1 $Cambridge: exim/doc/doc-txt/ChangeLog,v 1.568 2009/10/14 13:52:48 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 614: 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/18 Bugzilla 745: TLS version reporting
114 TK/01 Add preliminary DKIM support. Currently requires a forked version of
115 ALT-N's libdkim that I have put here:
116 http://duncanthrax.net/exim-experimental/
118 Note to Michael Haardt: I had to rename some vars in sieve.c. They
119 were called 'true' and it seems that C99 defines that as a reserved
120 keyword to be used with 'bool' variable types. That means you could
121 not include C99-style headers which use bools without triggering
122 build errors in sieve.c.
124 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
125 as mailq or other aliases. Changed the --help handling significantly
126 to do whats expected. exim_usage() emits usage/help information.
128 SC/01 Added the -bylocaldomain option to eximstats.
130 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr
132 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp
134 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall)
140 PH/01 Another patch from the Sieve maintainer.
142 PH/02 When an IPv6 address is converted to a string for single-key lookup
143 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
144 dots are used instead of colons so that keys in lsearch files need not
145 contain colons. This was done some time before quoting was made available
146 in lsearch files. However, iplsearch files do require colons in IPv6 keys
147 (notated using the quote facility) so as to distinguish them from IPv4
148 keys. This meant that lookups for IP addresses in host lists did not work
149 for iplsearch lookups.
151 This has been fixed by arranging for IPv6 addresses to be expressed with
152 colons if the lookup type is iplsearch. This is not incompatible, because
153 previously such lookups could never work.
155 The situation is now rather anomolous, since one *can* have colons in
156 ordinary lsearch keys. However, making the change in all cases is
157 incompatible and would probably break a number of configurations.
159 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
162 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
163 conversion specification without a maximum field width, thereby enabling
164 a rogue spamd server to cause a buffer overflow. While nobody in their
165 right mind would setup Exim to query an untrusted spamd server, an
166 attacker that gains access to a server running spamd could potentially
167 exploit this vulnerability to run arbitrary code as the Exim user.
169 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
170 $primary_hostname instead of what libspf2 thinks the hosts name is.
172 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
173 a directory entry by the name of the lookup key. Previously, if a
174 symlink pointed to a non-existing file or a file in a directory that
175 Exim lacked permissions to read, a lookup for a key matching that
176 symlink would fail. Now it is enough that a matching directory entry
177 exists, symlink or not. (Bugzilla 503.)
179 PH/03 The body_linecount and body_zerocount variables are now exported in the
182 PH/04 Added the $dnslist_matched variable.
184 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
185 This means they are set thereafter only if the connection becomes
188 PH/06 Added the client_condition to authenticators so that some can be skipped
189 by clients under certain conditions.
191 PH/07 The error message for a badly-placed control=no_multiline_responses left
192 "_responses" off the end of the name.
194 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
196 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
197 (without spaces) instead of just copying the configuration text.
199 PH/10 Added the /noupdate option to the ratelimit ACL condition.
201 PH/11 Added $max_received_linelength.
203 PH/12 Added +ignore_defer and +include_defer to host lists.
205 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
206 way in which PCRE > 7.0 is built.
208 PH/14 Implemented queue_only_load_latch.
210 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
211 MAIL command. The effect was to mangle the value on 64-bit systems.
213 PH/16 Another patch from the Sieve maintainer.
215 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
217 PH/18 If a system quota error occurred while trying to create the file for
218 a maildir delivery, the message "Mailbox is full" was not appended to the
219 bounce if the delivery eventually timed out. Change 4.67/27 below applied
220 only to a quota excession during the actual writing of the file.
222 PH/19 It seems that peer DN values may contain newlines (and other non-printing
223 characters?) which causes problems in log lines. The DN values are now
224 passed through string_printing() before being added to log lines.
226 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
227 and InterBase are left for another time.)
229 PH/21 Added message_body_newlines option.
231 PH/22 Guard against possible overflow in moan_check_errorcopy().
233 PH/23 POSIX allows open() to be a macro; guard against that.
235 PH/24 If the recipient of an error message contained an @ in the local part
236 (suitably quoted, of course), incorrect values were put in $domain and
237 $local_part during the evaluation of errors_copy.
243 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
244 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
247 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
248 issue a MAIL command.
250 PH/02 In an ACL statement such as
252 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
254 if a client was not listed at all, or was listed with a value other than
255 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
256 the condition was not true (as it should be), so access was not denied.
257 The bug was that the ! inversion was incorrectly passed on to the second
258 item. This has been fixed.
260 PH/03 Added additional dnslists conditions == and =& which are different from
261 = and & when the dns lookup returns more than one IP address.
263 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
264 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
266 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
267 FSYNC, which compiles an option called disable_fsync that allows for
268 bypassing fsync(). The documentation is heavily laced with warnings.
270 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
272 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
273 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
274 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
275 including adding "make clean"; (3) Added -fPIC when compiling the test
276 dynamically loaded module, to get rid of a warning.
278 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
279 message fails, move_frozen_messages = true and ignore_bounce_errors_after
280 = 0s. The bug is otherwise harmless.
282 PH/07 There was a bug in the dovecot authenticator such that the value of
283 $auth1 could be overwritten, and so not correctly preserved, after a
284 successful authentication. This usually meant that the value preserved by
285 the server_setid option was incorrect.
287 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
289 PH/09 Installed PCRE release 7.0.
291 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
292 run for batched SMTP input. It is now run at the start of every message
293 in the batch. While fixing this I discovered that the process information
294 (output by running exiwhat) was not always getting set for -bs and -bS
295 input. This is fixed, and it now also says "batched" for BSMTP.
297 PH/11 Added control=no_pipelining.
299 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
300 patch, slightly modified), and move the expansion of helo_data till after
301 the connection is made in the smtp transport (so it can use these
304 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
306 PH/14 Added log_selector = +pid.
308 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
310 PH/16 Add ${if forany and ${if forall.
312 PH/17 Added dsn_from option to vary the From: line in DSNs.
314 PH/18 Flush SMTP output before performing a callout, unless control =
315 no_callout_flush is set.
317 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
318 was true (the default) a successful delivery failed to delete the retry
319 item, thus causing premature timeout of the address. The bug is now
322 PH/20 Added hosts_avoid_pipelining to the smtp transport.
324 PH/21 Long custom messages for fakedefer and fakereject are now split up
325 into multiline reponses in the same way that messages for "deny" and
326 other ACL rejections are.
328 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
329 with slight modification.
331 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
332 draft, changing the syntax and factoring some duplicate code".
334 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
335 for deliveries of the second and subsequent messages over the same SMTP
338 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
339 ${reduce, with only minor "tidies".
341 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
343 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
344 expansion side effects.
346 PH/27 When a message times out after an over-quota error from an Exim-imposed
347 quota, the bounce message says "mailbox is full". This message was not
348 being given when it was a system quota that was exceeded. It now should
351 MH/03 Made $recipients available in local_scan(). local_scan() already has
352 better access to the recipient list through recipients_list[], but
353 $recipients can be useful in postmaster-provided expansion strings.
355 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
356 in the case of a MAIL command with additional options following the
357 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
358 were accidentally chopped off.
360 PH/29 SMTP synchronization checks are implemented when a command is read -
361 there is a check that no more input is waiting when there shouldn't be
362 any. However, for some commands, a delay in an ACL can mean that it is
363 some time before the response is written. In this time, more input might
364 arrive, invalidly. So now there are extra checks after an ACL has run for
365 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
366 pipelining has not been advertised.
368 PH/30 MH's patch to allow iscntrl() characters to be list separators.
370 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
371 returned in the SMTP response when smtp_return_error_details was false.
374 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
375 instead of the C I/O that was originally supplied, because problems were
378 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
379 Exim which did not show up earlier: it was assuming that a call to
380 SSL_CTX_set_info_callback() might give an error value. In fact, there is
381 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
382 was a macro that became an assignment, so it seemed to work. This has
383 changed to a proper function call with a void return, hence the compile
384 error. Exim's code has been fixed.
386 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
389 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
391 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
393 PH/37 If a message is not accepted after it has had an id assigned (e.g.
394 because it turns out to be too big or there is a timeout) there is no
395 "Completed" line in the log. When some messages of this type were
396 selected by exigrep, they were listed as "not completed". Others were
397 picked up by some special patterns. I have improved the selection
398 criteria to be more general.
400 PH/38 The host_find_failed option in the manualroute router can now be set
401 to "ignore", to completely ignore a host whose IP address cannot be
402 found. If all hosts are ignored, the behaviour is controlled by the new
403 host_all_ignored option.
405 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
406 homing or because of multiple MX records with /mx) generated more than
407 one IP address, and the following item turned out to be the local host,
408 all the secondary addresses of the first item were incorrectly removed
409 from the list, along with the local host and any following hosts (which
410 is what is supposed to happen).
412 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
413 whoever called Exim into the -H file. In the case of the daemon it was
414 behaving confusingly. When first started, it used values for whoever
415 started the daemon, but after a SIGHUP it used the Exim user (because it
416 calls itself on a restart). I have changed the code so that it now always
419 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
420 message are rejected with the same error (e.g. no authentication or bad
421 sender address), and a DATA command is nevertheless sent (as can happen
422 with PIPELINING or a stupid MUA), the error message that was given to the
423 RCPT commands is included in the rejection of the DATA command. This is
424 intended to be helpful for MUAs that show only the final error to their
427 PH/42 Another patch from the Sieve maintainer.
429 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
430 Eximstats - Fixed some broken HTML links and added missing column headers
432 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
433 columns for Rejects, Temp Rejects, Ham, and Spam rows.
435 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
437 PH/43 Yet another patch from the Sieve maintainer.
439 PH/44 I found a way to check for a TCP/IP connection going away before sending
440 the response to the final '.' that terminates a message, but only in the
441 case where the client has not sent further data following the '.'
442 (unfortunately, this is allowed). However, in many cases there won't be
443 any further data because there won't be any more messages to send. A call
444 to select() can be used: if it shows that the input is "ready", there is
445 either input waiting, or the socket has been closed. An attempt to read
446 the next input character can distinguish the two cases. Previously, Exim
447 would have sent an OK response which the client would never have see.
448 This could lead to message repetition. This fix should cure that, at
449 least in a lot of common cases.
451 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
452 advertised in response to EHLO.
458 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
459 fixed by 4.65/MH/01 (is this a record?) are fixed:
461 (i) An empty string was always treated as zero by the numeric comparison
462 operators. This behaviour has been restored.
464 (ii) It is documented that the numeric comparison operators always treat
465 their arguments as decimal numbers. This was broken in that numbers
466 starting with 0 were being interpreted as octal.
468 While fixing these problems I realized that there was another issue that
469 hadn't been noticed. Values of message_size_limit (both the global option
470 and the transport option) were treated as octal if they started with 0.
471 The documentation was vague. These values are now always treated as
472 decimal, and I will make that clear in the documentation.
478 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
479 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
482 MH/01 Don't check that the operands of numeric comparison operators are
483 integers when their expansion is in "skipping" mode (fixes bug
484 introduced by 4.64-PH/07).
486 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
487 child addresses, Exim now panics and dies. Previously, because the count
488 is held in a short int, deliveries were likely to be lost. As such a
489 large number of recipients for a single message is ridiculous
490 (performance will be very, very poor), I have chosen to impose a limit
491 rather than extend the field.
497 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
498 leftover -K file (the existence of which was triggered by #402).
499 While we were at it, introduced process PID as part of the -K
500 filename. This should rule out race conditions when creating
503 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
504 processing considerably. Previous code took too long for large mails,
505 triggering a timeout which in turn triggers #401.
507 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
508 in the DK code in transports.c. sendfile() is not really portable,
509 hence the _LINUX specificness.
511 TF/01 In the add_headers option to the mail command in an Exim filter,
512 there was a bug that Exim would claim a syntax error in any
513 header after the first one which had an odd number of characters
516 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
517 callout verification, Exim cached a "reject" for the entire domain. This
518 is correct for most verifications, but it is not correct for a recipient
519 verification with use_sender or use_postmaster set, because in that case
520 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
521 case of MAIL FROM:<> rejection from other early rejections (e.g.
522 rejection of HELO). When verifying a recipient using a non-null MAIL
523 address, the cache is ignored if it shows MAIL FROM:<> rejection.
524 Whatever the result of the callout, the value of the domain cache is
525 left unchanged (for any other kind of callout, getting as far as trying
526 RCPT means that the domain itself is ok).
528 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
531 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
532 manifest itself as EPIPE rather than ECONNECT. When tidying away a
533 session, the daemon ignores ECONNECT errors and logs others; it now
534 ignores EPIPE as well.
536 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
537 (quoted-printable decoding).
539 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
540 later the small subsequent patch to fix an introduced bug.
542 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
544 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
546 PH/08 An error is now given if message_size_limit is specified negative.
548 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
549 to be given (somewhat) arbitrary names.
551 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
554 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
555 miscellaneous code fixes
557 PH/10 Added the log_reject_target ACL modifier to specify where to log
560 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
561 hostname. This is wrong, because it relates to the incoming message (and
562 probably the interface on which it is arriving) and not to the outgoing
563 callout (which could be using a different interface). This has been
564 changed to use the value of the helo_data option from the smtp transport
565 instead - this is what is used when a message is actually being sent. If
566 there is no remote transport (possible with a router that sets up host
567 addresses), $smtp_active_hostname is used.
569 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
570 tweaks were necessary in order to get it to work (see also 21 below):
571 (a) The code assumed that strncpy() returns a negative number on buffer
572 overflow, which isn't the case. Replaced with Exim's string_format()
574 (b) There were several signed/unsigned issues. I just did the minimum
575 hacking in of casts. There is scope for a larger refactoring.
576 (c) The code used strcasecmp() which is not a standard C function.
577 Replaced with Exim's strcmpic() function.
578 (d) The code set only $1; it now sets $auth1 as well.
579 (e) A simple test gave the error "authentication client didn't specify
580 service in request". It would seem that Dovecot has changed its
581 interface. Fortunately there's a specification; I followed it and
582 changed what the client sends and it appears to be working now.
584 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
587 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
588 address A is aliased to B and C, where B exists and C does not. Without
589 -v the output is "A verified" because verification stops after a
590 successful redirection if more than one address is generated. However,
591 with -v the child addresses are also verified. Exim was outputting "A
592 failed to verify" and then showing the successful verification for C,
593 with its parentage. It now outputs "B failed to verify", showing B's
594 parentage before showing the successful verification of C.
596 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
597 look up a TXT record in a specific list after matching in a combined
600 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
601 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
602 they consult the DNS. I had assumed they would set it the way they
603 wanted; and indeed my experiments on Linux seem to show that in some
604 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
605 To be on the safe side, however, I have now made the interface to
606 host_find_byname() similar to host_find_bydns(), with an argument
607 containing the DNS resolver options. The host_find_byname() function now
608 sets these options at its start, just as host_find_bydns() does. The smtp
609 transport options dns_qualify_single and dns_search_parents are passed to
610 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
611 of host_find_byname() use the default settings of RES_DEFNAMES
612 (qualify_single) but not RES_DNSRCH (search_parents).
614 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
615 spool_read_header() do less string testing, by means of a preliminary
616 switch on the second character of optional "-foo" lines. (This is
617 overdue, caused by the large number of possibilities that now exist.
618 Originally there were few.) While I was there, I also converted the
619 str(n)cmp tests so they don't re-test the leading "-" and the first
620 character, in the hope this might squeeze out yet more improvement.
622 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
623 flag allowing group syntax was set by the header_syntax check but not
624 turned off, possible causing trouble later; (2) The flag was not being
625 set at all for the header_verify test, causing "group"-style headers to
626 be rejected. I have now set it in this case, and also caused header_
627 verify to ignore an empty address taken from a group. While doing this, I
628 came across some other cases where the code for allowing group syntax
629 while scanning a header line wasn't quite right (mostly, not resetting
630 the flag correctly in the right place). These bugs could have caused
631 trouble for malformed header lines. I hope it is now all correct.
633 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
634 with the "reply" argument non-NULL. The code, however (which originally
635 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
636 but it didn't always do it. This confused somebody who was copying the
637 code for some other use. I have removed all the tests.
639 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
640 feature that was used to support insecure browsers during the U.S. crypto
641 embargo. It requires special client support, and Exim is probably the
642 only MTA that supported it -- and would never use it because real RSA is
643 always available. This code has been removed, because it had the bad
644 effect of slowing Exim down by computing (never used) parameters for the
645 RSA_EXPORT functionality.
647 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
648 authenticator to fail if there's a tab character in the incoming data
649 (there should never be unless someone is messing about, as it's supposed
650 to be base64-encoded). Also added, on Timo's advice, the "secured" option
651 if the connection is using TLS or if the remote IP is the same as the
652 local IP, and the "valid-client-cert option" if a client certificate has
655 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
656 authenticators. This can be used for authorization after authentication
657 succeeds. (In the case of plaintext, it servers for both authentication
660 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
661 if any retry times were supplied.
663 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
664 connection, where there is no client IP address to check. In this
665 situation, the verify now always succeeds.
667 PH/25 Applied John Jetmore's -Mset patch.
669 PH/26 Added -bem to be like -Mset, but loading a message from a file.
671 PH/27 In a string expansion for a processed (not raw) header when multiple
672 headers of the same name were present, leading whitespace was being
673 removed from all of them, but trailing whitespace was being removed only
674 from the last one. Now trailing whitespace is removed from each header
675 before concatenation. Completely empty headers in a concatenation (as
678 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
679 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
681 PH/29 [Removed. This was a change that I later backed out, and forgot to
682 correct the ChangeLog entry (that I had efficiently created) before
683 committing the later change.]
685 PH/30 Exim was sometimes attempting to deliver messages that had suffered
686 address errors (4xx response to RCPT) over the same connection as other
687 messages routed to the same hosts. Such deliveries are always "forced",
688 so retry times are not inspected. This resulted in far too many retries
689 for the affected addresses. The effect occurred only when there were more
690 hosts than the hosts_max_try setting in the smtp transport when it had
691 the 4xx errors. Those hosts that it had tried were not added to the list
692 of hosts for which the message was waiting, so if all were tried, there
693 was no problem. Two fixes have been applied:
695 (i) If there are any address or message errors in an SMTP delivery, none
696 of the hosts (tried or untried) are now added to the list of hosts
697 for which the message is waiting, so the message should not be a
698 candidate for sending over the same connection that was used for a
699 successful delivery of some other message. This seems entirely
700 reasonable: after all the message is NOT "waiting for some host".
701 This is so "obvious" that I'm not sure why it wasn't done
702 previously. Hope I haven't missed anything, but it can't do any
703 harm, as the worst effect is to miss an optimization.
705 (ii) If, despite (i), such a delivery is accidentally attempted, the
706 routing retry time is respected, so at least it doesn't keep
707 hammering the server.
709 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
710 in ${readsocket because some servers need this prod.
712 PH/32 Added some extra debug output when updating a wait-xxx database.
714 PH/33 The hint "could be header name not terminated by colon", which has been
715 given for certain expansion errors for a long time, was not being given
716 for the ${if def:h_colon_omitted{... case.
718 PH/34 The spec says: "With one important exception, whenever a domain list is
719 being scanned, $domain contains the subject domain." There was at least
720 one case where this was not true.
722 PH/35 The error "getsockname() failed: connection reset by peer" was being
723 written to the panic log as well as the main log, but it isn't really
724 panic-worthy as it just means the connection died rather early on. I have
725 removed the panic log writing for the ECONNRESET error when getsockname()
728 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
729 runs only) independently of the message's sender address. This meant
730 that, if the 4xx error was in fact related to the sender, a different
731 message to the same recipient with a different sender could confuse
732 things. In particualar, this can happen when sending to a greylisting
733 server, but other circumstances could also provoke similar problems.
734 I have changed the default so that the retry time for these errors is now
735 based a combination of the sender and recipient addresses. This change
736 can be overridden by setting address_retry_include_sender=false in the
739 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
740 remote server are returned as part of bounce messages. This was not
741 happening for LMTP over a pipe (the lmtp transport), but now it is the
742 same for both kinds of LMTP.
744 PH/38 Despite being documented as not happening, Exim was rewriting addresses
745 in header lines that were in fact CNAMEs. This is no longer the case.
747 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
748 and queue runs started by the daemon processed all messages. This has
749 been fixed so that -R and -S can now usefully be given with -q<time>.
751 PH/40 Import PCRE release 6.7 (fixes some bugs).
753 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
755 PH/42 Give an error if -q is specified more than once.
757 PH/43 Renamed the variables $interface_address and $interface_port as
758 $received_ip_address and $received_port, to make it clear that these
759 values apply to message reception, and not to the outgoing interface when
760 a message is delivered. (The old names remain recognized, of course.)
762 PH/44 There was no timeout on the connect() call when using a Unix domain
763 socket in the ${readsocket expansion. There now is.
765 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
766 be meaningful with "accept".
768 SC/01 Eximstats V1.43
769 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
771 SC/02 Eximstats V1.44
772 Use a glob alias rather than an array ref in the generated
773 parser. This improves both readability and performance.
775 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
776 Collect SpamAssassin and rejection statistics.
777 Don't display local sender or destination tables unless
778 there is data to show.
779 Added average volumes into the top table text output.
781 SC/04 Eximstats V1.46
782 Collect data on the number of addresses (recipients)
783 as well as the number of messages.
785 SC/05 Eximstats V1.47
786 Added 'Message too big' to the list of mail rejection
787 reasons (thanks to Marco Gaiarin).
789 SC/06 Eximstats V1.48
790 Mainlog lines which have GMT offsets and are too short to
791 have a flag are now skipped.
793 SC/07 Eximstats V1.49 (Alain Williams)
794 Added the -emptyok flag.
796 SC/08 Eximstats V1.50
797 Fixes for obtaining the IP address from reject messages.
799 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
800 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
801 whitesspace changes from 4.64-PH/27
803 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
806 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
807 are found, allow negative numbers in numeric criteria)
809 JJ/06 exipick.20061117.2, added new $message_body_missing variable
811 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
812 to match changes made in 4.64-PH/43
814 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
816 PH/47 Put in an explicit test for a DNS lookup of an address record where the
817 "domain" is actually an IP address, and force a failure. This locks out
818 those revolvers/nameservers that support "A-for-A" lookups, in
819 contravention of the specifications.
821 PH/48 When a host name was looked up from an IP address, and the subsequent
822 forward lookup of the name timed out, the host name was left in
823 $sender_host_name, contrary to the specification.
825 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
826 restricted to single-key lookups, Exim was not diagnosing an error if
827 * or *@ was used with a query-style lookup.
829 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
831 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
832 long ago, but noone interested enough thought of it. Let's just say that
833 the "1.1" means that there are some new functions that weren't there at
834 some point in the past.
836 PH/51 Error processing for expansion failure of helo_data from an smtp
837 transport during callout processing was broken.
839 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
840 tested/used via the -bh/-bhc/-bs options.
842 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
843 bug, fixed in subsequent PCRE releases).
845 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
846 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
848 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
854 SC/01 Use a glob alias rather than an array ref in eximstats generated
855 parser. This improves both readability and performance.
857 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
858 Don't display local sender or destination tables in eximstats unless
859 there is data to show.
860 Added average volumes into the eximstats top table text output.
862 SC/03 Collect data on the number of addresses (recipients) as well
863 as the number of messages in eximstats.
865 TF/01 Correct an error in the documentation for the redirect router. Exim
866 does (usually) call initgroups() when daemonizing.
868 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
869 with consistent privilege compared to when running as a daemon.
871 TF/03 Note in the spec that $authenticated_id is not set for local
872 submissions from trusted users.
874 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
875 Thanks to Dean Brooks <dean@iglou.com> for the patch.
877 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
878 by adding some example configuration directives to the default
879 configuration file. A little bit of work is required to uncomment the
880 directives and define how usernames and passwords are checked, but
881 there is now a framework to start from.
883 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
884 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
885 without this. I don't know how relevant this is to other LDAP libraries.
887 PH/02 Add the verb name to the "unknown ACL verb" error.
889 PH/03 Magnus Holmgren's patch for filter_prepend_home.
891 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
893 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
894 directory not expanded when it should be if an expanded home directory
895 was set for the address (which is overridden by the transport).
897 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
900 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
901 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
902 because it is too late at that time, and has no effect.
904 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
905 security issue with \' (bugzilla #107). I could not use the
906 PQescapeStringConn() function, because it needs a PGconn value as one of
909 PH/08 When testing addresses using -bt, indicate those final addresses that
910 are duplicates that would not cause an additional delivery. At least one
911 person was confused, thinking that -bt output corresponded to deliveries.
912 (Suppressing duplicates isn't a good idea as you lose the information
913 about possibly different redirections that led to the duplicates.)
915 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
916 systems where poll() doesn't work, in particular OS X.
918 PH/10 Added more information to debugging output for retry time not reached.
920 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
921 operations in malware.c.
923 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
926 PH/13 If write_rejectlog was set false when logging was sent to syslog with
927 syslog_duplication set false, log lines that would normally be written
928 both the the main log and to the reject log were not written to syslog at
931 PH/14 In the default configuration, change the use of "message" in ACL warn
932 statements to "add_header".
934 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
935 not followed by a command (e.g. "seen endif").
937 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
938 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
941 PH/17 Added extra conditions to the default value of delay_warning_condition
945 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
946 { match{$h_precedence:}{(?i)bulk|list|junk} } \
947 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
950 The Auto-Submitted: and various List- headers are standardised, whereas I
951 don't think Precedence: ever was.
953 PH/18 Refactored debugging code in route_finduser() to show more information,
954 in particular, the error code if getpwnam() issues one.
956 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
957 This is apparently needed in addition to the PH/07 change above to avoid
958 any possible encoding problems.
960 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
961 but not after initializing Perl.
963 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
964 output them only if debugging. By default they are written stderr,
965 apparently, which is not desirable.
967 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
970 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
973 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
975 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
976 authenticated or an ident call has been made. Suppress the default
977 values for $authenticated_id and $authenticated_sender (but permit -oMai
978 and -oMas) when testing with -bh.
980 PH/24 Re-jigged the order of the tests in the default configuration so that the
981 tests for valid domains and recipients precede the DNS black list and CSA
982 tests, on the grounds that those ones are more expensive.
984 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
985 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
986 This bug exists in every version of Exim that I still have, right back to
989 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
990 However, an attempt to turn on case-sensitivity in a regex key by
991 including (?-i) didn't work because the subject string was already
992 lowercased, and the effects were non-intuitive. It turns out that a
993 one-line patch can be used to allow (?-i) to work as expected.
999 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
1000 other effects) broke the use of negated acl sub-conditions.
1002 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
1005 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
1006 "Deny" causes Exim to reject the incoming connection with a 554 error.
1007 Unfortunately, if there is a major crisis, such as a disk failure,
1008 tcp-wrappers gives "deny", whereas what one would like would be some
1009 kind of temporary error. A kludge has been added to help with this.
1010 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
1011 554 error is used if errno is still zero or contains ENOENT (which occurs
1012 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
1015 PH/03 Add -lutil to the default FreeBSD LIBS setting.
1017 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
1018 errors. Otherwise a message that provokes a temporary error (when other
1019 messages do not) can cause a whole host to time out.
1021 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
1022 addresses were routed directly to files or pipes from a redirect router.
1023 File deliveries just didn't batch; pipe deliveries might have suffered
1026 PH/06 A failure to get a lock for a hints database would erroneously always say
1027 "Failed to get write lock", even when it was really a read lock.
1029 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
1030 of 0600. This has been changed to use the value of the lockfile_mode
1031 option (which defaults to 0600).
1033 PH/08 Applied small patch from the Sieve maintainer.
1035 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
1036 folder from quota calculations, a direct delivery into this folder messed
1037 up the contents of the maildirsize file. This was because the regex was
1038 used only to exclude .Trash (or whatever) when the size of the mailbox
1039 was calculated. There was no check that a delivery was happening into an
1040 excluded directory. This bug has been fixed by ignoring all quota
1041 processing for deliveries into excluded directories.
1043 PH/10 Added the maildirfolder_create_regex option to appendfile.
1049 PH/01 The code for finding all the local interface addresses on a FreeBSD
1050 system running IPv6 was broken. This may well have applied to all BSD
1051 systems, as well as to others that have similar system calls. The broken
1052 code found IPv4 interfaces correctly, but gave incorrect values for the
1053 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
1054 that it would not match correctly against @[] and not recognize the IPv6
1057 PH/02 The ipliteral router was not recognizing addresses of the form user@
1058 [ipv6:....] because it didn't know about the "ipv6:" prefix.
1060 PH/03 Added disable_ipv6.
1062 PH/04 Changed $reply_address to use the raw form of the headers instead of the
1063 decoded form, because it is most often used to construct To: headers
1064 lines in autoreplies, and the decoded form may well be syntactically
1065 invalid. However, $reply_address has leading white space removed, and all
1066 newlines turned into spaces so that the autoreply transport does not
1069 PH/05 If group was specified without a user on a router, and no group or user
1070 was specified on a transport, the group from the router was ignored.
1072 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
1073 for visible compile-time settings that can be used to change these
1074 numbers, for those that want even more. Backwards compatibility with old
1075 spool files has been maintained. However, going back to a previous Exim
1076 release will lost any variables that are in spool files.
1078 PH/07 Two small changes when running in the test harness: increase delay when
1079 passing a TCP/IP connection to a new process, in case the original
1080 process has to generate a bounce, and remove special handling of
1081 127.0.0.2 (sic), which is no longer necessary.
1083 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
1084 be the same on different OS.
1086 PH/09 Moved a debug statement in filter processing to avoid a race problem when
1089 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
1090 whether --show-vars was specified or not
1092 JJ/02 exipick: Added support for new ACL variable spool format introduced
1095 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
1096 syntactically invalid From: or Reply-to: line, and a filter used this to
1097 generate an autoreply, and therefore failed to obtain an address for the
1098 autoreply, Exim could try to deliver to a non-existent relative file
1099 name, causing unrelated and misleading errors. What now happens is that
1100 it logs this as a hard delivery error, but does not attempt to create a
1103 PH/11 The exinext utility has a -C option for testing purposes, but although
1104 the given file was scanned by exinext itself; it wasn't being passed on
1105 when Exim was called.
1107 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
1108 an end-of-file indication when reading a command response.
1110 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
1111 compiled. In many other places in Exim, IPv6 addresses are always
1112 recognized, so I have changed this. It also means that IPv4 domain
1113 literals of the form [IPV4:n.n.n.n] are now always recognized.
1115 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
1116 used if the router is not running as root, for example, when verifying at
1117 ACL time, or when using -bh. The debugging output from this situation was
1118 non-existent - all you got was a failure to exec. I have made two
1121 (a) Failures to set uid/gid, the current directory, or a process leader
1122 in a subprocess such as that created by queryprogram now generate
1123 suitable debugging ouput when -d is set.
1125 (b) The queryprogram router detects when it is not running as root,
1126 outputs suitable debugging information if -d is set, and then runs
1127 the subprocess without attempting to change uid/gid.
1129 PH/15 Minor change to Makefile for building test_host (undocumented testing
1132 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
1133 additional section of a DNS packet that returns MX or SRV records.
1134 Instead, it always explicitly searches for A/AAAA records. This avoids
1135 major problems that occur when a DNS server includes only records of one
1136 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
1137 fixed another bug: if SRV records were looked up and the corresponding
1138 address records were *not* found in the additional section, the port
1139 values from the SRV records were lost.
1141 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
1142 using the correct key (the original address) when searching the retry
1143 rules in order to find which one to use for generating the retry hint.
1145 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
1146 adding the default one. Similarly, if it contains a Reply-To: header, the
1147 errors_reply_to option, if set, is not used.
1149 PH/19 When calculating a retry time, Exim used to measure the "time since
1150 failure" by looking at the "first failed" field in the retry record. Now
1151 it does not use this if it is later than than the arrival time of the
1152 message. Instead it uses the arrival time. This makes for better
1153 behaviour in cases where some deliveries succeed, thus re-setting the
1154 "first failed" field. An example is a quota failure for a huge message
1155 when small messages continue to be delivered. Without this change, the
1156 "time since failure" will always be short, possible causing more frequent
1157 delivery attempts for the huge message than are intended.
1158 [Note: This change was subsequently modified - see PH/04 for 4.62.]
1160 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
1161 $1, $2, $3) because the numerical variables can be reset during some
1162 expansion items (e.g. "match"), thereby losing the authentication data.
1164 PH/21 Make -bV show the size of off_t variables so that the test suite can
1165 decide whether to run tests for quotas > 2G.
1167 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
1168 mailbox_size, and mailbox_filecount in the appendfile transport. If a
1169 filecount value is greater than 2G or if a quota value is greater than 2G
1170 on a system where the size of off_t is not greater than 4, a panic error
1173 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
1174 never match. The debug and -bh output now contains an explicit error
1175 message indicating a malformed IPv4 address or mask.
1177 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
1178 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
1179 PH/23 above applies.
1181 PH/25 Do not write to syslog when running in the test harness. The only
1182 occasion when this arises is a failure to open the main or panic logs
1183 (for which there is an explicit test).
1185 PH/26 Added the /no_tell option to "control=freeze".
1187 PH/27 If a host name lookup failed very early in a connection, for example, if
1188 the IP address matched host_lookup and the reverse lookup yielded a name
1189 that did not have a forward lookup, an error message of the form "no IP
1190 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
1191 could be logged. Now it outputs the IP address instead of "NULL".
1193 PH/28 An enabling patch from MH: add new function child_open_exim2() which
1194 allows the sender and the authenticated sender to be set when
1195 submitting a message from within Exim. Since child_open_exim() is
1196 documented for local_scan(), the new function should be too.
1198 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
1199 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
1200 results in an empty string is now treated as unset.
1202 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
1204 PH/31 Added sender_verify_fail logging option.
1206 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
1207 needed by RFC 822 but not by RFC 2822 was commented out. I have now
1208 tidied the source and removed it altogether.
1210 PH/33 When a queue run was abandoned because the load average was too high, a
1211 log line was always written; now it is written only if the queue_run log
1212 selector is set. In addition, the log line for abandonment now contains
1213 information about the queue run such as the pid. This is always present
1214 in "start" and "stop" lines but was omitted from the "abandon" line.
1216 PH/34 Omit spaces between a header name and the colon in the error message that
1217 is given when verify = headers_syntax fails (if there are lots of them,
1218 the message gets confusing).
1220 PH/35 Change the default for dns_check_names_pattern to allow slashes within
1221 names, as there are now some PTR records that contain slashes. This check
1222 is only to protect against broken name servers that fall over on strange
1223 characters, so the fact that it applies to all lookups doesn't matter.
1225 PH/36 Now that the new test suite is complete, we can remove some of the
1226 special code in Exim that was needed for the old test suite. For example,
1227 sorting DNS records because real resolvers return them in an arbitrary
1228 order. The new test suite's fake resolver always returns records in the
1231 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
1232 bounces) except when queue_only is set, to avoid logging races between
1233 the different processes.
1235 PH/38 Panic-die if .include specifies a non-absolute path.
1237 PH/39 A tweak to the "H" retry rule from its user.
1239 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
1240 a label. They prevented compilation on older perls.
1242 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
1243 a warning to be raised on newish perls.
1245 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
1246 on queue. Changes to match documented behaviour of showing count of
1247 messages matching specified criteria.
1249 PH/40 Changed the default ident timeout from 30s to 5s.
1251 PH/41 Added support for the use of login_cap features, on those BSD systems
1252 that have them, for controlling the resources used by pipe deliveries.
1254 PH/42 The content-scanning code uses fopen() to create files in which to put
1255 message data. Previously it was not paying any attention to the mode of
1256 the files. Exim runs with umask(0) because the rest of the code creates
1257 files with open(), and sets the required mode explicitly. Thus, these
1258 files were ending up world-writeable. This was not a big issue, because,
1259 being within the spool directory, they were not world-accessible. I have
1260 created a function called modefopen, which takes an additional mode
1261 argument. It sets umask(777), creates the file, chmods it to the required
1262 mode, then resets the umask. All the relevant calls to fopen() in the
1263 content scanning code have been changed to use this function.
1265 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
1266 to 24 hours. This avoids potential overflow problems when processing G
1267 and H retry rules. I suspect nobody ever tinkers with this value.
1269 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
1271 PH/45 When the plaintext authenticator is running as a client, the server's
1272 challenges are checked to ensure they are valid base64 strings. By
1273 default, the authentication attempt is cancelled if an invalid string is
1274 received. Setting client_ignore_invalid_base64 true ignores these errors.
1275 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
1276 they are received. Thus, the responses can be made to depend on the
1277 challenges. If an invalid string is ignored, an empty string is placed in
1280 PH/46 Messages that are created by the autoreply transport now contains a
1281 References: header, in accordance with RFCs 2822 and 3834.
1283 PH/47 Added authenticated_sender_force to the smtp transport.
1285 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
1287 PH/49 Installed latest patch from the Sieve maintainer.
1289 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
1290 was also set, the appendfile transport was unnecessarily scanning a
1291 directory of message files (e.g. for maildir delivery) to find the count
1292 of files (along with the size), even though it did not need this
1293 information. It now does the scan only if it needs to find either the
1294 size of the count of files.
1296 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
1298 PH/52 Two bugs concerned with error handling when the smtp transport is
1301 (i) Exim was not creating retry information for temporary errors given
1302 for individual recipients after the DATA command when the smtp transport
1303 was used in LMTP mode. This meant that they could be retried too
1304 frequently, and not timed out correctly.
1306 (ii) Exim was setting the flag that allows error details to be returned
1307 for LMTP errors on RCPT commands, but not for LMTP errors for individual
1308 recipients that were returned after the DATA command.
1310 PH/53 This is related to PH/52, but is more general: for any failing address,
1311 when detailed error information was permitted to be returned to the
1312 sender, but the error was temporary, then after the final timeout, only
1313 "retry timeout exceeded" was returned. Now it returns the full error as
1314 well as "retry timeout exceeded".
1316 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
1317 do this, and (what is worse) MTAs that accept it.
1319 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
1320 will now be deprecated.
1322 PH/56 New os.c-cygwin from the Cygwin maintainer.
1324 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
1325 formats (previously only available in exim formats via -bpr, -bpru,
1326 and -bpra. Now also available in native and exiqgrep formats)
1328 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
1329 with very large, slow to parse queues
1331 JJ/08 exipick: added ! as generic prefix to negate any criteria format
1333 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
1335 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
1336 responses to authentication challenges, though it was showing the
1337 challenges; (ii) I've removed the CR characters from the debug output for
1340 PH/58 Allow for the insertion of a newline as well as a space when a string
1341 is turned into more than one encoded-word during RFC 2047 encoding. The
1342 Sieve code now uses this.
1344 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
1345 data_4xx, lost_connection, tls_required.
1347 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
1348 message was being sent as an SMTP response.
1350 PH/61 Add -l and -k options to exicyclog.
1352 PH/62 When verifying, if an address was redirected to one new address, so that
1353 verification continued, and the new address failed or deferred after
1354 having set something in $address_data, the value of $address_data was not
1355 passed back to the ACL. This was different to the case when no
1356 redirection occurred. The value is now passed back in both cases.
1358 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
1359 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
1360 use login_cap.h, so on its own it isn't the distinguishing feature. The
1361 new name refers directly to the setclassresources() function.
1363 PH/65 Added configuration files for NetBSD3.
1365 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
1367 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
1368 is preferred over IPv4.
1370 PH/68 The bounce_return_message and bounce_return_body options were not being
1371 honoured for bounces generated during the reception of non-SMTP messages.
1372 In particular, this applied to messages rejected by the ACL. This bug has
1373 been fixed. However, if bounce_return_message is true and bounce_return_
1374 body is false, the headers that are returned for a non-SMTP message
1375 include only those that have been read before the error was detected.
1376 (In the case of an ACL rejection, they have all been read.)
1378 PH/69 The HTML version of the specification is now built in a directory called
1379 spec_html instead of spec.html, because the latter looks like a path with
1380 a MIME-type, and this confuses some software.
1382 PH/70 Catch two compiler warnings in sieve.c.
1384 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
1385 function verify_get_ident() calls ip_connect() to connect a socket, but
1386 if the "connect()" function timed out, ip_connect() used to close the
1387 socket. However, verify_get_ident() also closes the socket later, and in
1388 between Exim writes to the log, which may get opened at this point. When
1389 the socket was closed in ip_connect(), the log could get the same file
1390 descriptor number as the socket. This naturally causes chaos. The fix is
1391 not to close the socket in ip_connect(); the socket should be closed by
1392 the function that creates it. There was only one place in the code where
1393 this was missing, in the iplookup router, which I don't think anybody now
1394 uses, but I've fixed it anyway.
1396 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
1397 well as to direct DNS lookups. Otherwise the handling of names in host
1398 lists is inconsistent and therefore confusing.
1404 PH/01 Two changes to the default runtime configuration:
1406 (1) Move the checks for relay_from_hosts and authenticated clients from
1407 after to before the (commented out) DNS black list checks.
1409 (2) Add control=submission to the relay_from_hosts and authenticated
1410 clients checks, on the grounds that messages accepted by these
1411 statements are most likely to be submissions.
1413 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
1415 (1) Generate an error if the third argument for the ${prvs expansion is
1418 (2) Treat a missing third argument of ${prvscheck as if it were an empty
1421 (3) Reset the variables that are obtained from the first argument of
1422 ${prvscheck and used in the second argument before leaving the code,
1423 because their memory is reclaimed, so using them afterwards may do
1426 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
1427 one (it's much easier than Tom thought :-).
1429 (5) Because of (4), we can now allow for the use of $prvscheck_result
1430 inside the third argument.
1432 PH/03 For some reason, the default setting of PATH when running a command from
1433 a pipe transport was just "/usr/bin". I have changed it to
1436 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
1437 anything to be listed in the output from -bV.
1439 PH/05 When a filter generated an autoreply, the entire To: header line was
1440 quoted in the delivery log line, like this:
1442 => >A.N.Other <ano@some.domain> <original@ddress> ...
1444 This has been changed so that it extracts the operative address. There
1445 may be more than one such address. If so, they are comma-separated, like
1448 => >ano@some.domain,ona@other.domain <original@ddress> ...
1450 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
1451 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
1452 not being looked up in the rDNS to get a host name, Exim was showing the
1453 IP address twice in Received: lines, even though the IP addresses were
1454 identical. For example:
1456 Received: from [1.2.3.4] (helo=[1.2.3.4])
1458 However, if the real host name was known, it was omitting the HELO data
1459 if it matched the actual IP address. This has been tidied up so that it
1460 doesn't show the same IP address twice.
1462 PH/07 When both +timestamp and +memory debugging was on, the value given by
1463 $tod_xxx expansions could be wrong, because the tod_stamp() function was
1464 called by the debug printing, thereby overwriting the timestamp buffer.
1465 Debugging no longer uses the tod_stamp() function when +timestamp is set.
1467 PH/08 When the original message was included in an autoreply transport, it
1468 always said "this is a copy of the message, including all the headers",
1469 even if body_only or headers_only was set. It now gives an appropriate
1472 PH/09 Applied a patch from the Sieve maintainer which:
1474 o fixes some comments
1475 o adds the (disabled) notify extension core
1476 o adds some debug output for the result of if/elsif tests
1477 o points to the current vacation draft in the documentation
1478 and documents the missing references header update
1482 o fixes a bug in processing the envelope test (when testing
1483 multiple envelope elements, the last element determinted the
1486 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
1487 Electronic Mail") by including:
1489 Auto-submitted: auto-generated
1491 in the messages that it generates (bounce messages and others, such as
1492 warnings). In the case of bounce messages for non-SMTP mesages, there was
1493 also a typo: it was using "Auto_submitted" (underscore instead of
1494 hyphen). Since every message generated by Exim is necessarily in response
1495 to another message, thes have all been changed to:
1497 Auto-Submitted: auto-replied
1499 in accordance with these statements in the RFC:
1501 The auto-replied keyword:
1503 - SHOULD be used on messages sent in direct response to another
1504 message by an automatic process,
1506 - MUST NOT be used on manually-generated messages,
1508 - MAY be used on Delivery Status Notifications (DSNs) and Message
1509 Disposition Notifications (MDNs),
1511 - MUST NOT be used on messages generated by automatic or periodic
1512 processes, except for messages which are automatic responses to
1515 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
1516 to the default Received: header definition.
1518 PH/12 Added log selector acl_warn_skipped (default on).
1520 PH/13 After a successful wildlsearch lookup, discard the values of numeric
1521 variables because (a) they are in the wrong storage pool and (b) even if
1522 they were copied, it wouldn't work properly because of the caching.
1524 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
1525 checking when decoding. Apparently there are clients that generate
1526 overlong encoded strings. Why am I not surprised?
1528 PH/15 If the first argument of "${if match_address" was not empty, but did not
1529 contain an "@" character, Exim crashed. Now it writes a panic log message
1530 and treats the condition as false.
1532 PH/16 In autoreply, treat an empty string for "once" the same as unset.
1534 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
1535 extension "envelope-auth". The code is finished and in agreement with
1536 other implementations, but there is no documentation so far and in fact,
1537 nobody wrote the draft yet. This extension is currently #undef'ed, thus
1538 not changing the active code.
1540 Print executed "if" and "elsif" statements when debugging is used. This
1541 helps a great deal to understand what a filter does.
1543 Document more things not specified clearly in RFC3028. I had all this
1544 sorted out, when out of a sudden new issues came to my mind. Oops."
1546 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
1549 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
1550 canonical form (as documented). However, after a host name lookup from
1551 the IP address, check_host() was doing a simple string comparison with
1552 addresses acquired from the DNS when checking that the found name did
1553 have the original IP as one of its addresses. Since any found IPv6
1554 addresses are likely to be in abbreviated form, the comparison could
1555 fail. Luckily, there already exists a function for doing the comparison
1556 by converting both addresses to binary, so now that is used instead of
1557 the text comparison.
1559 PH/20 There was another similar case to PH/19, when a complete host name was
1560 given in a host list; looking up its IP address could give an abbreviated
1561 form, whereas the current host's name might or might not be abbreviated.
1562 The same fix has been applied.
1568 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
1569 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
1572 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
1573 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
1575 PH/03 Typo: missing ".o" in src/pcre/Makefile.
1577 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
1578 header line, restrict the check to what is listed in RFCs 2369 and 2929.
1579 Also, for "Auto-Submitted", treat anything other than "no" as
1580 non-personal, in accordance with RFC 3834. (Previously it treated
1581 anything starting "auto-" as non-personal.)
1583 TF/01 The control=submission/name=... option had a problem with syntax
1584 errors if the name included a slash character. The /name= option
1585 now slurps the rest of the string, so it can include any characters
1586 but it must come last in the list of options (after /sender_retain
1589 PH/05 Some modifications to the interface to the fake nameserver for the new
1597 TK/01 Added the "success_on_redirect" address verification option. See
1598 NewStuff for rationale and an example.
1600 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
1602 PH/02 Patch to exigrep to allow it to work on syslog lines.
1604 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
1605 fread() to skip over the body file's header line, because in Cygwin the
1606 header line is locked and is inaccessible.
1608 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
1609 co-exist for some time) to make it clear that it is the Exim ID that is
1610 referenced, not the Message-ID: header line.
1612 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
1613 string_format() function, because snprintf() does not exist on all
1616 PH/06 The use of forbid_filter_existstest now also locks out the use of the
1617 ${stat: expansion item.
1619 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
1620 protocol synchronization error", to keep the pedants happy.
1622 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
1623 well as for IRIX systems, when gcc is being used. See the host.c source
1626 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
1628 PH/10 Named domain lists were not working if used in a queue_smtp_domains
1631 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
1632 transport and to the smtp transport in LMTP mode.
1634 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
1636 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
1637 run a filter in a subprocess. This could lead to confusion in subsequent
1638 lookups in the parent process. There should also be a search_tidyup() at
1639 the end of the subprocess.
1641 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
1642 only if the host matched helo_try_verify_hosts, which caused the
1643 verification to occur when the EHLO/HELO command was issued. The ACL just
1644 tested the remembered result. Now, if a previous verification attempt has
1645 not happened, "verify = helo" does it there and then.
1647 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
1649 TK/03 Fix log output including CR from clamd.
1651 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
1652 exist provoked a memory error which could cause a segfault.
1654 PH/15 Installed PCRE 6.2
1656 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
1658 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
1659 of the problem. Specifically, suggested +O2 rather than +O1 for the
1662 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
1664 PH/20 If a delivery was routed to a non-standard port by means of an SRV
1665 record, the port was not correctly logged when the outgoing_port log
1666 selector was set (it logged the transort's default port).
1668 PH/21 Added support for host-specific ports to manualroute, queryprogram,
1669 fallback_hosts, and "hosts" in the smtp transport.
1671 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
1672 host errors such as "Connection refused".
1674 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
1675 authentication with radiusclient 0.4.9:
1677 - Error returned from rc_read_config was caught wrongly
1678 - Username/password not passed on to radius server due to wrong length.
1680 The presumption is that some radiusclient API changes for 4.51/PH/17
1681 were not taken care of correctly. The code is still untested by me (my
1682 Linux distribution still has 0.3.2 of radiusclient), but it was
1683 contributed by a Radius user.
1685 PH/24 When doing a callout, the value of $domain wasn't set correctly when
1686 expanding the "port" option of the smtp transport.
1688 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
1689 while reading a MIME header. Thanks to Tom Hughes for a patch.
1691 PH/24 Include config.h inside local_scan.h so that configuration settings are
1694 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
1695 that in an ACL for RCPT (for example), you can examine exactly what was
1698 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
1699 commands, but it was not correctly comparing the address with the actual
1700 client host address. Thus, it would show the EHLO address in Received:
1701 header lines when this was not necessary.
1703 PH/27 Added the % operator to ${eval:}.
1705 PH/28 Exim tries to create and chdir to its spool directory when it starts;
1706 it should be ignoring failures (because with -C, for example, it has lost
1707 privilege). It wasn't ignoring creation failures other than "already
1710 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
1713 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
1714 because an input line was too long, either on its own, or by virtue of
1715 too many continuations, the temporary file was not being removed, and the
1716 return code was incorrect.
1718 PH/31 Missing "BOOL" in function definition in filtertest.c.
1720 PH/32 Applied Sieve patches from the maintainer.
1722 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
1724 PH/33 Added "verify = not_blind".
1726 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
1727 Local/Makefile (with some defaults set). These are used in built scripts
1728 such as exicyclog, but they have never been used in the exim_install
1729 script (though there are many overriding facilities there). I have
1730 arranged that the exim_install script now takes note of these two
1733 PH/35 Installed configuration files for Dragonfly.
1735 PH/36 When a locally submitted message by a trusted user did not contain a
1736 From: header, and the sender address was obtained from -f or from an SMTP
1737 MAIL command, and the trusted user did not use -F to supply a sender
1738 name, $originator_name was incorrectly used when constructing a From:
1739 header. Furthermore, $originator_name was used for submission mode
1740 messages from external hosts without From: headers in a similar way,
1741 which is clearly wrong.
1743 PH/37 Added control=suppress_local_fixups.
1745 PH/38 When log_selector = +received_sender was set, and the addition of the
1746 sender made the log line's construction buffer exactly full, or one byte
1747 less than full, an overflow happened when the terminating "\n" was
1750 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
1751 when the result of a list match is failure because a DNS lookup failed.
1753 PH/40 RM_COMMAND is now used in the building process.
1755 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
1756 the "build-* directories that it finds.
1758 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
1759 address in a domain literal was a prefix of an interface address.
1761 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
1762 when verifying a sender address, unless rewrite_headers is false.
1764 PH/44 Wrote a long comment about why errors_to addresses are verified as
1765 recipients, not senders.
1767 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
1768 the ratelimit ACL was added.
1770 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
1772 PH/46 Added extra information about PostgreSQL errors to the error string.
1774 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
1775 avoiding the need to install special zones in a real server. This is
1776 backwards compatible; if it can't find the fake resolver, it drops back.
1777 Thus, both old and new test suites can be run.
1779 TF/02 Added util/ratelimit.pl
1781 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
1782 clock is set back in time.
1784 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
1785 Candler <B.Candler@pobox.com>.
1787 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
1788 OK for addresses that are the result of redirections.
1790 PH/48 A number of further additions for the benefit of the new test suite,
1791 including a fake gethostbyname() that interfaces to the fake DNS resolver
1794 TF/06 The fix for widen_domains has also been applied to qualify_single and
1795 search_parents which are the other dnslookup options that can cause
1798 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
1801 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
1803 TF/07 Exim produced the error message "an SRV record indicated no SMTP
1804 service" if it encountered an MX record with an empty target hostname.
1805 The message is now "an MX or SRV record indicated no SMTP service".
1807 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
1808 if the DNS of the sending site is misconfigured. This is quite a
1809 common situation. This change restores the behaviour of treating a
1810 helo verification defer as a failure.
1812 PH/51 If self=fail was set on a router, the bounce message did not include the
1813 actual error message.
1819 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
1821 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
1822 command itself ended in error, the underlying message about the transport
1823 filter timeout was being overwritten with the pipe command error. Now the
1824 underlying error message should be appended to the second error message.
1826 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
1828 PH/02 Reduce the amount of output that "make" produces by default. Full output
1829 can still be requested.
1831 PH/03 The warning log line about a condition test deferring for a "warn" verb
1832 was being output only once per connection, rather than after each
1833 occurrence (because it was using the same function as for successful
1834 "warn" verbs). This seems wrong, so I have changed it.
1836 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
1837 it should not have, which might have caused a crash in the right
1838 circumstances, but probably never did.
1840 PH/04 Installed a modified version of Tony Finch's patch to make submission
1841 mode fix the return path as well as the Sender: header line, and to
1842 add a /name= option so that you can make the user's friendly name appear
1845 TF/03 Added the control = fakedefer ACL modifier.
1847 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
1848 Mark Lowes for thorough testing.
1850 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
1852 TK/03 Merged latest SRS patch from Miles Wilton.
1854 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
1855 with the definition in sysexits.h (which is #included earlier).
1856 Fortunately, Exim does not actually use EX_OK. The code used to try to
1857 preserve the sysexits.h value, by assumimg that macro definitions were
1858 scanned for macro replacements. I have been disabused of this notion,
1859 so now the code just undefines EX_OK before #including unistd.h.
1861 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
1862 in the smtp transport. When a block could not be written in a single
1863 write() function, the timeout was being re-applied to each part-write.
1864 This seems wrong - if the receiver was accepting one byte at a time it
1865 would take for ever. The timeout is now adjusted when this happens. It
1866 doesn't have to be particularly precise.
1868 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
1869 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
1871 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
1872 without a domain if the check to <postmaster@domain> fails.
1874 SC/01 Eximstats: added -xls and the ability to specify output files
1875 (patch written by Frank Heydlauf).
1877 SC/02 Eximstats: use FileHandles for outputing results.
1879 SC/03 Eximstats: allow any combination of xls, txt, and html output.
1881 SC/04 Eximstats: fixed display of large numbers with -nvr option
1883 SC/05 Eximstats: fixed merging of reports with empty tables.
1885 SC/06 Eximstats: added the -include_original_destination flag
1887 SC/07 Eximstats: removed tabs and trailing whitespace.
1889 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
1891 TK/06 MBOX spool code: Add real "From " MBOX separator line
1892 so the .eml file is really in mbox format (even though
1893 most programs do not really care). Patch from Alex Miller.
1895 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
1896 The latter is generated from $received_to and is only set if the
1897 message has one envelope recipient. SA can use these headers,
1898 obviously out-of-the-box. Patch from Alex Miller.
1900 PH/08 The ${def test on a variable was returning false if the variable's
1901 value was "0", contrary to what the specification has always said!
1902 The result should be true unless the variable is empty.
1904 PH/09 The syntax error of a character other than { following "${if
1905 def:variable_name" (after optional whitespace) was not being diagnosed.
1906 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
1907 accidental colon was present, for example, could give incorrect results.
1909 PH/10 Tidied the code in a number of places where the st_size field of a stat()
1910 result is used (not including appendfile, where other changes are about
1913 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
1914 This involved changing a lot of size variables from int to off_t. It
1915 should work with maildirs and everything.
1917 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
1918 spamd dying while we are connected to it.
1920 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
1923 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
1924 function for BDB 4.3.
1926 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
1928 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
1929 the sources for PCRE have been re-arranged and more files are now
1932 PH/15 The code I had for printing potentially long long variables in PH/11
1933 above was not the best (it lost precision). The length of off_t variables
1934 is now inspected at build time, and an appropriate printing format (%ld
1935 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
1936 to be "long long int" or "long int". This is needed for the internal
1937 formatting function string_vformat().
1939 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
1940 the configuration file to be ":syslog", then the script "guesses" where
1941 the logs files are, rather than using the compiled in default. In our
1942 case the guess is not the same as the compiled default, so the script
1943 suddenly stopped working when I started to use syslog. The patch checks
1944 to see if log_file_path is "". If so, it attempts to read it from exim
1945 with no configuration file to get the compiled in version, before it
1946 falls back to the previous guessing code."
1948 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
1949 implementing BATV in an Exim configuration. See NewStuff for the gory
1952 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
1953 Makefile that are specific to HP-UX.
1955 PH/18 If the "use_postmaster" option was set for a recipient callout together
1956 with the "random" option, the postmaster address was used as the MAIL
1957 FROM address for the random test, but not for the subsequent recipient
1958 test. It is now used for both.
1960 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
1961 patch removes a few documentation additions to RFC 3028, because the
1962 latest draft now contains them. It adds the new en;ascii-case comparator
1963 and a new error check for 8bit text in MIME parts. Comparator and
1964 require names are now matched exactly. I enabled the subaddress
1965 extension, but it is not well tested yet (read: it works for me)."
1967 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
1968 rework some of the code of TK/09 above to avoid the hardwired use of
1969 "%lld" and "long long". Replaced the call to snprintf() with a call to
1972 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
1973 records point to non-existent hosts", "retry timeout exceeded", and
1974 "retry time not reached for any host after a long failure period".
1976 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
1977 experimental DomainKeys support:
1979 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
1980 (2) On an error such as an illegally used "control", the wrong name for
1981 the control was given.
1983 These problems did NOT occur unless DomainKeys support was compiled.
1985 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
1987 PH/24 Added ${if match_ip condition.
1989 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
1990 databases so that it will be absolutely obvious if a crash occurs in the
1991 DB library. This is a regular occurrence (often caused by mis-matched
1994 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
1995 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
1996 picked up on a user's system that detects such things. There doesn't seem
1997 to be a gcc warning option for this - only an attribute that has to be
1998 put on the function's prototype. It seems that in Fedora Core 4 they have
1999 set this on a number of new functions. No doubt there will be more in due
2002 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
2003 specify a transport. However, if an address that was verified by such a
2004 router was the subject of a callout, Exim crashed because it tried to
2005 read the rcpt_include_affixes from the non-existent transport. Now it
2006 just assumes that the setting of that option is false. This bug was
2007 introduced by 4.51/PH/31.
2009 PH/28 Changed -d+all to exclude +memory, because that information is very
2010 rarely of interest, but it makes the output a lot bigger. People tend to
2011 do -d+all out of habit.
2013 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
2014 code in os-type was giving problems when libc.so lives in lib64, like on
2017 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
2018 aren't the modern standard, and it seems that some systems' include files
2019 don't always have them. Exim was already checking for some of the newer
2020 ones like T_AAAA, and defining it itself. I've added checks for all the
2021 record types that Exim uses.
2023 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
2024 not automatically generating a new one, as it is supposed to. This
2025 prevented TLS from working. If the file did exist, but contained invalid
2026 data, a new version was generated, as expected. It was only the case of a
2027 non-existent file that was broken.
2029 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
2030 with a change in libdomainkeys > 0.64.
2032 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
2033 from DNS. If the selector record carries the flag, it now has
2034 precedence over the domain-wide flag.
2036 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
2038 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
2039 the use of a port name that isn't defined in /etc/services) occurred, the
2040 message was deferred as in a normal delivery, and thus remained on the
2041 spool, instead of being failed because of the mua_wrapper setting. This
2042 is now fixed, and I tidied up some of the mua_wrapper messages at the
2045 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
2046 the messages in a hash of arrays rather than using individual hashes.
2047 This is a bit cleaner and results in dramatic memory savings, albeit
2048 at a slight CPU cost.
2050 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
2051 as requested by Marc Sherman.
2053 SC/10 Eximstats: added histograms for user specified patterns as requested
2056 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
2058 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
2059 fopen() in the content-scanning modules that did not already have it.
2065 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
2066 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
2068 TK/02 Fix ACL "control" statement not being available in MIME ACL.
2070 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
2072 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
2073 to test Sieve filters that use "vacation".
2075 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
2076 that changes the way the GnuTLS parameters are stored in the cache file.
2077 The new format can be generated externally. For backward compatibility,
2078 if the data in the cache doesn't make sense, Exim assumes it has read an
2079 old-format file, and it generates new data and writes a new file. This
2080 means that you can't go back to an older release without removing the
2083 PH/03 A redirect router that has both "unseen" and "one_time" set does not
2084 work if there are any delivery delays because "one_time" forces the
2085 parent to be marked "delivered", so its unseen clone is never tried
2086 again. For this reason, Exim now forbids the simultaneous setting of
2089 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
2090 redirected to themselves ("homonym" addresses). Read the long ChangeLog
2091 entry if you want to know the details. The fix, however, neglected to
2092 consider the case when local delivery batching is involved. The test for
2093 "previously delivered" was not happening when checking to see if an
2094 address could be batched with a previous (undelivered) one; under
2095 certain circumstances this could lead to multiple deliveries to the same
2098 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
2099 in its include files, and this causes problems building Exim.
2101 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
2102 header_syntax) but Exim was just ignoring anything given after a slash.
2103 In particular, this caused confusion with an attempt to use "verify =
2104 reverse_host_lookup/defer_ok". An error is now given when options are
2105 supplied for verify items that do not have them. (Maybe reverse_host_
2106 lookup should have a defer_ok option, but that's a different point.)
2108 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
2109 defined by RFC 821) to 2048, because there were problems with some AUTH
2110 commands, and RFC 1869 says the size should be increased for extended
2111 SMTP commands that take arguments.
2113 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
2116 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
2117 "unknown" error; now it says that the functionality isn't in the binary.
2119 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
2120 an address' error message when a string expansion fails (syntax or
2121 whatever). Otherwise the password may appear in the log. Following change
2122 PH/42 below, there is no longer a chance of it appearing in a bounce
2125 PH/11 Installed exipick version 20050225.0 from John Jetmore.
2127 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
2128 of its addresses was ever tried. (Bugzilla bug #2.)
2130 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
2131 the result incorrectly in the debug output. (It correctly added a newline
2132 to what was transported.)
2134 TF/01 Added $received_time.
2136 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
2137 commented out examples of how to interface to a virus scanner and to
2138 SpamAssassin. Also added commented examples of av_scanner and
2139 spamd_address settings.
2141 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
2142 and controls are allowed in which ACLs. There were a couple of minor
2143 errors. Some of the entries in the conditions table (which is a table of
2144 where they are NOT allowed) were getting very unwieldy; rewrote them as a
2145 negation of where the condition IS allowed.
2147 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
2149 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
2150 header file does not have a version number, so I've had to invent a new
2151 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
2152 API. The code is untested by me (my Linux distribution still has 0.3.2 of
2153 radiusclient), but it was contributed by a Radius user.
2155 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
2156 files or directories, for OpenSSL.
2158 PH/19 When an Exim process that is running as root has to create an Exim log
2159 file, it does so in a subprocess that runs as exim:exim so as to get the
2160 ownership right at creation (otherwise, other Exim processes might see
2161 the file with the wrong ownership). There was no test for failure of this
2162 fork() call, which would lead to the process getting stuck as it waited
2163 for a non-existent subprocess. Forks do occasionally fail when resources
2164 run out. I reviewed all the other calls to fork(); they all seem to check
2167 PH/20 When checking for unexpected SMTP input at connect time (before writing
2168 the banner), Exim was not dealing correctly with a non-positive return
2169 from the read() function. If the client had disconnected by this time,
2170 the result was a log entry for a synchronization error with an empty
2171 string after "input=" when read() returned zero. If read() returned -1
2172 (an event I could not check), uninitialized data bytes were printed.
2173 There were reports of junk text (parts of files, etc) appearing after
2176 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
2178 PH/22 Added support for macro redefinition, and (re)definition in between
2179 driver and ACL definitions.
2181 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
2182 forgetting to use the resulting value; it was using the unexpanded value.
2184 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
2185 hadn't been configured. The fix is from Juergen Kreileder, who
2186 understands it better than I do:
2188 "Here's what I see happening with three configured cyrus_sasl
2189 authenticators configured (plain, login, cram-md5):
2191 On startup auth_cyrus_sasl_init() gets called for each of these.
2192 This means three calls to sasl_listmech() without a specified mech_list.
2193 => SASL tests which mechs of all available mechs actually work
2194 => three warnings about OTP not working
2195 => the returned list contains: plain, login, cram-md5, digest-md5, ...
2197 With the patch, sasl_listmech() also gets called three times. But now
2198 SASL's mech_list option is set to the server_mech specified in the the
2199 authenticator. Or in other words, the answer from sasl_listmech()
2200 gets limited to just the mech you're testing for (which is different
2202 => the return list contains just 'plain' or 'login', 'cram-md5' or
2203 nothing depending on the value of ob->server_mech.
2205 I've just tested the patch: Authentication still works fine,
2206 unavailable mechs specified in the exim configuration are still
2207 caught, and the auth.log warnings about OTP are gone."
2209 PH/25 When debugging is enabled, the contents of the command line are added
2210 to the debugging output, even when log_selector=+arguments is not
2213 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
2214 answer is "GNU", and only if the return is "GNU/something" is the answer
2217 PH/27 $acl_verify_message is now set immediately after the failure of a
2218 verification in an ACL, and so is available in subsequent modifiers. In
2219 particular, the message can be preserved by coding like this:
2221 warn !verify = sender
2222 set acl_m0 = $acl_verify_message
2224 Previously, $acl_verify_message was set only while expanding "message"
2225 and "log_message" when a very denied access.
2227 PH/28 Modified OS/os.c-Linux with
2229 -#ifndef OS_LOAD_AVERAGE
2230 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
2232 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
2233 nomenclature these days.)
2235 PH/29 Installed patch from the Sieve maintainer that adds the options
2236 sieve_useraddress and sieve_subaddress to the redirect router.
2238 PH/30 In these circumstances:
2239 . Two addresses routed to the same list of hosts;
2240 . First host does not offer TLS;
2241 . First host accepts first address;
2242 . First host gives temporary error to second address;
2243 . Second host offers TLS and a TLS session is established;
2244 . Second host accepts second address.
2245 Exim incorrectly logged both deliveries with the TLS parameters (cipher
2246 and peerdn, if requested) that were in fact used only for the second
2249 PH/31 When doing a callout as part of verifying an address, Exim was not paying
2250 attention to any local part prefix or suffix that was matched by the
2251 router that accepted the address. It now behaves in the same way as it
2252 does for delivery: the affixes are removed from the local part unless
2253 rcpt_include_affixes is set on the transport.
2255 PH/32 Add the sender address, as F=<...>, to the log line when logging a
2256 timeout during the DATA phase of an incoming message.
2258 PH/33 Sieve envelope tests were broken for match types other than :is. I have
2259 applied a patch sanctioned by the Sieve maintainer.
2261 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
2262 the uid or gid is negative. A case of a negative gid caused this to be
2263 noticed. The fix allows for either to be negative.
2265 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
2266 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
2269 PH/36 The change PH/12 above was broken. Fixed it.
2271 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
2272 the grounds that routing the same address twice would always produce the
2273 same answer. This might have been true once, but it is certainly no
2274 longer true now. Routing a child address may depend on the previous
2275 routing that produced that child. Some complicated redirection strategies
2276 went wrong when messages had multiple recipients, and made Exim's
2277 behaviour dependent on the order in which the addresses were given.
2279 I have moved the duplicate checking until after the routing is complete.
2280 Exim scans the addresses that are assigned to local and remote
2281 transports, and removes any duplicates. This means that more work will be
2282 done, as duplicates will always all be routed, but duplicates are
2283 presumably rare, so I don't expect this is of any significance.
2285 For deliveries to pipes, files, and autoreplies, the duplicate checking
2286 still happens during the routing process, since they are not going to be
2289 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
2290 It corrects a timeout issue with spamd. This is Ian's comment: "The
2291 background is that sometimes spamd either never reads data from a
2292 connection it has accepted, or it never writes response data. The exiscan
2293 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
2294 blindly assumes that writes won't block so it may never time out."
2296 PH/39 Allow G after quota size as well as K and M.
2298 PH/40 The value set for $authenticated_id in an authenticator may not contain
2299 binary zeroes or newlines because the value is written to log lines and
2300 to spool files. There was no check on this. Now the value is run through
2301 the string_printing() function so that such characters are converted to
2302 printable escape sequences.
2304 PH/41 $message_linecount is a new variable that contains the total number of
2305 lines in the message. Compare $body_linecount, which is the count for the
2308 PH/42 Exim no longer gives details of delivery errors for specific addresses in
2309 bounce and delay warning messages, except in certain special cases, which
2312 (a) An SMTP error message from a remote host;
2313 (b) A message specified in a :fail: redirection;
2314 (c) A message specified in a "fail" command in a system filter;
2315 (d) A message specified in a FAIL return from the queryprogram router;
2316 (e) A message specified by the cannot_route_message router option.
2318 In these cases only, Exim does include the error details in bounce and
2319 warning messages. There are also a few cases where bland messages such
2320 as "unrouteable address" or "local delivery error" are given.
2322 PH/43 $value is now also set for the "else" part of a ${run expansion.
2324 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
2325 being worked on, but at least Exim now implements the latest version to
2328 PH/45 In a pipe transport, although a timeout while waiting for the pipe
2329 process to complete was treated as a delivery failure, a timeout while
2330 writing the message to the pipe was logged, but erroneously treated as a
2331 successful delivery. Such timeouts include transport filter timeouts. For
2332 consistency with the overall process timeout, these timeouts are now
2333 treated as errors, giving rise to delivery failures by default. However,
2334 there is now a new Boolean option for the pipe transport called
2335 timeout_defer, which, if set TRUE, converts the failures into defers for
2336 both kinds of timeout. A transport filter timeout is now identified in
2339 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
2340 systems where "make" and "gmake" are different, calling "gmake" at top
2341 level broke things. I've arranged for the value of $(MAKE) to be passed
2342 from the Makefile to this script so that it can call the same version of
2346 A note about Exim versions 4.44 and 4.50
2347 ----------------------------------------
2349 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
2350 changes of various kinds. As a consequence, a big documentation update was
2351 needed. This delayed the release for rather longer than seemed good, especially
2352 in the light of a couple of (minor) security issues. Therefore, the changes
2353 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
2354 release. So 4.44 and 4.50 are in effect two different branches that both start
2357 I have left the 4.50 change log unchanged; it contains all the changes since
2358 4.43. The change log for 4.44 is below; many of its items are identical to
2359 those for 4.50. This seems to be the most sensible way to preserve the
2360 historical information.
2366 1. Minor wording change to the doc/README.SIEVE file.
2368 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
2369 computation of the current number of files was incorrect.
2371 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2372 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2375 4. Give more explanation in the error message when the command for a transport
2376 filter fails to execute.
2378 5. There are several places where Exim runs a non-Exim command in a
2379 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2380 was being done only for the command run by the queryprogram router. It is
2381 now done for all such subprocesses. The other cases are: ${run, transport
2382 filters, and the commands run by the lmtp and pipe transports.
2384 6. Added CONFIGURE_GROUP build-time option.
2386 7. Some older OS have a limit of 256 on the maximum number of file
2387 descriptors. Exim was using setrlimit() to set 1000 as a large value
2388 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2389 systems. I've change it so that if it can't get 1000, it tries for 256.
2391 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2392 was an oversight, and furthermore, ever since the addition of extra
2393 controls (e.g. 4.43/32), the checks on when to allow different forms of
2394 "control" were broken. There should now be diagnostics for all cases when a
2395 control that does not make sense is encountered.
2397 9. Added the /retain_sender option to "control=submission".
2399 10. $recipients is now available in the predata ACL (oversight).
2401 11. Tidy the search cache before the fork to do a delivery from a message
2402 received from the command line. Otherwise the child will trigger a lookup
2403 failure and thereby defer the delivery if it tries to use (for example) a
2404 cached ldap connection that the parent has called unbind on.
2406 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2407 of $address_data from the recipient verification was clobbered by the
2408 sender verification.
2410 13. The value of address_data from a sender verification is now available in
2411 $sender_address_data in subsequent conditions in the ACL statement.
2413 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
2415 15. Added a new option "connect=<time>" to callout options, to set a different
2418 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2419 was its contents. (It was OK if the option was not defined at all.)
2421 17. A "Completed" log line is now written for messages that are removed from
2422 the spool by the -Mrm option.
2424 18. New variables $sender_verify_failure and $recipient_verify_failure contain
2425 information about exactly what failed.
2427 19. Added -dd to debug only the daemon process.
2429 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
2430 handles timeouts, both on the server side and network timeouts. Renamed the
2431 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
2433 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
2434 It is now set to "smtps".
2436 22. $host_address is now set to the target address during the checking of
2437 ignore_target_hosts.
2439 23. When checking ignore_target_hosts for an ipliteral router, no host name was
2440 being passed; this would have caused $sender_host_name to have been used if
2441 matching the list had actually called for a host name (not very likely,
2442 since this list is usually IP addresses). A host name is now passed as
2445 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2446 code that specifies a non-restarting handler (typically sigaction() in
2447 modern systems) in an attempt to fix a rare and obscure crash bug.
2449 25. Narrowed the window for a race in the daemon that could cause it to ignore
2450 SIGCHLD signals. This is not a major problem, because they are used only to
2451 wake it up if nothing else does.
2453 26. A malformed maildirsize file could cause Exim to calculate negative values
2454 for the mailbox size or file count. Odd effects could occur as a result.
2455 The maildirsize information is now recalculated if the size or filecount
2458 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2459 support for a long time. Removed HAVE_SYS_VFS_H.
2461 28. Installed the latest version of exipick from John Jetmore.
2463 29. In an address list, if the pattern was not a regular expression, an empty
2464 subject address (from a bounce message) matched only if the pattern was an
2465 empty string. Non-empty patterns were not even tested. This was the wrong
2466 because it is perfectly reasonable to use an empty address as part of a
2467 database query. An empty address is now tested by patterns that are
2468 lookups. However, all the other forms of pattern expect the subject to
2469 contain a local part and a domain, and therefore, for them, an empty
2470 address still always fails if the pattern is not itself empty.
2472 30. Exim went into a mad DNS loop when attempting to do a callout where the
2473 host was specified on an smtp transport, and looking it up yielded more
2474 than one IP address.
2476 31. Re-factored the code for checking spool and log partition space into a
2477 function that finds that data and another that does the check. The former
2478 is then used to implement four new variables: $spool_space, $log_space,
2479 $spool_inodes, and $log_inodes.
2481 32. The RFC2047 encoding function was originally intended for short strings
2482 such as real names; it was not keeping to the 75-character limit for
2483 encoded words that the RFC imposes. It now respects the limit, and
2484 generates multiple encoded words if necessary. To be on the safe side, I
2485 have increased the buffer size for the ${rfc2047: expansion operator from
2488 33. It is now permitted to omit both strings after an "if" condition; if the
2489 condition is true, the result is "true". As before, when the second string
2490 is omitted, a false condition yields an empty string. This makes it less
2491 cumbersome to write custom ACL and router conditions.
2493 34. Failure to deliver a bounce message always caused it to be frozen, even if
2494 there was an errors_to setting on the router. The errors_to setting is now
2497 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2498 canonical form (fully expanded) before being placed in
2499 $sender_host_address.
2501 36. The table in the code that translates DNS record types into text (T_A to
2502 "A" for instance) was missing entries for NS and CNAME. It is just possible
2503 that this could have caused confusion if both these types were looked up
2504 for the same domain, because the text type is used as part of Exim's
2505 per-process caching. But the chance of anyone hitting this buglet seems
2508 37. The dnsdb lookup has been extended in a number of ways.
2510 (1) There is a new type, "zns", which walks up the domain tree until it
2511 finds some nameserver records. It should be used with care.
2513 (2) There is a new type, "mxh", which is like "mx" except that it returns
2514 just the host names, not the priorities.
2516 (3) It is now possible to give a list of domains (or IP addresses) to be
2517 looked up. The behaviour when one of the lookups defers can be
2518 controlled by a keyword.
2520 (4) It is now possible to specify the separator character for use when
2521 multiple records are returned.
2523 38. The dnslists ACL condition has been extended: it is now possible to supply
2524 a list of IP addresses and/or domains to be looked up in a particular DNS
2527 39. Added log_selector=+queue_time_overall.
2529 40. When running the queue in the test harness, wait just a tad after forking a
2530 delivery process, to get repeatability of debugging output.
2532 41. Include certificate and key file names in error message when GnuTLS fails
2533 to set them up, because the GnuTLS error message doesn't include the name
2534 of the failing file when there is a problem reading it.
2536 42. Allow both -bf and -bF in the same test run.
2538 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
2540 44. The "Exiscan patch" is now merged into the mainline Exim source.
2542 45. Sometimes the final signoff response after QUIT could fail to get
2543 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2544 < 0 before doing a fflush(). This bug looks as though it goes back to the
2545 introduction of TLS in release 3.20, but "sometimes" must have been rare
2546 because the tests only now provoked it.
2548 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
2549 (this can affect the format of dates).
2551 47. exim_tidydb, when checking for the continued existence of a message for
2552 which it has found a message-specific retry record, was not finding
2553 messages that were in split spool directories. Consequently, it was
2554 deleting retry records that should have stayed in existence.
2556 48. Steve fixed some bugs in eximstats.
2558 49. The SPA authentication driver was not abandoning authentication and moving
2559 on to the next authenticator when an expansion was forced to fail,
2560 contradicting the general specification for all authenticators. Instead it
2561 was generating a temporary error. It now behaves as specified.
2563 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
2564 (the order specifies the preference for clients). The order is now AES256,
2565 AES128, 3DES, ARCFOUR128.
2567 51. Small patch to Sieve code - explicitly set From: when generating an
2570 52. Exim crashed if a remote delivery caused a very long error message to be
2571 recorded - for instance if somebody sent an entire SpamAssassin report back
2572 as a large number of 550 error lines. This bug was coincidentally fixed by
2573 increasing the size of one of Exim's internal buffers (big_buffer) that
2574 happened as part of the Exiscan merge. However, to be on the safe side, I
2575 have made the code more robust (and fixed the comments that describe what
2578 53. Now that there can be additional text after "Completed" in log lines (if
2579 the queue_time_overall log selector is set), a one-byte patch to exigrep
2580 was needed to allow it to recognize "Completed" as not the last thing in
2583 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
2584 patch that reportedly fixes this has been added. I am not expert enough to
2585 create a test for it. This is what the patch creator wrote:
2587 "I found a little strange behaviour of ldap code when working with
2588 Windows 2003 AD Domain, where users was placed in more than one
2589 Organization Units. When I tried to give exim partial DN, the exit code
2590 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
2591 But simultaneously result of request was absolutely normal ldap result,
2592 so I produce this patch..."
2594 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
2595 REFERENCE, so I have modified the code to exclude the patch when that macro
2598 55. Some experimental protocols are using DNS PTR records for new purposes. The
2599 keys for these records are domain names, not reversed IP addresses. The
2600 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2601 leaves it alone. Component reversal etc. now happens only for IP addresses.
2604 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
2606 57. Double the size of the debug message buffer (to 2048) so that more of very
2607 long debug lines gets shown.
2609 58. The exicyclog utility now does better if the number of log files to keep
2610 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
2612 59. Two changes related to the smtp_active_hostname option:
2614 (1) $smtp_active_hostname is now available as a variable.
2615 (2) The default for smtp_banner uses $smtp_active_hostname instead
2616 of $primary_hostname.
2618 60. The host_aton() function is supposed to be passed a string that is known
2619 to be a valid IP address. However, in the case of IPv6 addresses, it was
2620 not checking this. This is a hostage to fortune. Exim now panics and dies
2621 if the condition is not met. A case was found where this could be provoked
2622 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2623 components; fortuitously, this particular loophole had already been fixed
2624 by change 4.50/55 above.
2626 If there are any other similar loopholes, the new check in host_aton()
2627 itself should stop them being exploited. The report I received stated that
2628 data on the command line could provoke the exploit when Exim was running as
2629 exim, but did not say which command line option was involved. All I could
2630 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2631 running as the user.
2634 61. There was a buffer overflow vulnerability in the SPA authentication code
2635 (which came originally from the Samba project). I have added a test to the
2636 spa_base64_to_bits() function which I hope fixes it.
2639 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
2640 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
2642 63. The daemon start-up calls getloadavg() while still root for those OS that
2643 need the first call to be done as root, but it missed one case: when
2644 deliver_queue_load_max is set with deliver_drop_privilege. This is
2645 necessary for the benefit of the queue runner, because there is no re-exec
2646 when deliver_drop_privilege is set.
2648 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
2649 This has been fixed.
2651 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2652 was in use, was not putting the data itself into the right store pool;
2653 consequently, it could be overwritten for a subsequent message in the same
2654 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2657 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
2659 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
2660 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
2661 the function were treating the return as a boolean value, which happened to
2662 work because 0=false and not-0=true, but is not correct code.
2664 68. The host_aton() function was not handling scoped IPv6 addresses (those
2665 with, for example, "%eth0" on the end) correctly.
2667 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
2668 negated items (that is, ~something) in unsigned ints. Some compilers
2669 apparently mutter when there is no cast.
2671 70. If an address verification called from an ACL failed, and did not produce a
2672 user-specific message (i.e. there was only a "system" message), nothing was
2673 put in $acl_verify_message. In this situation, it now puts the system
2676 71. Change 4.23/11 added synchronization checking at the start of an SMTP
2677 session; change 4.31/43 added the unwanted input to the log line - except
2678 that it did not do this in the start of session case. It now does.
2680 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
2681 This is wrong and can cause the other end to generate a synchronization
2682 error if it is another Exim or anything else that does the synchronization
2683 check. A QUIT command is no longer sent after a timeout.
2685 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
2686 during host lookups.
2688 74. The defer_ok option of callout verification was not working if it was used
2689 when verifying addresses in header lines, that is, for this case:
2691 verify = header_sender/callout=defer_ok
2693 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
2694 those file descriptors could be used for SMTP connections. If anything
2695 wrote to stderr (the example that came up was "warn" in embedded Perl), it
2696 could be sent to the SMTP client, causing chaos. The daemon now opens
2697 stdin, stdout, and stderr to /dev/null when it puts itself into the
2700 76. Arrange for output from Perl's "warn" command to be written to Exim's main
2701 log by default. The user can override this with suitable Perl magic.
2703 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
2704 the log message when discard triggers, was not working for the DATA ACL or
2705 for the non-SMTP ACL.
2707 78. Error message wording change in sieve.c.
2709 79. If smtp_accept_max_per_host was set, the number of connections could be
2710 restricted to fewer than expected, because the daemon was trying to set up
2711 a new connection before checking whether the processes handling previous
2712 connections had finished. The check for completed processes is now done
2713 earlier. On busy systems, this bug wouldn't be noticed because something
2714 else would have woken the daemon, and it would have reaped the completed
2717 80. If a message was submitted locally by a user whose login name contained one
2718 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
2719 It caused a spool format error. I have fixed the spool reading code. A
2720 related problem was that the "from" clause in the Received: line became
2721 illegal because of the space(s). It is now covered by ${quote_local_part.
2723 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
2726 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
2728 83. Patch from Sieve maintainer to fix unterminated string problem in
2729 "vacation" handling.
2731 84. Some minor changes to the Linux configuration files to help with other
2732 OS variants using glibc.
2734 85. One more patch for Sieve to update vacation handling to latest spec.
2737 ----------------------------------------------------
2738 See the note above about the 4.44 and 4.50 releases.
2739 ----------------------------------------------------
2745 1. Change 4.43/35 introduced a bug that caused file counts to be
2746 incorrectly computed when quota_filecount was set in an appendfile
2749 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
2750 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
2753 3. Give more explanation in the error message when the command for a transport
2754 filter fails to execute.
2756 4. There are several places where Exim runs a non-Exim command in a
2757 subprocess. The SIGUSR1 signal should be disabled for these processes. This
2758 was being done only for the command run by the queryprogram router. It is
2759 now done for all such subprocesses. The other cases are: ${run, transport
2760 filters, and the commands run by the lmtp and pipe transports.
2762 5. Some older OS have a limit of 256 on the maximum number of file
2763 descriptors. Exim was using setrlimit() to set 1000 as a large value
2764 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
2765 systems. I've change it so that if it can't get 1000, it tries for 256.
2767 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
2768 was an oversight, and furthermore, ever since the addition of extra
2769 controls (e.g. 4.43/32), the checks on when to allow different forms of
2770 "control" were broken. There should now be diagnostics for all cases when a
2771 control that does not make sense is encountered.
2773 7. $recipients is now available in the predata ACL (oversight).
2775 8. Tidy the search cache before the fork to do a delivery from a message
2776 received from the command line. Otherwise the child will trigger a lookup
2777 failure and thereby defer the delivery if it tries to use (for example) a
2778 cached ldap connection that the parent has called unbind on.
2780 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
2781 of $address_data from the recipient verification was clobbered by the
2782 sender verification.
2784 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
2785 was its contents. (It was OK if the option was not defined at all.)
2787 11. A "Completed" log line is now written for messages that are removed from
2788 the spool by the -Mrm option.
2790 12. $host_address is now set to the target address during the checking of
2791 ignore_target_hosts.
2793 13. When checking ignore_target_hosts for an ipliteral router, no host name was
2794 being passed; this would have caused $sender_host_name to have been used if
2795 matching the list had actually called for a host name (not very likely,
2796 since this list is usually IP addresses). A host name is now passed as
2799 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
2800 code that specifies a non-restarting handler (typically sigaction() in
2801 modern systems) in an attempt to fix a rare and obscure crash bug.
2803 15. Narrowed the window for a race in the daemon that could cause it to ignore
2804 SIGCHLD signals. This is not a major problem, because they are used only to
2805 wake it up if nothing else does.
2807 16. A malformed maildirsize file could cause Exim to calculate negative values
2808 for the mailbox size or file count. Odd effects could occur as a result.
2809 The maildirsize information is now recalculated if the size or filecount
2812 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
2813 support for a long time. Removed HAVE_SYS_VFS_H.
2815 18. Updated exipick to current release from John Jetmore.
2817 19. Allow an empty sender to be matched against a lookup in an address list.
2818 Previously the only cases considered were a regular expression, or an
2821 20. Exim went into a mad DNS lookup loop when doing a callout where the
2822 host was specified on the transport, if the DNS lookup yielded more than
2825 21. The RFC2047 encoding function was originally intended for short strings
2826 such as real names; it was not keeping to the 75-character limit for
2827 encoded words that the RFC imposes. It now respects the limit, and
2828 generates multiple encoded words if necessary. To be on the safe side, I
2829 have increased the buffer size for the ${rfc2047: expansion operator from
2832 22. Failure to deliver a bounce message always caused it to be frozen, even if
2833 there was an errors_to setting on the router. The errors_to setting is now
2836 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
2837 canonical form (fully expanded) before being placed in
2838 $sender_host_address.
2840 24. Updated eximstats to version 1.33
2842 25. Include certificate and key file names in error message when GnuTLS fails
2843 to set them up, because the GnuTLS error message doesn't include the name
2844 of the failing file when there is a problem reading it.
2846 26. Expand error message when OpenSSL has problems setting up cert/key files.
2849 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
2850 (this can affect the format of dates).
2852 28. exim_tidydb, when checking for the continued existence of a message for
2853 which it has found a message-specific retry record, was not finding
2854 messages that were in split spool directories. Consequently, it was
2855 deleting retry records that should have stayed in existence.
2857 29. eximstats updated to version 1.35
2858 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
2859 1.35 - bugfix such that pie charts by volume are generated correctly
2861 30. The SPA authentication driver was not abandoning authentication and moving
2862 on to the next authenticator when an expansion was forced to fail,
2863 contradicting the general specification for all authenticators. Instead it
2864 was generating a temporary error. It now behaves as specified.
2866 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
2867 (the order specifies the preference for clients). The order is now AES256,
2868 AES128, 3DES, ARCFOUR128.
2870 31. Small patch to Sieve code - explicitly set From: when generating an
2873 32. Exim crashed if a remote delivery caused a very long error message to be
2874 recorded - for instance if somebody sent an entire SpamAssassin report back
2875 as a large number of 550 error lines. This bug was coincidentally fixed by
2876 increasing the size of one of Exim's internal buffers (big_buffer) that
2877 happened as part of the Exiscan merge. However, to be on the safe side, I
2878 have made the code more robust (and fixed the comments that describe what
2881 33. Some experimental protocols are using DNS PTR records for new purposes. The
2882 keys for these records are domain names, not reversed IP addresses. The
2883 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
2884 leaves it alone. Component reversal etc. now happens only for IP addresses.
2887 34. The host_aton() function is supposed to be passed a string that is known
2888 to be a valid IP address. However, in the case of IPv6 addresses, it was
2889 not checking this. This is a hostage to fortune. Exim now panics and dies
2890 if the condition is not met. A case was found where this could be provoked
2891 from a dnsdb PTR lookup with an IPv6 address that had more than 8
2892 components; fortuitously, this particular loophole had already been fixed
2893 by change 4.50/55 or 4.44/33 above.
2895 If there are any other similar loopholes, the new check in host_aton()
2896 itself should stop them being exploited. The report I received stated that
2897 data on the command line could provoke the exploit when Exim was running as
2898 exim, but did not say which command line option was involved. All I could
2899 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
2900 running as the user.
2903 35. There was a buffer overflow vulnerability in the SPA authentication code
2904 (which came originally from the Samba project). I have added a test to the
2905 spa_base64_to_bits() function which I hope fixes it.
2908 36. The daemon start-up calls getloadavg() while still root for those OS that
2909 need the first call to be done as root, but it missed one case: when
2910 deliver_queue_load_max is set with deliver_drop_privilege. This is
2911 necessary for the benefit of the queue runner, because there is no re-exec
2912 when deliver_drop_privilege is set.
2914 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
2915 was in use, was not putting the data itself into the right store pool;
2916 consequently, it could be overwritten for a subsequent message in the same
2917 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
2920 38. Sometimes the final signoff response after QUIT could fail to get
2921 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
2922 < 0 before doing a fflush(). This bug looks as though it goes back to the
2923 introduction of TLS in release 3.20, but "sometimes" must have been rare
2924 because the tests only now provoked it.
2930 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
2931 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
2932 Now it returns an integer. A number of calls were still expecting a T/F
2933 return. Fortuitously, in all cases, the tests worked in OK situations,
2934 which is the norm. However, things would have gone wrong on any write
2935 failures on the smtp file descriptor. This function is used when sending
2936 messages over SMTP and also when doing verify callouts.
2938 2. When Exim is called to do synchronous delivery of a locally submitted
2939 message (the -odf or -odi options), it no longer closes stderr before doing
2942 3. Implemented the mua_wrapper option.
2944 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
2946 5. Implemented the functions header_remove(), header_testname(),
2947 header_add_at_position(), and receive_remove_recipient(), and exported them
2950 6. If an ACL "warn" statement specified the addition of headers, Exim already
2951 inserted X-ACL-Warn: at the start if there was no header name. However, it
2952 was not making this test for the second and subsequent header lines if
2953 there were newlines in the string. This meant that an invalid header could
2954 be inserted if Exim was badly configured.
2956 7. Allow an ACL "warn" statement to add header lines at the start or after all
2957 the Received: headers, as well as at the end.
2959 8. Added the rcpt_4xx retry error code.
2961 9. Added postmaster_mailfrom=xxx to callout verification option.
2963 10. Added mailfrom=xxxx to the callout verification option, for verify=
2966 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
2967 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
2969 12. Inserted some casts to stop certain compilers warning when using pointer
2970 differences as field lengths or precisions in printf-type calls (mostly
2971 affecting debugging statements).
2973 13. Added optional readline() support for -be (dynamically loaded).
2975 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
2976 same clock tick as a message's arrival, so that its received time was the
2977 same as the "first fail" time on the retry record, and that message
2978 remained on the queue past the ultimate address timeout, every queue runner
2979 would try a delivery (because it was past the ultimate address timeout) but
2980 after another failure, the ultimate address timeout, which should have then
2981 bounced the address, did not kick in. This was a "< instead of <=" error;
2982 in most cases the first failure would have been in the next clock tick
2983 after the received time, and all would be well.
2985 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
2986 being recognized when the domain list was tested by the match_domain
2987 condition in an expansion string.
2989 16. Added the ${str2b64: operator.
2991 17. Exim was always calling setrlimit() to set a large limit for the number of
2992 processes, without checking whether the existing limit was already
2993 adequate. (It did check for the limit on file descriptors.) Furthermore,
2994 errors from getrlimit() and setrlimit() were being ignored. Now they are
2995 logged to the main and panic logs, but Exim does carry on, to try to do its
2996 job under whatever limits there are.
2998 18. Imported PCRE 5.0.
3000 19. Trivial typo in log message " temporarily refused connection" (the leading
3003 20. If the log selector return_path_on_delivery was set and an address was
3004 redirected to /dev/null, the delivery process crashed because it assumed
3005 that a return path would always be set for a "successful" delivery. In this
3006 case, the whole delivery is bypassed as an optimization, and therefore no
3009 21. Internal re-arrangement: the function for sending a challenge and reading
3010 a response while authentication was assuming a zero-terminated challenge
3011 string. It's now changed to take a pointer and a length, to allow for
3012 binary data in such strings.
3014 22. Added the cyrus_sasl authenticator (code supplied by MBM).
3016 23. Exim was not respecting finduser_retries when seeking the login of the
3017 uid under which it was called; it was always trying 10 times. (The default
3018 setting of finduser_retries is zero.) Also, it was sleeping after the final
3019 failure, which is pointless.
3021 24. Implemented tls_on_connect_ports.
3023 25. Implemented acl_smtp_predata.
3025 26. If the domain in control=submission is set empty, Exim assumes that the
3026 authenticated id is a complete email address when it generates From: or
3027 Sender: header lines.
3029 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
3030 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
3031 chown and chgrp in /bin and hostname in /usr/bin.
3033 28. Exim was keeping the "process log" file open after each use, just as it
3034 does for the main log. This opens the possibility of it remaining open for
3035 long periods when the USR1 signal hits a daemon. Occasional processlog
3036 errors were reported, that could have been caused by this. Anyway, it seems
3037 much more sensible not to leave this file open at all, so that is what now
3040 29. The long-running daemon process does not normally write to the log once it
3041 has entered its main loop, and it closes the log before doing so. This is
3042 so that log files can straightforwardly be renamed and moved. However,
3043 there are a couple of unusual error situations where the daemon does write
3044 log entries, and I had neglected to close the log afterwards.
3046 30. The text of an SMTP error response that was received during a remote
3047 delivery was being truncated at 512 bytes. This is too short for some of
3048 the long messages that one sometimes sees. I've increased the limit to
3051 31. It is now possible to make retry rules that apply only when a message has a
3052 specific sender, in particular, an empty sender.
3054 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
3055 it possible to be selective about when SMTP synchronization is enforced.
3057 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
3059 32. Implemented hosts_connection_nolog.
3061 33. Added an ACL for QUIT.
3063 34. Setting "delay_warning=" to disable warnings was not working; it gave a
3066 35. Added mailbox_size and mailbox_filecount to appendfile.
3068 36. Added control = no_multiline_responses to ACLs.
3070 37. There was a bug in the logic of the code that waits for the clock to tick
3071 in the case where the clock went backwards by a substantial amount such
3072 that the microsecond fraction of "now" was more than the microsecond
3073 fraction of "then" (but the whole seconds number was less).
3075 38. Added support for the libradius Radius client library this is found on
3076 FreeBSD (previously only the radiusclient library was supported).
3082 1. When certain lookups returned multiple values in the form name=value, the
3083 quoting of the values was not always being done properly. Specifically:
3084 (a) If the value started with a double quote, but contained no whitespace,
3086 (b) If the value contained whitespace other than a space character (i.e.
3087 tabs or newlines or carriage returns) it was not quoted.
3088 This fix has been applied to the mysql and pgsql lookups by writing a
3089 separate quoting function and calling it from the lookup code. The fix
3090 should probably also be applied to nisplus, ibase and oracle lookups, but
3091 since I cannot test any of those, I have not disturbed their existing code.
3093 2. A hit in the callout cache for a specific address caused a log line with no
3094 reason for rejecting RCPT. Now it says "Previous (cached) callout
3095 verification failure".
3097 3. There was an off-by-one bug in the queryprogram router. An over-long
3098 return line was truncated at 256 instead of 255 characters, thereby
3099 overflowing its buffer with the terminating zero. As well as fixing this, I
3100 have increased the buffer size to 1024 (and made a note to document this).
3102 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
3103 when Exim is waiting for an SMTP response from a remote server, Exim
3104 restarts its select() call on the socket, thereby resetting its timeout.
3105 This is not a problem when such interrupts are rare. Somebody set up a cron
3106 job to run exiwhat every 2 minutes, which is less than the normal select()
3107 timeout (5 or 10 minutes). This meant that the select() timeout never
3108 kicked in because it was always reset. I have fixed this by comparing the
3109 time when an interrupt arrives with the time at the start of the first call
3110 to select(). If more time than the timeout has elapsed, the interrupt is
3111 treated as a timeout.
3113 5. Some internal re-factoring in preparation for the addition of Sieve
3114 extensions (by MH). In particular, the "personal" test is moved to a
3115 separate function, and given an option for scanning Cc: and Bcc: (which is
3116 not set for Exim filters).
3118 6. When Exim created an email address using the login of the caller as the
3119 local part (e.g. when creating a From: or Sender: header line), it was not
3120 quoting the local part when it contained special characters such as @.
3122 7. Installed new OpenBSD configuration files.
3124 8. Reworded some messages for syntax errors in "and" and "or" conditions to
3125 try to make them clearer.
3127 9. Callout options, other than the timeout value, were being ignored when
3128 verifying sender addresses in header lines. For example, when using
3130 verify = header_sender/callout=no_cache
3132 the cache was (incorrectly) being used.
3134 10. Added a missing instance of ${EXE} to the exim_install script; this affects
3135 only the Cygwin environment.
3137 11. When return_path_on_delivery was set as a log selector, if different remote
3138 addresses in the same message used different return paths and parallel
3139 remote delivery occurred, the wrong values would sometimes be logged.
3140 (Whenever a remote delivery process finished, the return path value from
3141 the most recently started remote delivery process was logged.)
3143 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
3144 lines when AUTH and/or TLS are in use. This is the "received protocol"
3145 field. Exim used to use "asmtp" for authenticated SMTP, without any
3146 indication (in the protocol name) for TLS use. Now it follows the RFC and
3147 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
3148 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
3149 names appear in log lines as well as in Received: header lines.
3151 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
3152 extensions, and comparison tests, and to fix some bugs.
3154 14. Changes to the "personal" filter test:
3156 (1) The test was buggy in that it was just doing the equivalent of
3157 "contains" tests on header lines. For example, if a user's address was
3158 anne@some.where, the "personal" test would incorrectly be true for
3160 To: susanne@some.where
3162 This test is now done by extracting each address from the header in turn,
3163 and checking the entire address. Other tests that are part of "personal"
3164 are now done using regular expressions (for example, to check local parts
3165 of addresses in From: header lines).
3167 (2) The list of non-personal local parts in From: addresses has been
3168 extended to include "listserv", "majordomo", "*-request", and "owner-*",
3169 taken from the Sieve specification recommendations.
3171 (3) If the message contains any header line starting with "List-" it is
3172 treated as non-personal.
3174 (4) The test for "circular" in the Subject: header line has been removed
3175 because it now seems ill-conceived.
3177 15. Minor typos in src/EDITME comments corrected.
3179 16. Installed latest exipick from John Jetmore.
3181 17. If headers_add on a router specified a text string that was too long for
3182 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
3183 of string_sprintf() is now avoided.
3185 18. $message_body_size was not set (it was always zero) when running the DATA
3186 ACL and the local_scan() function.
3188 19. For the "mail" command in an Exim filter, no default was being set for
3189 the once_repeat time, causing a random time value to be used if "once" was
3190 specified. (If the value happened to be <= 0, no repeat happened.) The
3191 default is now 0s, meaning "never repeat". The "vacation" command was OK
3192 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
3193 (I found it when inspecting the code).
3195 20. There is now an overall timeout for performing a callout verification. It
3196 defaults to 4 times the callout timeout, which applies to individual SMTP
3197 commands during the callout. The overall timeout applies when there is more
3198 than one host that can be tried. The timeout is checked before trying the
3199 next host. This prevents very long delays if there are a large number of
3200 hosts and all are timing out (e.g. when the network connections are timing
3201 out). The value of the overall timeout can be changed by specifying an
3202 additional sub-option for "callout", called "maxwait". For example:
3204 verify = sender/callout=5s,maxwait=20s
3206 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
3207 to the end before writing, but this should make it even safer).
3209 22. Exim was forgetting that it had advertised PIPELINING for the second and
3210 subsequent messages on an SMTP connection. It was also not resetting its
3211 memory on STARTTLS and an internal HELO.
3213 23. When Exim logs an SMTP synchronization error within a session, it now
3214 records whether PIPELINING has been advertised or not.
3216 24. Added 3 instances of "(long int)" casts to time_t variables that were being
3217 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
3218 rather than long int.
3220 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
3222 26. Added the never_mail option to autoreply.
3228 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
3229 crash if the getsockname() call failed; this can happen if a connection is
3230 closed very soon after it is established. The problem was simply in the
3231 order in which certain operations were done, causing Exim to try to write
3232 to the SMTP stream before it had set up the file descriptor. The bug has
3233 been fixed by making things happen in the correct order.
3239 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
3240 before the connection was closed, thus losing the rejection response.
3242 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
3243 some early Solaris releases, but causes trouble in current releases where
3244 socklen_t is defined.
3246 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
3249 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
3252 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
3253 that releasing the top of it at the end releases what was used for sub-
3254 expansions (unless the block got too big). However, discard this block if
3255 the first thing is a variable or header, so that we can use its block when
3256 it is dynamic (useful for very large $message_headers, for example).
3258 6. Lookups now cache *every* query, not just the most recent. A new, separate
3259 store pool is used for this. It can be recovered when all lookup caches are
3260 flushed. Lookups now release memory at the end of their result strings.
3261 This has involved some general refactoring of the lookup sources.
3263 7. Some code has been added to the store_xxx() functions to reduce the amount
3264 of flapping under certain conditions.
3266 8. log_incoming_interface used to affect only the <= reception log lines. Now
3267 it causes the local interface and port to be added to several more SMTP log
3268 lines, for example "SMTP connection from", and rejection lines.
3270 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
3272 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
3274 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
3275 could be overwritten at the end of the current message (or the start of a
3276 new message if it was set in a HELO ACL). The value is now preserved for
3277 the duration of the SMTP connection.
3279 12. If a transport had a headers_rewrite setting, and a matching header line
3280 contained an unqualified address, that address was qualified, even if it
3281 did not match any rewriting rules. The underlying bug was that the values
3282 of the flags that permit the existence of unqualified sender and recipient
3283 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
3284 non-local messages, and by -bnq for local messages) were not being
3285 preserved with the message after it was received.
3287 13. When Exim was logging an SMTP synchronization error, it could sometimes log
3288 "next input=" as part of the text comprising the host identity instead of
3289 the correct text. The code was using the same buffer for two different
3290 strings. However, depending on which order the printing function evaluated
3291 its arguments, the bug did not always show up. Under Linux, for example, my
3292 test suite worked just fine.
3294 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
3295 doesn't work with some older versions of Perl. It has been changed to "my",
3296 which in any case is probably the better facility to use.
3298 15. A really picky compiler found some instances of statements for creating
3299 error messages that either had too many or two few arguments for the format
3302 16. The size of the buffer for calls to the DNS resolver has been increased
3303 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
3304 for addresses that have a lot of PTR records. This alleviates a problem; it
3305 does not fully solve it.
3307 17. A dnsdb lookup for PTR records that receives more data than will fit in the
3308 buffer now truncates the list and logs the incident, which is the same
3309 action as happens when Exim is looking up a host name and its aliases.
3310 Previously in this situation something unpredictable would happen;
3311 sometimes it was "internal error: store_reset failed".
3313 18. If a server dropped the connection unexpectedly when an Exim client was
3314 using GnuTLS and trying to read a response, the client delivery process
3315 crashed while trying to generate an error log message.
3317 19. If a "warn" verb in an ACL added multiple headers to a message in a single
3318 string, for example:
3320 warn message = H1: something\nH2: something
3322 the text was added as a single header line from Exim's point of view
3323 though it ended up OK in the delivered message. However, searching for the
3324 second and subsequent header lines using $h_h2: did not work. This has been
3325 fixed. Similarly, if a system filter added multiple headers in this way,
3326 the routers could not see them.
3328 20. Expanded the error message when iplsearch is called with an invalid key to
3329 suggest using net-iplsearch in a host list.
3331 21. When running tests using -bh, any delays imposed by "delay" modifiers in
3332 ACLs are no longer actually imposed (and a message to that effect is
3335 22. If a "gecos" field in a passwd entry contained escaped characters, in
3336 particular, if it contained a \" sequence, Exim got it wrong when building
3337 a From: or a Sender: header from that name. A second bug also caused
3338 incorrect handling when an unquoted " was present following a character
3339 that needed quoting.
3341 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
3342 was not being matched caselessly.
3344 24. Arranged for all hyphens in the exim.8 source to be escaped with
3347 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
3348 properly. Recipient callout cache records were still being keyed to include
3349 the sender, even when use_sender was set false. This led to far more
3350 callouts that were necessary. The sender is no longer included in the key
3351 when use_sender is false.
3353 26. Added "control = submission" modifier to ACLs.
3355 27. Added the ${base62d: operator to decode base 62 numbers.
3357 28. dnsdb lookups can now access SRV records.
3359 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
3360 the configuration file.
3362 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
3363 (-v) mode. This makes the output for a verbose queue run more intelligible.
3365 31. Added a use_postmaster feature to recipient callouts.
3367 32. Added the $body_zerocount variable, containing the number of binary zero
3368 bytes in the message body.
3370 33. The time of last modification of the "new" subdirectory is now used as the
3371 "mailbox time last read" when there is a quota error for a maildir
3374 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
3376 35. Added +ignore_unknown as a special item in host lists.
3378 36. Code for decoding IPv6 addresses in host lists is now included, even if
3379 IPv6 support is not being compiled. This fixes a bug in which an IPv6
3380 address was recognized as an IP address, but was then not correctly decoded
3381 into binary, causing unexpected and incorrect effects when compared with
3388 1. Very minor rewording of debugging text in manualroute to say "list of
3389 hosts" instead of "hostlist".
3391 2. If verify=header_syntax was set, and a header line with an unqualified
3392 address (no domain) and a large number of spaces between the end of the
3393 name and the colon was received, the reception process suffered a buffer
3394 overflow, and (when I tested it) crashed. This was caused by some obsolete
3395 code that should have been removed. The fix is to remove it!
3397 3. When running in the test harness, delay a bit after writing a bounce
3398 message to get a bit more predictability in the log output.
3400 4. Added a call to search_tidyup() just before forking a reception process. In
3401 theory, someone could use a lookup in the expansion of smtp_accept_max_
3402 per_host which, without the tidyup, could leave open a database connection.
3404 5. Added the variables $recipient_data and $sender_data which get set from a
3405 lookup success in an ACL "recipients" or "senders" condition, or a router
3406 "senders" option, similar to $domain_data and $local_part_data.
3408 6. Moved the writing of debug_print from before to after the "senders" test
3411 7. Change 4.31/66 (moving the time when the Received: is generated) caused
3412 problems for message scanning, either using a data ACL, or using
3413 local_scan() because the Received: header was not generated till after they
3414 were called (in order to set the time as the time of reception completion).
3415 I have revised the way this works. The header is now generated after the
3416 body is received, but before the ACL or local_scan() are called. After they
3417 are run, the timestamp in the header is updated.
3423 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
3424 before starting a queue runner without re-exec. This happened only when
3425 deliver_drop_privilege was set or when the Exim user was set to root. The
3426 effect of the bug was that timeouts during subsequent deliveries caused
3427 crashes instead of being properly handled. The handler is now left at its
3428 default (and expected) setting.
3430 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
3431 message, again when deliver_drop_privilege is set or Exim is run as root.
3432 The bug described in (1) was not present in this case, but the tidying up
3433 of the other signals was missing. I have made the two cases consistent.
3435 3. The ignore_target_hosts setting on a manualroute router was being ignored
3436 for hosts that were looked up using the /MX notation.
3438 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
3441 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
3442 operated on the sender address. After changing the $sender_address to <>
3443 for the sender address verify, Exim was re-instated it as the original
3444 (before rewriting) address, but remembering that it had rewritten it, so it
3445 wasn't rewriting it again. This bug also had the effect of breaking the
3446 sender address verification caching when the sender address was rewritten.
3448 6. The ignore_target_hosts option was being ignored by the ipliteral router.
3449 This has been changed so that if the ip literal address matches
3450 ignore_target_hosts, the router declines.
3452 7. Added expansion conditions match_domain, match_address, and match_local_
3453 part (NOT match_host).
3455 8. The placeholder for the Received: header didn't have a length field set.
3457 9. Added code to Exim itself and to exim_lock to test for a specific race
3458 condition that could lead to file corruption when using MBX delivery. The
3459 issue is with the lockfile that is created in /tmp. If this file is removed
3460 after a process has opened it but before that process has acquired a lock,
3461 there is the potential for a second process to recreate the file and also
3462 acquire a lock. This could lead to two Exim processes writing to the file
3463 at the same time. The added code performs the same test as UW imapd; it
3464 checks after acquiring the lock that its file descriptor still refers to
3465 the same named file.
3467 10. The buffer for building added header lines was of fixed size, 8192 bytes.
3468 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
3471 11. Added the smtp_active_hostname option. If used, this will typically be made
3472 to depend on the incoming interface address. Because $interface_address is
3473 not set up until the daemon has forked a reception process, error responses
3474 that can happen earlier (such as "too many connections") no longer contain
3477 12. If an expansion in a condition on a "warn" statement fails because a lookup
3478 defers, the "warn" statement is abandoned, and the next ACL statement is
3479 processed. Previously this caused the whole ACL to be aborted.
3481 13. Added the iplsearch lookup type.
3483 14. Added ident_timeout as a log selector.
3485 15. Added tls_certificate_verified as a log selector.
3487 16. Added a global option tls_require_ciphers (compare the smtp transport
3488 option of the same name). This controls incoming TLS connections.
3490 17. I finally figured out how to make tls_require_ciphers do a similar thing
3491 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
3492 before starting the TLS session.
3494 18. Tabs are now shown as \t in -bP output.
3496 19. If the log selector return_path_on_delivery was set, Exim crashed when
3497 bouncing a message because it had too many Received: header lines.
3499 20. If two routers both had headers_remove settings, and the first one included
3500 a superfluous trailing colon, the final name in the first list and the
3501 first name in the second list were incorrectly joined into one item (with a
3502 colon in the middle).
3508 1. Added -C and -D options to the exinext utility, mainly to make it easier
3509 to include in the automated testing, but these could be helpful when
3510 multiple configurations are in use.
3512 2. The exinext utility was not formatting the output nicely when there was
3513 an alternate port involved in the retry record key, nor when there was a
3514 message id as well (for retries that were specific to a specific message
3515 and a specific host). It was also confused by IPv6 addresses, because of
3516 the additional colons they contain. I have fixed the IPv4 problem, and
3517 patched it up to do a reasonable job for IPv6.
3519 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
3520 delivery, the log line now contains "pipelined" if PIPELINING was used.
3522 4. An SMTP transport process used to panic and die if the bind() call to set
3523 an explicit outgoing interface failed. This has been changed; it is now
3524 treated in the same way as a connect() failure.
3526 5. A reference to $sender_host_name in the part of a conditional expansion
3527 that was being skipped was still causing a DNS lookup. This no longer
3530 6. The def: expansion condition was not recognizing references to header lines
3531 that used bh_ and bheader_.
3533 7. Added the _cache feature to named lists.
3535 8. The code for checking quota_filecount in the appendfile transport was
3536 allowing one more file than it should have been.
3538 9. For compatibility with Sendmail, the command line option
3546 and sets the incoming protocol and host name (for trusted callers). The
3547 host name and its colon can be omitted when only the protocol is to be set.
3548 Note the Exim already has two private options, -pd and -ps, that refer to
3549 embedded Perl. It is therefore impossible to set a protocol value of "d" or
3550 "s", but I don't think that's a major issue.
3552 10. A number of refactoring changes to the code, none of which should affect
3555 (a) The number of logging options was getting close to filling up the
3556 32-bit word that was used as a bit map. I have split them into two classes:
3557 those that are passed in the argument to log_write(), and those that are
3558 only ever tested independently outside of that function. These are now in
3559 separate 32-bit words, so there is plenty of room for expansion again.
3560 There is no change in the user interface or the logging behaviour.
3562 (b) When building, for example, log lines, the code previously used a
3563 macro that called string_cat() twice, in order to add two strings. This is
3564 not really sufficiently general. Furthermore, there was one instance where
3565 it was actually wrong because one of the argument was used twice, and in
3566 one call a function was used. (As it happened, calling the function twice
3567 did not affect the overall behaviour.) The macro has been replaced by a
3568 function that can join an arbitrary number of extra strings onto a growing
3571 (c) The code for expansion conditions now uses a table and a binary chop
3572 instead of a serial search (which was left over from when there were very
3573 few conditions). Also, it now recognizes conditions like "pam" even when
3574 the relevant support is not compiled in: a suitably worded error message is
3575 given if an attempt is made to use such a condition.
3577 11. Added ${time_interval:xxxxx}.
3579 12. A bug was causing one of the ddress fields not to be passed back correctly
3580 from remote delivery subprocesses. The field in question was not being
3581 subsequently used, so this caused to problems in practice.
3583 13. Added new log selectors queue_time and deliver_time.
3585 14. Might have fixed a bug in maildirsizefile handling that threw up
3586 "unexpected character" debug warnings, and recalculated the data
3587 unnecessarily. In any case, I expanded the warning message to give more
3590 15. Added the message "Restricted characters in address" to the statements in
3591 the default ACL that block characters like @ and % in local parts.
3593 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
3594 Three changes have been made:
3596 (a) There was a serious bug; a negative response to MAIL caused the whole
3597 recipient domain to be cached as invalid, thereby blocking all messages
3598 to all local parts at the same domain, from all senders. This bug has
3599 been fixed. The domain is no longer cached after a negative response to
3600 MAIL if the sender used is not empty.
3602 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
3605 (c) A new callout option, "use_sender" has been added for people who want
3606 the modified behaviour.
3612 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
3615 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
3616 indeed breaks things for older releases.
3618 3. Added additional logging to the case where there is a problem reading data
3619 from a filter that is running in a subprocess using a pipe, in order to
3620 try to track down a specific problem.
3622 4. Testing facility fudge: when running in the test harness and attempting
3623 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
3624 getting "No route to host". Convert this to a timeout.
3626 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
3629 6. Some OS don't have socklen_t but use size_t instead. This affects the
3630 fifth argument of getsockopt() amongst other things. This is now
3631 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
3632 can be set for individual OS. I have set it for SunOS5, OSF1, and
3633 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
3634 some earlier ones do not.
3636 7. Change 4.30/15 was not doing the test caselessly.
3638 8. The standard form for an IPv6 address literal was being rejected by address
3639 parsing in, for example, MAIL and RCPT commands. An example of this kind of
3640 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
3641 this, as well as the form without the "IPv6" on the front (but only when
3642 address literals are enabled, of course).
3644 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
3646 10. Exim crashed if a message with an empty sender address specified by -f
3647 encountered a router with an errors_to setting. This could be provoked only
3648 by a command such as
3652 where an empty string was supplied; "<>" did not hit this bug.
3654 11. Installed PCRE release 4.5.
3656 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
3657 remained set. It is now erased.
3659 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
3660 times from message ids (which are base 36 rather than the normal 62).
3662 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
3663 were being counted as actual protocol errors, and logged if the log
3664 selector +smtp_protocol_error was set. One cannot be perfect in this test,
3665 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
3666 and DATA following a set of rejected RCPTs do not count as protocol errors.
3667 In other words, Exim assumes they were pipelined, though this may not
3668 actually be the case. Of course, in all cases the client gets an
3669 appropriate error code.
3671 15. If a lookup fails in an ACL condition, a message about the failure may
3672 be available; it is used if testing the ACL cannot continue, because most
3673 such messages specify what the cause of the deferral is. However, some
3674 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
3675 that caused an old message to be retained and used if a later statement
3676 caused a defer, replacing the real cause of the deferral.
3678 16. If an IP address had so many PTR records that the DNS lookup buffer
3679 was not large enough to hold them, Exim could crash while trying to process
3680 the truncated data. It now detects and logs this case.
3682 17. Further to 4.21/58, another change has been made: if (and only if) the
3683 first line of a message (the first header line) ends with CRLF, a bare LF
3684 in a subsequent header line has a space inserted after it, so as not to
3685 terminate the header.
3687 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
3688 unnecessarily, used atoi() instead of strtol(), and didn't check the
3689 termination when getting file sizes from file names by regex.
3691 19. Completely re-implemented the support for maildirsize files, in the light
3692 of a number of problems with the previous contributed implementation
3693 (4.30/29). In particular:
3695 . If the quota is zero, the maildirsize file is maintained, but no quota is
3698 . If the maildir directory does not exist, it is created before any attempt
3699 to write a maildirsize file.
3701 . The quota value in the file is just a cache; if the quota is changed in
3702 the transport, the new value overrides.
3704 . A regular expression is available for excluding directories from the
3707 20. The autoreply transport checks the characters in options that define the
3708 message's headers; it allows continued headers, but it was checking with
3709 isspace() after an embedded newline instead of explicitly looking for a
3712 21. If all the "regular" hosts to which an address was routed had passed their
3713 expiry times, and had not reached their retry times, the address was
3714 bounced, even if fallback hosts were defined. Now Exim should go on to try
3717 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
3718 equivalent code in the SMTP transport. Some hosts send humungous responses
3719 to HELO/EHLO, more than 1024 it seems.
3721 23. Refactoring: code in filter.c used (void *) for "any old type" but this
3722 gives compiler warnings in some environments. I've now done it "properly",
3725 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
3726 (because of problems with the built-in one) was declared to return uschar *
3727 instead of char *, causing compiler failure.
3729 25. Fixed a file descriptor leak when processing alias/forward files.
3731 26. Fixed a minor format string issue in dbfn.c.
3733 27. Typo in exim.c: ("dmbnz" for "dbmnz").
3735 28. If a filter file refered to $h_xxx or $message_headers, and the headers
3736 contained RFC 2047 "words", Exim's memory could, under certain conditions,
3739 29. When a sender address is verified, it is cached, to save repeating the test
3740 when there is more than one recipient in a message. However, when the
3741 verification involves a callout, it is possible for different callout
3742 options to be set for different recipients. It is too complicated to keep
3743 track of this in the cache, so now Exim always runs a verification when a
3744 callout is required, relying on the callout cache for the optimization.
3745 The overhead is duplication of the address routing, but this should not be
3748 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
3749 to be verified with callout=postmaster, and the main callout worked but the
3750 postmaster check failed, the verification correctly failed. However, if a
3751 subsequent RCPT command asked for sender verification *without* the
3752 postmaster check, incorrect caching caused this verification also to fail,
3755 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
3756 it was not caching the DNS options (qualify_single, search_parents) that
3757 were used when the lookup failed. A subsequent lookup with different
3758 options therefore always gave the same answer, though there were cases
3759 where it should not have. (Example: a "domains = !$mx_any" option on a
3760 dnslookup router: the "domains" option is always processed without any
3761 widening, but the router might have qualify_single set.) Now Exim uses the
3762 cached value only when the same options are set.
3764 32. Added John Jetmore's "exipick" utility to the distribution.
3766 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
3767 than a timeout (e.g. a certificate is required, and is not provided), an
3768 Exim server now closes the connection immediately. Previously it waited for
3769 the client to close - but if the client is SSL, it seems that they each
3770 wait for each other, leading to a delay before one of them times out.
3772 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
3773 maintained 0.8.x compatibility because I don't think many are using it, and
3774 it is clearly obsolete.
3776 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
3779 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
3780 client certificate was expired. A simple patch fixes this, though I don't
3781 understand the full logic of why the verify callback is called multiple
3784 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
3785 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
3786 which causes problems with some clients (such as the Certicom SSL Plus
3787 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
3788 disables the coutermeasure allowing Eudora to connect."
3790 38. Exim was not checking that a write() to a log file succeeded. This could
3791 lead to Bad Things if a log got too big, in particular if it hit a file
3792 size limit. Exim now panics and dies if it cannot write to a log file, just
3793 as it does if it cannot open a log file.
3795 39. Modified OS/Makefile-Linux so that it now contains
3797 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
3799 The two -D definitions ensure that Exim is compiled with large file
3800 support, which makes it possible to handle log files that are bigger than
3803 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
3804 instance) a domain was checked against a named list that involved a lookup,
3805 causing $domain_data to be set, then another domain was checked against the
3806 same list, then the first domain was re-checked, the value of $domain_data
3807 after the final check could be wrong. In particular, if the second check
3808 failed, it could be set empty. This bug probably also applied to
3811 41. The strip_trailing_dot option was not being applied to the address given
3812 with the -f command-line option.
3814 42. The code for reading a message's header from the spool was incrementing
3815 $received_count, but never initializing it. This meant that the value was
3816 incorrect (doubled) while delivering a message in the same process in which
3817 it was received. In the most common configuration of Exim, this never
3818 happens - a fresh exec is done - but it can happen when
3819 deliver_drop_privilege is set.
3821 43. When Exim logs an SMTP synchronization error - client data sent too soon -
3822 it now includes up to 150 characters of the unexpected data in the log
3825 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
3826 and building data strings. The size of both of these buffers was 10 000
3827 bytes - far larger than anybody would *ever* want, thought I. Needless to
3828 say, somebody hit the limit. I have increased the maximum line length to
3829 20 000 and the maximum data length of concatenated lines to 100 000. I have
3830 also fixed two bugs, because there was no checking on these buffers. Tsk,
3831 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
3832 buffer is too small.
3834 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
3835 lsearch lookups. Now it does.
3837 46. When parsing a route_list item in a manualroute router, a fixed-length
3838 buffer was used for the list of hosts. I made this 1024 bytes long,
3839 thinking that nobody would ever have a list of hosts that long. Wrong.
3840 Somebody had a whole pile of complicated expansion conditions, and the
3841 string was silently truncated, leading to an expansion error. It turns out
3842 that it is easier to change to an unlimited length (owing to other changes
3843 that have happened since this code was originally written) than to build
3844 structure for giving a limitation error. The length of the item that
3845 expands into the list of hosts is now unlimited.
3847 47. The lsearch lookup could not handle data where the length of text line was
3848 more than 4095 characters. Such lines were truncated, leading to shortened
3849 data being returned. It should now handle lines of any length.
3851 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
3852 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
3855 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
3856 replacing strings such as BIN_DIRECTORY in a source file: the replacement
3857 no longer happens in comment lines. A list of replacements is now placed
3858 at the head of all of the source files, except those whose only change is
3859 to replace PERL_COMMAND in the very first #! line.
3861 50. Replaced the slow insertion sort in queue.c, for sorting the list of
3862 messages on the queue, with a bottom-up merge sort, using code contributed
3863 by Michael Haardt. This should make operations like -bp somewhat faster on
3864 large queues. It won't affect queue runners, except when queue_run_in_order
3867 51. Installed eximstats 1.31 in the distribution.
3869 52. Added support for SRV lookups to the dnslookup router.
3871 53. If an ACL referred to $message_body or $message_body_end, the value was not
3872 reset for any messages that followed in the same SMTP session.
3874 54. The store-handling optimization for building very long strings was not
3875 differentiating between the different store pools. I don't think this
3876 actually made any difference in practice, but I've tidied it.
3878 55. While running the routers to verify a sender address, $sender_address
3879 was still set to the sender address. This is wrong, because when routing to
3880 send a bounce to the sender, it would be empty. Therefore, I have changed
3881 it so that, while verifying a sender address, $sender_address is set to <>.
3882 (There is no change to what happens when verifying a recipient address.)
3884 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
3885 target A or AAAA records (if not already returned) without resetting the
3886 qualify_single or search_parents options of the DNS resolver. These are
3887 inappropriate in this case because the targets of MX and SRV records must
3888 be FQDNs. A broken DNS record could cause trouble if it happened to have a
3889 target that, when qualified, matched something in the local domain. These
3890 two options are now turned off when doing these lookups.
3892 57. It seems that at least some releases of Reiserfs (which does not have the
3893 concept of a fixed number of inodes) returns zero and not -1 for the
3894 number of available inodes. This interacted badly with check_spool_inodes,
3895 which assumed that -1 was the "no such thing" setting. What I have done is
3896 to check that the total number of inodes is greater than zero before doing
3897 the test of how many are available.
3899 58. When a "warn" ACL statement has a log_message modifier, the message is
3900 remembered, and not repeated. This is to avoid a lot of repetition when a
3901 message has many recipients that cause the same warning to be written.
3902 Howewer, Exim was preserving the list of already written lines for an
3903 entire SMTP session, which doesn't seem right. The memory is now reset if a
3904 new message is started.
3906 59. The "rewrite" debugging flag was not showing the result of rewriting in the
3907 debugging output unless log_rewrite was also set.
3909 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
3910 of (int)(handle) when we know that handle contains (void *)(-1).
3912 61. The Exim daemon panic-logs an error return when it closes the incoming
3913 connection. However "connection reset by peer" seems to be common, and
3914 isn't really an error worthy of noting specially, so that particular error
3917 62. When Exim is trying to find all the local interfaces, it used to panic and
3918 die if the ioctl to get the interface flags failed. However, it seems that
3919 on at least one OS (Solaris 9) it is possible to have an interface that is
3920 included in the list of interfaces, but for which you get a failure error
3921 for this call. This happens when the interface is not "plumbed" into a
3922 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
3923 failure of the "get flags" call assumes that the interface is down.
3925 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
3926 makes life easier for people who are doing arithmetic on fields extracted
3927 from dates, where you often get leading zeros that should not be
3928 interpreted as octal.
3930 64. Added qualify_domain to the redirect router, to override the global
3933 65. If a pathologically long header line contained very many addresses (the
3934 report of this problem mentioned 10 000) and each of them was rewritten,
3935 Exim could use up a very large amount of memory. (It kept on making new
3936 copies of the header line as it rewrote, and never released the old ones.)
3937 At the expense of a bit more processing, the header rewriting function has
3938 been changed so that it no longer eats memory in this way.
3940 66. The generation of the Received: header has been moved from the time that a
3941 message starts to be received, to the time that it finishes. The timestamp
3942 in the Received: header should now be very close to that of the <= log
3943 line. There are two side-effects of this change:
3945 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
3946 logged header lines no longer include the local Received: line, because
3947 it has not yet been created. The same applies to a copy of the message
3948 that is returned to a non-SMTP sender when a message is rejected.
3950 (b) When a filter file is tested using -bf, no additional Received: header
3951 is added to the test message. After some thought, I decided that this
3954 This change does not affect the value of $received_for. It is still set
3955 after address rewriting, but before local_scan() is called.
3957 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
3959 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
3960 gave an unhelpful panic error message, and a defer error. I have managed to
3961 change this behaviour so that it now rejects any supplied certificate,
3962 which seems right, as the list of acceptable certificates is empty.
3964 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
3965 gave an unhelpful defer error. I have not managed to make this reject any
3966 supplied certificates, but the error message it gives is "no certificate
3967 supplied", which is not helpful.
3969 70. exigrep's output now also includes lines that are not associated with any
3970 message, but which match the given pattern. Implemented by a patch from
3971 Martin Sluka, which also tidied up the Perl a bit.
3973 71. Recipient callout verification, like sender verification, was using <> in
3974 the MAIL FROM command. This isn't really the right thing, since the actual
3975 sender may affect whether the remote host accepts the recipient or not. I
3976 have changed it to use the actual sender in the callout; this means that
3977 the cache record is now keyed on a recipient/sender pair, not just the
3978 recipient address. There doesn't seem to be a real danger of callout loops,
3979 since a callout by the remote host to check the sender would use <>.
3980 [SEE ABOVE: changed after hitting problems.]
3982 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
3983 temporary errors. However, in the case of such a code being given after
3984 the end of a data transmission (i.e. after ".") Exim was failing to write
3985 a retry record for the message. (Yes, there was some broken host that was
3986 actually sending 8xx at this point.)
3988 73. An unknown lookup type in a host list could cause Exim to panic-die when
3989 the list was checked. (An example that provoked this was putting <; in the
3990 middle of a list instead of at the start.) If this happened during a DATA
3991 ACL check, a -D file could be left lying around. This kind of configuration
3992 error no longer causes Exim to die; instead it causes a defer errror. The
3993 incident is still logged to the main and panic logs.
3995 74. Buglet left over from Exim 3 conversion. The message "too many messages
3996 in one connection" was written to the rejectlog but not the mainlog, except
3997 when address rewriting (yes!) was being logged.
3999 75. Added write_rejectlog option.
4001 76. When a system filter was run not as root (that is, when system_filter_user
4002 was set), the values of the $n variables were not being returned to the
4003 main process; thus, they were not subsequently available in the $sn
4006 77. Added +return_path_on_delivery log selector.
4008 78. A connection timeout was being treated differently from recipients deferred
4009 when testing hosts_max_try with a message that was older than the host's
4010 retry timeout. (The host should not be counted, thus allowing all hosts to
4011 be tried at least once before bouncing.) This may have been the cause of an
4012 occasionally reported bug whereby a message would remain on the queue
4013 longer than the retry timeout, but would be bounced if a delivery was
4014 forced. I say "may" because I never totally pinned down the problem;
4015 setting up timeout/retry tests is difficult. See also the next item.
4017 79. The ultimate address timeout was not being applied to errors that involved
4018 a combination of host plus message (for example, a timeout on a MAIL
4019 command). When an address resolved to a number of possible hosts, and they
4020 were not all tried for each delivery (e.g. because of hosts_max_try), a
4021 message could remain on the queue longer than the retry timeout.
4023 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
4026 81. Fixed an obscure SMTP outgoing bug which required at least the following
4027 conditions: (a) there was another message waiting for the same server;
4028 (b) the server returned 5xx to all RCPT commands in the first message so
4029 that the message was not completed; (c) the server dropped the connection
4030 or gave a negative response to the RSET that Exim sends to abort the
4031 transaction. The observed case was a dropped connection after DATA that had
4032 been sent in pipelining mode. That is, the server had advertised PIPELINING
4033 but was not implementing it correctly. The effect of the bug was incorrect
4034 behaviour, such as trying another host, and this could lead to a crash.
4040 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
4041 and daemon.c were passed as pointers to ints; they should have been
4042 pointers to socklen_t variables (which are typically unsigned ints).
4044 2. Some signed/unsigned type warnings in the os.c file for Linux have been
4047 3. Fixed a really odd bug that affected only the testing scheme; patching a
4048 certain fixed string in the binary changed the value of another string that
4049 happened to be identical to the end of the original first string.
4051 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
4052 name", it returns that address as the IP address. On some operating
4053 systems (e.g. Solaris), it also passes back the IP address string as the
4054 "host name". However, on others (e.g. Linux), it passes back an empty
4055 string. Exim wasn't checking for this, and was changing the host name to an
4056 empty string, assuming it had been canonicized.
4058 5. Although rare, it is permitted to have more than one PTR record for a given
4059 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
4060 all the names associated with an address, because they do in Solaris.
4061 However, it seems that they do not in Linux for data that comes from the
4062 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
4063 I found this out when I moved to a new Linux workstation and tried to run
4064 the Exim test suite.
4066 To get round this problem I have changed the code so that it now does its
4067 own call to the DNS to look up PTR records when searching for a host name.
4068 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
4069 addresses that are only in /etc/hosts are still found.
4071 This behaviour is, however, controlled by an option called host_lookup_
4072 order, which defaults to "bydns:byaddr". If people want to use the other
4073 order, or indeed, just use one or the other means of lookup, they can
4074 specify it in this variable.
4076 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
4077 some operating systems, this comes back from gethostbyaddr() as an empty
4078 string, and this is what Exim used to test for. However, it seems that in
4079 other systems, "." is yielded. Exim now tests for this case too.
4081 7. The values of check_spool_space and check_log_space are now held internally
4082 as a number of kilobytes instead of an absolute number of bytes. If a
4083 numbers is specified without 'K' or 'M', it is rounded up to the nearest
4084 kilobyte. This means that much larger values can be stored.
4086 8. Exim monitor: an attempt to get the action menu when not actually pointing
4087 at a message produces an empty menu entitled "No message selected". This
4088 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
4089 no entries in it ("Shell widget menu has zero width and/or height"). So I
4090 have added a single, blank menu entry in this case.
4092 9. Added ${quote_local_part.
4094 10. MIME decoding is now applied to the contents of Subject: header lines when
4097 11. Now that a reference to $sender_host_address automatically causes a reverse
4098 lookup to occur if necessary (4.13/18), there is no need to arrange for a
4099 host lookup before query-style lookups in lists that might use this
4100 variable. This has therefore been abolished, and the "net-" prefix is no
4101 longer necessary for query-style lookups.
4103 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
4104 have been a typo for LFLAGS, so it has been changed.
4106 13. The install script calls Exim with "-C /dev/null" in order to find the
4107 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
4108 to be output. Howeve, since Exim outputs its version number before the
4109 error, it didn't break the script. It just looked ugly. I fixed this by
4110 always allowing "-C /dev/null" if the caller is root.
4112 14. Ignore overlarge ACL variable number when reading spool file - insurance
4113 against a later release with more variables having written the file.
4115 15. The standard form for an IPv6 address literal was being rejected by EHLO.
4116 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
4117 this, as well as the form without the "IPv6" on the front.
4119 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
4120 OS/Makefile-Darwin file.
4122 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
4123 only to LDAP libraries that do not have LDAP_OPT_DEREF.
4125 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
4126 variable. However, some OS use ints for inodes. I've added cast to long int
4127 to get rid of the compiler warning.
4129 19. I had forgotten to lock out "/../" in configuration file names when
4130 ALT_CONFIG_PREFIX was set.
4132 20. Routers used for verification do not need to specify transports. However,
4133 if such a router generated a host list, and callout was configured, Exim
4134 crashed, because it could not find a port number from the (non-existent)
4135 transport. It now assumes port 25 in this circumstance.
4137 21. Added the -t option to exigrep.
4139 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
4140 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
4141 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
4142 them is set, LOOKUP_LSEARCH is forced.
4144 23. "exim -bV" now outputs a list of lookups that are included in the binary.
4146 24. Added sender and host information to the "rejected by local_scan()" log
4147 line; previously there was no indication of these.
4149 25. Added .include_if_exists.
4151 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
4152 Linux. It turns out that not all file systems support this. Apparently some
4153 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
4154 do it.) To cope with this, the error EINVAL, which means that sync-ing is
4155 not supported on the file descriptor, is now ignored when Exim is trying to
4156 sync a directory. This applies only to Linux.
4158 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
4160 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
4161 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
4164 29. Added support for maildirsize files from supplied patch (modified a bit).
4166 30. The use of :fail: followed by an empty string could lead Exim to respond to
4167 sender verification failures with (e.g.):
4169 550 Verification failed for <xxx>
4170 550 Sender verify failed
4172 where the first response line was missing the '-' that indicates it is not
4173 the final line of the response.
4175 31. The loop for finding the name of the user that called Exim had a hardwired
4176 limit of 10; it now uses the value of finduser_retries, which is used for
4177 all other user lookups.
4179 32. Added $received_count variable, available in data and not_smtp ACLs, and at
4182 33. Exim was neglecting to zero errno before one call of strtol() when
4183 expanding a string and expecting an integer value. On some systems this
4184 resulted in spurious "integer overflow" errors. Also, it was casting the
4185 result into an int without checking.
4187 34. Testing for a connection timeout using "timeout_connect" in the retry rules
4188 did not work. The code looks as if it has *never* worked, though it appears
4189 to have been documented since at least releast 1.62. I have made it work.
4191 35. The "timeout_DNS" error in retry rules, also documented since at least
4192 1.62, also never worked. As it isn't clear exactly what this means, and
4193 clearly it isn't a major issue, I have abolished the feature by treating it
4194 as "timeout", and writing a warning to the main and panic logs.
4196 36. The display of retry rules for -brt wasn't always showing the error code
4199 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
4200 timeout_connect_A, timeout_connect_MX.
4202 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
4203 to the empty sender.
4205 39. The daemon was not analysing the content of -oX till after it had closed
4206 stderr and disconnected from the controlling terminal. This meant that any
4207 syntax errors were only noted on the panic log, and the return code from
4208 the command was 0. By re-arranging the code a little, I've made the
4209 decoding happen first, so such errors now appear on stderr, and the return
4210 code is 1. However, the actual setting up of the sockets still happens in
4211 the disconnected process, so errors there are still only recorded on the
4214 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
4215 connections (as happens on some IP stacks) was logged at start up time as
4216 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
4217 it from "IPv6 and IPv4", which means that two separate sockets are being
4220 41. The debug output for gethostbyname2() or getipnodebyname() failures now
4221 says whether AF_INET or AF_INET6 was passed as an argument.
4223 42. Exiwhat output was messed up when time zones were included in log
4226 43. Exiwhat now gives more information about the daemon's listening ports,
4227 and whether -tls-on-connect was used.
4229 44. The "port" option of the smtp transport is now expanded.
4231 45. A "message" modifier in a "warn" statement in a non-message ACL was being
4232 silently ignored. Now an error message is written to the main and panic
4235 46. There's a new ACL modifier called "logwrite" which writes to a log file
4236 as soon as it is encountered.
4238 47. Added $local_user_uid and $local_user_gid at routing time.
4240 48. Exim crashed when trying to verify a sender address that was being
4243 49. Exim was recognizing only a space character after ".include". It now also
4244 recognizes a tab character.
4246 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
4247 extracting the relevant information from the specification. The man page no
4248 longer contains scrambled data for the -d option, and I've added a section
4249 at the front about calling Exim under different names.
4251 51. Added "extra_headers" argument to the "mail" command in filter files.
4253 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
4256 53. Installed eximstats 1.29.
4258 54. Added transport_filter_timeout as a generic transport option.
4260 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
4261 Cc: header lines. This was required by RFC 822, but it not required by RFC
4264 56. Exim used to add From:, Date:, and Message-Id: header lines to any
4265 incoming messages that did not have them. Now it does so only if the
4266 message originates locally, that is, if there is no associated remote host
4267 address. When Resent- header lines are present, this applies to the Resent-
4268 lines rather than the non-Resent- lines.
4270 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
4271 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
4273 58. Messages for configuration errors now include the name of the main
4274 configuration file - useful now that there may be more than one file in a
4275 list (.included file names were always shown).
4277 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
4278 for those rare installations that do not start the daemon as root or run it
4279 setuid root. I've cut out the call to initgroups() if the daemon is not
4282 60. The Exim user and group can now be bound into the binary as text strings
4283 that are looked up at the start of Exim's processing.
4285 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
4287 62. Added $mailstore_basename variable.
4289 63. Installed patch to sieve.c from Michael Haardt.
4291 64. When Exim failed to open the panic log after failing to open the main log,
4292 the original message it was trying to log was written to stderr and debug
4293 output, but if they were not available (the usual case in production), it
4294 was lost. Now it is written to syslog before the two lines that record the
4295 failures to open the logs.
4297 65. Users' Exim filters run in subprocesses under the user's uid. It is
4298 possible for a "deliver" command or an alias in a "personal" command to
4299 provoke an address rewrite. If logging of address rewriting is configured,
4300 this fails because the process is not running as root or exim. There may be
4301 a better way of dealing with this, but for the moment (because 4.30 needs
4302 to be released), I have disabled address rewrite logging when running a
4303 filter in a non-root, non-exim process.
4309 1. The buildconfig auxiliary program wasn't quoting the value set for
4310 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
4311 not defined. This bug was masked in 4.22 by the effect that was fixed in
4314 2. Some messages that were rejected after a message id was allocated were
4315 shown as "incomplete" by exigrep. It no longer does this for messages that
4316 are rejected by local_scan() or the DATA or non-SMTP ACLs.
4318 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
4319 have allow_domain_literals set, the ID did not get logged in the <= line.
4320 Domain literals are now always recognized in Message-ID: header lines.
4322 4. The first argument for a ${extract expansion item is the key name or field
4323 number. Leading and trailing spaces in this item were not being ignored,
4324 causing some misleading effects.
4326 5. When deliver_drop_privilege was set, single queue runner processes started
4327 manually (i.e. by the command "exim -q") or by the daemon (which uses the
4328 same command in the process it spins off) were not dropping privilege.
4330 6. When the daemon running as "exim" started a queue runner, it always
4331 re-executed Exim in the spun-off process. This is a waste of effort when
4332 deliver_drop_privilege is set. The new process now just calls the
4333 queue-runner function directly.
4339 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
4342 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
4343 ignored. Though the use of -oP was forcing the writing of a pid file, it
4344 was always written to the default place.
4346 3. If the message "no IP address found for host xxxx" is generated during
4347 incoming verification, it is now followed by identification of the incoming
4348 connection (so you can more easily find what provoked it).
4350 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
4352 5. Added some features to "harden" Exim a bit more against certain attacks:
4354 (a) There is now a build-time option called FIXED_NEVER_USERS that can
4355 be put in Local/Makefile. This is like the never_users runtime option,
4356 but it cannot be overridden. The default setting is "root".
4358 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
4359 prefix string with which any file named in a -C command line option
4362 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
4363 is retained for -C and -D only if the caller of Exim is root. Without
4364 it, the exim user may also use -C and -D and retain privilege.
4366 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
4367 command line option is disabled.
4369 6. Macro names set by the -D option must start with an upper case letter, just
4370 like macro names defined in the configuration file.
4372 7. Added "dereference=" facility to LDAP.
4374 8. Two instances of the typo "uknown" in the source files are fixed.
4376 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
4377 the Configure-Makefile script screwed up while processing it.
4379 10. Incorporated PCRE 4.4.
4381 11. The SMTP synchronization check was not operating right at the start of an
4382 SMTP session. For example, it could not catch a HELO sent before the client
4383 waited for the greeting. There is now a check for outstanding input at the
4384 point when the greeting is written. Because of the duplex, asynchronous
4385 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
4386 way, but not yet received, when the check is performed.
4388 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
4389 on TCP/IP sockets, because this apparently causes some broken clients to
4392 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
4393 unchanged) from the Cygwin maintainer.
4395 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
4396 was supported instead of testing for mbx. Effectively a typo.
4398 15. The spa authenticator server code was not checking that the input it
4399 received was valid base64.
4401 16. The debug output line for the "set" modifier in ACLs was not showing the
4402 name of the variable that was being set.
4404 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
4406 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
4407 lookup on its own. Something else was needed to trigger the lookup. For
4408 example, a match in host_lookup or the need for a host name in a host list.
4409 Now, if $sender_host_name is referenced and the host name has not yet been
4410 looked up, a lookup is performed. If the lookup fails, the variable remains
4411 empty, and $host_lookup_failed is set to "1".
4413 19. Added "eqi" as a case-independent comparison operator.
4415 20. The saslauthd authentication condition could segfault if neither service
4416 nor realm was specified.
4418 21. If an overflowing value such as "2048M" was set for message_size_limit, the
4419 error message that was logged was misleading, and incoming SMTP
4420 connections were dropped. The message is now more accurate, and temporary
4421 errors are given to SMTP connections.
4423 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
4424 (except RSET) with a 421 error, until QUIT is received. However, it was
4425 failing to send a response to QUIT.
4427 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
4428 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
4429 ACL is now run after the helo_try_verify_hosts code.
4431 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
4432 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
4433 case-independent, so other case variants are also recognized). Apparently
4434 some systems use these upper case variants.
4436 25. If more than two messages were waiting for the same host, and a transport
4437 filter was specified for the transport, Exim sent two messages over the
4438 same TCP/IP connection, and then failed with "socket operation on non-
4439 socket" when it tried to send the third.
4441 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
4443 27. The extern definition of crypt16() in expand.c was not being excluded when
4444 the OS had its own crypt16() function.
4446 28. Added bounce_return_body as a new option, and bounce_return_size_limit
4447 as a preferred synonym for return_size_limit, both as an option and as an
4450 29. Added LIBS=-liconv to OS/Makefile-OSF1.
4452 30. Changed the default configuration ACL to relax the local part checking rule
4453 for addresses that are not in any local domains. For these addresses,
4454 slashes and pipe symbols are allowed within local parts, but the sequence
4455 /../ is explicitly forbidden.
4457 31. SPA server authentication was not clearing the challenge buffer before
4460 32. log_message in a "warn" ACL statement was writing to the reject log as
4461 well as to the main log, which contradicts the documentation and doesn't
4462 seem right (because no rejection is happening). So I have stopped it.
4464 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
4465 However, I am unable to do any testing of this.
4467 34. Fixed an infelicity in the appendfile transport. When checking directories
4468 for a mailbox, to see if any needed to be created, it was accidentally
4469 using path names with one or more superfluous leading slashes; tracing
4470 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
4472 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
4473 contents are added to the log line that is written for every discarded
4474 recipient. (Previously a log_message setting was ignored.)
4476 36. The ${quote: operator now quotes the string if it is empty.
4478 37. The install script runs exim in order to find its version number. If for
4479 some reason other than non-existence or emptiness, which it checks, it
4480 could not run './exim', it was installing it with an empty version number,
4481 i.e. as "exim-". This error state is now caught, and the installation is
4484 38. An argument was missing from the function that creates an error message
4485 when Exim fails to connect to the socket for saslauthd authentication.
4486 This could cause Exim to crash, or give a corrupted message.
4488 39. Added isip, isip4, and isip6 to ${if conditions.
4490 40. The ACL variables $acl_xx are now saved with the message, and can be
4491 accessed later in routers, transports, and filters.
4493 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
4494 strings in the file are not string-expanded.
4496 42. If a MAIL command specified a SIZE value that was too large to fit into an
4497 int variable, the check against message_size_limit failed. Such values are
4498 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
4499 day this will have to be increased, but I don't think I want to be around
4500 when emails are that large.
4507 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
4508 iconv() is not standard in FreeBSD.
4510 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
4511 IPv6 enabled. The observed symptom was a segmentation fault on return from
4512 the function os_common_find_running_interfaces() in src/os.c.
4514 3. In the check_special_case() function in daemon.c I had used "errno" as an
4515 argument name, which causes warnings on some systems. This was basically a
4516 typo, since it was named "eno" in the comments!
4518 4. The code that waits for the clock to tick (at a resolution of some fraction
4519 of a second) so as to ensure message-id uniqueness was always waiting for
4520 at least one whole tick, when it could have waited for less. [This is
4521 almost certainly not relevant at current processor speeds, where it is
4522 unlikely to ever wait at all. But we try to future-proof.]
4524 5. The function that sleeps for a time interval that includes fractions of a
4525 second contained a race. It did not block SIGALRM between setting the
4526 timer, and suspending (a couple of lines later). If the interval was short
4527 and the sigsuspend() was delayed until after it had expired, the suspension
4528 never ended. On busy systems this could lead to processes getting stuck for
4531 6. Some uncommon configurations may cause a lookup to happen in a queue runner
4532 process, before it forks any delivery processes. The open lookup caching
4533 mechanism meant that the open file or database connection was passed into
4534 the delivery process. The problem was that delivery processes always tidy
4535 up cached lookup data. This could cause a problem for the next delivery
4536 process started by the queue runner, because the external queue runner
4537 process does not know about the closure. So the next delivery process
4538 still has data in the lookup cache. In the case of a file lookup, there was
4539 no problem because closing a file descriptor in a subprocess doesn't affect
4540 the parent. However, if the lookup was caching a connection to a database,
4541 the connection was closed, and the second delivery process was likely to
4542 see errors such as "PGSQL: query failed: server closed the connection
4543 unexpectedly". The problem has been fixed by closing all cached lookups
4544 in a queue runner before running a delivery process.
4546 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
4547 seem to have the "const" qualifier which it has on other OS. I've
4550 8. Change 4.21/2 was too strict. It is only if there are two authenticators
4551 *of the same type* (client or server) with the same public name that an
4552 error should be diagnosed.
4554 9. When Exim looked up a host name for an IP address, but failed to find the
4555 original IP address when looking up the host name (a safety check), it
4556 output the message "<ip address> does not match any IP for NULL", which was
4557 confusing, to say the least. The bug was that the host name should have
4558 appeared instead of "NULL".
4560 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
4561 user that is built into the binary, and the -C or -D options is used, root
4562 privilege is dropped before the configuration file is read. In addition,
4563 logging is switched to stderr instead of the normal log files. If the
4564 configuration then re-defines the Exim user, the unprivileged environment
4565 is probably not what is expected, so Exim logs a panic warning message (but
4568 However, if deliver_drop_privilege is set, the unprivileged state may well
4569 be exactly what is intended, so the warning has been cut out in that case,
4570 and Exim is allowed to try to write to its normal log files.
4576 1. smtp_return_error_details was not giving details for temporary sender
4577 or receiver verification errors.
4579 2. Diagnose a configuration error if two authenticators have the same public
4582 3. Exim used not to create the message log file for a message until the first
4583 delivery attempt. This could be confusing when incoming messages were held
4584 for policy or load reasons. The message log file is now created at the time
4585 the message is received, and an initial "Received" line is written to it.
4587 4. The automatically generated man page for command line options had a minor
4588 bug that caused no ill effects; however, a more serious problem was that
4589 the procedure for building the man page automatically didn't always
4590 operate. Consequently, release 4.20 contains an out-of-date version. This
4591 shouldn't happen again.
4593 5. When building Exim with embedded Perl support, the script that builds the
4594 Makefile was calling 'perl' to find its compile-time parameters, ignoring
4595 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
4597 6. The freeze_tell option was not being used for messages that were frozen on
4598 arrival, either by an ACL or by local_scan().
4600 7. Added the smtp_incomplete_transaction log selector.
4602 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
4603 was accepting AUTH without a new EHLO.
4605 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
4606 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
4609 10. Logging of TCP/IP connections (when configured) now happens in the main
4610 daemon process instead of the child process, so that the TCP/IP connection
4611 count is more accurate (but it can never be perfect).
4613 11. The use of "drop" in a nested ACL was not being handled correctly in the
4614 outer ACL. Now, if condition failure induced by the nested "drop" causes
4615 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
4616 or "require"), the connection is dropped.
4618 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
4619 that yield "discard" can now be used with an "accept" or a "discard" verb,
4620 but an error is generated for any others (because I can't see a useful way
4621 to define what should happen).
4623 13. When an ACL is read dynamically from a file (or anywhere else), the lines
4624 are now processed in the same way as lines in the Exim configuration file.
4625 In particular, continuation lines are supported.
4627 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
4629 15. Added -ti meaning -t -i.
4631 16. Check for letters, digits, hyphens, and dots in the names of dnslist
4632 domains, and warn by logging if others are found.
4634 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
4635 returned from GIFCONF when Exim is trying to find the list of interfaces on
4636 a host. The code in os.c has been modified to copy each ifreq to an aligned
4637 structure in all cases.
4639 Also, in some cases, the returned ifreq's were being copied to a 'struct
4640 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
4641 means the last couple of bytes of an IPv6 address could be chopped if the
4642 ifreq contained only a normal sockaddr (14 bytes storage).
4644 18. Named domain lists were not supported in the hosts_treat_as_local option.
4645 An entry such as +xxxx was not recognized, and was treated as a literal
4648 19. Ensure that header lines added by a DATA ACL are included in the reject log
4649 if the ACL subsequently rejects the message.
4651 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
4652 MD5 (which is deprecated).
4654 21. When testing a filter file using -bf, Exim was writing a message when it
4655 took the sender from a "From " line in the message, but it was not doing so
4656 when it took $return_path from a Return-Path: header line. It now does.
4658 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
4659 with a valid header line field name (a series of printing characters
4660 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
4662 23. Changed "disc" in the source to "disk" to conform to the documentation and
4663 the book and for uniformity.
4665 24. Ignore Sendmail's -Ooption=value command line item.
4667 25. When execve() failed while trying to run a command in a pipe transport,
4668 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
4669 could be confused with a return value of 69 from the command itself. This
4670 has been changed to 127, the value the shell returns if it is asked to run
4671 a non-existent command. The wording for the related log line suggests a
4672 non-existent command as the problem.
4674 26. If received_header_text expands to an empty string, do not add a Received:
4675 header line to the message. (Well, it adds a token one on the spool, but
4676 marks it "old" so that it doesn't get used or transmitted.)
4678 27. Installed eximstats 1.28 (addition of -nt option).
4680 28. There was no check for failure on the call to getsockname() in the daemon
4681 code. This can fail if there is a shortage of resources on the system, with
4682 ENOMEM, for example. A temporary error is now given on failure.
4684 29. Contrary to the C standard, it seems that in some environments, the
4685 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
4686 program. Exim now does this explicitly; it affects the formatting of
4687 timestamps using strftime().
4689 30. If exiqsumm was given junk data, it threw up some uninitialized variable
4690 complaints. I've now initialized all the variables, to avoid this.
4692 32. Header lines added by a system filter were not being "seen" during
4693 transport-time rewrites.
4695 33. The info_callback() function passed to OpenSSL is set up with type void
4696 (*)(SSL *, int, int), as described somewhere. However, when calling the
4697 function (actually a macro) that sets it up, the type void(*)() is
4698 expected. I've put in a cast to prevent warnings from picky compilers.
4700 34. If a DNS black list lookup found a CNAME record, but there were no A
4701 records associated with the domain it pointed at, Exim crashed.
4703 35. If a DNS black list lookup returned more than one A record, Exim ignored
4704 all but the first. It now scans all returned addresses if a particular IP
4705 value is being sought. In this situation, the contents of the
4706 $dnslist_value variable are a list of all the addresses, separated by a
4709 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
4710 to accept a host name and all its aliases if the forward lookup for any of
4711 them yielded the IP address of the incoming connection. Now it accepts only
4712 those names whose forward lookup yields the correct IP address. Any other
4713 names are discarded. This closes a loophole whereby a rogue DNS
4714 administrator could create reverse DNS records to break through a
4715 wildcarded host restriction in an ACL.
4717 37. If a user filter or a system filter that ran in a subprocess used any of
4718 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
4719 the wrong values were passed to the pipe command ($thisaddress had the
4720 value of $0, $0 had the value of $1, etc). This bug was introduced by
4721 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
4723 38. Improved the line breaking for long SMTP error messages from ACLs.
4724 Previously, if there was no break point between 40 and 75 characters, Exim
4725 left the rest of the message alone. Two changes have been made: (a) I've
4726 reduced the minimum length to 35 characters; (b) if it can't find a break
4727 point between 35 and 75 characters, it looks ahead and uses the first one
4728 that it finds. This may give the occasional overlong line, but at least the
4729 remaining text gets split now.
4731 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
4732 file descriptors might be low, and that setting 1000 would always raise it.
4733 It turns out that in some environments, the limit is already over 1000 and
4734 that lowering it causes trouble. So now Exim takes care not to decrease it.
4736 40. When delivering a message, the value of $return_path is set to $sender_
4737 address at the start of routing (routers may change the value). By an
4738 oversight, this default was not being set up when an address was tested by
4739 -bt or -bv, which affected the outcome if any router or filter referred to
4742 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
4743 log only when "message" or "log_message" are set. However, if one of the
4744 conditions was an address verification, or a call to a nested ACL, the
4745 messages generated by the underlying test were being passed through. This
4746 no longer happens. The underlying message is available in $acl_verify_
4747 message for both "message" and "log_message" expansions, so it can be
4748 passed through if needed.
4750 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
4751 new expansion $bh_ to give the encoded byte string without charset
4752 translation. Translation happens only if iconv() is available; HAVE_ICONV
4753 indicates this at build time. HEADERS_CHARSET gives the charset to
4754 translate to; headers_charset can change it in the configuration, and
4755 "headers charset" can change it in an individual filter file.
4757 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
4758 that creates RFC 2047 encoded "words" labels them as that charset instead
4759 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
4760 expansion operator; (ii) when Exim creates a From: line for a local
4761 message; (iii) when a header line is rewritten to include a "phrase" part.
4763 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
4764 buggy, causing it to skip the first lines of messages whose message ID
4765 ended in 'D'. This would not have bitten before Exim release 4.14, because
4766 message IDs were unlikely to end in 'D' before then. The effect was to have
4767 incorrect size information for certain domains.
4769 45. #include "config.h" was missing at the start of the crypt16.c module. This
4770 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
4773 46. If there was a timeout during a "random" callout check, Exim treated it as
4774 a failure of the random address, and carried on sending RSET and the real
4775 address. If the delay was just some slowness somewhere, the response to the
4776 original RCPT would be taken as a response to RSET and so on, causing
4777 mayhem of various kinds.
4779 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
4780 when I implemented it. It didn't allow for the fact that some option values
4781 may legitimatetly be negative (e.g. size_addition), and it didn't even do
4782 the right test for positive values.
4784 48. Domain names in DNS records are case-independent. Exim always looks them up
4785 in lower case. Some resolvers return domain names in exactly the case they
4786 appear in the zone file, that is, they may contain uppercase letters. Not
4787 all resolvers do this - some return always lower case. Exim was treating a
4788 change of case by a resolver as a change of domain, similar to a widening
4789 of a domain abbreviation. This triggered its re-routing code and so it was
4790 trying to route what was effectively the same domain again. This normally
4791 caused routing to fail (because the router wouldn't handle the domain
4792 twice). Now Exim checks for this case specially, and just changes the
4793 casing of the domain that it ultimately uses when it transmits the message
4796 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
4799 50. If a filter generated a file delivery with a non-absolute name (possible if
4800 no home directory exists for the router), the forbid_file option was not
4803 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
4804 the existing equality matching.
4806 52. Exim was using ints instead of ino_t variables in some places where it was
4807 dealing with inode numbers.
4809 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
4810 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
4811 called TMPDIR, and if it finds it is different, it changes its value.
4813 54. The smtp_printf() function is now made available to local_scan() so
4814 additional output lines can be written before returning. There is also an
4815 smtp_fflush() function to enable the detection of a dropped connection.
4816 The variables smtp_input and smtp_batched_input are exported to
4819 55. Changed the default runtime configuration: the message "Unknown user"
4820 has been removed from the ACL, and instead placed on the localuser router,
4821 using the cannot_route_message feature. This means that any verification
4822 failures that generate their own messages won't get overridden. Similarly,
4823 the "Unrouteable address" message that was in the ACL for unverifiable
4824 relay addresses has also been removed.
4826 56. Added hosts_avoid_esmtp to the smtp transport.
4828 57. The exicyclog script was not checking for the esoteric option
4829 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
4830 will work only if exicyclog is run under the appropriate euid.
4832 58. Following a discussion on the list, the rules by which Exim recognises line
4833 endings on incoming messages have been changed. The -dropcr and drop_cr
4834 options are now no-ops, retained only for backwards compatibility. The
4835 following line terminators are recognized: LF CRLF CR. However, special
4836 processing applies to CR:
4838 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
4839 nor a local message in the state where . is a terminator.
4841 (ii) If a bare CR is encountered in a header line, an extra space is added
4842 after the line terminator so as not to end the header. The reasoning
4843 behind this is that bare CRs in header lines are most likely either
4844 to be mistakes, or people trying to play silly games.
4846 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
4847 was being incorrectly given as 18 bytes larger than it should have been.
4848 This is a VOB (very old bug).
4850 60. This may never have affected anything current, but just in case it has:
4851 When the local host is found other than at the start of a list of hosts,
4852 the local host, those with the same MX, and any that follow, are discarded.
4853 When the list in question was part of a longer list of hosts, the following
4854 hosts (not currently being processed) were also being discarded. This no
4855 longer happens. I'm not sure if this situation could ever has previously
4858 61. Added the "/MX" feature to lists of hosts in the manualroute and query
4861 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
4862 header. This is something that is recommended in a new Internet Draft, and
4863 is something that is documented as being done by Sendmail. There are two
4864 possible values. For messages generated by the autoreply transport, Exim
4867 Auto-Submitted: auto-replied
4869 whereas for all other generated messages (e.g. bounces) it adds
4871 Auto-Submitted: auto-generated
4873 63. The "personal" condition in filters now includes a test for the
4874 Auto-Submitted: header. If it contains the string "auto-" the message it
4875 not considered personal.
4877 64. Added rcpt_include_affixes as a generic transport option.
4879 65. Added queue_only_override (default true).
4881 66. Added the syslog_duplication option.
4883 67. If what should have been the first header line of a message consisted of
4884 a space followed by a colon, Exim was mis-interpreting it as a header line.
4885 It isn't of course - it is syntactically invalid and should therefore be
4886 treated as the start of the message body. The misbehaviour could have
4887 caused a number of strange effects, including loss of data in subsequent
4888 header lines, and spool format errors.
4890 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
4891 client host had authenticated. This control can now be exercised by an ACL
4892 for more flexibility.
4894 69. By default, callouts do not happen when testing with -bh. There is now a
4895 variant, -bhc, which does actually run the callout code, including
4896 consulting and updating the callout cache.
4898 70. Added support for saslauthd authentication, courtesy of Alexander
4901 71. If statvfs() failed on the spool or log directories while checking their
4902 size for availability, Exim confusingly gave the error "space shortage".
4903 Furthermore, in debugging mode it crashed with a floating point exception.
4904 These checks are done if check_{spool,log}_{space,inodes} are set, and when
4905 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
4906 serious problem, Exim now writes to the main and panic logs when this
4907 happens, with details of the failure. It then refuses to accept the
4908 incoming message, giving the message "spool directory problem" or "log
4909 directory problem" with a 421 code for SMTP messages.
4911 72. When Exim is about to re-exec itself, it ensures that the file descriptors
4912 0, 1, and 2 exist, because some OS complain for execs without them (see
4913 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
4914 descriptors. However, the code omitted to check that the open succeeded,
4915 causing mysterious errors if for some reason the permissions on /dev/null
4916 got screwed. Now Exim writes a message to the main and panic logs, and
4917 bombs out if it can't open /dev/null.
4919 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
4920 interact so that it is all more flexible. It is supposed to remain
4921 backwards compatible. Also added extra_local_interfaces.
4923 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
4924 to bomb out with an assertion failure - to the client this appears as a
4925 connection drop. This problem occurs in the part of the code that was taken
4926 from the Samba project. Fortunately, the assertion is in a very simple
4927 function, so I have fixed this by reproducing the function inline in the
4928 one place where it is called, and arranging for authentication to fail
4929 instead of killing the process with assert().
4931 75. The SPA client code was not working when the server requested OEM rather
4932 than Unicode encoding.
4934 76. Added code to make require_files with a specific uid setting more usable in
4935 the case where statting the file as root fails - usually a non-root-mounted
4936 NFS file system. When this happens and the failure is EACCES, Exim now
4937 forks a subprocess and does the per-uid checking as the relevant uid.
4939 77. Added process_log_path.
4941 78. If log_file_path was not explicitly set, a setting of check_log_space or
4942 check_log_inodes was ignored.
4944 79. If a space check for the spool or log partitions fails, the incident is now
4945 logged. Of course, in the latter case the data may get lost...
4947 80. Added the %p formatting code to string_format() so that it can be used to
4948 print addresses in debug_print(). Adjusted all the address printing in the
4949 debugging in store.c to use %p rather than %d.
4951 81. There was a concern that a line of code in smtp_in.c could overflow a
4952 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
4953 initially expressed, the concern was not well-founded, because trailing
4954 spaces are removed early. However, if the trailing spaces were followed by
4955 a NULL, they did not get removed, so the overflow was possible. Two fixes
4958 (a) I re-wrote the offending code in a cleaner fashion.
4959 (b) If an incoming SMTP command contains a NULL character, it is rejected
4962 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
4963 runs initgroups(), so that if the Exim user is in any additional groups,
4964 they will be used during message reception.
4970 The change log for 4.20 and earlier releases has been archived.