddr". If people want to use the other
4114 order, or indeed, just use one or the other means of lookup, they can
4115 specify it in this variable.
4117 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4118 some operating systems, this comes back from gethostbyaddr() as an empty
4119 string, and this is what Exim used to test for. However, it seems that in
4120 other systems, "." is yielded. Exim now tests for this case too.
4122 7. The values of check_spool_space and check_log_space are now held internally
4123 as a number of kilobytes instead of an absolute number of bytes. If a
4124 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4125 kilobyte. This means that much larger values can be stored.
4127 8. Exim monitor: an attempt to get the action menu when not actually pointing
4128 at a message produces an empty menu entitled "No message selected". This
4129 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4130 no entries in it ("Shell widget menu has zero width and/or height"). So I
4131 have added a single, blank menu entry in this case.
4133 9. Added ${quote_local_part.
4135 10. MIME decoding is now applied to the contents of Subject: header lines when
4138 11. Now that a reference to $sender_host_address automatically causes a reverse
4139 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4140 host lookup before query-style lookups in lists that might use this
4141 variable. This has therefore been abolished, and the "net-" prefix is no
4142 longer necessary for query-style lookups.
4144 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4145 have been a typo for LFLAGS, so it has been changed.
4147 13. The install script calls Exim with "-C /dev/null" in order to find the
4148 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4149 to be output. Howeve, since Exim outputs its version number before the
4150 error, it didn't break the script. It just looked ugly. I fixed this by
4151 always allowing "-C /dev/null" if the caller is root.
4153 14. Ignore overlarge ACL variable number when reading spool file - insurance
4154 against a later release with more variables having written the file.
4156 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4157 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4158 this, as well as the form without the "IPv6" on the front.
4160 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4161 OS/Makefile-Darwin file.
4163 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4164 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4166 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4167 variable. However, some OS use ints for inodes. I've added cast to long int
4168 to get rid of the compiler warning.
4170 19. I had forgotten to lock out "/../" in configuration file names when
4171 ALT_CONFIG_PREFIX was set.
4173 20. Routers used for verification do not need to specify transports. However,
4174 if such a router generated a host list, and callout was configured, Exim
4175 crashed, because it could not find a port number from the (non-existent)
4176 transport. It now assumes port 25 in this circumstance.
4178 21. Added the -t option to exigrep.
4180 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4181 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4182 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4183 them is set, LOOKUP_LSEARCH is forced.
4185 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4187 24. Added sender and host information to the "rejected by local_scan()" log
4188 line; previously there was no indication of these.
4190 25. Added .include_if_exists.
4192 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4193 Linux. It turns out that not all file systems support this. Apparently some
4194 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4195 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4196 not supported on the file descriptor, is now ignored when Exim is trying to
4197 sync a directory. This applies only to Linux.
4199 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4201 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4202 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4205 29. Added support for maildirsize files from supplied patch (modified a bit).
4207 30. The use of :fail: followed by an empty string could lead Exim to respond to
4208 sender verification failures with (e.g.):
4210 550 Verification failed for <xxx>
4211 550 Sender verify failed
4213 where the first response line was missing the '-' that indicates it is not
4214 the final line of the response.
4216 31. The loop for finding the name of the user that called Exim had a hardwired
4217 limit of 10; it now uses the value of finduser_retries, which is used for
4218 all other user lookups.
4220 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4223 33. Exim was neglecting to zero errno before one call of strtol() when
4224 expanding a string and expecting an integer value. On some systems this
4225 resulted in spurious "integer overflow" errors. Also, it was casting the
4226 result into an int without checking.
4228 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4229 did not work. The code looks as if it has *never* worked, though it appears
4230 to have been documented since at least releast 1.62. I have made it work.
4232 35. The "timeout_DNS" error in retry rules, also documented since at least
4233 1.62, also never worked. As it isn't clear exactly what this means, and
4234 clearly it isn't a major issue, I have abolished the feature by treating it
4235 as "timeout", and writing a warning to the main and panic logs.
4237 36. The display of retry rules for -brt wasn't always showing the error code
4240 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4241 timeout_connect_A, timeout_connect_MX.
4243 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4244 to the empty sender.
4246 39. The daemon was not analysing the content of -oX till after it had closed
4247 stderr and disconnected from the controlling terminal. This meant that any
4248 syntax errors were only noted on the panic log, and the return code from
4249 the command was 0. By re-arranging the code a little, I've made the
4250 decoding happen first, so such errors now appear on stderr, and the return
4251 code is 1. However, the actual setting up of the sockets still happens in
4252 the disconnected process, so errors there are still only recorded on the
4255 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4256 connections (as happens on some IP stacks) was logged at start up time as
4257 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4258 it from "IPv6 and IPv4", which means that two separate sockets are being
4261 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4262 says whether AF_INET or AF_INET6 was passed as an argument.
4264 42. Exiwhat output was messed up when time zones were included in log
4267 43. Exiwhat now gives more information about the daemon's listening ports,
4268 and whether -tls-on-connect was used.
4270 44. The "port" option of the smtp transport is now expanded.
4272 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4273 silently ignored. Now an error message is written to the main and panic
4276 46. There's a new ACL modifier called "logwrite" which writes to a log file
4277 as soon as it is encountered.
4279 47. Added $local_user_uid and $local_user_gid at routing time.
4281 48. Exim crashed when trying to verify a sender address that was being
4284 49. Exim was recognizing only a space character after ".include". It now also
4285 recognizes a tab character.
4287 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4288 extracting the relevant information from the specification. The man page no
4289 longer contains scrambled data for the -d option, and I've added a section
4290 at the front about calling Exim under different names.
4292 51. Added "extra_headers" argument to the "mail" command in filter files.
4294 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4297 53. Installed eximstats 1.29.
4299 54. Added transport_filter_timeout as a generic transport option.
4301 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4302 Cc: header lines. This was required by RFC 822, but it not required by RFC
4305 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4306 incoming messages that did not have them. Now it does so only if the
4307 message originates locally, that is, if there is no associated remote host
4308 address. When Resent- header lines are present, this applies to the Resent-
4309 lines rather than the non-Resent- lines.
4311 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4312 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4314 58. Messages for configuration errors now include the name of the main
4315 configuration file - useful now that there may be more than one file in a
4316 list (.included file names were always shown).
4318 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4319 for those rare installations that do not start the daemon as root or run it
4320 setuid root. I've cut out the call to initgroups() if the daemon is not
4323 60. The Exim user and group can now be bound into the binary as text strings
4324 that are looked up at the start of Exim's processing.
4326 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4328 62. Added $mailstore_basename variable.
4330 63. Installed patch to sieve.c from Michael Haardt.
4332 64. When Exim failed to open the panic log after failing to open the main log,
4333 the original message it was trying to log was written to stderr and debug
4334 output, but if they were not available (the usual case in production), it
4335 was lost. Now it is written to syslog before the two lines that record the
4336 failures to open the logs.
4338 65. Users' Exim filters run in subprocesses under the user's uid. It is
4339 possible for a "deliver" command or an alias in a "personal" command to
4340 provoke an address rewrite. If logging of address rewriting is configured,
4341 this fails because the process is not running as root or exim. There may be
4342 a better way of dealing with this, but for the moment (because 4.30 needs
4343 to be released), I have disabled address rewrite logging when running a
4344 filter in a non-root, non-exim process.
4350 1. The buildconfig auxiliary program wasn't quoting the value set for
4351 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4352 not defined. This bug was masked in 4.22 by the effect that was fixed in
4355 2. Some messages that were rejected after a message id was allocated were
4356 shown as "incomplete" by exigrep. It no longer does this for messages that
4357 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4359 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4360 have allow_domain_literals set, the ID did not get logged in the <= line.
4361 Domain literals are now always recognized in Message-ID: header lines.
4363 4. The first argument for a ${extract expansion item is the key name or field
4364 number. Leading and trailing spaces in this item were not being ignored,
4365 causing some misleading effects.
4367 5. When deliver_drop_privilege was set, single queue runner processes started
4368 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4369 same command in the process it spins off) were not dropping privilege.
4371 6. When the daemon running as "exim" started a queue runner, it always
4372 re-executed Exim in the spun-off process. This is a waste of effort when
4373 deliver_drop_privilege is set. The new process now just calls the
4374 queue-runner function directly.
4380 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4383 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4384 ignored. Though the use of -oP was forcing the writing of a pid file, it
4385 was always written to the default place.
4387 3. If the message "no IP address found for host xxxx" is generated during
4388 incoming verification, it is now followed by identification of the incoming
4389 connection (so you can more easily find what provoked it).
4391 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4393 5. Added some features to "harden" Exim a bit more against certain attacks:
4395 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4396 be put in Local/Makefile. This is like the never_users runtime option,
4397 but it cannot be overridden. The default setting is "root".
4399 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4400 prefix string with which any file named in a -C command line option
4403 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4404 is retained for -C and -D only if the caller of Exim is root. Without
4405 it, the exim user may also use -C and -D and retain privilege.
4407 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4408 command line option is disabled.
4410 6. Macro names set by the -D option must start with an upper case letter, just
4411 like macro names defined in the configuration file.
4413 7. Added "dereference=" facility to LDAP.
4415 8. Two instances of the typo "uknown" in the source files are fixed.
4417 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4418 the Configure-Makefile script screwed up while processing it.
4420 10. Incorporated PCRE 4.4.
4422 11. The SMTP synchronization check was not operating right at the start of an
4423 SMTP session. For example, it could not catch a HELO sent before the client
4424 waited for the greeting. There is now a check for outstanding input at the
4425 point when the greeting is written. Because of the duplex, asynchronous
4426 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4427 way, but not yet received, when the check is performed.
4429 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4430 on TCP/IP sockets, because this apparently causes some broken clients to
4433 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4434 unchanged) from the Cygwin maintainer.
4436 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4437 was supported instead of testing for mbx. Effectively a typo.
4439 15. The spa authenticator server code was not checking that the input it
4440 received was valid base64.
4442 16. The debug output line for the "set" modifier in ACLs was not showing the
4443 name of the variable that was being set.
4445 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4447 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4448 lookup on its own. Something else was needed to trigger the lookup. For
4449 example, a match in host_lookup or the need for a host name in a host list.
4450 Now, if $sender_host_name is referenced and the host name has not yet been
4451 looked up, a lookup is performed. If the lookup fails, the variable remains
4452 empty, and $host_lookup_failed is set to "1".
4454 19. Added "eqi" as a case-independent comparison operator.
4456 20. The saslauthd authentication condition could segfault if neither service
4457 nor realm was specified.
4459 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4460 error message that was logged was misleading, and incoming SMTP
4461 connections were dropped. The message is now more accurate, and temporary
4462 errors are given to SMTP connections.
4464 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4465 (except RSET) with a 421 error, until QUIT is received. However, it was
4466 failing to send a response to QUIT.
4468 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4469 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4470 ACL is now run after the helo_try_verify_hosts code.
4472 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4473 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4474 case-independent, so other case variants are also recognized). Apparently
4475 some systems use these upper case variants.
4477 25. If more than two messages were waiting for the same host, and a transport
4478 filter was specified for the transport, Exim sent two messages over the
4479 same TCP/IP connection, and then failed with "socket operation on non-
4480 socket" when it tried to send the third.
4482 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4484 27. The extern definition of crypt16() in expand.c was not being excluded when
4485 the OS had its own crypt16() function.
4487 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4488 as a preferred synonym for return_size_limit, both as an option and as an
4491 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4493 30. Changed the default configuration ACL to relax the local part checking rule
4494 for addresses that are not in any local domains. For these addresses,
4495 slashes and pipe symbols are allowed within local parts, but the sequence
4496 /../ is explicitly forbidden.
4498 31. SPA server authentication was not clearing the challenge buffer before
4501 32. log_message in a "warn" ACL statement was writing to the reject log as
4502 well as to the main log, which contradicts the documentation and doesn't
4503 seem right (because no rejection is happening). So I have stopped it.
4505 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4506 However, I am unable to do any testing of this.
4508 34. Fixed an infelicity in the appendfile transport. When checking directories
4509 for a mailbox, to see if any needed to be created, it was accidentally
4510 using path names with one or more superfluous leading slashes; tracing
4511 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4513 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4514 contents are added to the log line that is written for every discarded
4515 recipient. (Previously a log_message setting was ignored.)
4517 36. The ${quote: operator now quotes the string if it is empty.
4519 37. The install script runs exim in order to find its version number. If for
4520 some reason other than non-existence or emptiness, which it checks, it
4521 could not run './exim', it was installing it with an empty version number,
4522 i.e. as "exim-". This error state is now caught, and the installation is
4525 38. An argument was missing from the function that creates an error message
4526 when Exim fails to connect to the socket for saslauthd authentication.
4527 This could cause Exim to crash, or give a corrupted message.
4529 39. Added isip, isip4, and isip6 to ${if conditions.
4531 40. The ACL variables $acl_xx are now saved with the message, and can be
4532 accessed later in routers, transports, and filters.
4534 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4535 strings in the file are not string-expanded.
4537 42. If a MAIL command specified a SIZE value that was too large to fit into an
4538 int variable, the check against message_size_limit failed. Such values are
4539 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4540 day this will have to be increased, but I don't think I want to be around
4541 when emails are that large.
4548 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4549 iconv() is not standard in FreeBSD.
4551 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4552 IPv6 enabled. The observed symptom was a segmentation fault on return from
4553 the function os_common_find_running_interfaces() in src/os.c.
4555 3. In the check_special_case() function in daemon.c I had used "errno" as an
4556 argument name, which causes warnings on some systems. This was basically a
4557 typo, since it was named "eno" in the comments!
4559 4. The code that waits for the clock to tick (at a resolution of some fraction
4560 of a second) so as to ensure message-id uniqueness was always waiting for
4561 at least one whole tick, when it could have waited for less. [This is
4562 almost certainly not relevant at current processor speeds, where it is
4563 unlikely to ever wait at all. But we try to future-proof.]
4565 5. The function that sleeps for a time interval that includes fractions of a
4566 second contained a race. It did not block SIGALRM between setting the
4567 timer, and suspending (a couple of lines later). If the interval was short
4568 and the sigsuspend() was delayed until after it had expired, the suspension
4569 never ended. On busy systems this could lead to processes getting stuck for
4572 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4573 process, before it forks any delivery processes. The open lookup caching
4574 mechanism meant that the open file or database connection was passed into
4575 the delivery process. The problem was that delivery processes always tidy
4576 up cached lookup data. This could cause a problem for the next delivery
4577 process started by the queue runner, because the external queue runner
4578 process does not know about the closure. So the next delivery process
4579 still has data in the lookup cache. In the case of a file lookup, there was
4580 no problem because closing a file descriptor in a subprocess doesn't affect
4581 the parent. However, if the lookup was caching a connection to a database,
4582 the connection was closed, and the second delivery process was likely to
4583 see errors such as "PGSQL: query failed: server closed the connection
4584 unexpectedly". The problem has been fixed by closing all cached lookups
4585 in a queue runner before running a delivery process.
4587 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4588 seem to have the "const" qualifier which it has on other OS. I've
4591 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4592 *of the same type* (client or server) with the same public name that an
4593 error should be diagnosed.
4595 9. When Exim looked up a host name for an IP address, but failed to find the
4596 original IP address when looking up the host name (a safety check), it
4597 output the message "<ip address> does not match any IP for NULL", which was
4598 confusing, to say the least. The bug was that the host name should have
4599 appeared instead of "NULL".
4601 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4602 user that is built into the binary, and the -C or -D options is used, root
4603 privilege is dropped before the configuration file is read. In addition,
4604 logging is switched to stderr instead of the normal log files. If the
4605 configuration then re-defines the Exim user, the unprivileged environment
4606 is probably not what is expected, so Exim logs a panic warning message (but
4609 However, if deliver_drop_privilege is set, the unprivileged state may well
4610 be exactly what is intended, so the warning has been cut out in that case,
4611 and Exim is allowed to try to write to its normal log files.
4617 1. smtp_return_error_details was not giving details for temporary sender
4618 or receiver verification errors.
4620 2. Diagnose a configuration error if two authenticators have the same public
4623 3. Exim used not to create the message log file for a message until the first
4624 delivery attempt. This could be confusing when incoming messages were held
4625 for policy or load reasons. The message log file is now created at the time
4626 the message is received, and an initial "Received" line is written to it.
4628 4. The automatically generated man page for command line options had a minor
4629 bug that caused no ill effects; however, a more serious problem was that
4630 the procedure for building the man page automatically didn't always
4631 operate. Consequently, release 4.20 contains an out-of-date version. This
4632 shouldn't happen again.
4634 5. When building Exim with embedded Perl support, the script that builds the
4635 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4636 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4638 6. The freeze_tell option was not being used for messages that were frozen on
4639 arrival, either by an ACL or by local_scan().
4641 7. Added the smtp_incomplete_transaction log selector.
4643 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4644 was accepting AUTH without a new EHLO.
4646 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4647 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4650 10. Logging of TCP/IP connections (when configured) now happens in the main
4651 daemon process instead of the child process, so that the TCP/IP connection
4652 count is more accurate (but it can never be perfect).
4654 11. The use of "drop" in a nested ACL was not being handled correctly in the
4655 outer ACL. Now, if condition failure induced by the nested "drop" causes
4656 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4657 or "require"), the connection is dropped.
4659 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4660 that yield "discard" can now be used with an "accept" or a "discard" verb,
4661 but an error is generated for any others (because I can't see a useful way
4662 to define what should happen).
4664 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4665 are now processed in the same way as lines in the Exim configuration file.
4666 In particular, continuation lines are supported.
4668 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4670 15. Added -ti meaning -t -i.
4672 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4673 domains, and warn by logging if others are found.
4675 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4676 returned from GIFCONF when Exim is trying to find the list of interfaces on
4677 a host. The code in os.c has been modified to copy each ifreq to an aligned
4678 structure in all cases.
4680 Also, in some cases, the returned ifreq's were being copied to a 'struct
4681 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4682 means the last couple of bytes of an IPv6 address could be chopped if the
4683 ifreq contained only a normal sockaddr (14 bytes storage).
4685 18. Named domain lists were not supported in the hosts_treat_as_local option.
4686 An entry such as +xxxx was not recognized, and was treated as a literal
4689 19. Ensure that header lines added by a DATA ACL are included in the reject log
4690 if the ACL subsequently rejects the message.
4692 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4693 MD5 (which is deprecated).
4695 21. When testing a filter file using -bf, Exim was writing a message when it
4696 took the sender from a "From " line in the message, but it was not doing so
4697 when it took $return_path from a Return-Path: header line. It now does.
4699 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4700 with a valid header line field name (a series of printing characters
4701 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4703 23. Changed "disc" in the source to "disk" to conform to the documentation and
4704 the book and for uniformity.
4706 24. Ignore Sendmail's -Ooption=value command line item.
4708 25. When execve() failed while trying to run a command in a pipe transport,
4709 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4710 could be confused with a return value of 69 from the command itself. This
4711 has been changed to 127, the value the shell returns if it is asked to run
4712 a non-existent command. The wording for the related log line suggests a
4713 non-existent command as the problem.
4715 26. If received_header_text expands to an empty string, do not add a Received:
4716 header line to the message. (Well, it adds a token one on the spool, but
4717 marks it "old" so that it doesn't get used or transmitted.)
4719 27. Installed eximstats 1.28 (addition of -nt option).
4721 28. There was no check for failure on the call to getsockname() in the daemon
4722 code. This can fail if there is a shortage of resources on the system, with
4723 ENOMEM, for example. A temporary error is now given on failure.
4725 29. Contrary to the C standard, it seems that in some environments, the
4726 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4727 program. Exim now does this explicitly; it affects the formatting of
4728 timestamps using strftime().
4730 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4731 complaints. I've now initialized all the variables, to avoid this.
4733 32. Header lines added by a system filter were not being "seen" during
4734 transport-time rewrites.
4736 33. The info_callback() function passed to OpenSSL is set up with type void
4737 (*)(SSL *, int, int), as described somewhere. However, when calling the
4738 function (actually a macro) that sets it up, the type void(*)() is
4739 expected. I've put in a cast to prevent warnings from picky compilers.
4741 34. If a DNS black list lookup found a CNAME record, but there were no A
4742 records associated with the domain it pointed at, Exim crashed.
4744 35. If a DNS black list lookup returned more than one A record, Exim ignored
4745 all but the first. It now scans all returned addresses if a particular IP
4746 value is being sought. In this situation, the contents of the
4747 $dnslist_value variable are a list of all the addresses, separated by a
4750 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4751 to accept a host name and all its aliases if the forward lookup for any of
4752 them yielded the IP address of the incoming connection. Now it accepts only
4753 those names whose forward lookup yields the correct IP address. Any other
4754 names are discarded. This closes a loophole whereby a rogue DNS
4755 administrator could create reverse DNS records to break through a
4756 wildcarded host restriction in an ACL.
4758 37. If a user filter or a system filter that ran in a subprocess used any of
4759 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4760 the wrong values were passed to the pipe command ($thisaddress had the
4761 value of $0, $0 had the value of $1, etc). This bug was introduced by
4762 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4764 38. Improved the line breaking for long SMTP error messages from ACLs.
4765 Previously, if there was no break point between 40 and 75 characters, Exim
4766 left the rest of the message alone. Two changes have been made: (a) I've
4767 reduced the minimum length to 35 characters; (b) if it can't find a break
4768 point between 35 and 75 characters, it looks ahead and uses the first one
4769 that it finds. This may give the occasional overlong line, but at least the
4770 remaining text gets split now.
4772 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4773 file descriptors might be low, and that setting 1000 would always raise it.
4774 It turns out that in some environments, the limit is already over 1000 and
4775 that lowering it causes trouble. So now Exim takes care not to decrease it.
4777 40. When delivering a message, the value of $return_path is set to $sender_
4778 address at the start of routing (routers may change the value). By an
4779 oversight, this default was not being set up when an address was tested by
4780 -bt or -bv, which affected the outcome if any router or filter referred to
4783 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4784 log only when "message" or "log_message" are set. However, if one of the
4785 conditions was an address verification, or a call to a nested ACL, the
4786 messages generated by the underlying test were being passed through. This
4787 no longer happens. The underlying message is available in $acl_verify_
4788 message for both "message" and "log_message" expansions, so it can be
4789 passed through if needed.
4791 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4792 new expansion $bh_ to give the encoded byte string without charset
4793 translation. Translation happens only if iconv() is available; HAVE_ICONV
4794 indicates this at build time. HEADERS_CHARSET gives the charset to
4795 translate to; headers_charset can change it in the configuration, and
4796 "headers charset" can change it in an individual filter file.
4798 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4799 that creates RFC 2047 encoded "words" labels them as that charset instead
4800 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4801 expansion operator; (ii) when Exim creates a From: line for a local
4802 message; (iii) when a header line is rewritten to include a "phrase" part.
4804 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4805 buggy, causing it to skip the first lines of messages whose message ID
4806 ended in 'D'. This would not have bitten before Exim release 4.14, because
4807 message IDs were unlikely to end in 'D' before then. The effect was to have
4808 incorrect size information for certain domains.
4810 45. #include "config.h" was missing at the start of the crypt16.c module. This
4811 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4814 46. If there was a timeout during a "random" callout check, Exim treated it as
4815 a failure of the random address, and carried on sending RSET and the real
4816 address. If the delay was just some slowness somewhere, the response to the
4817 original RCPT would be taken as a response to RSET and so on, causing
4818 mayhem of various kinds.
4820 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4821 when I implemented it. It didn't allow for the fact that some option values
4822 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4823 the right test for positive values.
4825 48. Domain names in DNS records are case-independent. Exim always looks them up
4826 in lower case. Some resolvers return domain names in exactly the case they
4827 appear in the zone file, that is, they may contain uppercase letters. Not
4828 all resolvers do this - some return always lower case. Exim was treating a
4829 change of case by a resolver as a change of domain, similar to a widening
4830 of a domain abbreviation. This triggered its re-routing code and so it was
4831 trying to route what was effectively the same domain again. This normally
4832 caused routing to fail (because the router wouldn't handle the domain
4833 twice). Now Exim checks for this case specially, and just changes the
4834 casing of the domain that it ultimately uses when it transmits the message
4837 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4840 50. If a filter generated a file delivery with a non-absolute name (possible if
4841 no home directory exists for the router), the forbid_file option was not
4844 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4845 the existing equality matching.
4847 52. Exim was using ints instead of ino_t variables in some places where it was
4848 dealing with inode numbers.
4850 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4851 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4852 called TMPDIR, and if it finds it is different, it changes its value.
4854 54. The smtp_printf() function is now made available to local_scan() so
4855 additional output lines can be written before returning. There is also an
4856 smtp_fflush() function to enable the detection of a dropped connection.
4857 The variables smtp_input and smtp_batched_input are exported to
4860 55. Changed the default runtime configuration: the message "Unknown user"
4861 has been removed from the ACL, and instead placed on the localuser router,
4862 using the cannot_route_message feature. This means that any verification
4863 failures that generate their own messages won't get overridden. Similarly,
4864 the "Unrouteable address" message that was in the ACL for unverifiable
4865 relay addresses has also been removed.
4867 56. Added hosts_avoid_esmtp to the smtp transport.
4869 57. The exicyclog script was not checking for the esoteric option
4870 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4871 will work only if exicyclog is run under the appropriate euid.
4873 58. Following a discussion on the list, the rules by which Exim recognises line
4874 endings on incoming messages have been changed. The -dropcr and drop_cr
4875 options are now no-ops, retained only for backwards compatibility. The
4876 following line terminators are recognized: LF CRLF CR. However, special
4877 processing applies to CR:
4879 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4880 nor a local message in the state where . is a terminator.
4882 (ii) If a bare CR is encountered in a header line, an extra space is added
4883 after the line terminator so as not to end the header. The reasoning
4884 behind this is that bare CRs in header lines are most likely either
4885 to be mistakes, or people trying to play silly games.
4887 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4888 was being incorrectly given as 18 bytes larger than it should have been.
4889 This is a VOB (very old bug).
4891 60. This may never have affected anything current, but just in case it has:
4892 When the local host is found other than at the start of a list of hosts,
4893 the local host, those with the same MX, and any that follow, are discarded.
4894 When the list in question was part of a longer list of hosts, the following
4895 hosts (not currently being processed) were also being discarded. This no
4896 longer happens. I'm not sure if this situation could ever has previously
4899 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4902 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4903 header. This is something that is recommended in a new Internet Draft, and
4904 is something that is documented as being done by Sendmail. There are two
4905 possible values. For messages generated by the autoreply transport, Exim
4908 Auto-Submitted: auto-replied
4910 whereas for all other generated messages (e.g. bounces) it adds
4912 Auto-Submitted: auto-generated
4914 63. The "personal" condition in filters now includes a test for the
4915 Auto-Submitted: header. If it contains the string "auto-" the message it
4916 not considered personal.
4918 64. Added rcpt_include_affixes as a generic transport option.
4920 65. Added queue_only_override (default true).
4922 66. Added the syslog_duplication option.
4924 67. If what should have been the first header line of a message consisted of
4925 a space followed by a colon, Exim was mis-interpreting it as a header line.
4926 It isn't of course - it is syntactically invalid and should therefore be
4927 treated as the start of the message body. The misbehaviour could have
4928 caused a number of strange effects, including loss of data in subsequent
4929 header lines, and spool format errors.
4931 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4932 client host had authenticated. This control can now be exercised by an ACL
4933 for more flexibility.
4935 69. By default, callouts do not happen when testing with -bh. There is now a
4936 variant, -bhc, which does actually run the callout code, including
4937 consulting and updating the callout cache.
4939 70. Added support for saslauthd authentication, courtesy of Alexander
4942 71. If statvfs() failed on the spool or log directories while checking their
4943 size for availability, Exim confusingly gave the error "space shortage".
4944 Furthermore, in debugging mode it crashed with a floating point exception.
4945 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4946 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4947 serious problem, Exim now writes to the main and panic logs when this
4948 happens, with details of the failure. It then refuses to accept the
4949 incoming message, giving the message "spool directory problem" or "log
4950 directory problem" with a 421 code for SMTP messages.
4952 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4953 0, 1, and 2 exist, because some OS complain for execs without them (see
4954 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4955 descriptors. However, the code omitted to check that the open succeeded,
4956 causing mysterious errors if for some reason the permissions on /dev/null
4957 got screwed. Now Exim writes a message to the main and panic logs, and
4958 bombs out if it can't open /dev/null.
4960 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4961 interact so that it is all more flexible. It is supposed to remain
4962 backwards compatible. Also added extra_local_interfaces.
4964 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4965 to bomb out with an assertion failure - to the client this appears as a
4966 connection drop. This problem occurs in the part of the code that was taken
4967 from the Samba project. Fortunately, the assertion is in a very simple
4968 function, so I have fixed this by reproducing the function inline in the
4969 one place where it is called, and arranging for authentication to fail
4970 instead of killing the process with assert().
4972 75. The SPA client code was not working when the server requested OEM rather
4973 than Unicode encoding.
4975 76. Added code to make require_files with a specific uid setting more usable in
4976 the case where statting the file as root fails - usually a non-root-mounted
4977 NFS file system. When this happens and the failure is EACCES, Exim now
4978 forks a subprocess and does the per-uid checking as the relevant uid.
4980 77. Added process_log_path.
4982 78. If log_file_path was not explicitly set, a setting of check_log_space or
4983 check_log_inodes was ignored.
4985 79. If a space check for the spool or log partitions fails, the incident is now
4986 logged. Of course, in the latter case the data may get lost...
4988 80. Added the %p formatting code to string_format() so that it can be used to
4989 print addresses in debug_print(). Adjusted all the address printing in the
4990 debugging in store.c to use %p rather than %d.
4992 81. There was a concern that a line of code in smtp_in.c could overflow a
4993 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4994 initially expressed, the concern was not well-founded, because trailing
4995 spaces are removed early. However, if the trailing spaces were followed by
4996 a NULL, they did not get removed, so the overflow was possible. Two fixes
4999 (a) I re-wrote the offending code in a cleaner fashion.
5000 (b) If an incoming SMTP command contains a NULL character, it is rejected
5003 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
5004 runs initgroups(), so that if the Exim user is in any additional groups,
5005 they will be used during message reception.
5011 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob